The COVID-19 Pandemic, India and The World - Economic and Social Policy Perspectives-Routledge India (2021)

The Covid-19 pandemic has had far-reaching social and economic consequences
that are almost certainly greater in magnitude than earlier disruptive events such
as the financial crisis of 2008 and the fall of the former Soviet bloc which dot
our collective memory. This book is a timely look into the economic and social
impact of the pandemic in the Indian context which also includes some forward
looking analysis.
— Sumon Bhaumik, Chair in Finance at the University of
Sheffield, UK and Research Fellow at IZA – Institute of Labor
Economics, Bonn, Germany
Including in-depth studies on the impact of the Covid-19 crisis on health, econ-
omy, education, jobs, gender relations and society, this timely book is greatly
welcomed for its uniquely exhaustive and contemporary analysis of the crisis in
India, the complex transformations underway, their implications for the future
and related policy challenges.
— Sukti Dasgupta, Chief, Employment, Labour Markets
and Youth Branch, International Labour Organization
(ILO), Geneva
The carefully curated set of articles in this book provides a comprehensive analy-
sis of the socio-economic impact of the pandemic. The chapters written by lead-
ing field experts do a deep dive into many of the pandemic-related questions the
world is grappling with. It is a must have for both researchers and policymakers.
— Chetan Subramanian, Professor in Economics & Social
Sciences & Dean (Faculty), IIM Bangalore
THE COVID-19 PANDEMIC, INDIA
AND THE WORLD
This book analyses the economic and social impact of the Covid-19 crisis, with
special focus on India. It examines the economic disruption caused by the pan-
demic, policy responses to it and the prospect of a severe global recession. It also
covers how the pandemic has contributed to considerable suffering among the
masses and affected socio-cultural relationships, behavioural patterns and psy-
chological attitudes governing human interaction.
A topical and timely collection on the pandemic, the chapters in the volume
discuss several key themes which include the following:
• The corona pandemic and the changing global economy; growth, trade and
macroeconomic recovery;
• Public health and policy failures; appropriate policy response;
• Impact on education; guidelines for the future;
• Idea of economic herd immunity; impact of India’s lockdown; crisis of the
migrant labourers;
• Impact on agriculture, industry, firms, households and the informal sector;
• Implications of digital technology for production, labour and labour relations;
• Violence amidst the virus; Covid-19 and Hindu-Muslim conflict in In-
dia, domestic violence, questions of occupation, identity, gender and
vulnerability;
• De-globalization and environmental challenges in the post-Covid era.
Engagingly written, this comprehensive volume compiles original research by

leading economists from India and abroad. It will be useful for scholars and
researchers of economics, the Indian economy, development economics, devel-
opment studies, labour studies, public policy, public administration, governance,
sociology and political economy.
Rajib Bhattacharyya is an associate professor in economics, Department of
Economics, Goenka College of Commerce and Business Administration, Kol-
kata, India.
Ananya Ghosh Dastidar is a professor in economics, Department of Finance

and Business Economics, the University of Delhi, India.
Soumyen Sikdar is a professor in economics, Indian Institute of Management

Calcutta, Kolkata, India.
THE COVID-19
PANDEMIC, INDIA AND
THE WORLD
Economic and Social
Policy Perspectives
Edited by
Rajib Bhattacharyya, Ananya Ghosh Dastidar
and Soumyen Sikdar
First published 2022
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
605 Third Avenue, New York, NY 10158
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2022 selection and editorial matter, Rajib Bhattacharyya,
Ananya Ghosh Dastidar and Soumyen Sikdar; individual chapters,
the contributors
The right of Rajib Bhattacharyya, Ananya Ghosh Dastidar and Soumyen
Sikdar to be identified as the authors of the editorial material, and of the
authors for their individual chapters, has been asserted in accordance
with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other
means, now known or hereafter invented, including photocopying and
recording, or in any information storage or retrieval system, without
permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record has been requested for this book
ISBN: 978-0-367-68886-8 (hbk)

ISBN: 978-1-032-11496-5 (pbk)
ISBN: 978-1-003-22014-5 (ebk)
DOI: 10.4324/9781003220145
Typeset in Bembo
by codeMantra
CONTENTS
List of figures xi
List of tables xv
Notes on contributors xix
Preface xxix
Acknowledgements xxxi
1 Corona pandemic and the changing global economy 1

Rajib Bhattacharyya, Ananya Ghosh Dastidar and
Soumyen Sikdar
PART I
Conf licts, trust and society 29
2 The Covid-19 pandemic and its impact on global

political economy 31
Rajesh Bhattacharya and Ujjwal Kango
3 Riding the wave or going under? The Covid-19

pandemic and trust in governments 50
Catherine Bros
4 India’s lockdown 64
Debraj Ray, S. Subramanian and Lore Vandewalle
viii Contents
5 Violence amidst virus: an essay on conflicts in the

world and at home in times of a pandemic 75
Soumyanetra Munshi
6 Brothers in arms? Covid-19 and Hindu-Muslim

conflict in India 92
Anirban Mitra and Arnab Mukherji
PART II
Macroeconomic issues and environment 107
7 The long shadow of epidemics 109

Shankha Chakraborty and Mausumi Das
8 The macroeconomic impact of Covid-19 on the

Indian economy: an exploratory framework 129
Ajitava Raychaudhuri
9 Developing economic herd immunity in the

face of a pandemic 146
Parikshit Ghosh
10 Covid-19 and an assessment of some policy actions

adopted by India 163
Hiranya Lahiri
11 Impact of the coronavirus pandemic on

de-globalization 178
Ranajoy Bhattacharyya, Anupriya Gangopadhyay
and Abhilasha Pandey
12 Green or brown: global environment in the

post-COVIDian era, and challenges for India 194
Runa Sarkar
PART III
Sectoral perspectives 211
13 Coping with Covid-19 and its consequences:

household adjustments in India 213
Arnab Mukherji and Arjun Shatrunjay
Contents ix
14 Firm fragility: a re-interpretation in the times of the

Covid-19 outbreak 231
Debdatta Saha
15 Impact of Covid-19 on manufacturing industries in

India: networks and trade dependence 260
Partha Chatterjee and Shweta Jain
16 Measurement of efficiency loss in Indian

agricultural sector due to Covid-19: evidence
from rice production 278
Dipyaman Pal, Arpita Ghose and Chandrima Chakraborty
17 The pandemic and the informal sector of India:

probable impacts and strategic intervention 289
Anirban Kundu, Manojit Bhattacharjee, Pratip Kumar Datta,
Kasturi Sadhu and Saumya Chakrabarti
18 Covid-19 pandemic: implications for widowed and

elderly women in India 309
Sutirtha Bandyopadhyay and Bipasha Maity
PART IV
Health, education and labour 321
19 Policy failure before Covid-19 pandemic and options

for the government in near future 323
Arijita Dutta and Montu Bose
20 Higher education in the Post-Covid era:

India and the world 337
Saumen Chattopadhyay
21 Lockdowns, migrants, and the spatial distribution of

Covid-19 cases in India: consequences of a “tough and
timely decision” 353
Zakir Husain and Richa Kothari
22 Labour and the pandemic: a study on work,

employment, and work situation 376
Lopamudra Banerjee and Snehashish Bhattacharya
x Contents
23 Covid-19 pandemic and migrant workers: India and

the world 391
Srobonti Chattopadhyay
24 The future of labour segmentation after Covid-19 408

Mélika Ben Salem
Index 421
FIGURES
1.1 Economic and social impact of corona pandemic 2

1.2 Interaction of the AD and AS model 5
1.3 Pre- and post-corona unemployment rates in India 11
1.4 World merchandise trade volume (quarterly), 2008–20
(USD million) 12
1.5 India’s foreign trade (as on January 2021) 13
1.6 Employment with and without social security in selected
Asian countries 19
1.7 Extent of fiscal and debt sustainability 19
1.8 Probable strategies and policies for revival of economies 20
1.9 Fiscal stimulus package in selected Asian countries (as % of GDP) 22
2.1 Share of economies (%) in recession, 1871–2021 33
2.2 Real GDP growth (%) for world, developed and emerging
economies 35
2.3 Growth in global trade 35
2.4 Average annual % change in employment share, 1995–2012 39
2.5 GVCs as a share of global trade 41
3.1 Approval rates of ten governments’ heads between February 1
and May 31, 2020 51
3.2 Change in approval ratings between January 1 and August 18,
2020 versus stringency measures and number of new Covid cases 54
7.1 Daily new cases of Covid-19: Brazil, China, India and the
USA, from 22 January 2020 to 20 February 2021. Seven-day
moving average 110
7.2 Life expectancies in Brazil, China, India and the USA 110
7.3 Balanced growth rate 116
7.4 Growth effect of mortality shock 121
xii Figures
7.5 Asymmetric effect of morbidity shock 125

7.6 Morbidity effects when sectoral output composition differs 126
8.1 IS-LM-BP adjustments post Covid 137
8.2 Post-Covid shift in the AS curve along with AD curve 139
9.1 Percentage change in the prices of select stocks,
January 1–April 14, 2020 151
9.2 Standard deviation of annual percentage changes in stock
prices for BSE Sensex companies, 2011–21 152
9.3 Standard deviation of annual percentage changes in stock
prices for S&P100 companies grouped by industry 152
10.1 Determination of Y and e 169
10.2 Effect of a rate cut 170
10.3 Effect of an increase in t 173
11.1 World exports (2001–19) 179
11.2 World FDI (2001–19) 180
13.1 Daily counts of Covid-19 new confirmed cases, additional
active cases and recoveries as reported on 28th October 2020 214
13.2 Oxford C19 government response tracker 214
13.3 Monthly demand for work under MGNREGA 226
14.1 Alternative supplier network configurations 242
15.1 Output share and import share in 2015–16 266
15.2 Output share and dependency 268
15.3 YoY change in IIP for the months April–July 2020 sorted by
import share 271
15.4 YoY change in IIP for the months April–July 2020 sorted by
dependency 272
15.5 Dependency networks 277
16.1 The input- and output-oriented measure of technical efficiency 281
17.1 Impact on income 308
19.1 Relationship among cases per thousand of population, health
infrastructure and per capita Gross State Domestic
Product (GSDP) 331
19.2 Relationship among test coverage (tests per hundred of
population), health infrastructure and per capita Gross State
Domestic Product (GSDP) 332
19.3 Test positivity rate vis-à-vis health infrastructure rank and
GSDP PC rank 333
19.4 Case fatality rate per 100 positive cases vis-à-vis health
infrastructure rank and composite index of aged population
share-wise rank and NCD-related hospitalization-wise rank 334
21.1 Regional distribution of active cases per million over time 360
21.2 District-wise distribution of active cases per million population 361
21.3 LISA maps for active cases per million population 362
21.4 Change in hotspots and cold spots between May and August 363
Figures xiii
21.5 Choropleth maps for Getis-Ord gi*(d) statistics 363

21.6 Graphs for ITS model – selected states 364
21.7 Spatial variation of coefficient of MIGRANT 369
23.1 Region-wise distribution of migrant workers 394
23.2 Country-wise distribution of remittances 396
24.1 Employment and growth – India vs. the world 410
TABLES
1.1 India’s growth predictions in GDP, GVA and its components (%) 4
1.2 Differential impact on workers of various sectors 8
1.3 Health system, basic access to infrastructure and connectivity
indicators in South Asian economies and world 17
2.1 Global GDP growth forecasts: percentage change in GDP 34
3.1 Change in approval ratings between January 1 and April 30, 2020 53
3.2(a) Correlates of the approval ratings of the heads of state 58
3.2(b) Correlates of the approval ratings of the heads of state 59
3.3 Correlates of the approval ratings of the heads of state
between day 50 (February 19) and day 150 (May 29) 60
3.4 Correlates of the approval ratings of the heads of state
between day 150 (May 29) and day 230 (August 18) 61
6.1 The effect of Hindu-Muslim conflict (Casualties) on
Covid-19 confirmed cases: OLS district-level regressions 98
6.2 The effect of Hindu-Muslim conflict (Casualties, Killed and
Outbreak) on Covid-19 deaths: OLS district-level regressions 100
10.1 Selected macroeconomic indicators from 2000–01 to 2018–19 175
11.1 Policies undertaken by the Government of India (apart from
international trade and investment policies) 183
11.2 Policies undertaken by the Reserve Bank of India 185
11.A1 Timeline of US-China trade war 192
13.1 Quarterly trends in income and expenditure 217
13.2 Quarterly trends in the Gini Coefficient 219
13.3 Income trends in households, by educational categories 220
13.4 Income trends for households by employment categories who
faced income losses 221
xvi Tables
13.5 Households by employment categories who experienced

income gains 222
14.1 Definition and expression of fragility in economic systems in
the existing literature 236
14.2 Classification of firm-level fragility in existing
macroeconomics literature 238
15.1 Import share 265
15.2 Output share 265
15.3 Correlation between import share and output share 266
15.4 Panel regression between output share and import share 267
15.5 Ranking (dependency) 267
15.6 Correlation between output share and dependency 268
15.7 Panel regression between output share and dependency 269
15.8 Correlation between centrality and dependency 269
15.9 Panel regression between centrality and dependency 269
15.10 Correlation between import share and dependency 270
15.11 Panel regression between import share and dependency 270
15.12 Correlation between change in IIP, import share and dependency 272
15.13 Difference in difference: IIP and import share 273
15.14 Difference in difference: IIP and dependency 274
15.15 Difference in difference: IIP, import share and dependency 274
15.16 Vulnerability index 275
16.1 Change in efficiency due to Covid-19 285
16.2 Change in input slack due to Covid-19 286
16.3 Percentage change in output due to Covid-19 286
17.1 Total employment (million) and share (%) in different sectors
in India over time 291
17.2 Gross value added per worker (in Rs.) in 2017–18 prices 293
17.3 Total employment and vulnerable employment across labour-
size class among self-employed during 2010–11 294
17.4 Total employment and vulnerable employment across labour-
size class among self-employed during 2015–16 295
17.5 Share (%) of households owning financial assets (in Rs.) other
than shares and debentures as of 30.06.2012 in rural areas 296
17.6 Share (%) of households owning financial assets (in Rs.) other
than shares and debentures as of 30.06.2012 in urban areas 296
17.7 Monthly per-capita consumption expenditure of households
(in Rs.) by broad item groups, All India, 2011–12 297
17.8 Leontief Input-Output Coefficient 2003–04 300
17.9 Percentage of Manufacturing Enterprises Facing Severe
Problems during 2015–16 301
18.1 Land and livestock ownership status of households of
women who were married in 2005 and who are widowed or
remained married in 2012 315
Tables xvii
18.2 If suffering from any chronic ailment (self-reported) among

individuals aged 60 years or older 316
19.1 Summary table for selected states and their indicators 334
20.1 Global and national ranking of Indian universities and top
world universities as per world ranking (2020) 349
21.1 Results of ITS model for selected states 365
21.2 Share of migrants among affected persons – West Bengal 367
21.3 Correlation between migration indicators and
incidence of Covid-19 368
21.4 Results of multi-level regression models 368
23.1 Region-wise distribution of migrant workers in 2017 393
24.1 Education and labour market outcomes in India 411
24.2 Informal employment in India (2012 Indian surveys) 411
24.3 Informal employment by economic sector in India 412
CONTRIBUTORS
Sutirtha Bandyopadhyay is currently an assistant professor in economics at

the Indian Institute of Management, Indore. His primary research interests in-
clude applied welfare economics, development economics, food and agricultural
trade. Sutirtha obtained his PhD in quantitative economics from Indian Statisti-
cal Institute (ISI), Delhi, in 2017. Before joining the PhD program, he did MA
and BA (Hons) both in Economics from Jadavpur University, Kolkata.
Lopamudra Banerjee teaches economics and on the environment at Benning-

ton College, Vermont, USA. Her work is at the interface of disaster studies and
distributional asymmetries in an economy.
Mélika Ben Salem is a full professor of economics at the Université Gustave

Eiffel (France) and Deputy-Head of ERUDITE (Research Team on the Use
of Individual Data in relation to Economic Theory). She is also an associate
researcher at the Paris School of Economics (France). She is a member of the
editorial committee of the Revue d’Économie Politique and an elected member
of the Academic Board of the Université Gustave Eiffel and of the board of its
Economics and Management department. Her areas of specialization are labour
economics, macroeconomics and time series econometrics.
Manojit Bhattacharjee is presently working as an assistant professor at Ma-

hatma Gandhi Government Arts College, Mahe. He has completed his PhD
from Institute for Social and Economic Change, Bangalore and has also worked
in Jain University and St. Joseph’s College (Autonomous), Bangalore. His re-
search focuses on household finance, rural farm and nonfarm sector, and issues
related to development. He has published several articles in reputed journals such
xx Contributors
as Applied Economic Perspectives and Policy, Journal of Asian Economics, Margin: The
Journal of Applied Economic Research, and Indian Economic Journal.
Rajesh Bhattacharya is an associate professor in the Public Policy and Man-

agement Group at Indian Institute of Management Calcutta, Kolkata. He has
also taught at South Asian University, New Delhi and University of Calcutta,
Kolkata. He obtained his masters in economics from the University of Calcutta
and his PhD in economics from the University of Massachusetts, Amherst. His
research areas include informal economy, urban political economy, social sector
policies, planning and Indian economic development and Indian perspectives in
social sciences.
Snehashish Bhattacharya teaches at the Faculty of Economics, South Asian

University, New Delhi, India. His research interests are in the fields of Marxian
political economy and political economy of development.
Rajib Bhattacharyya is an associate professor in the Department of Economics,

Goenka College of Commerce and Business Administration, Kolkata, India. He
did his PhD in economics from the University of Calcutta. His fields of inter-
ests are international trade, finance, Indian economic development and women
empowerment. He has contributed articles in reputed national and international
journals/books and has delivered invited lectures in many UGC-sponsored sem-
inars/conferences. He has also worked as a member of the editorial board of var-
ious national- and international-level refereed journals. He has recently jointly
edited a book entitled Role of IT-ITES in Economic Development of Asia: Issues of
Growth, Sustainability and Governance (2020).
Ranajoy Bhattacharyya is a professor and head of Economics Division, Indian

Institute of Foreign Trade, Kolkata Campus. Prior to this, he was reader at the
Department of Business Management, Calcutta University. He has also taught in
several other institutions including the Department of Economics (Calcutta Uni-
versity) and Roanoke College (Virginia, USA). His main research fields are em-
pirical international trade and environment economics. He has published many
papers in reputed journals.
Montu Bose is an assistant professor of TERI School of Advanced Studies,

New Delhi. He is a health economist with PhD from the Department of Eco-
nomics, University of Calcutta. His areas of research include health economics,
health financing, environmental health, sustainable health systems and social de-
terminants of health. Dr Bose is also associated with National Health Systems
Resource Centre (NHSRC), Government of India and Dr B. R. Ambedkar
University (AUD). He has also served Public Health Foundation of India (PHFI),
the University of Delhi and the University of Calcutta under various capacities
before joining TERI School of Advanced Studies.
Contributors xxi
Catherine Bros holds a master’s degree in international and development eco-

nomics and a PhD in economics, both from the University Paris 1 Pantheon-
Sorbonne. She is an associate professor of economics at University Gustave Eif-
fel, Paris, France and an affiliated researcher to IRD-Dial Paris-Dauphine PSL.
Catherine’s research inquiries focus on political economy and more specifically
on the provision of public goods, trust in institutions and social conflicts as well
as the study of informal institutions in developing countries, especially India. She
has also worked on issues pertaining to social stratification and identity in the
context of India and China.
Saumya Chakrabarti is a professor and head of the Department of Economics

and Politics at Visva-Bharati (University), Santiniketan, India. He has also taught
at St. Xavier’s College Kolkata, the University of Calcutta, and at Presidency
University Kolkata. He has been a visiting fellow at Brown University, USA. He
was an honorary director at Agro-Economic Research Centre, Santiniketan. He
has published in journals such as Cambridge Journal of Economics, Review of Radical
Political Economics, and Economic and Labour Relations Review and has written books
published by Prentice Hall and Oxford University Press.
Chandrima Chakraborty is currently working as an assistant professor in the

Department of Economics at Vidyasagar University, India. She has been awarded
a PhD in economics from Jadavpur University. She has attended a number of
seminars and conferences both at national and international levels. She has pub-
lications in many books and journals of reputation. Her research areas include
industrial economics, crime, social sector economics, agricultural economics,
etc. She has also authored two books.
Shankha Chakraborty is a professor of economics at the University of Oregon.

He specializes in the macroeconomics of economic development, with interests
in population health, demography, culture and inequality. His research has been
published in leading academic journals including the Journal of Economic Theory,
International Economic Review, Economic Journal, Journal of Monetary Economics and
Journal of Economic Growth. He has also held visiting positions at the International
Monetary Fund, Indian Statistical Institute and the University of New South
Wales.
Partha Chatterjee is currently a professor and head of the Economics Depart-

ment at Shiv Nadar University. His research interests are broadly in macroeco-
nomics, international economics and issues related to emerging economies. He
has a PhD in economics from the University of Minnesota, USA. He also has
an MPhil in economics from IGIDR, Mumbai and an MSc and a BSc in physics
from Jadavpur University. Previously, Professor Chatterjee was a faculty member
at NUS Business School, National University of Singapore and Faculty of Man-
agement Studies, the University of Delhi.
xxii Contributors
Saumen Chattopadhyay is a faculty at the Zakir Husain Centre for Educa-

tional Studies, School of Social Sciences, Jawaharlal Nehru University ( JNU).
He obtained his PhD from JNU. Earlier he worked at the National Institute of
Public Finance and Policy (NIPFP), New Delhi during 1995–2004. His research
areas include economics of education focusing particularly on policymaking in
higher education, and public finance specializing in the areas of tax evasion and
tax compliance. His publications include Education and Economics: Disciplinary Evo-
lution and Policy Discourse published by the Oxford University Press, New Delhi
(2012); and Macroeconomics of the Black Economy, by Orient BlackSwan (2018), as
well as numerous papers and chapters.
Srobonti Chattopadhyay is an assistant professor in the Department of Eco-

nomics at Rabindra Bharati University, Kolkata, West Bengal. She did her MPhil
and PhD at Jawaharlal Nehru University, New Delhi. She had worked as a visit-
ing scientist at Indian Statistical Institute, Kolkata and as a postdoctoral research
fellow at Indian Institute of Management Calcutta. Her area of research primarily
includes theories of Auction and Industrial Organization. She also co-authored
a book entitled Understanding Auctions as part of the Routledge Focus Series that
got published in 2020.
Mausumi Das is a professor at the Institute of Economic Growth, Delhi. Before

joining the Institute of Economic Growth, she held faculty positions at the Delhi
School of Economics and Jawaharlal Nehru University, New Delhi and also held
visiting position at the Indian Statistical Institute, Delhi. Her areas of interest are
economic growth and development macroeconomics. She has published in top
field journals such as Journal of Economic Growth, Journal of Development Economics
and Journal of Economic Dynamics and Control.
Ananya Ghosh Dastidar is a professor in the Department of Finance and Busi-

ness Economics at the University of Delhi, South Campus, New Delhi, India.
She has a doctorate in economics from the Centre for Economic Studies and
Planning, Jawaharlal Nehru University, New Delhi, and started her teaching ca-
reer at the Department of Economics, Delhi School of Economics. Her research
interests are in the area of international economics and development economics.
Her recent publication includes the edited volume (co-editors: Rajeev Malho-
tra and VivekSuneja) Economic Theory and Policy amidst Global Discontent, a fest-
schrift volume in honour of Professor Deepak Nayyar, published by Routledge,
London.
Pratip Kumar Datta teaches economics and mathematics at Rajatpur I.N.

Vidyapith, Bolpur, West Bengal, India. Apart from teaching, he is engaged in
theory- and policy-oriented research work. Mr Datta has written research papers
published by Review of Radical Political Economics, The World Financial Review,
GlobalDev, Merit Research Journal, Researcher’s Tandem, etc. He has contributed
Contributors xxiii
book chapters with Routledge, Taylor and Francis, IGI Global, etc. The author
also presented research papers in different institutes and universities of India.
Arijita Dutta is a professor and head in the Department of Economics at the

University of Calcutta. An alumnus of Centre for Economic Studies and Plan-
ning, Jawaharlal Nehru University, New Delhi and the University of Calcutta,
her research interests include health economics, quantitative development eco-
nomics and applied econometrics. She has many publications in national and
international journals to her credit and has completed many research projects
funded by Indian and international agencies, including IGC (London), ICSSR,
New Delhi, UGC, Asiatic Society, etc. She has been closely associated with
health policymaking of the state of West Bengal. She has been a visiting professor
at the University of Le Havre, France; UNU Merit, Maastricht, the Netherlands;
ZEF, the University of Bonn, Germany; and Yamaguchi University, Japan.
Anupriya Gangopadhyay is currently a research associate at the Centre for

WTO studies, New Delhi. She has done her MA in economics from the Indian
Institute of Foreign Trade, Kolkata Campus and her BSc in economics from
Presidency University, Kolkata.
Arpita Ghose is currently a professor in the Department of Economics, Ja-

davpur University; chair-professor (Honorary) at Planning and Development
Unit, Jadavpur University (NITI Aayog), Government of India; and also the
former head in the Department of Economics, Jadavpur University. She obtained
her PhD from Indian Statistical Institute, Kolkata. She has authored and edited
books from renowned international publishers, published many papers in re-
puted national and international journals and supervised a number of PhDs. She
has research interest in studies on econometrics, macroeconomics, productivity,
efficiency, empirical issues relating to international trade, applied development
economics and social sector.
Parikshit Ghosh is a professor of economics at the Delhi School of Economics.

He was educated in Presidency College, Kolkata, the Delhi School of Econom-
ics and Boston University. He has also held faculty positions at Texas A&M
University and the University of British Columbia. Ghosh’s research interests
lie in the fields of information economics, political economy and economic
development. His op-ed commentary on Indian economic policy has appeared
in the print and electronic media. He is a member of the editorial board of Ideas
for India.
Zakir Husain is a graduate from the Economics Department, Presidency Col-

lege. He has worked in institutes such as Indian Institute of Technology Kharag-
pur and Institute of Economic Growth Delhi, and is currently a professor of
economics at Presidency University. He has also been a senior consultant in the
xxiv Contributors
Prime Minister’s High-Level Committee for preparing a status report on the

Muslim community and has served as a member in the West Bengal State Plan-
ning Board. His research interests are gender, education and health.
Shweta Jain is pursuing PhD from Shiv Nadar University (SNU), India. She
completed her MA in economics from Jawaharlal Nehru University and BA
(Hons) in Economics from Delhi University. Her primary research interests in-
clude macroeconomics, international economics and econometrics. She has pub-
lished her work in Economic and Political Weekly (EPW). She has also presented her
work at various national and international conferences.
Ujjwal Kango is currently a doctoral student in the Public Policy and Man-
agement group at IIM Calcutta. Prior to joining the doctoral program, he had
completed his undergraduate education in mechanical engineering and worked
in the automobile and power industry for over four years. His research interests
include platform economy, organization of work and Indian business history
from a political economy perspective.
Richa Kothari is a research scholar at the Economics Department, Presidency

University. She has worked as an assistant professor at the BES College, Kolkata,
and as a data scientist and business analyst at Evalueserve, Bengaluru. She has
been awarded the UGC-NET Junior Research Fellowship. Her research inter-
ests are in the field of development economics, especially pertaining to issues of
gender, education and health.
Anirban Kundu is an assistant professor at Presidency University, Bengaluru,

India. He holds a PhD in economics from Centre for Development Studies,
Trivandrum, which is affiliated under Jawaharlal Nehru University, New Delhi,
India. He has written several research papers in journals such as Economic and Po-
litical Weekly, Journal of Economic Development, Journal of Economic Structures, Indian
Journal of Labour Economics, Journal of Land and Rural Studies and Indian Journal of
Agricultural Economics. His research interest lies in informal sector and macroeco-
nomics, and policy analysis in relation to informal sector in India, with a focus
on industry-agriculture interlinkages.
Hiranya Lahiri has completed his PhD from Jadavpur University after complet-
ing his MA in economics from JNU, where he ranked first. His areas of interest
are open-economy macroeconomics and Indian economy. He is an avid reader
of Tagore’s works and is presently engaged in research on various aspects of eco-
nomic theories that can be evidenced in the bard’s literary creations.
Bipasha Maity is an assistant professor of economics at Ashoka University in

Sonepat, India. Her research interests lie in the areas of development economics,
Contributors xxv
economics of gender, labour economics and economic history. Prior to joining

Ashoka University, she was a research fellow at the Osaka School of International
Public Policy, Osaka University, Japan. Bipasha received her PhD in econom-
ics from the University of British Columbia, Canada, in 2016. She also holds a
master of science in quantitative economics from the Indian Statistical Institute,
Delhi.
Anirban Mitra is a senior lecturer (associate professor) in economics at the Uni-

versity of Kent, (Canterbury, UK). His research interests lie in the areas of Devel-
opment Economics and Political Economy – often in their overlap. His research
articles have been published in journals such as the Journal of Political Economy,
Economic Journal, Economica and the Journal of Comparative Economics. Anirban has
been a member of the CESifo Research Network (Munich) since 2014. Prior to
joining Kent, he was employed at the University of Oslo. Anirban received his
PhD in economics from New York University (USA) in 2012 and an MA in
economics from JNU (New Delhi, India) in 2005.
Arnab Mukherji is a professor of public policy at the Center for Public Policy,
IIM Bangalore, and the founding program director for the Mahatma Gandhi
National Fellowship program – an academia-government partnership to build
state capacity in district administration. His work seeks to combine administra-
tive and survey data to inform public policy, particularly in the area of health and
development economics. From time to time, he has consulted with different state
governments in India, union government ministries and multilateral institutions
on policy design, their implementation and evaluation.
Soumyanetra Munshi is currently an associate professor at the Economic Re-

search Unit (ERU), Indian Statistical Institute (ISI), Kolkata. Previously, she
held faculty positions at the Indira Gandhi Institute of Development Research
(IGIDR), Mumbai and the Indian Institute of Management Bangalore (IIMB).
She completed her Bacherlor’s from Presidency College, Kolkata, Master’s from
ISI, Kolkata and earned her doctorate in Economics from Rutgers, The State
University of New Jersey, USA. She broadly works in the fields of political econ-
omy and applied game theory and has explored varied research areas ranging
from conflict and violence, democratization and suffrage rights for women, and
dowry.
Dipyaman Pal is currently working as an assistant professor in the Department

of Economics at Bethune College, India. He has been awarded a PhD degree in
economics from Jadavpur University, Kolkata, India. He has attended a number
of seminars and conferences both at national and international levels. He has
publications in many books and journals of reputation. His areas of interests in-
clude agricultural economics, crime, social sector economics, etc.
xxvi Contributors
Abhilasha Pandey is a former student of the Economics Division of the Indian

Institute of Foreign Trade, Kolkata Campus. She has done her BSc in econom-
ics from the University of Calcutta. She is the recipient of the Junior Research
Fellow scholarship of the University Grants Commission, Government of India
in 2020 and is currently applying to different institutions to register as a PhD
student in the field of economics.
Debraj Ray is Silver Professor in the Faculty of Arts and Science, professor of
economics at New York University and part-time professor at the University of
Warwick. Ray works in the areas of development economics and game theory.
He is a fellow of the Econometric Society, a Guggenheim Fellow, and a fellow
of the Society for Advancement in Economic Theory. He is on the board of
directors of the Bureau for Research in the Economic Analysis of Development
(BREAD). He received the Deans Award for Distinguished Teaching at Stanford
and the Gittner Award for Teaching Excellence at Boston University. He holds
an honorary degree from the University of Oslo and is a co-editor of the Amer-
ican Economic Review.
Ajitava Raychaudhuri received his PhD from American University (USA) and

is a professor and former head and coordinator of Centre for Advanced Studies,
Department of Economics, Jadavpur University, Kolkata, India. He was a sen-
ior Fulbright Scholar at Economic Growth Center, Yale University; Reid Hall
Scholar in Residence of Columbia University, Paris Campus as well as visitor to
SOAS, University of London. He was a visiting faculty at Indian Statistical Insti-
tute (Calcutta), Calcutta University and Indian Institute of Foreign Trade (Cal-
cutta). He had undertaken several important projects from UNESCAP, ADB,
World Bank, UNDP, PEP, GDN, IPPG, SANEI, Finance Commission, Plan-
ning Commission, etc. He has also written numerous papers in international and
national journals and edited books. Among his notable books are India’s Exports:
An Analytical Study (with SugataMarjit) and Myanmar’s Integration with the World
(edited with Prabir De). He has also authored two District Human Development
Reports for West Bengal (with Sushil Haldar) published by UNDP and Planning
Commission of India.
Kasturi Sadhu is a postdoctoral fellow of the Department of Industrial and

Management Engineering, Indian Institute of Technology, Kanpur, India. She
did her PhD degree from the Department of Economics and Politics, Visva-
Bharati (University), Santiniketan, India. She is engaged in empirical and
policy-oriented research on Indian informal sector. She has presented research
papers at different institutions and universities at the national and international
levels. She has written a book-chapter published by Oxford University Press and
a blog-article in GlobalDev. Presently, her research interest is centred on the
condition of informal entrepreneurs and workers in India.
Contributors xxvii
Debdatta Saha is an economist by training. She has completed her doctoral

work in applied game theory and industrial organization from the Indian Sta-
tistical Institute, New Delhi. She currently teaches advanced microeconomics
and industrial organization at the Faculty of Economics, South Asian University,
New Delhi. She has experience of working in some aspects of industrialization
in developing countries, including a brief stint as an expert with the Competition
Commission of India. Her research interests include organizational problems of
small enterprises, competition issues such as abuse of dominance by dominant
firms and coordination games.
Runa Sarkar is a professor with the Economics Group at the Indian Institute
of Management Calcutta. Her interests lie in sustainable development where
business interests are in consonance with environmental and social interests. She
is the chairperson of the Basix Social Enterprise Group and Ctran Consulting
Services, a leading climate change consulting business in India, and on the board
of Basix Consulting and Technology Services. In addition to authoring several
journal and conference papers, Prof. Sarkar has been one of the co-editors of the
India Infrastructure Report (IIR) 2010 on Infrastructure Development in a Low
Carbon Economy and IIR 2009 on Land – A Critical Resource for Infrastruc-
ture, published by the 3i network.
Arjun Shatrunjay is a graduate student at Oxford Department of International

Development, University of Oxford. Prior to joining the University of Oxford,
he was a research assistant in the Indian Institute of Management, Bangalore.
Soumyen Sikdar has been a professor of Indian Institute of Management, Cal-

cutta since 2009. He was awarded PhD degree (Economics) from the University
of Minnesota, USA. He served as a research associate and teaching associate at
the University of Minnesota, USA (Sept 1979–Dec 1983); associate professor
at Centre for Economic Studies, Presidency College, Calcutta ( Jan 1986–Dec
1990); professor at Burdwan University ( Jan 1991–Mar 1995); visiting fellow
at IGIDR, Mumbai (Aug–Sept, 1993); professor at University of Calcutta (Apr
1995–Mar 2009); visiting professor at Institute of Social and Economic Research,
Osaka University, Japan (Sept 1998–Feb1999, Aug–Nov, 2008); and visiting pro-
fessor at Shenzhen University, China (Nov 1–25, 2009). He has contributed in
many national and international journals of reputation and has also published
several valuable books from Oxford, Springer and others. He has recently edited
a book entitled Role of IT-ITES in Economic Development of Asia: Issues of Growth,
Sustainability and Governance (2020) – jointly with Ramesh Chandra Das and
Rajib Bhattacharyya.
S. Subramanian is a retired professor from the Madras Institute of Develop-

ment Studies (MIDS), and a former Indian Council of Social Science Research
xxviii Contributors
National Fellow. He is an elected fellow of the Human Development and Ca-

pabilities Association, and he was a member of the advisory board of the World
Bank’s Commission on Global Poverty. His work has been on aspects of social
and economic measurement, collective choice theory and development econom-
ics. He is the author of, among other books, Rights, Deprivation, and Disparity: Es-
says in Concepts and Measurement; The Poverty Line; and Economic Offences (Delhi:
2006, 2012 and 2013 respectively).
Lore Vandewalle is an associate professor in the Department of Economics at

the Graduate Institute, Geneva where she holds the Pictet Chair for Finance and
Development. Previously, she was a part-time senior researcher at the University
of Oslo. An applied micro-economist, specialized in development and politi-
cal economics, her research mainly focuses on financial inclusion and micro-
enterprise development in India, Bangladesh and Uganda. She has also been
working on political reservations and public good provision in India.
PREFACE
Barring man-made disasters like war, the corona pandemic is the greatest nat-
ural shock to hit the world in a century. As no aspect of life could escape its
lethal touch, the purely economic cost in terms of lost jobs, production, lives and
livelihoods, impaired health, crippled education and the consequent lowering
of productivity has been incalculable. Adding to it the psychological costs of
enforced isolation over prolonged periods, unemployment, postponed or can-
celled plans and shattered dreams, the magnitude of the catastrophe becomes
truly mind-numbing. Several countries, including India, resorted to drastic lock-
downs as measure to contain the spread of the disease. Whether the resultant
benefit has been proportionate to the cost imposed on the society at large contin-
ues to be debated. At the time of writing the menace is far from over, although
the severity seems to be going down. And most positively of all, vaccines have
finally become available.
Naturally, such a devastating event has elicited an enormous amount of reflec-
tion and response from doctors, health workers, educationists, psychologists and
scientists in every field of human endeavour and research. Though controversy
continues to swirl in a foggy environment of ignorance and uncertainty, much
useful knowledge has also emerged. A fairly large number of our friends and
colleagues in the profession have indeed been active in the generation of that
knowledge. As time passed and a clearer perspective slowly began to emerge, the
idea for this book germinated in our minds and it was decided to contact some of
these people for their possible contribution.
Very happily for us the response was both prompt and positive. Almost
everybody agreed to contribute. Timelines were set and for the most part met
without delay. This is indeed extremely praiseworthy, given that conditions –
inside households and in workplaces – are still quite a long way off normalcy.
xxx Preface
The submitted chapters were reviewed and the authors came back with carefully
done revisions, once again within reasonable time limits.
To give some idea about the all-encompassing nature of the shock, the chap-
ters discuss a broad range of issues. As a sample, some of the topics covered are
the relation between epidemic shocks and growth dynamics, lessons for envi-
ronment, education and health policy, income inequality; effect of the pandemic
on firm fragility, on households, on migrant and informal workers, on domestic
violence and marginalized communities like widows, elderly women; India’s ex-
perience with lockdown, conflicts and the question of trust in governments in
pandemic times; and impact of the pandemic on trade, technological changes,
reorganization of global value chains and broad labour market trends.
As should be clear to the reader, every chapter offers careful analysis sup-
ported, wherever possible, by available data.
Rajib Bhattacharyya
Ananya Ghosh Dastidar
Soumyen Sikdar
ACKNOWLEDGEMENTS
We take this opportunity to record our deep gratitude to all the contributors and
the anonymous reviewers for their very sincere efforts. A special word of thanks
goes to the entire editorial team at Routledge, especially Shoma Choudhury for
her help and cooperation. The editors would also like to thank their families for
their patience and support throughout this endeavour.
Editors:
Rajib Bhattacharyya
Ananya Ghosh Dastidar
Soumyen Sikdar
1
CORONA PANDEMIC AND THE
CHANGING GLOBAL ECONOMY
Rajib Bhattacharyya, Ananya Ghosh Dastidar
and Soumyen Sikdar
1.1 Introduction
The largest unprecedented and devastating health shock of the twenty-first cen-
tury appeared with the outbreak of Covid-19 virus, which was first detected by
the end of December 2019 in Wuhan city of China. The reported cases started
increasing from the end of January 2020 outside mainland China. By the end of
March 2020, this outbreak had affected so many countries that WHO declared
it a ‘pandemic’. Indeed, it has already been proved that this kind of health pan-
demic may be far more fatal than devastations wreaked by a war. It has brought
the world to a standstill and thrown before it the challenge of ‘life or livelihood’.
It has already generated tremendous human suffering and major economic dis-
ruption. The devastation of this health pandemic has made economists predict
that the second phase of the battle would be even more severe and may take the
world back to the days of ‘The Great Depression’ of the 1930s.
On the macroeconomic front lockdown and other containment measures ini-
tially resulted in disruption of supply chain and production of goods and services
and hence had an adverse effect on gross output. The most severe impact of this
pandemic is the drastic rise in unemployment, particularly in the unorganized
informal sector, which accounts for a major proportion of employment in emerg-
ing countries like India, Brazil and Mexico, to name only a few. Manufacturing
activities other than those of essential items have been virtually stalled. Other
than manufacturing, industries such as wholesale and retail trade, transport,
tourism, storage, communication, construction, real estate, hotels and restau-
rants, art, entertainment, recreation and accommodation were among the worst
hit segments of the economy. Work from home has become a regular schedule
of many employees during lockdown. The restrictions imposed on production
and transport systems had a positive impact on environmental quality across the
DOI: 10.4324/9781003220145-1
2 Rajib Bhattacharyya et al.
world especially during the lockdown. It has also significantly affected the socio-
cultural relationships, behavioural patterns and psychological attitudes governing
human interactions. It has heavily underscored the importance of digital plat-
forms in the provision of services, especially in the fields of education and bank-
ing. Currently the entire global economy is trying to combat the pandemic in
terms of region- or country-specific revival policies amid extreme uncertainty.
The flowchart in Figure 1.1 attempts to capture the impact of Covid-19 pan-
demic on economy and society.
The corona pandemic has affected every aspect of life across the entire world.
Entire ways of life and livelihoods are being forced to change and adapt to the new
normal, the shape and substance of which is yet unclear. As the ‘utterly abnormal’
becomes routine and widely accepted as the new normal, this book brings to-
gether original essays on a wide range of development issues. The authors identify
areas for concern, and provide deep insights into how these issues may be analysed
as well as directions for policy, especially for emerging nations like India.
A variety of issues have been discussed herein, ranging from economic growth,
trade and overall macroeconomic recovery to conflicts between nations, and be-
tween communities and trust placed by societies in governments in pandemic
times. Appropriate policy response to the economic pandemic, implications of
digital technologies for production, labour and labour relations as well as issues
concerning specific groups such as widowed women and migrant workers are
explored in various chapters in the present volume.
FIGURE 1.1 Economic and social impact of corona pandemic.

Source: Authors’ construction.
Corona pandemic & changing global economy 3
1.2 Economic impact: growth, employment and trade

Here we briefly discuss the impact of the pandemic on some major macroeco-
nomic variables, viz., aggregate output, employment and international trade.
1.2.1 Economic growth

There will be a major setback to output growth as per estimates of major inter-
national organizations like the OECD, Asian Development Bank (ADB), UNC-
TAD and WEO. The disruption of global demand and supply chain coupled with
the adverse impact on the financial markets translates into major contraction of
economic growth and revisions in the global economic growth projections.
According to the OECD (March, 2020; June 2020), the world economy is
expected to grow at 2.4% (instead of the initial estimate of 2.9%), while the
growth of the European nations is expected to slow down from 1.2% to 0.8%.
The adverse impact on confidence, financial markets and the travel sector and
disruptions to supply chains all contribute to the downward revision of growth
estimates in all G20 economies in 2020, especially those with strong links with
China, such as Japan, Korea and Australia. According to the UNCTAD forecast
(9th March, 2020), ‘Covid-19’ shock will trigger a recession in some countries and
a deceleration of global annual growth to below 2.5%, often taken as the reces-
sionary threshold for the world economy. Abiad et al. (2020), Dev and Sengupta
(2020) have portrayed three different situations – best, moderate and worst case –
depending on the intensity of future spread of the pandemic. Following the
downward revision in consumption growth, reflecting a sharper than antici-
pated disruption activity, the WEO (World Economic Outlook, June 2020) has
projected a global GDP growth of −4.9%. Weaker private consumptions, due to
lockdowns and social distancing, coupled with rise in motives for precautionary
savings along with subdued investment plans of firms are mainly responsible for
this adverse shock to aggregate demand. According to Global Economic Pros-
pects, June 2020, World Bank Group, the growth predictions are even gloomier.
It estimates negative growth for the world economy (−5.2%) and across major
economies around the globe, including the USA (−7%), Euro Area (−9.1%), Japan
(6.1%) and India (−3.2%), to name only a few. Estimates in the Monthly Eco-
nomic Review (August, 2020), Government of India, for the April–June quarter
also point to significant global contraction of output on a year-on-year basis.
In the case of India, RBI estimates (Table 1.1) indicate a sharp contraction in
key macroeconomic variables like GDP, private final consumption expenditure
(PFCE) and gross fixed capital formation (GFCF) in real terms in 2020–21; it is
likely to be followed by some recovery in 2021–22 as restrictions are eased. Only
agriculture and its allied activities are expected to register positive growth, while
both industry and services are likely to shrink in 2020–21.
Indeed, India’s GDP contraction (23.9%) is the highest when compared
with other nations such as the USA (9.1%), UK (21.7%), France (18.9%),
TABLE 1.1 India’s growth predictions in GDP, GVA and its components (%)
2020–21 2021–22
1. Real GDP growth rate –7.6 9.5

Real PFCE growth rate –9.4 11.9
Real GFCF growth rate –14.5 12.1
Real GVA growth rate –7.3 9.3
Agriculture and allied activities 3.4 3.1
Industry growth rate –7.7 11
Services growth rate –9.6 10.1
Gross saving rate 29 28
Gross capital formation rate 26.9 28.1
Source: Authors’ calculations based on RBI – Survey of Professional Forecasters on

Macroeconomic Indicators – 68th round ( January 12–29 2021).
Spain (22.1%), Italy (17.7%), Germany (11.3%) and Japan (9.9%). This severe
contraction has been attributed to the most stringent lockdown measure im-
plemented in India (as reflected by the Government Response Stringency Index, of
the Oxford University, which assigned a value of 85.6, the highest in the world). It has
been argued that the lockdown has enabled India to achieve the lowest fatality
rate (1.78%, as on 31 August 2020) and has given an opportunity to restructure
the health and testing infrastructure in the country. However, the move has also
been widely criticized for its impact on livelihoods especially in the informal
sector and on migrant workers. In this context, Debraj Ray, S. Subramanian
and Lore Vandewalle discuss the contours of an alternate policy response in the
chapter ‘India’s lockdown’, based on age-graded, rather than blanket restrictions
on the working population. Despite the lockdown India has occupied the first
spot for a continuous period of 50 days in terms of the highest daily infected
from Covid-19 in the world. Currently, even as the surge in Covid cases con-
tinues, economic priorities have taken the centre stage, especially in emerging
nations, and lockdown measures are being relaxed in gradual steps. The strategy
suggested by Ray et al. is especially relevant at the current conjuncture. It would
benefit the self-employed in the informal sector, who are most vulnerable as their
meagre incomes and savings offer little scope for consumption smoothing in the
face of such disruptive shocks. Another aspect of India’s lockdown relates to its
impact on the population of migrant workers. Given hardly any time to plan,
the total shutdown of economic activity left many of these workers stranded,
without the means to sustain themselves in cities. The government’s response in
the initial days of the lockdown reflected a lack of anticipation regarding their
plight. As hordes of such workers took to the streets in desperation, a series of
special trains were arranged for their journey back to rural hinterlands. Did the
movement of migrants actually enhance the spread of Covid across India? This
question is explored empirically in the chapter ‘Lockdowns, migrants and the
spatial distribution of Covid-19 cases in India: consequences of a “tough and
timely decision”’ by Zakir Hussain and Richa Kothari, on the basis of extensive
spatial analysis on the movement of ‘shramik’ trains and spread of the disease.
Given that Covid can happen again, the response, especially if it involves drastic
measures such as lockdown, has to be well thought out and carefully planned
keeping in mind the likely impact on migrants. This is merited on public health
and humanitarian grounds and also for minimizing disruptions to production
processes and economic activities.
1.2.2 Simple AD and AS framework

Simple aggregate demand (AD) and aggregate supply (AS) analysis can be used
to understand the macroeconomic impact of the corona pandemic. Let the ini-
tial equilibrium be at the output level Y0 (Figure 1.2). The initial supply shock
(although different from the global crisis in 2008 or war or natural disaster) by
means of contraction in world production, due to lockdown and disruption in
international trade and the global supply chain, moves the supply curve leftwards
to AS1, and equilibrium income falls to Y2.
Thereafter, the demand-side contraction will start materializing. Initially,
there is complete uncertainty about the progress of the disease as this is totally
new and yet to be fully understood even by eminent medical scientists. More-
over, there are uncertainties regarding the economic policies that will mitigate
and alleviate the disease. Due to lockdown and phenomenon like ‘social distanc-
ing’ disrupting normal economic activities, non-permanent workers (especially
those employed in the manufacturing sector and services like travel and tourism,
AS2
AS1
Price
AS0
AD0
AD1
AD2
0 Y4 Y3 Y2 Y1 Y0 Output
FIGURE 1.2 Interaction of the AD and AS model.

restaurants, entertainment, etc.) will lose jobs. Meanwhile, households anticipat-

ing a financial crunch increase precautionary savings, while firms become wary
to invest until the situation normalizes and also face a liquidity crisis. So, the
demand shifts inward to AD1. This further feeds into the supply loop.
With shrinkage in demand, many firms face severe liquidity crunch and are
compelled to declare bankruptcy. So, the secondary contraction in supply shifts
the AS curve further to AS2. Again, as workers lose jobs owing to business
closures there is downward pressure on consumption, and thereby aggregate de-
mand and the AD curve again shift inwards to AD2. This race between the de-
mand and the supply continues, resulting in large-scale destruction of economic
surplus. So the first round supply shock due to disease and death results in panic
and uncertainty in the second round, which leads to a drop in consumption,
investment and hence aggregate demand. This contraction in demand dries up
corporate cash flows and triggers bankruptcies among firms. This leads to huge
lay-offs and sharp rise in unemployment (particularly among temporary workers
in the unorganized sector). As workers’ purchasing power falls there is a rise in
non-performing loans which weakens the financial sector. The above analysis
indicates how a massive supply shock can trigger deep recessionary trends start-
ing from a ‘normal’ equilibrium position (Figure 1.2). Recently, Dasgupta and
Rajeev (2020) pointed out that India was already facing a growth slowdown
due to demand-side factors but the supply shock has now turned it into a supply
constrained economy.
In the chapter ‘The macroeconomic impact of Covid-19 on the Indian econ-
omy: an exploratory framework’, Ajitava Raychaudhuri also points out that the
corona pandemic worsened an ongoing demand slowdown as it added a massive
supply shock. He analyses the phenomenon in an AS-AD model and explains
recovery is likely to be slow due to excessive uncertainty as consumption expend-
iture remains low, international trade is depressed, income inequality is widen-
ing and the government’s response is focused on monetary measures rather than
short-term income transfers to the poor. With fall in share of the poor (who have
higher marginal propensity to consume), a massive fiscal push would be required
for growth revival, especially in case aggregate demand does not show adequate
interest sensitivity in response to the credit policy incentives of the RBI. Recov-
ery in domestic business investments may be held back due to uncertainty and dis-
ruptions in the supply chain. However, there is evidence that India may be gaining
FDI inflows as many firms reassess investment locations in the Asian region and
also due to spill-over effects of fiscal expansion in the OECD. If FDI is used for
building infrastructure and boosting India’s connections to the global value chain,
the MSME sector which generates a sizeable share of manufacturing employment
and output stands to gain the most in such a scenario. A boost to this sector would
also overcome the stagnation in India’s private corporate investments.
The scale of the supply chain disruption caused in India’s manufacturing sec-
tor by the pandemic and consequent global lockdown is analysed in the chapter
‘Impact of Covid-19 on manufacturing industries in India: networks and trade
dependence’ by Partha Chatterjee and Shweta Jain. In particular, the authors

examine two specific channels of the supply-side shock, viz., dependence on
multiple sources for inputs and reliance on imported inputs, using network and
regression analysis and data from the Annual Surveys of Industry (ASI), and on
Index of Industrial Production (IIP). The authors find that mostly firms that
source inputs from many suppliers are also themselves suppliers to many other
firms; hence, supply disruptions in India are likely to have large ripple effects. By
and large, firms procuring inputs from many suppliers also have relatively higher
share of imported inputs. However, firms that buy inputs from a large number of
suppliers are not necessarily the ones most reliant on imported inputs. As such the
trade shock is likely to have an independent effect from disruptions in domestic
input supply. However, the impact of the two channels (import and dependence
on many input sources) differs between labour- and capital-intensive firms. A
vulnerability index constructed by the authors shows that maximum disruption
will be faced by firms that are more dependent (source inputs from multiple
sources) and import-intensive.
The importance of understanding the nature of supply chain networks within
which firms operate is stressed in the chapter ‘Firm fragility: a re-interpretation
in the times of the Covid-19 outbreak’ by Debdatta Saha wherein she rede-
fines the concept of firm fragility and tries to develop a generalized theoretical
framework for assessing processes that render firms fragile. This issue is espe-
cially relevant in the context of the pandemic, as even ex ante non-fragile firms
(on the basis of balance sheet fundamentals) may be rendered fragile owing to
network-driven inter-firm linkages in the face of a massive shock in the sup-
ply chain. Saha argues that existing definitions of fragility, being entirely based
on outcomes, especially indicators of financial performance, completely bypass
the processes that render firms fragile in the first place. The chapter argues that
firms are bound by a complex web of contracts embedded in the supply network
within which they operate. Therefore, an understanding of how these contracts
govern firm performance is essential to appreciate the processes at work that
may render firms fragile. For instance, the reasons for fragility may often arise
outside the firm, elsewhere within the network as the risks faced by other firms
with which it has a contract matters a lot. In particular, the type of contract it
has is important (e.g., whether it involves immediate (post-delivery), or deferred
payment). Policy implications and empirical measurement of this process and
associated challenges are also discussed in this chapter.
1.2.3 Employment
Based on different scenarios for the impact of corona on global GDP growth
(McKibbin and Fernando, 2020), preliminary ILO estimates indicate a rise in
global unemployment of between 5.3 million (‘low’ scenario) and 24.7 million
(‘high’ scenario) from a base level of 188 million in 2019. In the ‘low’ scenario,
where GDP growth drops by just around 2%, global unemployment would
increase by 5.3 million (with an uncertainty of 3.5–7 million). In the ‘mid’ sce-
nario, where GDP growth drops by 4%, global unemployment would increase by
13 million (7.4 million in high-income countries), with an uncertainty of 7.7–18.3
million. In the ‘high’ scenario, where Covid-19 has serious disruptive effects,
reducing GDP growth by around 8%, global unemployment would increase by
24.7 million (with an uncertainty ranging from 13 million to 36 million).
As of January 2021, estimates indicate the comparison of growth of em-
ployment in Q2 and Q3 of 2020.1 The largest decline is expected in upper-
middle-income countries, but the impact is comparable across all income groups.
However, ILO estimates also predict an uneven impact on workers – in some
sectors workers will be much more affected than in others (Table 1.2).
Another segment of the labour market especially affected by the pandemic
comprises migrant workers. This issue which held centre stage especially in the
TABLE.1.2 Differential impact on workers of various sectors
Economic Sector Impact on Output Growth of Growth of

Employment Employment
Year-on-Year Year-on-Year
2020 Q2 (%) 2020 Q3 (%)
Accommodation and food services High Impact –20.3 –13.6

Arts, entertainment and recreation, High Impact –13.4 –6.3
and other services
Transport and storage Medium High –6.2 –6.1
Impact
Agriculture, forestry and fishing Medium High –3.9 –3.1
Impact
Wholesale and retail trade, Medium High –7.2 –2.8
repair of motor vehicles and Impact
motorcycles
Manufacturing Medium High –5.6 –2.5
Impact
Construction Medium High –8.4 –2.2
Impact
Real estate. business and Medium High –2.5 –2.1
administrative activities Impact
Education Low Impact –1.4 0.1
Human health and social work Low Impact –0.8 0.5
activities
Utilities Low Impact 0.1 1.1
Public administration and defence; Low Impact –1.2 1.8
compulsory social security
Mining and quarrying Low Impact 3.6 2.8
Financial and insurance activities Low Impact 3.4 3.5
Information and Communication Low Impact 5.0 7.3
Source: Authors’ calculations based on the ILO Estimates January 2021.

Indian discourse on policy response to the Corona crisis, has been discussed in
the Chapter by Hussain and Kothari. However, the problems faced by migrant
workers play out on a much larger canvas, as analysed in the chapter ‘Covid-19
pandemic and migrant workers: India and the world’ by Srobonti Chattopadhyay,
which characterizes broad trends in international migration, wherein migrant
flows have mostly been from developing countries in the South to Europe, the
UK and the USA in the North. Chattopadhyay argues that migrant workers have
borne the brunt of the corona crisis as frontline workers, given their high share
in low-skilled jobs and essential services that continued despite lockdown across
Europe. Indeed, migrants from outside the EU, concentrated in low-educated,
low-skilled segments of the labour market, are among the most vulnerable seg-
ments of the labour market as they are likely to lose out to native workers in
the competition for jobs even as economic recovery takes place. While policy
attention has concentrated on attracting high-skilled workers in developed econ-
omies, not enough is being done to mitigate the problems faced by migrants in
low-skilled jobs who also play a crucial role.
Chattopadhyay also highlights the plight of Indian migrants abroad, both
those stranded in foreign lands (especially the Gulf ) amid immense economic
hardships and those who were repatriated after losing jobs abroad. Indeed, the
economic problems faced by international migrants add to negative demand-side
shocks as remittance inflows, a crucial source of balance of payments support,
take a hit. Overall, from a broader perspective, migrants play a crucial role both
in the local economy where they are employed and in their regions of origin.
There is urgent need for policy support to them, not only on humanitarian
grounds but also to mitigate negative demand-side effects and aid revival of pro-
duction activities.
The ILO ( June, 2020) report further focuses on two important aspects of
global employment, viz., (i) compared to the first quarter of 2020, the loss of
working hours of almost all countries affected is more rapid in the second quar-
ter; and (ii) year-on-year changes (from April 2019 to April 2020) in employment
for women and men (aged more than 15 years) reveal that it is more severe for
women than men. Therefore, Covid-19 has disproportionately affected women
workers and exacerbated gender inequalities in the labour market.
In the chapter ‘Labor and the pandemic: a study on work, employment, and
work situation’, Lopamudra Banerjee and Snehashish Bhattacharya argue that the
pandemic has brought out in sharp relief certain particulars about current con-
ditions of work and employment that often remain substratal. Using the ‘work
situation’ of individuals as an analytical lens, they study the differences in Covid
experiences of working populations in the Global North and the Global South,
as exemplified in the cases of the USA and India. The authors record certain fun-
damental and wide-ranging changes in the nature of work and relations of em-
ployment that have been taking place over the past few decades and develop their
idea of work situation. They reflect on how the nature of work performed, the
circumstance under which work is performed, and the position that a working
individual occupies within the economic processes of production and distribu-

tion shape the conditions of precarity in livelihood conditions and chances of
disease exposure observed during the pandemic.
The chapter by Melika Ben Salem, ‘The future of labor segmentation after
Covid-19’, also discusses how broad labour market trends are likely to be affected
by the pandemic. According to Ben Salem, the effects of the corona pandemic on
labour market outcomes will be felt through two channels, directly via increase
in digitalization (e.g., the work-from-home phenomenon) and through disrup-
tions in global value chains. How the labour market adapts to these changes
depends on micro-factors like the extent of social interaction (involved in the
work) and flexibility (whether or not it is amenable to working from home)
inherent in the employment. Under these circumstances, the self-employed in
informal sectors are especially vulnerable as they are often engaged in activities
(small-scale production, sales (vendors), etc.) (Bouey, 2020) with limited scope
for digitalization and have little or no social protection in terms of access to social
safety nets. Also being low skilled and less educated, they are most vulnerable
to being replaced by labour-saving technologies that are least able to adapt by
re-skilling and switching to alternative occupations. The problem is acute in
developing countries like India where the informal sector absorbs a significant
proportion of the total workforce.
The vulnerabilities faced by informal workers are brought out in the chapter
‘The pandemic and the informal sector of India: probable impacts and strategic
intervention’ by Anirban Kundu, Manojit Bhattacharjee, Pratip Kumar Datta,
Kasturi Sadhu and Saumya Chakrabarti specifically in context of India. The
analysis by Kundu et al., based on NSSO data, highlights the magnitude of the
problems faced by the self-employed. The authors call for large-scale institu-
tional support by the government to mitigate the unprecedented challenge for
inclusive development precipitated by this crisis. They recommend provision of
food support in the short run, unconditional cash transfers in the medium run
and fostering growth of clusters in the long run for addressing the problems of
effective demand as well as elasticity of supply. This chapter clearly brings out
the extent of vulnerability of the self-employed in the informal sector in terms
of their meagre incomes and savings that severely impair capacity to absorb
shocks.
Prior to the corona pandemic, the Indian economy was already facing un-
precedented decline in youth employment, mainly due to decline in agricul-
ture and manufacturing and sluggishness in construction jobs and the stalling
of the process of structural transformation (Mehrotra and Parida, October
2019). There was a sudden jump in the unemployment rate from 8.75% (March
2020) to 23.52% (April 2020) due to the lockdown and a sharp fall in labour
force participation rate (Figure 1.3). In the case of India, salaried jobs seem to
have taken the biggest sustained hit due to the current Covid-19-induced lock-
down – salaried jobs, estimated at 86.1 million in 2019–20, fell to 67.2 million
in July 2020 (CMIE Report, August, 2020). Moreover, 52% of job losses were
FIGURE 1.3 Pre- and post-corona unemployment rates in India.

Source: Authors’ calculations based on Centre for Monitoring Indian Economy Pvt Ltd (CMIE).
among those under 40 years of age (CMIE, Consumer Pyramids Household Sur-
vey data). Clearly, job losses concentrated among the young population would
have serious implications for consumer demand and savings.
In the chapter ‘Coping with Covid-19 and its consequences: household adjust-
ments in India’, Arnab Mukherji and Arjun Shatrunjay use the CMIE Consumer
Pyramids data (for April 2020 and for two prior quarters: January–March 2020
and October–December 2019) to analyse household-level adjustment mecha-
nisms at work during the Covid crisis. The authors find a very sharp increase in
inequality and a drop in average incomes and consumption expenditures across
all educational groups. However, they observe marked differences among house-
holds, across occupation groups (with some groups showing significant income
gains vis-à-vis others) and between labour-surplus (from where migrant workers
originate) and labour-deficient (destination states for migrant workers) states.
This chapter also utilizes MG-NREGA data on household demand for labour,
which shows a spike during the lockdown period especially in labour surplus
states, indicating rural households’ attempt to augment incomes in the face of
job loss and economic distress. The authors call for expanding the public dis-
tribution system and reframing labour policy to strengthen social security and
clearing of the labour market, to help households cope with a negative shock of
such magnitude.
In the chapter ‘Measurement of efficiency loss in Indian agricultural sector
due to Covid-19: evidence from rice production’, Dipyaman Pal, Arpita Ghose
and Chandrima Chakraborty highlight another fallout of the migrants returning
to labour surplus states due to job losses and suspension of economic activities
during the lockdown. They argue that the return of migrant workers to their
home states would accentuate the problem of surplus labour in the agricultural
sector as the bulk of such migrants hail from rural India. Using data on paddy
cultivation across 16 major states, the authors show that absorbing migrants in
the agricultural sector would reduce efficiency and roll back productivity gains
in a significant manner.
1.2.4 International trade

The unavoidable declines in trade and output will have painful conse-
quences for households and businesses, on top of the human suffering
caused by the disease itself.
[Statement made by WTO Director-General Roberto Azevêdo in
context of the corona pandemic]
Trade was already slowing in 2019 before the virus struck, weighed down by trade
tensions, especially between the USA and China, and slowing economic growth.
The corona pandemic shock has only led to further deterioration. The WTO has
predicted a decline of 13–32% in trade volumes, which is likely to exceed the trade
slump seen during the Global Financial Crisis of 2008–09 (Figure 1.4).
Maliszewska et al. (April, 2020) suggest that two scenarios may be expected:
(i) a global pandemic and (ii) an amplified global pandemic. In the case of a global
pandemic, it is assumed that only one-half of the countries bear the full impact of
the shock. In the case of the amplified global pandemic, the shocks are uniform
across all countries. Accordingly, two probable scenarios may be predicted in
2020 and 2021: (a) a relatively optimistic one, with a sharp drop in trade followed
FIGURE 1.4 World merchandise trade volume (quarterly), 2008–20 (USD million).
Source: Authors’ calculations based on WTO Quarterly Database.
by recovery from the second half of 2020; and (b) a more pessimistic scenario,
with steeper initial decline and a more prolonged and incomplete recovery.
One may even conceptualize a third situation whereby a sharp decline in
2020 trade volumes along the lines of the pessimistic scenario is followed by an
equally dramatic rebound, bringing trade much closer to the optimistic scenario
by 2021 or 2022. But whatever be the case, the depth of the negative structural
break is prominent in both the cases. Two other aspects that distinguish the cur-
rent downturn from the financial crisis are the role of value chains and trade in
services. Trade is likely to fall more steeply in sectors characterized by complex
value chain linkages, particularly in electronics and automotive products. In con-
text of India, the chapter by Chatterjee and Jain shows how consequent supply
chain disruptions may be problematic especially for manufacturing firms that are
highly dependent on imported inputs.
India’s overall exports (merchandise and services combined) in April–January
2020–21 are estimated to be USD 390.60 billion, exhibiting a negative growth
of (−) 10.63% over the same period last year. Overall, imports are estimated to
be USD 398.47 billion, exhibiting a negative growth of (−) 22.80% over the
same period last year (Figure 1.5). Taking merchandise and services together, the
overall trade surplus for April–January 2020–21 (estimated at USD 1.87 billion
compared to the deficit of USD 72.40 billion in April–January 2019–20), despite
the fall in exports, is indicative of the depth of the recessionary pressures un-
leashed by the pandemic.
Will the corona pandemic act as a catalyst for the process of de-globalization?
The US-China trade war amid the call for putting ‘America First’; closer home,
FIGURE 1.5 India’s foreign trade (as on January 2021).

Source: Authors’ calculations based on New Delhi (press release), dated 15 February 2021;
Ministry of Commerce & Industry.
the Indian government’s push for Atmanirbhar Bharat (self-reliant India) and for
going ‘vocal for local’ in the post-Covid scenario – are these evidence of the same
trend, wherein countries will start disengaging from the international economy
and roll back tariff reductions? Ranajoy Bhattacharyya, Anupriya Gangopadhyay
and Abhilasha Pandey explore this issue in the chapter ‘Impact of the coronavirus
pandemic on de-globalization’. The authors argue that in the post-Covid world,
countries’ bargaining power in trade negotiations, employment intensity of their
import-competing sectors and the weights assigned by governments to social
welfare will be key factors shaping their stance towards globalization. Indeed,
the Indian government’s call for self-reliance seems to address the concerns of
workers in its labour-intensive production sectors where jobs have been lost. It
is unlikely to lead to de-globalization both because tariffs reduce social welfare
and because India’s bargaining power in trade negotiations has remained low – so
that hike in tariffs may eventually have to be rolled back in the longer run.
In the chapter ‘The Covid-19 pandemic and its impact on global political
economy’, Rajesh Bhattacharya and Ujjwal Kango also explore the issue of
de-globalization, in the context of disruptions in the global supply chain and
emergence of populist trends within nations in the post-Covid scenario. The
authors argue that there is a tangible threat to the global production system, aris-
ing especially due to technological factors and falling cost of labour amid rising
unemployment within countries. Indeed, it is pointed out that the economic
fallout of the pandemic, especially if these are long lasting, might eventually take
the world back to the post-Second World War days.
1.3 Social impact: social relations, gender and education

The spread of the Covid-19 pandemic has brought about drastic changes in so-
cial interactions, norms, gender relations and in the education sector all over the
world.
The pandemic has a deep-rooted impact in social relations, customs and be-
havioural patterns in our daily human interactions (Ling and Ho, 2020; Sen,
2020). Wearing face masks, using hand sanitizers, maintaining social distance
and prioritizing health and cleanliness like never before have become a part and
parcel of the daily life of all rational and sensible individuals in the fight against
Covid until the discovery of a proven and effective vaccine which can assure
a return to normal life. Although the unlocking phase has already begun and
some vaccines have been invented and applied in some countries, still it is a long
way till past, normal social practices are restored. After fighting more than eight
months against a deadly Covid-19 virus and receiving the news update of a sec-
ond wave of the health pandemic from various corners of the world, coupled with
uncertainty regarding the ‘true vaccine’, we are still trying to find light in the
unknown darkness. The Government of India has officially permitted the appli-
cation of two vaccines Covaxin and Covishield to its citizens in a phased manner.
Social interactions in public places, ceremonies and religious gatherings are being
either partially controlled or allowed with strict adherence to social distancing

norms after assessment on a case-by-case basis. Regular classes in schools, col-
leges and universities are being substituted by online digital e-classrooms. Use
of online payments through cards, e-wallets and e-banking has been on the rise.
Physical presence in markets, shopping malls, restaurants and movie multiplexes
is being replaced by teleshopping, online marketing and home delivery of food,
medicines, groceries and so on. This has also enhanced the use of digital devices
such as computers, laptops, tablets and mobiles and the use of various ‘apps’ in
daily life. Work from home has become the new normal across various catego-
ries of employees, directly or indirectly associated with the application of IT and
ITES in their lines of work. The use of WhatsApp, Facebook, Messenger, online
movies and games has increased manifold, mostly replacing face-to-face inter-
actions. This has serious implications for health, both mental and physical, food
habits, lifestyles and evolution of social values.
Limited social interactions and large periods of confinement in closed spaces
in the home have also fostered discrimination against women and in many cases
sharp gender divides in roles, responsibilities and work participation norms. In
many cases, women, shouldering a disproportionately high share domestic work
(including increased childcare responsibilities as schools go online), are being
forced to drop out of the workforce. In many poor and backward societies, girl
children are being forced to drop out of school and help with domestic work,
as parental incomes drop and preference is given to the male child. Further, as
Soumyanetra Munshi points out in her chapter ‘Violence amidst virus: an essay
on conflict in times of a pandemic’, women are also bearing the brunt of domes-
tic violence, so much so that rampant violence against women reported in both
developed and developing countries has been termed a ‘shadow pandemic’. In
this context, Sutirtha Bandyopadhyay and Bipasha Maity point out in the chap-
ter ‘Covid-19 pandemic: implications for widowed and elderly women in India’
that widowhood is a separate channel of vulnerability for women in India. The
authors’ analysis based on household-level data from NSSO finds evidence of
discrimination against elderly widowed women within the household (in terms
of consumption differentials with married women of the same age group). Wid-
owhood is also associated with lower asset ownership (land and livestock in rural
areas) and incomes and higher incidence of health problems. As such this group
of women are likely to bear the brunt of the Covid crisis, being more vulnerable
to contract the disease and less able to afford healthcare.
Coming to education, due to lockdown, social distancing and various con-
tainment measures, the future of the school and higher education is facing tre-
mendous uncertainty. Some countries like the USA, France, Nigeria, etc. tried
to reopen schools physically during the months of July and August, but they were
compelled to abandon the attempt and close down due to tremendous increase in
the spread of infection among children. Such instances have brought into sharp
focus the issue of ‘traditional vs online education’ – which one is more effective,
accessible as well safe from a public health point of view. According to UNESCO
Director-General Audrey Azoulay, ‘we are entering uncharted territory and working
with countries to find hi-tech, low-tech and no-tech solutions to assure the continuity of
learning’.
The education sector displayed immense flexibility during the pandemic and
was among the first to fully adapt to the changed circumstances. Motivation
towards e-learning has already gained momentum across the globe as well as in
India. However, serious questions of access have also been raised as the presence
of a digital divide has been recorded in both developed and developing countries
like India. In India, where a vast majority of the population lacks access to inter-
net or smartphones, it is very difficult to implement a mass e-learning strategy
that would be inclusive. This would require not only a reorientation of teachers,
students, parents and other education workers but also a major restructuring of
the physical and technological infrastructure in the sector and the all-important
question of providing mass access to quality education.
Saumen Chattopadhyay writes on the global challenges facing higher educa-
tion in the chapter ‘Higher education in the post-Covid era: India and the world’,
with respect to inclusion and access to quality higher education, the demand for
which has increased manifold with growing importance of the knowledge econ-
omy. The author draws attention to a wide range of extremely pertinent issues
and challenges facing the higher education sector in India, in light of its Na-
tional Education Policy. He comments on problems associated with the growing
tendency for privatization, commodification and unbundling of the teaching-
learning process along with market-based delivery of education which is essen-
tially a public/merit good.
1.4 Towards a revival policy

In what follows, the contours of possible strategies for economic revival in the
post-corona period are discussed especially for India and similar emerging econ-
omies. At the outset, a few specific areas with a direct bearing on the success of
revival strategies are highlighted.
1.4.1 Assessing preparedness: health, social security and fiscal space

A discussion on a strategy for revival requires thorough assessment of basic indi-
cators of human development, prior to the pandemic. Table 1.3 shows the level
of human development of South Asian countries in terms of indicators of health,
access to basic infrastructure, and improvement in ICT connectivity and their
potential to combat the corona crisis. In terms of all the sub-indices of health
(i.e., government health expenditure, hospital beds, nurses, physicians), India is
lagging far behind the world average. Almost the same picture emerges in ICT
connectivity. As the health system is the first line of defence, the drawbacks were
exposed as it was hit hard by the pandemic.
TABLE 1.3 Health system, basic access to infrastructure and connectivity indicators in South Asian economies and world
Health System Access to Basic Infrastructure Connectivity
Govt. Hospital Nurses and Physicians Basic Drinking Basic Open Fixed Mobile Cellular
Health Beds (per Midwives (per 1,000 Water Sanitation Defection Broadband Subscription
Expenditure 1,000 (per 1,000 people) Services (% of Services (% (% of Subscription (per 100
(as % of people) people) population) of population) population) (per 100 people)
GDP) people)
2. Afghanistan 0.49 0.47 0.17 0.27 57.59 38.75 15.49 0.02 53.41
Bangladesh 0.47 0.67 0.26 0.46 96.72 43.78 2.88 2.47 76.69
Bhutan 2.49 1.75 1.28 0.31 95.38 65.67 1.56 2.41 83.3
India 0.91 0.7 1.37 0.74 89.81 50.48 35.17 1.21 75.4
Maldives 5.21 4.3 5.65 2.57 98.42 96.2 1.13 5.76 155.59
Nepal 1 0.3 2.31 0.67 87.13 51.28 30.54 1.12 88.85
Pakistan 0.72 0.6 0.52 0.9 90.31 54.45 16.31 0.8 65.26
Sri Lanka 1.62 3.55 1.92 0.84 87.39 93.41 1.29 3.3 109.72
South Asia 0.89 0.67 1.14 0.74 89.81 50.65 29.28 1.28 74.57
East Asia & Pacific 4.54 3.67 2.85 1.48 90.9 79.43 3.18 15.37 99.96
World 5.8 2.7 3.44 1.45 88.05 69.94 11.54 10.92 93.72
Source: Authors’ calculations based on UNESCAP (Covid 19 and South Asia), 2020 and World Development Indicators 2018, accessed at https://databank.worldbank.
org/source/world-development-indicators.
The importance of health infrastructure is brought out by Shankha

Chakraborty and Mausumi Das in the chapter ‘The long shadow of epidemics’
as they explore the impact of a health shock on economic growth in developed
countries (where highly developed health infrastructure keeps mortality rates
low) vis-à-vis developing nations (where mortality rates are high owing to, inter
alia, poor health facilities). The theoretical model of endogenous growth in an
overlapping generations framework demonstrates the likely outcome when both
sets of countries are affected by a health epidemic. By and large in developing
countries, production processes tend to rely more on physical rather than human
capital. This creates inflexibilities that hinder adaptation to new (especially dig-
ital) technologies and may lead to growth collapses. Improvements in health and
educational infrastructure help mitigate such negative outcomes by reducing the
severity of the mortality shock associated with epidemic episodes. The difference
in growth outcomes between India and Bangladesh in the post-Covid situation
supports this conclusion, especially in view of the strides made by Bangladesh in
improving mass access to social infrastructure in recent years.
In the chapter ‘Policy failure before Covid and options for the government
in near future’, Arijita Dutta and Montu Bose analyse available data on mor-
bidity and mortality due to Covid-19 across states in India as they examine the
extent of policy failure in response to the epidemic on the health front. The
analysis indicates that neglect of public health, juxtaposed with a heavy burden
of non-communicable diseases, can impose huge social costs in terms of loss of
lives. The authors advocate a far more holistic approach as they emphasize the
need to combine disease surveillance with a strong preventive mechanism in the
aftermath of Covid to strengthen and revamp the public health system altogether.
The second factor which has surfaced starkly due to the pandemic is the vul-
nerability of jobs in the informal sector and lack of social security (Figure 1.6).
The health shock of corona has ultimately turned into a livelihood and labour
market shock. Due to the lockdown and other containment measures, millions
of migrant workers, casual labourers and daily wage earners lost their jobs, and
were left devoid of income, food and other essentials exposing the glaring la-
cunae in India’s social security measures. This point has surfaced repeatedly in
various chapters in this volume, including the ones by Ray et al., Kundu et al.
and Hussain and Kothari.
Many countries of South Asia like India, Pakistan, Sri Lanka and Maldives
have high level fiscal deficits, coupled with relatively high levels of public debt
(Figure 1.7). As the corona pandemic is expected to squeeze additional public
revenues due to depressed business activity, it will be very difficult to sustain
fiscal stability with very low levels of tax-GDP ratio and low public revenues.
In this context, possible repercussions of various policy measures require
careful deliberation, as Hiranya Lahiri argues in the chapter ‘Covid-19 and an
assessment of some policy actions adopted by India’. This chapter evaluates the
Indian government’s policies of reducing interest rates and raising fuel taxes in
the wake of the pandemic. While raising indirect taxes enhances fiscal space, it is
FIGURE 1.6 Employment with and without social security in selected Asian countries.
Source: Authors’ calculations based on UNESCAP SSWA (2018) World Bank data.
Notes: Employment with social security refers to the share of the labour force actively contrib-
uting to old-age pension schemes.
120
100
80
60
40
20
0.6
0
-1 -5.2 -7.4 -5.6 -4.6 -8.8 -6.8
a
ka
-20
al
n
an
sh
s
di
ta
ve
ep
n
an
In
de
st
ta
di
N
Bh
iL
ki
is
la
al
Pa
an
Sr
ng
M
gh
Ba
Af
Fiscal Balance (% of GDP) Public Debt (% of GDP External Debt (% of GDP)
FIGURE 1.7 Extent of fiscal and debt sustainability.

Source: Authors’ calculations based on UNESCAP, IMF Fiscal Monitor (April 2020), World
Development Indicators and Fiscal Space Dataset (April 2020).
also inflationary and may hinder demand recovery. Further, in an open economy
setup, low interest rates can lead to capital outflows and exchange rate depreci-
ation, which may negatively impact supply-side recovery by raising the cost of
imported inputs.
1.4.2 Possible strategies for mitigation and revival

Apart from restructuring and expanding health infrastructure with participation
of both government and private sectors, several other measures are also needed
to mitigate the adverse impact of the pandemic (PWC – April, 2020; August,
PROBABLE STRATEGIES AND POLICIES FOR REVIVAL OF ECONOMIES
REMOVAL OF
DEVELOPMENT DEMAND SECTOR STRATEGIES OF
SUPPLY
OF HEALTH MANAGEMENT SPECIFIC INTERNATIONAL
BOTTLENECK COOPERATION
CARE SYSTEM POLICIES MEASURES
STRATEGY
1. Enhancing 1. Fiscal and 1. Strategy of self 1.Labour-Informal 1.Building globally

Tesng Facility Monetary reliance- workers and sustainable food
2. Tracking smulus developing migrant security
package indigenous labourers programmes.
3. Increase in
supply chain
Health 2. Cash transfers, 2.Micro, Small and 2.Internaonal
and reduce
infrastructure Unemployment Medium cooperaon for
foreign
4. Vaccinaon & insurance and dependence Enterprises combang Health
immunizaon PDS or food (MSMEs) Emergencies
programmes raoning.
2. Exploring new 3.Industry 3.Posive role of
5. Quaranne, 3. Extension of tax areas of 4.Trade Internaonal
Isolaon and and debt comparave 5.Finance instuons-like
social payments on advantage 7& IMF, WB, WHO,
6.Environment
distancing loans and EMIs product WTO for
diversificaon 7.Digital Educaon accelerang global
measures and reducon
& knowledge recovery
in
3. Lowering tariffs Upgradaon
interest(asinves
tment incenve) to boost exports 8.Investment in
Infrastructure
.
FIGURE 1.8 Probable strategies and policies for revival of economies.
2020). Indeed, an appropriate policy framework must be planned to reverse the

recessionary trend. Key areas for attention are discussed below (and summarized
in Figure 1.8) under the heads of health care, demand management, supply man-
agement and sector-specific measures. Thereafter, this chapter concludes with
comments on the limits to international cooperation, an imperative for the cur-
rent times but fraught with underlying counter-tensions.
1.4.2.1 Strengthening health-care system

In view of the possibly long-lasting and extremely infectious nature of Covid-19,
not only must health-care management systems improve efficacy of testing,
tracking, quarantine and isolation at present, but also focus attention on the fol-
lowing issues in the immediate future:
a Harnessing the use of technology for health care and fast-track the devel-
opment and adoption of accessible tele-medicine portals and online med-
ical apps. Technology can be effectively used for a coordinated response
to health emergencies (e.g., through regular check-ups for vulnerable and
asymptomatic population based on hotspot mapping, virtual health-care

portals, effective matching of supply of medical kits, medicines, ambulances,
etc., with end use). This calls for creatively channelizing India’s existing
strength in the IT and ITES sectors for efficient heath management systems.
b Keeping telecom rates affordable to ensure inclusive access to self-quarantine
and medical apps via mobile and GPS facilities.
c Extension of medical heath infrastructure so that the health systems can
respond to such epidemics (via testing, tracking and treating patients) as well
as on the needs of other patients. For this the average expenditure in public
health infrastructure in South Asian countries, which is around 0.89% of
GDP (compared to global average of 5.8%), has to be enhanced substantially
at the earliest.
d Cost of private health-care services that are filling gaps in public health-care
systems must be strictly regulated. The Indian government’s Ayushman Bharat
scheme, one of the world’s largest health insurance plans, can potentially play
a crucial role in inclusive access to healthcare in this context. Reports of de-
lays in payments, refusal in reimbursement etc. should be addressed urgently
and smooth functioning of this programme ensured at the earliest.
1.4.2.2 Demand management policies

Standard macroeconomic analysis suggests that demand management policies
can be effectively used to improve capacity utilization and to increase income,
output and employment in the short run. Fiscal measures aimed at increasing
resources for the health sector, relief and social assistance through income and
cash transfers, wage support and in-kind assistance and food vouchers, provi-
sions for deferred payments for taxes, rents, interest payments and public utilities
along with monetary policy easing for liquidity infusion to safeguard production
capacity and economic activity – these are some of the steps taken by various
governments so far. Monetary authorities have also taken measures to ensure
adequate liquidity and stability of financial system through reductions in policy
rates and lowering of reserve requirements.
There is wide variation in size of the fiscal stimulus among Asian countries.
While Japan announced a USD 2.18 trillion package in two tranches, which was
over 40% of GDP, China’s fiscal support was a meagre 3.5% of GDP (Figure 1.9).
To some extent, this also reflects the scale and nature of the economic impact
of the pandemic – even though Covid-19 originated in China, it was one of the
first countries to show signs of growth recovery. The Government of India had
announced a comprehensive relief package of about 10% of GDP (including both
fiscal and monetary measures) for migrant workers, small vendors and farmers
and to boost credit off take especially in the MSME sector.
So far, the Indian government has made several policy declarations aimed at
providing relief to those affected and reviving growth. For instance, on 26 March
2020 the Finance Minister announced a package of ₹1.7 lakh crores which largely
FIGURE 1.9 Fiscal stimulus package in selected Asian countries (as % of GDP).
Source: Authors’ calculations based on UNESCAP Covid-19 Stimulus Tracker (https://www.
unescap.org/covid19/policy-responses).
aimed at providing a safety net for those worst affected by Covid-19 crisis. The
government further announced an economic stimulus package of ₹20 lakh crores
on 13 May. An additional 50,000 crores for MSMEs were announced through the
‘Atmanirbhar Bharat’ package, which seeks to promote self-reliance by strength-
ening Indian enterprises and supply chains. Overall, the Indian government’s pol-
icies seem to rely more on monetary measures, with much greater stress on debt
relief and credit access for businesses, than on fiscal measures based on direct short-
term cash transfers. However, government expenditure on wage-labour–based
poverty alleviation schemes such as MGNREGA has proven extremely effective
during the pandemic. The analysis by Mukherji and Shatrunjay in this volume
underscores the role of this scheme in providing a safety net in rural India amid
economic distress caused by widespread job losses, especially among migrants.
The countrywide lockdown and consequent disruption in economic activities
had serious impact on government revenue collection, at a time when increased
health expenditure to tackle Covid-19 and fiscal stimulus for economic revival
was needed. Clearly, therefore, the fiscal deficit is likely to shoot up considerably.
There are simply no easy choices as massive public borrowing will essentially
‘shift the crisis to the government’ and pave the way for a future debt crisis. The
government could consider monetizing a substantial part of the fiscal deficit to
avoid this – an intervention commonly known as ‘helicopter money’.
Money-financed fiscal interventions are a powerful tool … policymakers

should resort to them only in emergency situations …. Unfortunately, that
emergency is currently upon us ….If ever, the time for helicopter money
is now …
[Centre for Economic Policy Research Cukierman (2020), Paul (2020)]
India’s fiscal deficit touched 115% of the budgeted target in the first half of 2020–
21 as the Covid-19 pandemic continued to hurt revenue generation while ex-
penditure high. The gap between revenue and expenditure stood at ₹9.14 lakh
crores during April–September.2 This has really been a matter of grave concern
to India and many other developing countries that were already facing deep
fiscal problems prior to the corona pandemic. Given the prospects of a deep
recession and possible need for a Keynesian fiscal response in that context, the
task for the future is cut out and a search for effective policy decisions is the need
of the hour.
In the chapter ‘Developing economic herd immunity in the face of a pan-
demic’, Parikshit Ghosh points out that to address the income distribution shock
delivered by Covid-19, existing tax codes may be inadequate to rise to the chal-
lenge. The virus creates a classic negative externality: greater economic activity
spreads the disease faster but agents do not internalize this beyond personal health
risk. This chapter analyses how these twin problems can be addressed using tax
transfer policies and economic incentives. It shows that there is no conflict be-
tween efficiency and equity – policies which reduce the disease burden optimally
must also redistribute from agents and sectors that have suffered relatively smaller
losses to those that have been beset with big shocks. Lags in income assessment
can be overcome by paying a universal basic income upfront and postponing tax
liabilities. Even from a public health perspective, an egalitarian response to the
heterogeneous impact of Covid-19 can be more effective than blanket measures
like lockdowns coupled with fiscal conservatism.
1.4.2.3 Addressing supply bottlenecks

The supply-side impact of the pandemic, as emphasized especially by Chatterjee
and Jain and Saha in this volume, needs immediate policy attention. For Indian,
the magnitude and depth of the crisis is already evident in preliminary data.
Barring coal, electricity and steel, all sectors – crude oil, natural gas, refinery
products, fertiliser and cement – recorded negative growth in September 2020.
During April–September, these sectors’ output dropped by 14.9%, compared to
a growth of 1.3% in the same period a year ago. The output of eight core in-
frastructure sectors dropped by 0.8% in September, contracting for the seventh
consecutive month, mainly due to decline in production of crude oil, natural gas,
refinery products and cement.3
The RBI’s 90th round of Industrial Outlook Survey of companies in the man-
ufacturing sector (for Q1:2020–21) indicates faltering of major indicators like
production, order books and employment conditions. The Business Assessment
Index (BAI) fell sharply to an all-time low of 55.3 in Q1:2020–21 and the Busi-
ness Expectations Index (BEI) also slipped into contraction zone as it stood at
99.5 for Q2:2020–21. However, as total lockdown measures were eased and many
parts of the country opened up, the Monthly Economic Review, Government
of India (GOI) (August, 2020) showed broad-based resurgent growth in several
areas such as PMI manufacturing, e-way bills, kharif sowing, power consump-
tion, railway freight, cargo traffic and passenger vehicle sales. Manufacturing
purchasing managers’ index (PMI), after marginally slipping in July, moved into
expansionary zone in August for the first time, while the global manufacturing
index reached a 21-month high of 51.8 in August, 2020.
With its emphasis on self-reliance the Indian government’s revival policy is
trying to strengthen indigenous supply chains and reduce dependence on im-
ported intermediates, especially from China. This is clearly evident from the
importance being given to the ‘Make in India’ and ‘Vocal for Local’ strategies
at the highest level in the government. While these are focused on developing
local manufacturing bases and supply chains, India must also concentrate on
producing for the export market, both for earning valuable foreign exchange and
for enhancing production efficiencies (producing qualitatively better products
at competitive prices). Exploring new areas of comparative advantage, product
diversification and removal of supply bottlenecks are key for a successful export
strategy. Finally, for effectiveness of the central government’s development strat-
egy, ‘Economic Resurgence Plans’ at state level have to be in sync. Government
programmes and projects for revival must be designed to reflect ground realities
with regard to current status of resources, productivity and capabilities and im-
plemented accordingly.
1.4.2.4 Some key issues

A few specific policy interventions that should be prioritized are now outlined.
Workers and businesses have borne the brunt of the pandemic, especially because
of the lockdown and associated disruptions. Therefore, labour market measures
for protecting jobs, stimulating the labour demand and supporting employment
and incomes must receive top priority. Schemes such as MGNREGA must be
strengthened further, given their importance for temporary and informal sector
workers in rural India. For such people at the bottom of the pyramid, the gov-
ernment must seriously mull placing a short-term universal basic income scheme
at the very heart of a comprehensive economic relief plan. So far as businesses are
concerned, the MSME sector forms the very backbone of the Indian economy.
Therefore, special economic relief packages for this sector (including measures
such as loan interest waivers, exemption of electricity/water charges, etc.) must
be implemented efficiently to ensure that the benefits actually reach this sector.
On the global front peaceful relations with trade partners in general, and with
neighbouring countries in particular, have critical implications for economic
prosperity. In this context, the chapter by Munshi emphasizes that the pandemic
may have contributed worldwide to the resolution of long-standing conflicts
between nations. With regard to the conflicts and the pandemic, the chapter
‘Brothers in arms: Covid-19 and Hindu-Muslim conflict in India’ by Anirban
Mitra and Arnab Mukherji presents interesting insight. The authors argue that
outbreaks of conflict essentially reflect a fundamental lack of trust between com-
munities. As such, it may be expected that lack of mutual trust and limited so-
cietal cooperation will lead to poor outcomes with respect to outbreak of the
epidemic in conflict-prone areas. However, their analysis of district-wise data
on incidence of Covid in districts where Hindu-Muslim conflicts have occurred
in the past shows contrary results; such districts actually performed better with
respect to Covid-related outcomes compared to others, ceteris paribus. The au-
thors attribute this to better societal cooperation resulting from enhanced trust
within communities created by the possibilities of conflict between communities.
For a while now, India has been engaged in talks to resolve border disputes
with China. In this context, economic ties with China (both trade and FDI) have
been under the scanner, especially in light of China’s military aggression and its
‘debt trap diplomacy’ that has affected neighbouring countries like Sri Lanka.
Currently, the country relies heavily on imported inputs from China across a
wide range of products including chemicals, pharmaceutical products, electron-
ics, electrical parts, to mention a few, and available data points to huge asym-
metries in trade between India and China (UNCTAD, 2019). Manufactured
goods constitute less than 50% of India’s merchandise exports; in contrast, the
bulk of Chinese exports comprises manufactures. India should concentrate on
software exports (by importing necessary hardware) and production of generic
drugs (by reducing import duty on ingredients), and on the automobile industry
(by imposing tariff on automobile imports but allowing FDI).These are some
of the sectors that can be prioritized within the ‘Atmanirbhar Bharat’ policy to
develop domestic industries and reduce dependence on imports.
Ultimately, green, inclusive and gender-sensitive programmes that address
multiple and intersecting deprivations are fundamental for revival. It has been
said that the year 2020 has taught us to appreciate all that we have. When it
comes to the environment, it has certainly shown us the immense value what
we could have had – the blue skies, clear waters and rejuvenated forestry during
the lockdown starkly demonstrated the impact of economic activities on the
environment. In the chapter ‘Green or brown? global environment in the post
Covidian era and challenges for India’, Runa Sarkar emphasizes this point as she
argues that these lessons must be kept firmly in view and placed at the very heart
of plans for economic recovery. In this context, Germany and Spain have shown
the way and India too should use this opportunity to incentivise and build in a
‘green’ component in its restructuring efforts for economy and business.
The pandemic has accentuated inequalities between the rich and the poor as
never before, threatening all achievements made with respect to inclusive devel-
opment till date. Perhaps, it is nowhere as clearly evident as in access to online
educational platforms, all over the world and India is no exception. Education is
a powerful tool for economic empowerment and it is imperative to enable access
to quality education in today’s knowledge economy. In this context, the pan-
demic has highlighted the issue of ‘traditional vs online education’ and the role of
public funding, issues discussed in context of higher education in the chapter by
Saumen Chattopadhyay in this volume. This calls for urgent policy focusing on
affordability of digital education and related issues of low-cost access to internet,
smartphones and other such devices. This is a major issue among the poor and in
many backward and rural areas across India.
While the pandemic is leading to massive job losses and has particularly af-
fected certain sectors like tourism and hospitality relatively more adversely, it has
also created new job opportunities in sectors such as e-commerce, robotics, health
services, and IT and IT-enabled services. The new areas are mostly high-skilled
intensive, which enhances further the value of an educated workforce and under-
scores the need to harness new opportunities through re-skilling of the workforce.
Finally, the importance of maintaining gender parity within development
processes can hardly be overemphasized. Here too the pandemic and the lock-
down have dealt a body blow. As emphasized in the chapters by Bandyopa-
dhyay and Maity and by Munshi, women in general and certain especially groups
within them in particular have borne the brunt of Covid-19 on multiple fronts;
be it girl children being forced to drop out of school, or women dropping out of
the workforce for childcare, or vulnerable elderly women missing out on health-
care or domestic violence, just to name a few. The only way to address this issue
would be to firmly incorporate a gender parity component within each and
every development measure put in place for economic revival.
1.5 Conclusion
The Covid-19 pandemic has shown us how very much citizens trust their gov-
ernments. Basic rights such as the right to practise one’s profession, the liberty
to socialize, to move about were all suspended for a significant period of time
during the lockdown, all over the globe. As Catherine Bros points out in the
chapter ‘Riding the wave or going under? The Covid-19 pandemic and trust
in governments’, there was a surge in nationalistic emotions, as all over the
world, citizens reposed their trust in leaders who were doing their best under
the circumstances. However, as the surge in Covid-19 cases continues unabated,
halting economic recovery, increasingly leaders will have to do a more difficult
balancing act between health and economy to preserve political stability and
prevent social unrest.
This is absolutely crucial. The need for global cooperation at the current con-
juncture is greater than ever before. This is the time for nations to set aside
global politics of rich and poor nations, geo-political blocks, trade competition
and come together and stand by each other, providing physical (e.g., access to
vaccines, supply of medical kits, medical teams, masks, sanitizers, etc.) as well as
monetary (e.g., setting up ‘Covid-19 Solidarity Response Fund’) help for crisis-
hit countries. Indeed, such efforts have been observed across the world among
developed and developing countries and country groups (such as the BRICS, the
SAARC, etc.), with world bodies such as the WHO, WFP, ILO, UNHCR, etc.
helping to coordinate and supplement medical and relief efforts across nations.
However, these are difficult times, with economic distress and nationalistic
feelings running high within nations. As such, determination to put ‘our nation
first’ could undermine strong tendencies for global cooperation, especially when
it comes to cross-border movements of goods and people. This issue is explored
by Bhattacharya and Kango as they point out, given the ways in which the pan-
demic has changed the relation between political leaders and citizens, its effects
in defining the political economy within nations and its fallout for global coop-
eration may actually be quite long-lasting.
Ultimately, nations are not just about borders and geographical features, but
about the people who inhabit them. It is faith in human nature, which makes us
hopeful that indeed the ‘world will come together as one’, set aside geopolitical
and geo-economic differences to overcome the present crisis and welcome the
new normal.
Notes
1 The OECD unemployment rate is calculated as ‘harmonized unemployment rates’
which define the unemployed as people of working age who are without work, are
available for work and have taken specific steps to find work. The uniform application
of this definition results in estimates of unemployment rates that are more interna-
tionally comparable than estimates based on national definitions of unemployment.
This indicator is measured in numbers of unemployed people as a percentage of the
labour force and it is seasonally adjusted. The labour force is defined as the total num-
ber of unemployed people plus those in civilian employment. The latest estimates are
based on a new ILO ‘Nowcasting’ model, which relies on real-time economic and
labour market data to predict the loss in working hours in the second quarter of 2020
(on the basis of data available on 1 April).
2 According to the Controller General of Accounts. [https://www.bloombergquint.
com/economy-f inance/indias-f iscal-def icit-in-f irst-half-of-fy21-rises-to-115-
of-budgeted-target- Bloomberg Quint].
3 [https://w w w.bloombergquint.com/business/eight-core-industries-output-
contracts-0-8-pc-in-sep-Bloomberg Quint].
References
Abiad, A., Arao, M., Dagli, S., Ferrarini, B., Noy, I., Osewe, P., Pagaduan, J., Park, D.
and Platitas, R. (2020). The Economic Impact of the Covid-19 Outbreak on Develop-
ing Asia. Asian Development Bank (ADB) Brief, No. 128 [Available at: http://dx.doi.
org/10.22617/BRF200096.]
Bouey, J. (2020). Assessment of Covid-19’s Impact on Small and Medium-Sized Enter-
prises: Implications from China, RAND Corporation [Available at: www.rand.org/
pubs/testimonies/CT524.html].
CMIE Reports, March 2020 Onwards.
Cukierman, A. (2020). Covid-19, Helicopter Money & the Fiscal-Monetary Nexus,
Centre for Policy Research, Discussion Paper-DP14734 [Available at: cepr.org/active/
publications/discussion_papers/dp.php?dpno=14734].
Dasgupta, D. and Rajeev, M. (2020). Paradox of a Supply Constrained Keynesian Equi-
librium: The Covid-19 Case. Economic and Political Weekly, 55(22), May 30.
Global Economic Prospects - World Bank, Chapter-3, Lasting Scars of Covid-19 Pan-
demic, June, 2020.
IMF (April 2020) Fiscal Monitor Policies to Support People during the Covid-19
Pandemic.
Impact of Covid-19 on India Economy, March, 20, 2020, FICCI.
International Labour Organization (ILO) Note, Covid-19 and the World of Work: Impact
and Policy Responses, 18th March, 2020; ILO Monitor – June 2020; January 2021.
Ling, G.H.T. and Ho, C.M.C. (2020). Effects of the Corona Virus (Covid-19) Pandemic
on Social Behaviours: From a Social Dilemma Perspective, Technium Social Sciences
Journal, 7, 312–320.
Maliszewska, M., Mattoo, A. and Mensbrugghe, D v d. (April, 2020). The Potential Impact
of Covid-19 on GDP and Trade: A Preliminary Assessment. Policy Research Work-
ing Paper; No. 9211, East Asia and the Pacific Region Office of the Chief Economist
& Macroeconomics, Trade and Investment Global Practice World Bank, Washington,
DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/33605.
McKibbin, W. and Fernando, R. (2020). The Global Macroeconomic Impacts of Covid-19:
Seven Scenarios. [https://www.brookings.edu/research/the-global-macroeconomic-
impacts-of-covid-19-seven-scenarios/.]
Mehrotra, S. and Parida, J.K. (2019). India’s Employment Crisis: Rising Education Levels
and Falling Non-agricultural Job Growth, CSE, working paper, 2019-04, Centre for
Sustainable Employment, Azim Premji University.
Monthly Economic Review (August, 2020). Department of Economic Affairs, Economic
Division, Govt. of India.
OECD Interim Economic Assessment (2020). Corona Virus: The World Economy at
Risk, 2nd March.
OECD ( June 2020). Covid-19 and International Trade: Issues and Actions.
Paul, M. (2020). Explaining Monetary Policy in the Time of Covid-19. Accountability
Initiative, Centre for Policy Research [Available at: https://accountabilityindia.in/
blog/explaining-monetary-policy-in-the-time-of-covid-19/].
Press Release- India’s Foreign Trade: 15th February 2021, Government of India, Minis-
try of Commerce & Industry, Economic Division.
PWC Report (April, 2020). Managing Covid-19 and Post-lockdown Planning.
PWC (August, 2020). Full Potential Revival and Growth- Charting India’s Medium-
Term Journey.
RBI - Survey of Professional Forecasters on Macroeconomic Indicators – 68th Round
( January 12–29, 2021).
RBI ( July, 2020) Industrial Outlook Survey of the Manufacturing Sector for Q1:2020–21.
Sen, K. (2020). Covid-19 and Socioeconomic Impact in Asia- The Case of India, United
Nations University World Institute for Development Economics Research [https://
doi.org/10.35188/UNU-WIDER/WBN/2020-1].
Trade and Development Report Update, UNCTAD, 4th March, 2020.
UNCTAD- trade and Development Report Update, 9th March, 2020.
UNCTAD- World Investment Report, 2019.
UNESCAP- Covid-19 and South Asia: National Strategies and Sub-regional Coopera-
tion for Accelerating Inclusive, Sustainable and Resilient Recovery, 2020
UNESCAP Covid-19 Stimulus Tracker (https://www.unescap.org/covid19/policy-
responses).
UNESCAP SSWA (2018) World Bank Data.
United Nations (August, 2020). Policy Brief: Education during Covid-19 and Beyond.
World Bank Group - Policy Note (April, 2020). Gender Dimension of the Covid -1
Pandemic.
World Development Indicators 2018 [Accessed at: https://databank.worldbank.org/
source/world-development-indicators].
World Economic Prospects (2020). United Nations New York.
World Economic Outlook Update, IMF, June, 2020.
WTO Quarterly Database [Accessed on 20th February, 2021 through http://data.wto.
org/en].
PART I
Conflicts, trust and society
2
THE COVID-19 PANDEMIC AND
ITS IMPACT ON GLOBAL
POLITICAL ECONOMY
Rajesh Bhattacharya and Ujjwal Kango
2.1 Introduction
The Covid-19 pandemic is likely to have a significant impact on the global polit-
ical economy through acceleration of incipient trends. Even before the pandemic
brought the global economy to a screeching halt and hurled it into the worst eco-
nomic recession since the Second World War, major global trends were already
pointing towards the emergence of significant fault lines in the domestic politi-
cal economies of richer countries, and, as a consequence, potentially significant
shifts in the global economic order. The health crisis unleashed by the Covid-19
pandemic has prompted responses from governments on a scale that is unprece-
dented. Harsh lockdowns, social distancing norms and other public health meas-
ures plunged societies into economic, political and social crises. As businesses
suffered losses and/or closures, workers lost jobs and consumers stopped spend-
ing, desperate attempts to stay afloat during the pandemic prompted responses
from businesses that may fundamentally alter the way they are conducted in
the long run. Similarly, the fear of contagion has in turn fuelled a contagion of
fear, leading to support for inward-looking, insular and even autocratic forms of
government. Such political trends are likely to have significant impact on inter-
national economic and geopolitical relations.
In this chapter, we focus on three such economic trends—the changes in
the labour market brought about by technological changes, the re-organisation
of global value chains and changes in social welfare regimes. These economic
trends are also likely to rhyme well with a fourth political trend—the rise of pop-
ulism across the world. Together, these trends can pose significant disruption to
the model of globalisation that had consolidated in the last four decades. We ar-
gue that, going beyond its immediate impact on health, output and employment,
DOI: 10.4324/9781003220145-3
32 Rajesh Bhattacharya and Ujjwal Kango
the Covid-19 pandemic may well turn out to be the point of inflection in global
political economy.
However, our views could be subject to the critique of a possible overemphasis
on the pandemic’s influence on the global political economy due to recency bias.1
Thus, it can be argued that the effects of the Covid-19 crisis would be temporary;
with progress of vaccination and or with herd immunity setting in, a “business as
usual scenario” might be restored. We have two arguments to make here. First,
while Covid-19 is not the first disease outbreak across the world, it is the first
one in which the response to the pandemic by almost all governments across the
globe has been on an unprecedented scale (Reinhart & Reinhart, 2020; Saad-
Filho, 2020). Our second point is more important. Our argument is not that the
pandemic per se will leave a lasting, transformative impact on global economy,
though it might very well do so (Allen et al., 2020). Our argument is that several
transformations were already under way and we were possibly in the middle
of moving away from “business as usual” scenario, when the pandemic hit the
world. The unprecedented responses of governments to this pandemic are likely
to have important impact on precisely those ongoing transformations that we
have identified in this chapter and which we argue will reshape the global politi-
cal economy in a significant way. It would be reckless to try to predict where the
global political economy is headed in the near future—particularly since it has
already been in the churn following the global financial crisis of 2008, and now
with its upending by once-in-a-century global health crisis. Instead, we would
like to make sense of the current conjuncture and suggest a frame for discussing
inter-connected, but distinct trends that have already been visible before the
onset of the pandemic—so that we may think of the impact of the pandemic on
each of these trends, contributing to a possibly different emergent global political
economic order.
The rest of the chapter is organised as follows. The next section traces the
economic impact of the pandemic across the world, setting the stage for one of
the greatest economic recessions since the Second World War and consequently
the return of the big government. It is followed by four sections, each presenting
a distinct contemporary trend—automation and future of jobs; stagnation or
reversal in global value chains (GVCs) in particular and globalisation, in general;
the urgency of novel welfare systems and the rise of populism across the world.
In each of these four sections, the likely impact of the pandemic on the specific
trends is discussed. The concluding section summarises the main arguments,
emphasising the inter-connected nature of these changes.
2.2 The pandemic and the great economic recession

Since the first recorded infection in late December 2019, the Covid-19 disease
has infected more than 50 million and killed over a million people around the
world. While the Covid-19 pandemic began as a health crisis, it had quickly
developed into an economic as well as a socio-political emergency, tearing hard
Covid-19 pandemic and its impact 33
at the very fabric of societies across the world. Before this, never had the world
witnessed such a policy-induced, severe and abrupt suspension of the circulation
of capital, labour and goods on such a scale, with global economies completely
shut down and national borders sealed by the enforced lockdowns and social
distancing restrictions.
The pandemic, and the response to it, has wrecked economies and has trig-
gered the sharpest economic contraction of our lifetimes. Suddenly, millions of
people lost their jobs and livelihoods (International Labour Organization (ILO),
2020). The pandemic is projected to increase the levels of extreme poverty by
hundreds of millions (Sumner, Hoy, & Ortiz-Juarez, 2020).
The recession, due to the pandemic, is also due to synchronised shocks across
all economies. According to the World Bank,2 never before in the recorded
history, not even during the Great Depression (1929–1939) in the last century
and the more recent global financial crisis (2007–2008) have so many countries
hurtled towards a recession simultaneously (Figure 2.1).
The unprecedented response to the pandemic resulted in a massive contrac-
tion of world GDP and global trade (Table 2.1). In its October World Economic
Outlook,3 the International Monetary Fund (IMF) projected that the global
economy would shrink by 4.4% in 2020. For 2021, the IMF forecasts global
growth of 5.2%. In comparison, the World Bank indicates a 5.2% contraction
in global GDP in 2020, with a 4.2% rebound in 2021.4 The S&P forecast is a
100
80
60
40
20
0
1871
1931
1961
1982
1991
2009
GlobalRecessions Percentage of economies in recession
FIGURE 2.1 Share of economies (%) in recession, 1871–2021.

Source: World Bank.
Note: Data for 2020–21 are forecasts. Shaded areas refer to global recessions.
TABLE 2.1 Global GDP growth forecasts: percentage change in GDP
2019 2020 2021
S&P IMF BMI S&P IMF BMI
USA 2.3 −5 −4.4 −4 5.2 5.2 3.9

China 6.1 1.2 1.8 2.2 7.4 8.2 5.8
Eurozone 1.2 −7.8 −8.2 −8.1 5.5 5.1 5.4
UK. 1.4 −8.1 −9.8 −9 6.5 5.9 5.5
Japan 0.7 −4.9 −5.2 −8.8 3.4 2.3 2.1
India 4.2 −5 −10.3 −8.6 8.5 8.8 6.2
Brazil 1.1 −7 −5.8 −5.3 3.5 2.8 3.2
Advanced 1.7 — −5.8 −5.8 — 3.9 4.2
Economies
Emerging 3.7 — −3.3 −2 — 6 4.9
Economies
World 2.8 −3.8 −4.4 −4.2 5.3 5.2 4.5
Sources: IMF World Economic Outlook; BMI Research database; S&P Global Economic forecasts.
little more optimistic, and it predicts that the world GDP will contract by 4.1%
in 2020.5 Overall, we can observe that in case we do not have any further major
outbreak, the global economy will shrink in the range of 3–5.5% in real terms
in 2020, which would make this the steepest slowdown since the Great Depres-
sion of the 1930s. A fall in output in this range makes it far more severe than the
2007 global financial crisis when the world economy contracted by around 1%
between 2008 and 2009.
Developed countries have been hit harder, and are expected to shrink nearly
by 6%—ranging from a contraction of about 4% in America to almost 10% in
Britain. Developing countries are expected to contract around 2.5%, and if
China is omitted from the list of emerging economies, the contraction for 2020
is expected to be much higher (Figure 2.2). Moreover, even this bleak outlook is
subject to significant uncertainty and downside risks, with economic forecasters
revising their data frequently.
The lockdowns and the subsequent slowdown in the global economy have
also impacted the movement of goods and people, causing international trade
to fall sharply—registering the slowest growth in global trade in post-Second
World War history (Figure 2.3). According to the WTO, global merchandise
trade fell by 14.3% in the second quarter of 2020, recording the sharpest ever
one-period decline compared to the previous quarter. The steepest declines in
exports were witnessed in Europe and North America, where exports fell by
24.5% and 21.8%, respectively.6
The Covid-19-induced economic crisis raises several questions regarding the
extent of the recession, the recovery and whether the shock to the global econ-
omy will be “temporary” or it will leave “permanent scars” on the economies.
FIGURE 2.2 Real GDP growth (%) for world, developed and emerging economies.
Source: BMI Research.
Lowest in post-WWII history (2009)

15
10
-5
-10
-15
2017
2018
2019
2020
2021
FIGURE 2.3 Growth in global trade.

Source: World Bank.7
Note: Trade growth for the years 2020 and 2021 are forecasts. Trade is the average of import
and export volumes.
When the recovery begins, it might take several years before the overall GDP of
most countries returns to their pre-pandemic levels. This is due to long-lasting
changes in the global economy due to demand- and supply-side shocks. On the
demand side, a shift in consumer spending preferences, weak consumer confi-
dence and high unemployment can open the door to stagnation traps. However,
the reallocation of capital and labour from shrinking industries to growing ones
would result in lower levels of output for a substantial period.
According to Mann (2020), the intertwined nature of international com-
modity markets, financial markets, public sentiment and the economy is likely
to inflict long-term damages to the global economy. The household savings
rate would be an interesting indicator to track in terms of consumer sentiment
and future growth. In April, the US household savings rate climbed to 33%,
while it fell to nearly 14% in August 2020; it is still much above the usual
7–8% range.8 In the UK, the household savings rate increased to 28.1% in the
second quarter of 2020 from 9.1% in the first quarter of 2020.9 The household
savings will drop in the future, but it is unlikely that it would go back to the
pre-pandemic levels in the short term, as consumers will stay away from any
avoidable consumption that involves personal exposure to the infection (as in
dining, shopping, travelling, vacationing, attending entertainment and social
events, etc.).
Besides its economic impact, the outbreak of coronavirus has magnified the
fault lines between globalisation and uneven development. This crisis has also
exposed the fragility of GVCs, has led to calls for industrial sovereignty (in the
USA and India, for example) and underlined the relevance of regional or do-
mestic value chains. Micklethwait and Wooldridge (2020) call this the era of the
return of the government. According to them,
the COVID-19 pandemic has made government important again. Not just
powerful again (look at those once-mighty companies begging for help),
but also vital again: It matters enormously whether your country has a
good health service, competent bureaucrats and sound finances. Good gov-
ernment is the difference between living and dying.
(ibid)
In a similar vein, Saad-Filho (2020) has signalled an end of neoliberalism, and a

transformed post-pandemic world where,
all that was solid melted into air: ‘globalisation’ went into reverse; long
supply chains, that were previously the only ‘rational’ way to organise pro-
duction, collapsed and hard borders returned; trade declined drastically;
and international travel was severely constrained. In a matter of days, tens
of millions of workers became unemployed and millions of businesses lost
their employees, customers, suppliers and credit lines.
(Saad-Filho, 2020: p. 1)
However, we question this idea that the Covid-19 pandemic has unleashed these
radical changes and unprecedented shifts in the global political-economic order.
Instead, we argue that the global shock inflicted by the coronavirus is leaving
its imprint on several already discernible and incipient trends. These relate to a
re-organisation of the changes in the labour market brought about by technolog-
ical changes, the re-organisation of global value chains, changes in social welfare
regimes and the rise of populism across the world. The next sections discuss
how the Covid-19 might lead to enduring shifts in the global political economy
through an acceleration of these incipient trends.
2.3 Automation and jobs

The pandemic has had a devastating impact on the labour market. A historic
number of people suddenly became unemployed as a result of partial or total
lockdown of businesses and restrictions on movement and social gatherings. The
International Labour Organization (ILO) warned of the estimated 17.3% loss
in working hours worldwide—equivalent to 495 million full-time jobs by the
second quarter of 2020.10 These estimated job losses are gigantic, when com-
pared with nearly 22 million full-time job losses during the economic recession
of 2008–09.11 Jobs in sectors like services and retail, where physical presence is
unavoidable, were severely affected by blanket lockdowns (Aum, Lee, & Shin,
2021). Particularly, low-skilled services—in many cases, involving physical prox-
imity of provider and consumer in the act of consumption—were the worst hit
as these carried the greater likelihood of transmitting infection and faced diffi-
culties of being transitioned to remote work. Additionally, in developing econ-
omies, where a significant portion of employment is in the informal economy,
lockdowns have caused significant migrations, from the urban regions to the
villages and towns in home states.
Overall, the pandemic has significantly transformed the world of work for
almost everyone: millions of people suddenly became unemployed; many people
had to work from home for the first time, and several others had to adjust to new
working conditions with numerous constraints and risks. The pandemic also ac-
celerated the digital transformation of work, as the physical movement of people
was restricted to keep infections low. Hotels, highway toll booths and airports
replaced human workers with automatic kiosks, and call centres started using AI
chatbots to service customer requests. Several police departments across India
(Kerala, Gujarat, Karnataka, Delhi and Mumbai) used drones and other means
of electronic surveillance to enforce lockdown and social distancing restrictions
with reduced officers in the field. Cadillac’s factory in China begun fully au-
tomating its car welding and painting production lines (Baker McKenzie and
Oxford Economics, 2020). In all these transformations, technology played a very
critical part—such that technology companies, like Microsoft, Apple, Zoom and
Alphabet, reached their greatest share of the S&P 500’s total stock market value
at nearly 40%, topping the dot-com era peak.
With the increasing adoption of digital technologies by numerous businesses

during the pandemic, it seems likely that many in-person service jobs, especially
those provided by small firms, might not return even after the eventual recovery.
Several in-person jobs such as cashiers, hotel staffers, toll collectors and others
might lose to technology as digitalisation and automation get a significant boost.
Not just automation of work, even telework can potentially lead to a reduction
in the number of overall jobs. For instance, the pandemic forced adoption of
technology in sectors like medicine and education. Telemedicine and online
education might reduce the overall requirement of doctors and teachers, as fewer
professionals can serve more patients and students. Moreover, this would signif-
icantly impact the employment of several blue-collar workers, such as clerical
staff and teaching assistants whose services would become less important with the
reduced physical footprint of these services.
Several economists have highlighted the non-linear nature of automation and
pointed that automation of work doesn’t happen at a steady pace. Instead, auto-
mation generally occurs in the wake of economic crisis or other shocks when
there are external pressures such as the rapid decline in business revenues. Studies
by Jaimovich and Siu (2020) and Hershbein and Kahn (2018) have highlighted
that while automation occurs at all times, economic downturns result in a surge
in levels of automation. The coronavirus pandemic, due to its scale and spe-
cial restrictions on mobility, could lead to even more automation. Therefore, it
seems plausible that the pandemic would lead to increasing adoption of labour-
substituting automation. While in the pre-pandemic era, artificial intelligence
and automation were gradually being adopted, this crisis has rapidly accelerated
the process of technological change and made it inevitable.
A consequence of this accelerated digitalisation and automation of work is
more acute polarisation. Scholars had found that over the past decades, the share
of employment in the middle-skill jobs had declined rapidly in a range of ad-
vanced economies (Autor & Dorn, 2013; OECD, 2017). They termed this phe-
nomenon as polarisation of the labour market into high-skill, high-pay jobs and
low-skill, low-pay jobs, conveying images of a dumbbell or hourglass labour
market (Figure 2.4). High-skilled jobs typically require abstract reasoning, an-
alytical abilities and creativity. These jobs generally require people to be highly
educated and employed in technical or managerial roles in finance, engineer-
ing, medicine, etc. Low-skilled jobs either involve manual labour or require
interpersonal care or communications that cannot be easily automated—e.g.,
cooking, cleaning, security. Traditionally, low-skilled jobs such as these have
also been low-paid jobs. Middle-skilled workers with decent salaries consti-
tute the large middle class in any country. These jobs included sales, customer
support, middle management, administrative support, construction, manufac-
turing, transportation, maintenance and repair and so on. It is a large chunk of
these middle-skilled jobs that is being automated due to its procedural or rou-
tine nature. Additionally, the phenomenon of job polarisation is not restricted
1.5
0.5
Low-skilled jobs
0 Middle-skilled jobs
ce
High-skilled jobs
n
ce
nd
ly
a
y
t
n
m
ico
es
l
li a
ee
a
ai
yp
ga
s
ke
di
in
an
o
Ita
an
ad
nd
do
la
at
Sp
tra
Gr
In
Ch
ex
Eg
rtu
r
rm
ra
ai
Fr
n
Tu
St
la
ng
M
Ca
Th
de
Po
er
Ge
Au
d
Ki
ite
th
Fe
d
Ne
ite
Un
n
ia
Un
ss
Ru
-0.5
-1
-1.5
FIGURE 2.4 Average annual % change in employment share, 1995–2012.

Source: World Development Report 2016.12
Note: The figure displays changes in employment shares between circa 1995 and circa 2012.
The classification follows Autor (2014). High-skilled occupations include legislators, senior of-
ficials and managers, professionals, and technicians and associate professionals. Middle-skilled
occupations comprise clerks, craft and related trades workers, plant and machine operators,
and assemblers. Low-skilled occupations refer to service and sales workers and elementary
occupations.
to solely the advanced economies. Maloney and Molina (2016) have shown that
even developing countries have registered a contraction in the middle-skilled
occupations.
The pandemic is likely to accelerate job polarisation as businesses shift to
automation and digital solutions. While automation and telework will result in
the creation of new jobs, such as computer programmers, cyber security experts,
user interface designers, information technology support personnel and others,
these would mostly be jobs at the upper end of the spectrum and require high
level of skills. At the bottom end, in addition to negatively affecting the demand
for several low-skilled service jobs, Covid-19 will accelerate the ongoing trend
of the growth of non-standard, precarious employment: part-time workers, gig
workers and contractual workers. For instance, the proliferation of on-demand
labour platforms, such as Uber, Ola, Swiggy, Dunzo and Urban Company, will
contribute to significant growth of low-skilled precarious jobs requiring last-
mile delivery of services.
2.4 Globalisation and global value chains

A global value chain (GVC) is a chain of separate but inter-linked and coordi-
nated activities in the production of a good or service—each stage in the process
adding value to the end product. Globalisation has been synonymous with emer-
gence of GVCs, which have grown increasingly longer and complex over time.
The Covid-19 pandemic led to significant disruptions in the GVCs.
This disruption was observed from the very beginning of the pandemic, as
factories closed in China, the hub of global GVC production, in January 2020.
Value chains were suddenly disrupted, triggering a supply-side shock to sev-
eral manufacturing centres concentrated in China and its neighbouring coun-
tries. From March 2020 onwards, as Covid-19 turned into a raging pandemic, a
number of countries closed down their borders. The resulting drop in demand
and the imposed restrictions on movement of people and goods led to factory
closures which adversely affected the functioning of GVCs. This resulted in de-
creases in production or changes to sourcing strategies of many organisations,
leading to shortages, hoarding and price manipulations. Several companies (and
governments) faced shortages of critical goods and services, as countries began
to restrict value chain flows to facilitate the urgent needs of domestic custom-
ers. This discrimination was particularly evident in the supply of medicines and
medical equipment such as hydroxychloroquine, ventilators and personal protec-
tive equipment. As a result, the disruption of supply chains during the pandemic
also highlighted the industries in which economies lacked domestic production
capabilities.
The lockdowns, border closures and travel bans which were implemented all
across the world gave rise to a GVC “concussion” (Baldwin, 2020) and exposed
the very vulnerabilities of the model of fragmented international production, in
particular, and globalisation, in general. Some scholars and policymakers ques-
tioned the model of GVCs and argued for making production systems more re-
silient by either localising production or diversifying the supplier base ( Javorcik,
2020). Politicians also asserted the re-nationalising of GVCs in order to insulate
countries from future trade disruptions. For instance, the US President Don-
ald Trump stressed upon decoupling the US economy from China and bring-
ing manufacturing operations back to the USA. Scott Morrison, the Australian
Prime Minister, emphasised nurturing local manufacturing facilities to ensure
lesser reliance on GVCs in future. Narendra Modi, the Indian Prime Minister,
has emphasised self-reliance as the economic fulcrum of the government’s re-
sponse to the pandemic.
It is important to note that the stagnation of the GVCs predates the pandemic.
While GVCs had formed the bedrock of global trade and presented several ad-
vantages to businesses—more efficient sourcing of inputs, access to knowledge
and capital beyond the domestic markets, and expansion of business into new
markets—there was ample evidence of a decline in fragmentation of production
across borders even before the pandemic (Figure 2.5). Since 2011, the process of
international production fragmentation had stagnated. Indicators that measure
55
50
Percent of globbal at rade
45
40
35
30
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
FIGURE 2.5 GVCs as a share of global trade.
Source: World Bank data from World Development Report 2020.13
the fragmentation of production across the world were decreasing since 2011,
pointing out that the GVCs are becoming shorter (Miroudot & Nordström,
2019). In the case of China, which acts as the hub global production networks,
the foreign value-added content of its exports declined from 26.3% in 2005 to
16.6% in 2016.
There are several reasons behind this decline in GVCs, i.e., why firms had begun
to produce locally and relied less on offshoring (De Backer & Flaig, 2017). It has
been associated with the post-2009 slowdown of global growth and investment,
and mounting public and political pressure to reduce carbon emissions through
more local and sustainable production. Also, many GVCs had matured, creating
additional difficulties for incumbents from developing countries to enter those
value chains. Some discern a shift from mass production to mass customisation—
a movement from catering to mass markets to supplying for millions of niche
markets—requiring production to be brought closer to the consumer. Robotics,
automation, computerised manufacturing, artificial intelligence, etc. all could re-
duce the labour costs in developed countries, and largely cancel out the advantages
of low-labour-cost emerging economies. This might indeed also suit businesses
well in the context of growing protectionism and concomitant policy uncertainty
on long-term possibilities of sourcing from certain countries. Now Covid-19 has
revealed the inherent risks in long and complex GVCs stretched across continents
in providing essential commodities when international trade is disrupted.
The Covid-19 pandemic has come at a time when the model of globalisa-
tion was already under mounting criticism (Aguilera et al., 2018; Rodrik, 2018).
The current cycle of globalisation had met several setbacks since the economic
recession of 2009: the sovereign debt crisis in Europe, Brexit, and the US-China
trade war. The rise of China as a global economic and military power had already
triggered the US-China trade war, with both USA and China imposing tariffs
on one another’s goods and trying to force allies to follow suit. The concomitant
rise of populism across a number of countries further intensified calls for prior-
itising national sovereignty. The outbreak of coronavirus has magnified the fault
lines between globalisation and national development.
In short, technology makes it attractive to shorten GVCs and bring produc-
tion home (without significant gains in employment) and the political sentiments
and the disruption due to the pandemic emphasise the dangers of keeping pro-
duction abroad. As domestic economies take centre stage once again, the recent
phase of globalisation, as we have witnessed it over the last four decades, might
well be coming to an end.
2.5 Social entitlements

Globalisation had made any domestic social accord with labour untenable. With
the slowdown in globalisation, acceleration in job polarisation, rise in unemploy-
ment, explosion in inequality, increase in political restlessness and triumphant
populism, a new space for a radical welfare regime might be in the offing—
distinct from both its emaciated form in the heydays of globalisation in the last
four decades and its more robust “welfare state” form after the Second World
War. The Covid-19 pandemic has presented governments across the world with
new challenges as it has impacted the functioning of the economy and has also
led to an unprecedented rise in the levels of unemployment. The pandemic and
the subsequent government response have severely impacted people’s ability to
earn a living. This was even more so for the informally employed workers, who
work without any social protection.
Across the world, the governments have responded to the crisis by providing
increased levels of support to those who lost their incomes and by rolling out
fiscal stimulus packages to spur demand. Japan rolled a massive stimulus package
which was 21.1% of its GDP, USA 13%, Sweden 12%, Germany 10.7%, India 10%,
France 9.3%, Spain 7.3% and Italy 5.7%.14 Several researchers and policymakers
saw Covid-19 as a potential pivot that could lead to a return of the welfare state
(Figari & Fiorio, 2020; Tzagkarakis et al., 2020). Some drew parallels with the Great
Depression of 1930s, which was closely linked with the rise of the welfare state.
However, we argue that (a) we are witnessing not a return to the welfare
state of the twentieth century, but something radically different and (b) the re-
turn to welfare is not only, or even mainly, due to the disruption caused by the
Covid-19 pandemic. The Great Depression and the Post-Second World War era
had ushered in the rise of the welfare state. The late 1970s marked the end of
the era of the welfare state. During the 1970s, welfare states began to be seen as
severely cash strapped and constrained in their typical ways of their operation.
This gave way to fiscal austerity, which shaped the operations of government in
ways that were more business-like. However, the reforms that began in the wake
of the economic stagflation of the 1970s contributed to increasing inequality,
de-unionisation and informalisation of labour. Around the same time, the pro-
cesses that led to the displacement of mid-skilled occupations in the advanced
economies gathered momentum, resulting in job polarisation that we have dis-
cussed in the above section. This also gave rise to a new class structure, with the
rich elites holding significant wealth, and with a rapidly increasing “precariat”,
continuously insecure, and often at the edge of unsustainable debt. At the same
time, in the developing countries, the consistent growth of the informal sec-
tor led to forced exclusion from the labour market itself. The dual process of
marginalisation of labour—through job polarisation and expansion of precarious
wage-employment (in majorly developed countries) and exclusion from the la-
bour market itself or forced self-employment (in developing countries)—had led
to a uniquely fragile global economic system. The new conditions of labour had
rendered large segments of the labour force redundant or substitutable—pointing
to a new reality where welfare systems designed around “temporary unemploy-
ment” have lost meaning.
One radical policy suggestion that has been making rounds over the last few
years and which offers a response to the prevalent marginalisation of labour in
the current cycle of capitalism is to implement a universal basic income (UBI)
scheme. The basic premise of universal basic income is a guaranteed income to
all citizens irrespective of work and income status. The low-skilled jobs that
have been forthcoming in recent times under the gig economy are even more
precarious, as the traditional social security systems do not work for such non-
employment categories. Therefore, under the current conditions of capitalism,
UBI can appear attractive to workers, capitalists and political parties alike—with
the government able to strike a “New Deal” with citizens, the workers able to
derive a modicum of income security in the face of uncertain labour markets and
businesses able to achieve a politically feasible transition out of the extant “dis-
tortionary” and “inefficient” welfare schemes. However, as the State ostensibly
seeks to redefine its relationship with citizens through UBI (or something simi-
lar), the language of insular populism provides the political means of forging that
relationship. While the idea of UBI has been bubbling around the edges of public
discussion in recent times, the Covid-19 crisis has provided a suitable opening
to build public consensus and popular sentiment for such a transition. The lat-
ter has also forced governments to erect overnight the necessary technological
infrastructure and administrative system for the management of the pandemic,
which can potentially provide the blueprint of the supportive structure needed
for running a UBI-like system.
2.6 Populism
In recent years, populism has been on the rise across the global political land-
scape. Growing populism has been considered the prime factor “behind the rise
of Trump in US, Orban in Hungary, Brexiteers in UK, Putin in Russia, Kaczyn-

ski in Poland, Babis in the Czech Republic, Erdogan in Turkey, Modi in India or
Duterte in the Philippines” (Bardhan, 2018). It is interesting to see whether the
pandemic makes the populist wave stronger or weaker.
Making sense of populism has always been a notoriously vexed exercise.
However, the articulation of all forms of populist discourses usually involves
the creation of an internal frontier fragmenting the social space, i.e., it pits “the
people” against “the other” (Laclau, 2005: p. 74). The political construction of
“the people” involves construction of equivalence between diverse unmet social
demands, such that it results in an inner homogeneity of differently positioned
subjects and a constitutive “other”, which then becomes the enemy of “the peo-
ple”. This is the crux of the well-known “populist” articulation of opposition
between “the people and the elite” (Laclau, 2005; Panizza, 2005).
Antagonism is central to populist politics, since it is through antagonism that
the political identities are formed (ibid). The contemporary populist leaders em-
phasise cultural, national, religious or racial identities of the “people” against
the “outsiders” or “elites” or “the establishment” who allegedly pose a threat
to the popular will. In the USA, Donald Trump has portrayed the Chinese,
Mexicans and Muslims as well as the American “deep state” as the “other”. In
Europe, right-wing populists have demonised the Muslim immigrants and the
faceless bureaucrats of Brussels as the “other”. Populism cannot be understood in
terms of a definite ideology—rather, its salience is a function of the degree and
power of articulation of equivalence and, hence, the force of the construction of
the internal social frontier between “the people” and “the elite”. However, the
contemporary populist wave can be ideologically classified as largely, right-wing.
It has been argued that the Covid-19 pandemic can expose the Achilles heel
of (particularly right-wing) populist politics, with the possibility that populism
might come out much weaker in the post-pandemic world (Bufacchi, 2020;
Chhibber, 2020). Two arguments are generally advanced in support of that po-
sition. First, the Covid-19 crisis has exposed the “incompetence” of the populist
leaders. In countries where populism was on the rise, such as USA, Britain, India
and Mexico, the pandemic had its greatest tolls on lives and livelihoods. The pan-
demic has highlighted the necessity of effective governance and evidence-based
policy responses, not rhetoric nor imagined threats. The health and economic
crisis unleashed by the pandemic has pointed to the importance of health experts
and skilled policy professionals. The crisis has led to a surge in the popularity
of technocratic or pro-establishment leaders, such as Germany’s Angela Merkel,
Italy’s Giuseppe Conte and New Zealand’s Jacinda Ardern, due to their effective
handling of the situation. Second, as the response to the pandemic rests on a
concerted global response, nationalistic, religious or racial divisions leveraged by
the populist leaders for their own rise may become untenable during and post the
Covid-19 crisis. “No-one will ever dare to resent the immigrant nurses, doctors,
and hospital cleaners heroically working in our health services, risking their lives
on a daily basis”.15
But the belief that populism is in peril due to the mismanagement of the
Covid-19 crisis is premature, and overlooks how the coronavirus crisis has been
a boon for this brand of politics. Moreover, given that the causes for the rise of
populism are structural rather than incidental, the current populist wave is un-
likely to be halted by the pandemic.
What are the structural reasons that had led to the rise of populism in recent
years? The rise of populism reflects cultural, social and economic grievances,
which have increased over the last decades—so many diverse unmet social de-
mands (De Vries, 2018). Several researchers have argued that advanced stages of
globalisation are prone to populist backlash (Rodrik, 2018). Globalisation drove
multiple cleavages in society “between capital and labor, skilled and unskilled
workers, employers and employees, globally mobile professionals and local pro-
ducers, industries/regions with comparative advantage and those without, cities
and the countryside, cosmopolitans versus communitarians, elites and ordinary
people” (Rodrik, 2018: p. 23). The economic anxiety and distributional strug-
gles exacerbated by globalisation generate a base for populism—where the iden-
tity of “the people” is often forged along ethno-national/cultural lines.
Besides globalisation, automation and new digital technologies have played
a critical role in de-industrialisation and intensification of income inequalities
and political cleavages, thereby aiding the rise of populism. As discussed ear-
lier, automation and technological advancement have led to job polarisation and
glaring labour-market inequalities. Furthermore, the Web 2.0 technologies and
the spread of social media have contributed to the transformation of the political
sphere. Social media sites have emerged as sites of divisive political arguments
and facilitated the rise of a new form of “hyper-representation”. Scholars argue
that social media increasingly creates digital spaces that are more aligned with an
individual’s worldview, which they refer to as online filter bubbles (Marks et al.,
2019). This exacerbates political polarisation.
The populist response to the pandemic has been a swift and predictable rhet-
oric of blaming “the other”. In the USA, Trump gave a racial makeover to
Covid-19 by labelling it as the “Chinese virus”. In India, several media houses
and ruling-party officials rushed to blame Muslims for the spread of coronavirus
cases in the country. There were several incidents of anti-Asian racism almost
throughout the world. Several populist political leaders also diverted the blame
towards the WHO. Furthermore, the populists have used the pandemic as an
opportunity to concentrate more power in their hands. Naomi Klein (2007)
has warned of the “shock doctrine”, according to which people’s disorientation
during disasters, such as wars, coups, terrorist attacks, market crashes or natural
disasters, provides opportune occasion for the most powerful actors (such as gov-
ernments or capitalists) to push through radical measures and further entrench
their influence in the society. This can be seen in Hungary, where Viktor Orbán
dismantled the last few remnants of democracy by duly closing down Parliament
and concentrating even more power in his own hands. In Serbia, President Alek-
sandar Vučić has announced an open-ended state of emergency, side-lining the
Parliament and enforcing some of Europe’s strictest measures. In India, the cen-
tral government used the pandemic to push several contentious policy reforms
affecting farmers and workers.
To sum up, in view of our arguments we have presented so far, we argue that
the current populist wave might be strengthened by a more inward-looking pol-
icy regime, withdrawal from globalisation and the unstoppable rise in precarity
of labour and income inequality, requiring a drastic transformation of the way
political parties and leaders seek to re-articulate their politics.
2.7 Conclusion
The purpose of this chapter is not to offer speculative arguments about how
the Covid-19 pandemic will alter the nature of the global political economy in
its wake. On the contrary, the chapter is motivated by the understanding that
the global political economic order has already been changing in an intelligible
way. The interesting question to ask then is how the pandemic is likely to influ-
ence these processes of incipient changes. This is particularly important since we
have seen that the impact of the pandemic on most countries has been dramatic,
calling for unprecedented interventions by most governments not only in man-
agement of the health crisis, but also in addressing the economic fallout of that
management.
Of all the significant changes at the global level, we have identified four visible
trends that are at the heart of recent political debates on the future of economy
and people’s livelihoods in most countries. These are distinct, yet inter-connected
processes that also happen to be central to politico-economic management of the
ongoing pandemic. First, the profound changes in the labour market brought
about by technological changes in the last few decades were predicted to enter
a phase of acceleration even before the pandemic hit us. The pandemic has not
only precipitated a crisis of unemployment and loss of livelihoods on a gigantic
scale, but also catalysed these processes of change by forcing businesses, consum-
ers and workers, in their health-protecting behaviour, to opt for those same tech-
nological solutions that were driving the changes. Consequently, applications of
AI, automation, robotics, etc. are likely to be adopted much more widely and
rapidly, sharpening job polarisation, accentuating precarity and uncertainty of
employment and worsening economic inequality.
Second, as a consequence of these changes, socio-political tensions are likely
to call forth two related policy responses—a retreat from globalisation to a lesser
or greater degree, so that governments have the policy space to promote domestic
employment and design novel welfare systems for those facing increasingly uncer-
tain employment prospects and sudden loss of jobs due the pandemic. Universal
basic income or variants thereof are no longer in the realms of speculation. Some
form of income transfer to large sections of the population was necessitated by the
pandemic itself, as part of the largest relief packages in history in many countries,
in response to the economic and humanitarian crisis precipitated by the pandemic.
The techno-administrative infrastructure necessary for running such welfare sys-

tems had to be developed almost overnight during the emergency created by the
pandemic—creating the basis for an easier transition to the new system.
Third, as political demands for protectionism and domestic job creation gath-
ered momentum over the last decade, GVCs, the drivers of globalisation, had
already begun to stagnate. The shocks to the supply chain caused by the pan-
demic, coupled with the possibilities offered by new technologies (AI, robotics,
“big data”, etc.) in reduction of labour costs, make it politically and economically
attractive to accelerate the transition out of the previous globalised production
systems. Travel restrictions and border shutdowns, caused by the pandemic, have
contributed to the salience of national economic boundaries in popular imagi-
nation. The emerging global politico-economic order is thus likely to resemble
more the world in the decades after the Second World War (albeit with very
different characteristics) than in the last four decades of globalisation.
Finally, all these changes are already being articulated in complex political
discourses with pronounced populist characteristics. Nationalism, self-reliance
and “nation first” sentiments, coupled with distinct trends of “otherisation” of
population groups, form the bases for the emergence of populist leaders across the
world. The pandemic has created rare opportunities for novel political manoeu-
vres and extra-ordinary actions that enable concentration of political power on
one hand and the forging of stronger political connection between the “leader”
and the “people” on the other hand. The pandemic might well turn out to be
more long-lasting in its effects than currently assumed.
Notes
1 Recency bias is a cognitive bias wherein individuals tend to favour recent events over
historic ones in making decisions.
2 Global Economic Prospects. Chapter 1. p. 5. URL https://openknowledge.
worldbank.org/bitstream/handle/10986/33748/211553-Ch01.pdf. Last Accessed on:
12 March 2021.
3 URL: https://www.imf.org/en/Publications/WEO/Issues/2020/09/30/world-
economic-outlook-october-2020. Last Accessed on: 12 March 2021.
4 URL: https://openknowledge.worldbank.org/bitstream/handle/10986/33748/
211553-Ch01.pdf. Last Accessed on: 12 March 2021.
5 URL: https://www.spglobal.com/ratings/en/research-insights/topics/economic-
reports-september-2020. Last Accessed on: 12 March 2021.
6 URL: https://www.wto.org/english/news_e/pres20_e/pr862_e.htm#:~:text=The%20
steepest%20declines%20were%20in,but%20just%207.1%25%20in%20Asia. Last Ac-
cessed on: 12 March 2021.
7 URL: http://pubdocs.worldbank.org/en/341901591464622054/GEP-June-2020-
Chapter1-Fig1-7.xlsx. Last Accessed on: 12 March 2021.
8 URL: https://www.statista.com/statistics/246268/personal-savings-rate-in-the-united-
states-by-month/#:~:text=of %20the%20Treasury,The%20personal%20saving%20
r ate%2 0 i n%2 0 t he%2 0Un ited%2 0 St ate s%2 0 a mou nted%2 0 to,to%2 011%2 0
percent%20in%201960. Last Accessed on: 12 March 2021.
9 URL: https://tradingeconomics.com/united-kingdom/personal-savings. Last Ac-
cessed on: 12 March 2021.
10 URL: https://www.ilo.org/global/topics/coronavirus/impacts-and-responses/WCMS_
755910/lang--en/index.htm. Last Accessed on: 12 March 2021.
11 URL: https://www.ilo.org/global/topics/coronavirus/impacts-and-responses/WCMS_
12 URL: http://bit.do/WDR2016-Fig2_15. Last Accessed on: 12 March 2021.
13 URL: http://pubdocs.worldbank.org/en/210831591464618025/GEP-June-2020-
Chapter1-Fig1-16.xlsx. Last Accessed on: 12 March 2021.
14 URL: https://www.statista.com/statistics/1107572/covid-19-value-g20-stimulus-
packages-share-gdp/. Last Accessed on: 12 March 2021.
15 URL: https://www.21global.ucsb.edu/global-e/april-2020/coronavirus-bad-populism.
Last Accessed on: 12 March 2021.
References
Aguilera, R., Henisz, W., Oxley, J. E., & Shaver, J. M. (2018). International strategy in
an era of global flux. Strategy Science, 4(2), 61–69.
Allen, J., Burns, N., Garrett, L., Haass, R.N., Ikenberry, G.J., Mahbubani, K., Menon,
S., Niblett, R., Nye Jr, J.S., O’neil, S.K., Schake, K., & S. M. Walt (2020). How
the world will look after the coronavirus pandemic. Foreign Policy, March 20, 2020.
https://foreignpolicy.com/2020/03/20/world-order-after-coroanvirus-pandemic/.
Aum, S., Lee, S. Y. T., & Shin, Y. (2021). Inequality of fear and self-quarantine: Is there
a trade-off between GDP and public health? Journal of Public Economics, 194(1), 104354.
Autor, D. H. (2014). Polanyi’s paradox and the shape of employment growth (Report No.
20485). Cambridge, MA: National Bureau of Economic Research.
Autor, D. H., & Dorn, D. (2013). The growth of low skill service jobs and the polarisation
of the US labor market. American Economic Review, 103(5), 1553–1597.
Baker McKenzie and Oxford Economics. 2020. Beyond COVID-19: Supply chain resil-
ience holds key to recovery. Available at: https://www.bakermckenzie.com/-/media/
files/insight/publications/2020/04/covid19-global-economy.pdf. Last Accessed on:
12 March 2021.
Baldwin, R. (2020). The greater trade collapse of 2020: Learnings from the 2008–09
great trade collapse. VoxEU.org. Available at: https://voxeu.org/article/greater-trade-
collapse-2020. Last Accessed on: 12 March 2021.
Bardhan, P. (2018). Populism in less developed countries is somewhat different. Business
Standard. Available at: https://www.business-standard.com/article/opinion/populism-
in-less-developed-countries-is-somewhat-different-118081200657_1.html. Last Ac-
cessed on : 12 March 2021.
Bufacchi, V. (2020). Is coronavirus bad for populism? 21st century global dynamics initi-
ative. Available at: https://www.21global.ucsb.edu/global-e/april-2020/coronavirus-
bad-populism. Last Accessed on: 12 March 2021.
Chhibber, A. (2020). The virus has seriously exposed the limits of populism. But will the
pandemic push it against the wall? The Economic Times, August 11, 2020. Available at:
https://economictimes.indiatimes.com/blogs/et-editorials/the-virus-has-seriously-
exposed-the-limits-of-populism-but-will-the-pandemic-push-it-against-the-wall/.
De Backer, K., & Flaig, D. (2017). The future of global value chains: Business as usual or “a new
normal”? OECD Science, Technology and Industry Policy Papers, No. 41.
De Vries, C. E. (2018). The cosmopolitan-parochial divide: Changing patterns of party
and electoral competition in the Netherlands and beyond. Journal of European Public
Policy, 25(11), 1541–1565.
Figari, F., & Fiorio, C. V. (2020). Welfare resilience in the immediate aftermath of the covid-19
outbreak in Italy. EUROMOD at the Institute for Social and Economic Research,
Tech. Rep.
Hershbein, B., & Kahn, L. B. (2018). Do recessions accelerate routine-biased techno-
logical change? Evidence from vacancy postings. American Economic Review, 108(7),
1737–1772.
International Labour Organization. (2020). ILO Monitor: COVID-19 and the world of work
(6th ed.). International Labour Organization, Geneva (2020), Sep 2020. Available
at: https://www.ilo.org/global/topics/coronavirus/impacts-and-responses/WCMS_
Jaimovich, N., & Siu, H. E. (2020). Job polarisation and jobless recoveries. Review of Eco-
nomics and Statistics, 102(1), 129–147.
Javorcik, B. (2020). Global supply chains will not be the same in the post-COVID-19
world. In Baldwin, R. E., & Evenett, S. J. (Eds.), COVID-19 and trade policy: Why
turning inward won’t work. CEPR Press, pp. 111–116.
Klein, N. (2007). The shock doctrine: The rise of disaster capitalism. Macmillan.
Laclau, E. (2005). On populist reason. Verso.
Maloney, W. F., & Molina, C. (2016). Are automation and trade polarising developing country
labor markets, too? The World Bank.
Mann, C. L. (2020). Real and financial lenses to assess the economic consequences of COVID-19.
A VoxEU.org Book. London: Centre for Economic Policy Research Press.
Marks, J., Copland, E., Loh, E., Sunstein, C. R., & Sharot, T. (2019). Epistemic spillo-
vers: Learning others’ political views reduces the ability to assess and use their exper-
tise in nonpolitical domains. Cognition, 188(7), 74–84.
Micklethwait, J., & Wooldridge, A. (2020). The virus should wake up the west.
Bloomberg, 13 April 2020. Available at: https://www.bloomberg.com/opinion/
articles/2020-04-13/coronaviruspandemic-is-wake-up-call-to-reinvent-the-state.
Miroudot, S., & Nordström, H. S. (2019). Made in the world revisited. Robert Schuman
Centre for Advanced Studies Research, Paper No. RSCAS 2019/84, 33.
OECD. (2017). OECD employment outlook 2017. OECD Publishing. Paris. Available at:
http://dx.doi.org/10.1787/empl_outlook-2017-en. Last Accessed on: 12 March 2021.
Panizza, F. (2005). Populism and the mirror of democracy. Verso, 2005.
Reinhart, C., & Reinhart, V. (2020). The pandemic depression: The global economy
will never be the same. Foreign Affairs, 99(5), 84–95.
Rodrik, D. (2018). Populism and the economics of globalization. Journal of International
Business Policy, 1(1–2), 12–33.
Saad-Filho, A. (2020). From COVID-19 to the end of neoliberalism. Critical Sociology,
46(4–5), 477–485.
Sumner, A., Hoy, C., & Ortiz-Juarez, E. (2020) Estimates of the impact of COVID-19 on
global poverty. WIDER Working Paper 2020/43. Helsinki: UNU-WIDER. Availa-
ble at: https://www.wider.unu.edu/publication/estimates-impact-covid-19-global-
poverty. Last Accessed on: 12 March 2021.
Tzagkarakis, S. I., Pappas, I., & Kritas, D. (2020). Covid-19 has brought back the neces-
sity of the welfare state: The Greek case. HAPSc Policy Briefs Series, 1(1), 67–73.
3
RIDING THE WAVE OR GOING
UNDER? THE COVID-19 PANDEMIC
AND TRUST IN GOVERNMENTS
Catherine Bros
3.1 Introduction
The coronavirus crisis sheds a crude light on the central role played by gov-
ernments in protecting and coordinating members of societies. Measures sig-
nificantly curtailing individual freedom have been taken in order to prevent
hospitals and public health infrastructures from being overwhelmed. Such meas-
ures, almost unseen in times of peace, have required citizens’ assent and a great
deal of trust in their governments. Surprisingly, even in countries where citizens
are known for being distrustful of their political leaders and vocal against them,
no stark opposition to these radical measures has been heard. There has even
been in most countries an increased level of adhesion to public policies, which
proved opportune as a show of trust is of paramount importance for the swift and
rapid implementation of decisive political action when faced with a crisis.
Mueller (1970, 1973) has evidenced what came to be known as a “rally-
around-the-flag” effect that describes a sudden surge in approval ratings of a
country’s head of state when faced with specific, dramatic and sharply focused
crisis that directly involves the government. These rallies have been documented
in the case of the September, 11 attacks (Chanley, 2002; Hetherington & Nel-
son, 2003) or following other terror attacks (Dinesen & Jaeger, 2013). Yet their
triggers remain unclear and need to be identified in order to gauge the longevity
of the rally and hence its potential long-term political consequences. On the
one hand, according to the “patriotism school of thought” the rally is an emo-
tional response to a perceived threat as citizens seek comfort into a feeling of
unity embodied by the head of state, either the president or the chief minister.
On the other hand, the “Opinion leadership school” holds that in times of cri-
sis, members of the opposition either refrain from commenting on governmen-
tal action or “make cautiously supportive statement” (Brody & Shapiro, 1991)
DOI: 10.4324/9781003220145-4
Riding the wave or going under? 51
leaving journalists with little to report and citizens only exposed to supportive
comments. Hetherington and Nelson (2003) argue that although both views are
useful to explain rallies, yet the patriotism school provides a better explanation
of why these rallies occur, while the opinion leadership school tends to better
explain their duration.
There is little doubt that many leaders around the world have seen improve-
ment in their approval ratings as the virus started its worldwide expansion. For
instance, India’s Prime Minister Narendra Modi saw his approval rating increase
by 12 points, from 71% in early March to 83% by the end of April. Canada’s Prime
Minister Justin Trudeau also witnessed a 22-point increase over the same period.
Even the popularity of the out-of-favour French President Macron had increased
by eight points to reach its highest point of these last two years. Figure 3.1 dis-
plays the approval rates of ten government leaders in the world between February
1 and May 31, 2020.
Scholars hold differing views on what caused this rapid increase in popularity.
Broadly speaking, some studies hold that the gains in popularity do represent a
“rally-around-the-flag” effect directly linked to the crisis,1 while other authors
argue that the crisis created a critical juncture that has changed for good the
relationship between citizens and their governments by increasing the demand
90%
80%
70%
60%
50%
40%
30%
20%
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
4/
4/
2/
3/
2/
4/
/
2/
2/
4/
3/
3/
3/
02
/0
/0
/0
/0
/0
/0
/0
/0
/0
/0
/0
/0
/
04
18
29
28
08
25
01
22
15
11
14
21
07
Modi (India) Lopez Obrador (Mexico) Bolsonaro (Brazil) Johnson (UK)

Trudeau (Canada) Trump (USA) Merkel (Germany) Abe (Japan)
Morrison (Australia) Macron (France)
FIGURE 3.1 pproval rates of ten governments’ heads between February 1 and May
A
31, 2020.
Source: These data come from Morning Consult Political Intelligence and are based on an av-
erage of over 3,000 weekly interviews per non-US leader, and over 49,000 weekly interviews
for Trump. Seven-day moving average.
52 Catherine Bros
for expertise and technocratic government at the expense of populism (Daniele

et al., 2020). Thus, increased popularity is the reward for governments’ swift
implementation of necessary measures.
Baekgaard et al. (2020) compare answers given to a questionnaire on trust in
the government and public bodies in Denmark before and after lockdown has
been ordered. Their results show that trust in the government increased directly
after the beginning of the lockdown period and that this enhanced trust in the
government spread to other public institutions. They interpret these results as
a rally-around-the-flag effect. However, Merkley et al. (2020) highlighted that
the spread of the virus had generated in Canada cross-partisanship consensus,
thereby muting opposition. These studies argue in favour of a rally-around-the-
flag effect – be it based on citizens’ emotions or on political debates being put
on hold.
Bol et al. (2021) provide a different perspective as they argue, based on a
web-based survey fielded before and after the lockdown, that the increased trust
in government is not due to the epidemic itself but rather due to the measures
taken and that they come more as a reward for their expertise and handling of the
situation. Daniele et al. (2020) find, based on a survey in four Western European
countries, that when individuals are being reminded of the epidemic they tend
to express higher trust in experts and scientists, provide less support to populist
views while maintaining the trust in government. The perceived threat posed by
the pandemic would therefore favour the demand for expertise.
There are therefore two main views: one that clearly associates this increased
approval with a “rally-around-the-flag” effect – be it motivated by an emo-
tional response or the mutism of the opposition, and another that argues that
the pandemic plays the role of a critical juncture that transforms socio-political
attitudes favouring technocratic forms of government at the expense of pop-
ulism. However, most of these studies exhibit three shortcomings: first they are
based on surveys at the individual level and the questions asked largely vary in
their wordings, probably owing to the fact that “trust in government” is a multi-
faceted concept, leaving various aspects subject to a single interpretation (Devine
et al., 2021). Second, they are focused on a handful of closely related countries
(e.g. Western Europe) where the spread of the epidemic and stay-at-home orders
were concurrent, making the task of disentangling the effect of the epidemic and
political action challenging. It is therefore difficult to isolate whether the surges
in popularity that occurred in March were indeed a “rally-around-the-flag” ef-
fect that is destined to be short lived or a profound change in political attitudes
and demand. Third, most studies have relied on short series of data, while longer
data series allow us to consider whether a structural change occurred, or whether
the crisis had triggered a “rally-around-the-flag” effect that should not last long
if it is only rooted in an emotional response or perhaps longer if it is due to the
mutism of the opposition.
In this chapter, we will make use of longer time series by looking approval
rates of ten heads of states between January 1, 2020 and August 18, 20202 and
question whether these jumps in popularity were linked to the spread and the
backflow of the epidemic, which presumably would indicate a concern by citi-
zens about the governments’ performance in managing the epidemic, or whether
they are correlated with measures that had been taken, after having taken into
account the evolution of the epidemic. A special attention will be given to coun-
tries where both the propagation and political action had not have taken place at
the same time. Finally, the long-lasting effect of this increased approval will be
questioned.
3.2 The evolution of the epidemic and the timing of the

implementation of stringent social distancing measures
Figure 3.1 shows that increase in approval ratings occurred around the month of
March and benefited almost all the considered leaders. Table 3.1 shows the gains
in approval ratings between January 1, 2020 and April 30, 2020.
It is important to note first that some heads of state such as Messrs Abe, Bol-
sonaro and Lopez-Obrador have not seen their ratings increase over the period.
Second, jumps in approval ratings all occurred at the same time, around the
month of March throughout the world, including in countries where the ep-
idemic was not in full swing, such as India for instance, yet where lockdown
measures had been taken. During a short period of time, many governments,
presumably facing different epidemic situations, have nonetheless issued stay-
at-home orders such that of France (March 16), India (March 24), Great Britain
(March 23) and Mexico (March 30) or passed emergency bills aiming at enforc-
ing social distancing measures and temporarily restricting personal freedoms.
The case of India is striking as very severe measures have been enacted as early as
March when the number of cases was very few given the size of the population,
yet the Prime Minister did benefit from an increased popularity. Figure 3.2 doc-
uments the timing of the spread of the virus, the severity of the measures taken
and approval ratings for each country. Graphs in column (1) plot on one axis
TABLE 3.1 Change in approval ratings

between January 1 and
April 30, 2020
Abe ( Japan) −7%

Bolsonaro (Brazil) −5%
Johnson (UK) +14%
Lopez-Obrador (Mexico) −2%
Macron (France) +6%
Modi (India) +6%
Morrison (Australia) +27%
Trump (USA) +3%
Trudeau (Canada) +17%
54 Catherine Bros
approval rates as displayed in Figure 3.1 and on the other axis a stringency index
measuring the severity3 of the measures taken between January 1, 2020 and Au-
gust 18, 2020, while graphs in column (2) of Figure 3.2 plot approval ratings and
number of cases over the same period.
Two facts stand out from these figures. First, severe social distancing measures
had been taken in all the sampled countries in only a week (between March 17
and March 24) except Japan.4 Second, selected countries have not been hit by the
epidemics at the same time. Generally speaking, these countries could be clubbed
into two groups: those where the epidemic started spreading as soon as mid-
March (day number 75) such as Australia, Canada, Japan and Western European
Column (1) Column (2)

approval ratings and stringency measures approval ratings and number of Covid cases
a Australia
80
.7
.7
20
new cases smoothed per million inhab.
60
.6
.6
15
Stringency index
Approval Rate
Approval Rate
40
.5
.5
10
20
.4
.4
5
.3
.3
0
0 50 100 150 200 250 0 50 100 150 200 250

Day Number Day Number
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
b Brazil
.6
.6
80
200
.55
.55
60
150
Stringency index
Approval Rate
Approval Rate
.5
.5
40
100
.45
.45
20
50
.4
.4
0
0 50 100 150 200 250 0 50 100 150 200 250

FIGURE 3.2 hange in approval ratings between January 1 and August 18, 2020 ver-
C
sus stringency measures and number of new Covid cases.
Source: Approval ratings are from Morning Consult Political Intelligence, stringency index
from the Oxford Covid-19 Government Response Tracker (University of Oxford) and num-
ber of new cases from the European Centre for Disease Prevention and Control.
c Canada
80
50
.7
.7
40
60
Approval Rate
.6
.6
Stringency index
Approval Rate
Approval Rate
30
40
.5
.5
20
20
.4
.4
10
.3
.3
0
0
0 50 100 150 200 250 0 50 100 150 200 250
d France
80
.4
.4
80
.35
.35
60
60
Stringency index
Approval Rate
Approval Rate
40
.3
.3
40
.25
.25
20
20
.2
.2
0
0 50 100 150 200 250 0 50 100 150 200 250

e Germany
80
80
.6
.6
.55
.55
60
60
Stringency index
Approval Rate
Approval Rate
.5
.5
40
40
.45
.45
20
20
.4
.4
0
0 50 100 150 200 250 0 50 100 150 200 250

FIGURE 3.2 (continued)
Countries on one side and on the other side those where the expansion of the
disease started in May (day number 120–150) such as Brazil, India and Mexico.
In India the government imposed the most severe restrictions by mid-March,
as the index reached the highest level of sample, nearly two months before the
virus significantly started to circulate (Mid-May or day 150). Yet, by mid-May
restrictions that had already been in place for two months were not sustainable
and loosened. A similar comment could be applied to either Brazil, where the
number of cases rapidly increased two months after the implementation of strict
56 Catherine Bros
f India
100
.85
.85
50
80
40
Stringency index
.8
.8
Approval Rate
Approval Rate
60
30
40
20
.75
.75
20
10
.7
.7
0
0
0 50 100 150 200 250 0 50 100 150 200 250
g Japan
50
15
.35
.35
40
10
Stringency index
Approval Rate
Approval Rate
.3
.3
20 30
.25
.25
10
.2
.2
0
0 50 100 150 200 250 0 50 100 150 200 250

h Mexico
60
.7
.7
80

60
.65
.65
40
Stringency index
Approval Rate
Approval Rate
40
.6
.6
20
20
.55
.55
0
0 50 100 150 200 250 0 50 100 150 200 250

measures, or Mexico. However, as far as Mexico and India are concerned, their
respective heads of state gained in popularity at the time they implemented the
restrictions, despite the fact that these measures appear ill-timed, as the epidemic
had not started.
Overall, increases in approval rate appear correlated in most countries with
the surge in the number of Covid cases, undermining the hypothesis that citizens
reward their governments for en efficient management of the crisis. However,
leaders’ gains in popularity appear directly tied to the severity of the measures
i USA
.46
.46
80
200
.44
.44
150
60
Stringency index
Approval Rate
Approval Rate
100
.42
.42
4020
50
.4
.4
.38
.38
0
0
0 50 100 150 200 250 0 50 100 150 200 250
j United Kingdom
80
80
.65
.65
.6
.6
60
60
Stringency index
Approval Rate
Approval Rate
.55
.55
40
40
.5
.5
20
20
.45
.45
.4
.4
0
0 50 100 150 200 250 0 50 100 150 200 250

taken, including measures taken when the spread of the virus had not started,
as in the Indian case for instance. Graphs in Figure 3.2 therefore suggest that a
perceived menace and a need for reassurance are the triggers of these surges in
popularity. In that context, one may wonder whether these reactions have not
been reinforced by an overall need for strong leaders and a renewed demand for
populism. From the graphs set out in Figure 3.2, no specific pattern seems to
emerge between leaders of governments who have adopted a more technocratic
approach and those with a populist stance.
3.3
A “rally-around-the-flag” effect or a reward for crisis
management?
This preliminary analysis suggests that jumps in approval ratings and popular-
ity may have more to do with a sense of threat and comfort-seeking behaviour
that is typical of a “rally-around-the-flag” than with an efficient management of
the epidemics. Should this hypothesis be true, we would observe a correlation
between stringency measure and approval ratings, even after having taken into
account the spread of the disease as proxied by the number of cases. Moreover,
if heads of state were rewarded for an efficient management of the crisis then
58 Catherine Bros
we should observe a negative correlation between new cases of Covid-19 and

approval ratings.
In order to establish these correlations, the following equation is estimated:
Approvalt = b0 + b1New Casest + b2Stringency Indext + b3 Approvalt −1 + b4t + εt (3.1)
where Approvalt , Casest and Stringency Indext are the seven-day moving average for
respectively approval rate, as measured by Morning Consult Political Intelligence
between January 1 and August 18, 2020, the number of new cases as measured by
the European Center for Disease Prevention and Control over the same period
and an index for the stringency of the restriction measures taken as computed by
Hale et al. (2020). Variable t captures the effect of time while pre-existing trends
in popularity are captured by the variable Approvalt−1 which is the approval rate
at the previous period t − 1. For instance, a general demand for strong leaders
would be captured by this trend variable so that the effect identified only per-
tains to changes over time. Tables 3.2(a) and 3.2(b) present the results from the
estimations.
Results from Tables 3.2(a) and (b) show that in all the selected countries but
Japan, Mexico and Brazil, popularity is significantly tied to the actions taken
by the government rather than to the evolution of the pandemic and positively
related to the severity of the measure. From this preliminary evidence, it seems
difficult to argue that governments are being rewarded for the management of
the epidemic, as its evolution, as proxied by the number of new cases, does
not seem to be significantly correlated with government’s popularity. Rather,
it seems citizens are grateful to their governments for their attempts to act to
protect their population, whether successful or not. Results suggest more a
TABLE 3.2(A) Correlates of the approval ratings of the heads of state
Australia Brazil Canada France Germany
New cases −0.00947 0.00103 −0.00392 0.00780** −0.000849

(0.0152) (0.00273) (0.00677) (0.00361) (0.00518)
Stringency 0.0253*** −0.0104* 0.0310*** 0.00743*** 0.0141**
Index
(0.00535) (0.00562) (0.00410) (0.00261) (0.00635)
Previous 95.74*** 94.66*** 92.32*** 89.11*** 93.90***
approval
rating (1.277) (1.727) (1.524) (2.660) (2.106)
Day number −0.000775 −0.000351 −0.00632*** 0.00145* 0.00108
(0.00224) (0.00463) (0.00133) (0.000848) (0.00121)
Observations 230 230 230 230 230
R-squared 0.997 0.977 0.992 0.969 0.993
Robust standard errors in parentheses. *** denotes a 1% level of significance, ** denotes a 5% level
of significance and * denotes a 10% level of significance. Estimations also include approval ratings
in the previous period and time.
TABLE 3.2(B) Correlates of the approval ratings of the heads of state
India Japan Mexico USA UK
New cases 0.0113 −0.00630 0.00935 0.000446 −0.00362

(0.00767) (0.0229) (0.0114) (0.000983) (0.0107)
Stringency 0.0158*** 0.00169 0.00185 0.00511** 0.0211**
Index
(0.00393) (0.00341) (0.00380) (0.00222) (0.00830)
Previous 89.57*** 92.09*** 88.58*** 92.51*** 94.91***
approval
rating (2.467) (2.559) (2.813) (2.512) (3.332)
Day number −0.00659*** −0.00271** −0.00660 −0.00356*** −0.00981***
(0.00230) (0.00111) (0.00471) (0.00124) (0.00325)
Observations 230 230 230 230 230
R-squared 0.949 0.948 0.888 0.959 0.988
Robust standard errors in parentheses. *** denotes a 1% level of significance, ** denotes a 5% level of
significance and * denotes a 10% level of significance.
“rally-around-the-flag” phenomenon than a reward for properly managing the

situation. Moreover, no pattern seems to emerge between technocratic and more
populists forms of government.
If the surge in popularity translates a “rally-around-the-flag” phenomenon,
then one could wonder whether it lasted. Graphs in Figure 3.2 suggest that in
countries that faced a first surge in coronavirus cases in March, the popularity of
the heads of state has not significantly declined. For instance, the approval rates
of S. Morrison, A. Merkel, J. Trudeau and E. Macron between January 1 and
August 18, 2020 gained respectively by 25%, 16%, 9% and 5%. Only B. Johnson’s
popularity was down from January 1 by 6% by mid-August. Approval ratings in
countries that faced a delayed pandemic behave differently. To further investigate
that question, correlates similar to that of Tables 3.2 and equation 3.1 are com-
puted on two sub-samples: before the month of May, a time when the virus was
not actively circulating in some countries while in full swing in others and when
measures were in place, and between May and mid-August when most countries
loosened their social distancing measures and faced heterogeneous epidemic sit-
uations. Tables 3.3 present the results.
One the one hand, the correlations for the first phase displayed in Table 3.3
largely echo those found in Tables 3.2. Brazil and Mexico set aside, the surges
in popularity are positively related to the stringency of the measures imposed
on the population and are not significantly related to the number of cases of
Covid-19.5 It is important to note that when the equation was estimated over the
whole period, no significant correlation between popularity and measures taken
was found in Japan, while one appears during the early time of the epidemic. On
the other hand, as far as the second phase is concerned, approval ratings appear
uncorrelated with the measures taken and rarely correlated with the number of
60 Catherine Bros
TABLE 3.3 Correlates of the approval ratings of the heads of state between day 50 (February 19) and day 150 (May 29)
Australia Brazil Canada France Germany India Japan Mexico USA UK
New cases −0.0575 0.00265 0.00103 0.00437 0.00349 −0.0399 −0.107* −0.0322 0.00492* 0.00363
(0.0669) (0.00892) (0.0152) (0.00617) (0.00792) (0.209) (0.0555) (0.0515) (0.00254) (0.0116)
Stringency Index 0.0911*** −0.00294 0.0339*** 0.0210*** 0.0192* 0.0236*** 0.0598*** 0.00306 0.0214*** 0.0654***
(0.0326) (0.00949) (0.00578) (0.00775) (0.0114) (0.00727) (0.0165) (0.00779) (0.00526) (0.0177)
Previous approval 79.83*** 92.82*** 88.80*** 84.11*** 87.36*** 77.30*** 88.31*** 81.95*** 59.07*** 80.93***
rating (6.784) (3.945) (4.854) (5.316) (3.343) (4.958) (4.279) (5.632) (10.34) (5.982)
Day number 0.0242** −0.0137 0.000851 −0.000546 0.0149* 0.00916 −0.0122*** 0.00155 −0.0171*** −0.0400***
(0.0113) (0.0165) (0.00738) (0.00528) (0.00803) (0.0133) (0.00344) (0.0184) (0.00479) (0.00936)
Observations 99 99 99 99 99 99 99 99 99 99
R-squared 0.997 0.979 0.994 0.972 0.992 0.978 0.899 0.721 0.910 0.986
Robust standard errors in parentheses. *** denotes a 1% level of significance, ** denotes a 5% level of significance and * denotes a 10% level of significance.
TABLE 3.4 Correlates of the approval ratings of the heads of state between day 150 (May 29) and day 230 (August 18)
Australia Brazil Canada France Germany India Japan Mexico USA UK
New cases −0.0365 −0.000476 −0.00178 0.00470 −0.0932** −0.00890 −0.0680 5.61e-06 −0.00106 −0.0451**
(0.0241) (0.00744) (0.0190) (0.0207) (0.0413) (0.0365) (0.0730) (0.0300) (0.00199) (0.0211)
Stringency Index 0.00654 0.0561* 0.0645 0.0149 −0.00367 0.0204 −0.00438 −0.0374 0.0112 0.0547
(0.0112) (0.0307) (0.137) (0.0163) (0.0230) (0.0941) (0.0368) (0.0650) (0.0670) (0.0343)
Previous 72.31*** 67.87*** 86.53*** 78.44*** 65.61*** 78.54*** 85.08*** 75.43*** 83.53*** 76.31***
Approval
Rating (7.387) (7.454) (6.478) (6.997) (11.07) (6.593) (6.076) (7.432) (6.063) (5.779)
Day number 0.00171 0.0388** −0.0112 0.00975 −6.82e-05 0.00764 0.00380 0.0124 0.00290 −0.0133*
(0.00867) (0.0168) (0.0146) (0.0113) (0.00412) (0.0185) (0.00929) (0.00922) (0.00440) (0.00684)
Observations 81 81 81 81 81 81 81 81 81 81
R-squared 0.653 0.914 0.961 0.728 0.717 0.665 0.879 0.733 0.783 0.878
Robust standard errors in parentheses. *** denotes a 1% level of significance, ** denotes a 5% level of significance and * denotes a 10% level of significance.
62 Catherine Bros
cases. As the epidemic became sort of “old news”, it did not exert too much in-
fluence on approval of heads of state. This can be explained by the arguments put
forward by the “Opinion Leadership School” that as the crisis loses its novelty,
other topics catch voters’ attention and the “rally-around-the-flag” effect of the
crisis slowly fades away. This is even more striking as the second phase identified
here corresponds for many countries such as India, Mexico and Brazil to the full
spread of the disease, and for many like Japan, France and Australia to a second
wave of the epidemic. Thus, if approval ratings were really an assessment of the
management of the crisis, a significant correlation should be found between the
number of cases and the popularity of the heads of state. Such a correlation is
only found in the UK and Germany, which might prefigure a third phase during
which the epidemic becomes less of an extraordinary phenomenon and govern-
ments will start being held accountable by the political opposition as well as the
citizens for its management.
3.4 Conclusion: Covid-19 as a critical juncture?

Some scholars have argued that the Covid-19 crisis created a critical juncture
that altered political expectations in favour of political expertise or that it could
ultimately provide the state with additional legitimacy (Khemani, 2020). Thus,
increased popularity was seen as rewarding appropriate crisis management. How-
ever, many studies have identified a “rally-around-the-flag” effect as the trigger
for surging approval ratings. This chapter provides some evidence that indeed
these increases in the short run have more to do with an emotional response to
a potential threat than with the evolution of the disease itself and hence with its
management. However, data show that the correlation between measures taken
and popularity is limited to the first five months of 2020. However, in both the
UK and Germany approval ratings started being negatively correlated with the
number of cases. It can be hypothesized that this year 2020, which was special
in many respects, can be subdivided into three phases: one where citizens felt
aghast and looked for protection from their heads of state, one where citizens
became accustomed to the threat and other issues, such as economic ones or
internal security came back to the forefront, and perhaps a third one where
citizens will start holding their governments accountable for the evolution of
the epidemic just as they do with other more common issues. Such a hypothesis
needs to be confirmed with longer data series. However, it would reconcile all
interpretations put forward so far by scholars: an exceptional event may trigger
an emotional response and a surge in approval rate. As the issue loses its novelty,
it becomes one of the items of the political debate and citizens eventually hold
their government accountable and reward proper management and expertise.
Notes
1 Leininger and Schaub (2020) show that the Covid-19 epidemic was partially respon-
sible for the success of the incumbent party at the March 15 municipal elections in
Bavaria.
2 These data were gathered by the author from the website of Morning Consult Polit-
ical Intelligence based on 3,000 interviews per week per non-US leader and 49,000
individuals for the USA.
3 Hale et al. (2020) have built a “Stringency Index” that assesses the severity of the
measures taken by governments to enforce social distancing, such as school or work-
place closing, restrictions on movements, etc. For instance, the value of the index
reached 20 in Australia, when the governments imposed quarantine if one is entering
the country from highly contaminated countries. The index reached 87 when France
passed stay-at-home orders and restricted individual movements.
4 The Stringency Index started exceeding 60 in all selected countries during this
seven-day period.
5 Popularity seems related to the number of cases during the first phase in Japan and the
USA although these are only significant at a 10% level of significance.
References
Baekgaard, M., Christensen, J., Madsen, J. K., & Mikkelsen, K. S. (2020). Rallying
around the flag in times of Covid-19: Societal lockdown and trust in democratic in-
stitutions. Journal of Behavioral Public Administration, 3(2).
Bol, D., Giani, M., Blais, A., & Loewen, P. J. (2021). The effect of Covid‐19 lockdowns
on political support: Some good news for democracy? European Journal of Political Re-
search 60(2), 497–505.
Brody, R. A., & Shapiro, C. R. (1991). The rally phenomenon in public opinion. In
R. A. Brody (Ed.). Assessing the president: The media, elite opinion, and public support
(pp. 45–78). Stanford, CA: Stanford University Press.
Chanley, V. (2002). Trust in government in the aftermath of 9/11: Determinants and
consequences. Political Psychology, 23(3), 469–483.
Daniele, G., Martinangeli, A. F., Passarelli, F., Sas, W., & Windsteiger, L. (2020). Wind
of change? Experimental survey evidence on the Covid-19 shock and socio-political
attitudes in Europe. Working Paper of the Max Planck Institute for Tax Law and Public
Finance, No. 2020-10.
Devine, D., Gaskell, J., Jennings, W., & Stoker, G. (2021). Trust and the coronavirus pan-
demic: What are the consequences of and for trust? An early review of the literature.
Political Studies Review 19(2), 274–285.
Dinesen, P. T., & Jæger, M. M. (2013). The effect of terror on institutional trust: New
evidence from the 3/11 Madrid terrorist attack. Political Psychology, 34(6), 917–926.
Hale, T., Petherick, A., Phillips, T., & Webster, S. (2020). Variation in government re-
sponses to Covid-19. Blavatnik School of Government Working Paper, 31, 2020-11.
Hetherington, M. J., & Nelson, M. (2003). Anatomy of a rally effect: George W. Bush
and the war on terrorism. PS: Political Science and Politics, 36(1), 37–42.
Khemani, S. (2020). An opportunity to build legitimacy and trust in public institutions
in the time of Covid-19. Research and Policy Briefs n°32, World Bank.
Leininger, A., & Schaub, M. (2020). Voting at the dawn of a global pandemic. Working
Paper SocArXiv Papers doi:10.31235/osf.io/a32r7.
Merkley, E., Bridgman, A., Loewen, P. J., Owen, T., Ruths, D., & Zhilin, O. (2020).
A rare moment of cross-partisan consensus: Elite and public response to the Covid-19
pandemic in Canada. Canadian Journal of Political Science/Revue canadienne de science
politique, 53(2), 311–318.
Mueller, J. E. (1970). Presidential popularity from Truman to Johnson. The American
Political Science Review, 64(1), 18–34.
Mueller, J. E. (1973). War, presidents, and public opinion. New York: Wiley.
4
INDIA’S LOCKDOWN*
Debraj Ray, S. Subramanian and Lore Vandewalle
4.1 Introduction
On March 24, 2020, the Government of India ordered a nationwide lockdown
for 21 days as a preventive measure against the spread of the coronavirus. The
lockdown – in full force as we write – restricts 1.3 billion people from leaving
their homes. Transport services are suspended, educational institutions are closed
and factories are shut down. This is in line with the measures imposed in most
European countries and in the United States, but the sheer scale of the measure –
as in the case of most policies in India – is intimidating. Add to this the grim
truth of Indian occupational structure and poverty, and you would likely predict
what we now see: unending streams of migrants trying to find their way home,
the fear of loss of all income, deep privations and even (in the space of days) hun-
ger, starvation and death.
The Indian experience highlights, in a visceral way, both the Scylla of wide-
spread viral infection and the Charybdis of socio-economic lockdown. It is not
a choice between lives on the one hand and loss of economic production on the
other. Because India is so poor and because her occupational structure so una-
menable to being shifted online, it is a question of lives versus lives: the nightmar-
ish culmination of all those philosophical trolley problems that we so wish were
innocently confined to the classroom.
* This is an early contribution to the analysis of India’s policy response to the Covid-19 pan-
demic, which appeared as CEPR Policy Insight 102, April 2020 (https://cepr.org/sites/default/
files/policy_insights/PolicyInsight102.pdf ). Since then, a large literature has appeared on policy
questions during the pandemic, to which we have also made more recent contributions. How-
ever, we choose to let this chapter appear as it was originally written, and continue to stand by
the main theme it outlines. Reprinted with permission from CEPR.
DOI: 10.4324/9781003220145-5
India’s lockdown 65
We want to be explicit about this ethical approach: lives versus lives. One
view – implicit, for instance, in arguments made in certain quarters in the
United States – is that there must be some allowable tradeoff between economic
well-being and human lives. We do not subscribe to that view, and entertain no
such tradeoff. Therefore, our first choice in the fight against Covid-19 would be
for governments to implement a comprehensive general lockdown, provided that
this is accompanied by comprehensive State support for compensating welfare
measures aimed at protecting the health, nutrition and psychological well-being
of all households. Many have called for such measures. But calling for them is one
thing, and implementing them another. What if the State is unable or unwilling
to provide that support to all families under lockdown? Then households are
exposed to the profound morbidity and mortality risks that stem from the loss of
incomes and jobs, not to mention their constricted freedom of mobility, height-
ened psychological stress, enhanced prospects of domestic violence and, indeed,
perhaps greater vulnerability to the virus itself. What then?
4.2 Vulnerability to lockdown

The vulnerability of the population – especially the poor – to a generalized
lockdown is accentuated in India by three specific structural features of the
population. The first has to do with occupational configurations. The latest re-
port based on the Periodic Labour Force Survey is instructive in this regard.1 Over
2017–18, about 52% of rural households received their main income from self-
employment. The share of rural households with a major source of income from
casual labor was 25%. Regular wage/salary earners accounted for 12.7% of rural
households. For observers in developed countries, the corresponding figures for
urban India’s households are less surreal − 32.4%, 11.8% and 41.4% respectively –
though it is worth bearing in mind that close to two-thirds of India’s population
is rural. Many of these self-employed rural households are marginal cultivators
or small artisans. In the urban areas, many are engaged in running local shops,
small businesses or minor intermediation activities. It would therefore be very
conservative to treat only the casual labor category of households as especially
vulnerable. Even if we did, then a simple combination of the numbers above
would reveal that casual households alone account for over 20% of all Indian
households. We are thus speaking of at least a fifth of the household population
that is particularly fragile in its ability to cope with an indefinitely imposed
economic lockdown, and most likely many more than that. This heightened
vulnerability is, of course, common in all poor countries, but it is particularly
worrisome in a society of 1.3 billion people.
A corollary of this sort of occupational structure is that it is highly “infor-
mal”; that is, it belongs to that penumbra of the economy in which transactions
are largely outside the fiscal reach of the government, in terms of both taxes (a
concern, perhaps, in normal times) and transfers (the overwhelming concern
now). Well over half of India’s GDP is produced in this informal sector. In
66 Debraj Ray et al.
contrast, countries in Western Europe have 10–20% of their national output

produced in the informal sector.2 The Indian informal sector isn’t a sector that
can transit to online existence with ease – it can’t start holding business dealings
on Zoom or provide internet-based services to replace their ongoing activity. It
is possible that some fraction of them have long-term relationships with other
better-placed Indian households, and can rely on them for assistance or a con-
tinuation of their economic relationships. We don’t have an estimate of how
widespread these ties are.
Another correlate of vulnerability is low savings. From a study by one of us in
rural Chhattisgarh,3 we observe that the median household spends INR 345 per
week on food items alone (this does not include other essentials such as soap or
kerosene). That median household has liquid savings of INR 881 only, though.
If we divide these savings by the weekly expenditure on food, we see that half
the households do not have enough liquid savings to get through a lockdown of
21 days. If we add the value of grain stocks, livestock and jewelry, the median
climbs to INR 1818, but it is still the case that 38% of households do not have the
wherewithal to get through a 21-day lockdown. Moreover, such forms of savings
are very hard to liquidate under a correlated shock, because the price of the assets
plunges at the same time that they need to be sold.
4.3 Reaching the vulnerable

How might we leverage the existing arms of the State to reach these households?
Even for the best-motivated Indian government, this is a tricky proposition. Cash
transfers come first and foremost to mind, but transfers to whom? We would
need to credibly separate the vulnerable from the protected, the low-income
earners from the high. We don’t have the income data to implement this separa-
tion. (For instance, no more than 3% of Indian citizens pay income taxes.)
In response, one might throw one’s hands up and ask for universal transfers.
This is related to the UBI debates that have recently occurred the world over,
including India. The Rangarajan committee’s 2014 thresholds ask for INR 972
(rural) and INR 1,407 (urban) per person per month, a number that appears to
be pitiably outdated. Nevertheless, with these low estimates, a universal transfer
would call for a budget a bit north of INR 1.43 trillion. Annual GDP for India is
around INR 200 trillion, so that comes to over 0.7% of annual GDP per month,
a not infeasible but still hefty price tag with bad leverage properties – everybody
would need to be paid to avert a crisis for those truly in need. To place the num-
ber in context, the Indian government has budgeted 1% of annual GDP under
all expenditure headings for the coronavirus crisis – more on this below. Finally,
note that the lockdown will come with significantly lower output – the United
States has been staring down the barrel of a 25% drop in GDP quarter over
quarter. While a universal transfer would still have a progressive component,
the associated real transfer would be significantly diminished by the prospects of
widespread hoarding, speculation and inflation.
Better, then, to exploit channels that are already self-selected by the poorer
public. Options such as India’s massive rural employment guarantee scheme
(MNREGA) and the public distribution system operating through India’s ration
shops come to mind. As Reetika Khera has argued, MNREGA could expand
its work guarantees well beyond 100 days per annum, and she has called for 20
days per month during the crisis. India has a buffer stock surplus: now is the time
to run that surplus down by increasing rations through the public distribution
system.4 These are eminently sensible suggestions, and they take advantage of the
self-targeting that’s built into these networks.
As MNREGA is implemented in rural areas only, it has to be suitably com-
plemented in cities in order not to prolong the human tragedy of the first week
of the lockdown. Thousands of poor migrant workers have been frustrated by
the shut-down of transport facilities and closed state borders in their effort to
return to their village homes. Many have been forced to undertake foot journeys
of several hundreds of kilometers, with limited cash to hand, and little access to
food, water or shelter. Trekking laborers have been rounded up and locked up
in crowded enclosures. In one shocking incident, a contingent of migrants was
showered with corrosive disinfectant. Significant reports of deaths are beginning
to come in, including the deaths of children. There is little doubt that these will
multiply as the coming difficult days go by.
The relief measures proposed by the Government of India to help its citizens
over the initial three-week period of the lockdown have been both delayed and
relatively scanty. In a recently announced “relief package” the government has
allocated just a little under 1% of GDP for corona-related assistance. In addition,
some components of the Indian relief package refer to provisions already made in
the Union Budget, while others are scarce in relation to their intended purpose.
Consider, for instance, the proposed allowance of INR 500/month/family: the
average monthly per person consumption expenditure is itself about two-and-a-
half times this amount; and, indeed, even the average per person expenditure of
the poorest 20% of the population exceeds the allowance.5 The specific purposes
for which assistance has been earmarked do not necessarily tally – in terms of
either quantum or priority – with painstaking lists which various commentators
with specialized knowledge in the field have drawn up.6
4.4 Confronting the tradeoff

It is the job of epidemiologists and specialists in infectious disease to warn about
the dangers of a virus. The novel coronavirus SARS-CoV-2 belongs to that cat-
egory of viruses which are highly contagious, and often lethal in their effects. It
can (and will) cause many deaths. Because contagion is front and center in these
considerations, social externalities are a matter of first-order importance. It is
little wonder that in countries that have enforced a lockdown, the plea to “self-
isolate” or engage in “social distancing” – household phrases now – is based more
on the implied threat to one’s own health, rather than the implied threat to others
of one’s own actions. These phrases – and the associated priorities of containing
the viral spread over all else – have percolated to India. They inform the thought
processes that underlie India’s own lockdown. They inform the general sense of
threat and fear that guide India’s policymakers. They also inform the terror that
appears to swamp the public when faced with a Covid-19-positive individual, or
even with health-care workers and medical personnel on the front lines of this
crisis. Such individuals have been ill-treated in the most demeaning ways.
In India, these effects are particularly palpable because much of the coun-
try lives side by side in the closest possible quarters. People are being advised
to maintain a distance of six feet from one another in a setting where slum
dwellers and inhabitants of shelters live in cramped and unsanitary conditions.
The context, too, is ominous: the requirement of physical distance as a sanitary
precaution often translates, in Indian society, to the maintenance and accentua-
tion of hierarchies based on notions of ritual purity and caste- and gender-based
disparities. The potential for abuse and maltreatment is enormous. (We hasten to
add that the same potentials exist all over the world, but population densities and
cultural context can amplify them.)
It is with this background in mind that it is worth taking a second look at the
current crisis. We ask you to forget Covid-19 for a second and imagine an incred-
ibly nasty season of the common influenza, one in which (say) half the population
is affected in a given year rather than the customary 10–15%.7 Look at this first
from the “micro” perspective of an average individual who gets the flu in this
particularly virulent year. The conversation between friends and acquaintances
can readily be imagined – “oh, it’s a terrible season this year, everyone’s getting
the flu”. It would be unfortunate, but from our average individual’s point of view
this is scarcely a cause for panic, even though the flu has a known fatality rate of
around 0.1% or perhaps slightly more. Except for the very old, we don’t give it
the time of day.
Now switch mirrors and consider the same scenario in “macro” perspective.
The number of influenza-related deaths in our imaginary scenario would triple.
In the United States, for instance, the 2017–18 flu season was on the high side: ac-
cording to the Center for Disease Control, an estimated 45 million people came
down with influenza, and 61,000 died. Scale this up for our imaginary monster
flu year – then 180,000 people would die. The additional 120,000 would place
enormous stress on the health system – there would be a shortage of hospital beds
and ventilators, or personal protective equipment, and the need for triage would
skyrocket. This “macro” nightmare is to be contrasted with the individual or
“micro” viewpoint which is no more threatening than getting the flu in a given
season.
It is possible to argue, and in fact it has been argued (by one of us, among
others), that Covid-19 may not be substantially different from our imaginary
setting.8 It is surely far more contagious than the flu, but conditional on being
affected, the mortality rate may be not that much higher than that of the flu.
However, widespread understanding of this fact by the general public is not
in the interests of a social planner. In the hypothetical scenario of a contagious

flu, it is just as important (as it is now) to maintain social distancing, to practice
self-isolation and to take all steps to reduce contagion. The strain on overall
society would be just as immense. However, all individual actions would need
to be taken with society in mind, and not with a sense of great personal threat.
That is much harder to implement. As a result, it is far more useful – from a social
perspective – to convey the impression that Covid-19 is a highly lethal disease, so
that self-preservation can be invoked to implement socially desirable outcomes.
These are understandable strategies from the point of view of an econom-
ically advanced country, where the economic fallout from lockdown can (and
should) be shouldered without complaint. After all, lives come first. But in the
“lives versus lives” setting in which India now finds herself, and given the panic-
induced social attitudes that now fester among the public, it is imperative that
there should be a more nuanced understanding of these truths – at least on the
part of policymakers.
With this in mind, consider again the moral dilemma we come up against
when we contemplate alternative approaches to the pandemic – namely, more
or less stringent forms of lockdown, with associated packages of transfers in cash
or kind. Especially in poor and unequal societies such as India’s, both the lock-
down and no-lockdown extremes entail triage. A comprehensive lockdown will
differentially assist those whose fallback options in a time of confinement are
better, and those who can afford the pecuniary and spatial luxuries of social
distancing. Effectively, the lockdown picks out the non-poor for favorable treat-
ment vis-à-vis the poor. In a relatively relaxed environment with freer mobility,
the virus attack would be more evenly distributed across socio-economic groups.
The criterion for differential outcomes under the relaxed alternative is more
likely to be age, with the elderly, implicitly, now becoming the prime candidates
for adverse selection. In the absence of a complete, first-best package of social
protection, neither a comprehensive lockdown nor a relaxed lockdown is free of ethical
complications. An uncompromising insistence on a thoroughgoing lockdown un-
der any circumstances is far from being a morally straightforward stance. These
complications are heightened when we add the micro and macro considerations
described above.
4.5 A possible compromise for India

In such a time of crisis, it is absolutely necessary (a) to concentrate our effort on
policies that are feasible; (b) to avoid measures which are unaffordable for most
citizens; (c) not to criminalize private actions triggered by the need for survival
and (d) to communicate State intent in credible, unambiguous and specific terms.
Against this background, we would like to place the following proposal on the
table for critical evaluation.9 We are not dogmatic about it. Like every other
proposal, this has its limitations; see especially the discussion surrounding item
(iv) below.
i The micro perspective adopted above, and the accompanying discussion

together suggest that the death rate from Covid-19 for people between 20
and 40 is significantly lower than the overall death rate for all ages from influ-
enza. Indeed, as can be argued, even the overall mortality rates (averaging
over age) could be comparable across the two diseases – though Covid-19
is far more contagious. Therefore, it is possible to cautiously advance the
proposition that if it is acceptable for people of all ages to move around
freely in the presence of influenza, it should be acceptable to allow (not
force) all adults under 40 in India to work freely at the present time. Again,
this is not an ideal outcome, but it is a reasonable compromise to work
with, and it can be readily monitored (for instance, an Aadhaar card states
the year of birth).10
ii This measure can – and must – be supplemented by serological testing as
such testing becomes widely available, and as antibody stocks in the popula-
tion build up. This must be seen as being crucial and indispensible. Everyone
certified under an antibody test should be permitted to work, in addition to
the 20–40 population.
i ii Later, as the infection rate subsides, new measures can be taken to move “up
the age distribution” for work permits.
iv Protection of the elderly must be left (though not entirely) to households,
who will possess the incentives and motivation to provide for and monitor
such protections. We need to rely on intra-household motivation, where some
degree of altruism is to be reasonably hoped for, even expected, as opposed
to inter-household altruism, on which it would be foolhardy to repose an equal
burden of faith.
v Transfer massive funds to the public health system, with an explicit emphasis
on the testing, isolating and treatment of older patients. This should be aided
by State provisioning of home-site visits by health workers. It should be un-
derstood that the core crisis is a public health crisis, and not necessarily an
individual one. This appears paradoxical, but in the light of our discussion
in the previous section, it should not be.
What is involved is a form of self-selection: people who can afford the luxury of
staying at home will do so, while those who cannot will opt to go to work. As
with many self-selection outcomes, this one too is unfortunately mediated by the
inherently inequitable prospects confronting our citizens: the laboring poor will
exercise the option of working, unlike those with more secure fallback options.
But at least the former will not be confronted with the involuntary contingencies
of hunger, under-nutrition and life-threatening economic shortfall.
The immense advantage of this proposal is obvious enough: it allows most
Indian families to keep a lifeline open, and in so doing it suggests a measure that
is equitable, balanced and usefully implementable. We highlight here two points
on which a final evaluation must rest.
4.5.1 What guarantees that the elderly will be

adequately protected under this policy?
The answer is that there are no guarantees. Intergenerational contact cannot be
fully avoided under any policy. Widespread poverty and inequality, both distinc-
tive features of India, constitute a poor environment for realizing the anti-disease
objectives of a lockdown. The bulk of the population live in cramped conditions
with full interpersonal exposure, with large families often confined to a single
room.
That said, consider an outright lockdown. No one among the poor can afford
to obey it. Barring draconian orders such as shoot-on-sight, shops and small busi-
nesses will reopen, with both young and old working. Compare this to a situation in
which the young are legally permitted to work. Then the locus of transmission
shifts to the household and the family.
How the two loci compare is a matter of debate. We do not have the data to
assuredly predict which measure poses the greater transmission risk to the elderly.
But here is what we can say. Our proposal gets the older individuals out of the
direct workforce, and it asks for protective measures primarily within the family,
supplemented by State-assistance across families. Once again, going by recent
reactions in the community to medical personnel and to individuals suspected
of having been infected, we would be extremely wary of any steps that rely on
the social altruism of private agents across families. There could be feasible, de-
centralized measures that families could adopt, perhaps with the help of a small
cluster of neighboring and friendly families. For instance, there is the option of
using two families to reallocate living quarters between old and young. But the
entire strategy should rely on small, voluntarily formed groups to internalize
externalities. While no measure is ideal, ours relies on the self-interest of house-
holds rather than a generalized notion of the social good. Finally, we should also
note that in India, the elderly constitute a relatively small fraction of the popu-
lation. The population of India aged 65 or older is just 5.1% of the total, a figure
nearly five times as small as the corresponding number for Italy, which is 23%.11
4.5.2 In any case, young versus old aside, won’t this proposal
increase the overall incidence of Covid-19, relative to a
comprehensive lockdown?
Alas, it will. The best of all worlds cannot be achieved. The one assured way
to minimize the incidence of the disease is to have a complete and fully imple-
mented lockdown. But – given the impossibility of the highlighted phrase in
the previous sentence – is that really what we want to do? Must we neglect
the immense burden – not merely economic, but in terms of human lives and
suffering – that a comprehensive lockdown must place on the majority of the
Indian population?
The considerations above must be supplemented by the micro and macro per-
spectives introduced earlier. The added burden placed on the health system at an
aggregative level could coexist with a much smaller burden at the individual or
micro level, particularly if it turns out - as some evidence suggests that it might -
that fatality rates from the coronavirus are of the same order of magnitude as the
flu. There is certainly a case to be made for massively funding the health system
in this time of crisis, while loosening the lockdown in the manner proposed
above. In the years to come, this improved health system will be a blessing.
4.6 Concluding observations

Epidemiologists and infectious-disease specialists correctly emphasize the dan-
gers of a lethal virus spread. It is their job to do so. With the prisoner’s dilemma
that pervades any such situation, it is also their job to warn the public that the
personal dangers of an infection are high.12 But that should not obscure the vision
of a broader social ethic. The objective is to save lives on the whole, not just to
reduce pandemic deaths.
In some societies with widespread and comprehensive safety nets, the two al-
ternatives in the previous sentence coincide. A lockdown will not kill, especially
with a well-articulated and well-targeted strategy of economic compensation.
(There could be vociferous objections from business groups, but that’s another
matter.) But in societies like India, a lockdown can kill: via job loss, increased
vulnerability to economic shocks, and social stigma and misinformation. Then
the objective of saving lives as a whole may or may not be achieved by a draco-
nian lockdown.
We’ve contended here that the problem of which strategy to invoke to com-
bat the pandemic is not only a complicated one, but one which does not nec-
essarily, and always, resolve itself in favor of the option that seems currently to
command a considerable measure of international consensus. In the context of
a large developing country like India, where both poverty and inequality are
ingrained structural features of the economy and society, there may be a case for
a relaxed version of the stringent universal lockdown approach. Stated without
nuance, this is an approach that would – over the long haul – allow the relatively
less vulnerable working-age population some latitude in the matter of going to
work, in order to minimize the costs of unemployment, starvation and income
loss. Responsibility for the welfare of the elderly could be decentralized to the
level of individual households, but must also be very robustly shared by the State
through an organized, old-age-targeted policy of special care – through testing,
quarantining and treatment.
We conclude on a final note of philosophical interest, one in which the polit-
ical aspect of the problem also acquires a certain salience. The debate of “lock-
down versus economy” (crudely put) assumes very different political hues in the
“West” and in India. In Europe and in the United States, one would be inclined
to support a lockdown without much hesitation, because Covid-19 is highly

contagious – much more so than the flu. Even if an individual’s risk of mortality
is about the same across Covid-19 and the flu (conditional on being infected, that
is), the strain on the health system is enormous because Covid-19 is far more con-
tagious. While the economic effects are large – certainly so in a high-inequality,
restricted-social-net country like the United States – one has the informational
and targeting capacities needed to make people whole: starvation is not first
order. In India, it is first order. The same progressive support for a lockdown in
the “West” may need to be entirely reallocated in a different context. It is in this
context that our suggestions are offered – in a genuine spirit of shared enquiry
that invites opinion and discussion, and not with assertive confidence or dog-
matic certainty.
Notes
1 National Statistical Office of the Government of India (2019), Annual Report, PLFS
2017–18.
2 Kelmanson et al. (2019).
3 Authors’ calculations using data from a 2015 project on “Savings Behaviour and the
Introduction of Mobile Banking in India” in which V. Somerville and L. Vandewalle
were involved.
4 Khera (2020).
5 Subramanian (2019).
6 For instance, Drèze (2020).
7 For instance, the Center for Disease Control in the United States estimated that
between 37.4 million and 42.9 million people contracted the flu during the 2018–19
season.
8 Ray (2020).
9 This section repeats, almost verbatim, parts of an article published in Ideas for India by
a subset of the present authors: Ray and Subramanian (2020).
10 As in the case of employees still at work in other parts of the world, it is important to
keep them informed through public awareness campaigns and to encourage employ-
ers to take precautions (e.g. through the provision of hand sanitizer and respecting
social distance).
11 Statistics from the Census of India and the World Bank.
12 For instance, it is not uncommon to make statements that the young could also be
lethally affected by Covid-19. Such arguments are often best made by highlighting
the absolute numbers of the young that could be affected, especially in highly populated
metropolitan areas.
References
Drèze, J. (2020), View: The Finance Minister’s Covid-19 Relief Package Is Helpful, But
There Are Gaping Holes in It. The Economic Times, 28 March.
Kelmanson, B., Kirabaeva, K., Medina, L., Mircheva, B. and Weiss, J (2019). Explaining
the Shadow Economy in Europe: Size, Causes and Policy Options, IMF Working Paper
19/278.
Khera, R. (2020). Covid-19: What Can Be Done Immediately to Help Vulnerable Pop-
ulation. Ideas for India, 25 March.
Ray, D. (2020). The Micro and the Macro of Covid-19, or the Case of the Invisible De-
nominator. Chhota Pegs, 30 March.
Ray, D. and Subramanian, S. (2020). Covid-19: Is There a Reasonable Alternative to a
Comprehensive Lockdown? Ideas for India, 28 March.
Subramanian, S. (2019). Letting the Data Speak: Consumer Spending, Rural Distress,
Urban Slow-Down, and Overall Stagnation. Arena Papers of The Hindu Centre for Poli-
tics and Public Policy, 11 December.
5
VIOLENCE AMIDST VIRUS
An essay on conflicts in the world and at
home in times of a pandemic
Soumyanetra Munshi
5.1 Introduction
And such peaceful and unthinking tranquillity almost effortlessly contra-
dicted the old images of pestilence: Athens stricken, abandoned by its birds;
Chinese towns full of people dying in silence; the convicts of Marseille pil-
ing dripping corpses into holes; the building of the great wall in Provence
in the hope of holding back the raging wind of plague; Jaffa and its ghastly
beggars; beds, damp and rotten, sticking to the earth floor of the hospital
in Constantinople; sick people dragged along by hooks; the carnival of
masked doctors during the Black Death; the living copulating in the cem-
eteries of Milan; the carts of the dead in London paralysed by fear; and days
and nights filled, everywhere and always, with the endless cries of men.
(The Plague, Albert Camus)
Episodes of pandemics,1 bringing entire civilisations to their knees, are as old

as civilisations themselves – the Black (bubonic) Plague of the Middle Ages;
smallpox in the seventeenth and eighteenth centuries; the cholera pandemics of
the nineteenth century; the influenza pandemics of 1880, 1918, 1957 and 1968;
childhood infectious disease epidemics of the early twentieth century, including
diphtheria, polio and scarlet fever; small pox epidemics of the nineteenth and
twentieth centuries; and also contemporary crises involving HIV/AIDS, tuber-
culosis, SARS and other newly emerging infectious diseases are some of them.
And every time, mankind has traversed yawning distances, snaked silent path-
ways and scaled eerie heights to overcome the apparently insurmountable hurdles
to emerge more knowledgeable and victorious.
DOI: 10.4324/9781003220145-6
76 Soumyanetra Munshi
As with any pandemic, the current Covid-19 pandemic has brought about
sweeping changes in existing economic, social and cultural paradigms. Greetings
no longer require handshakes, academic seminars have been replaced by webi-
nars, businesses often run in online modes, countries that proffered globalisation
have given a clarion call to “atmanirbharta” (self-reliance) – these are only a
handful of outright changes that have come about with the pandemic. It is hardly
surprising therefore that just the way the pandemic has deeply affected every
other aspect of human life, its existence and its mode of conduct, it will also have
substantially affected human disposition towards violence, both at a collective
level and at an individual level, thereby having repercussions on ongoing conflict
situations the world over, as well as intra-household. In which direction is the
change likely is however the moot question.
In a recent journalistic account, the apprehensions were well expressed by the
following:
Syria, Libya, Yemen, Afghanistan, the Sahel… with the great powers fo-
cused intently on the COVID-19 virus, will armed conflicts across the
world decrease in the severity or intensify? Experts, as well as diplomats at
the United Nations, say there is a serious risk of the latter.2
They go on to explain the reason why diplomats and experts fear an increase in
conflicts – when governments and international organisations are all preoccupied
managing the outbreak of a pandemic of such proportions, the level of diplomatic
attention and potential retaliation to incidences of violence will be seriously
compromised to say the least. Incidences of violence that would otherwise have
drawn widespread international criticism and condemnation are likely to gener-
ate lesser attention and flak, now that greater perils are to be dealt with. Evoking
the “potentially devastating impact of Covid-19 in Idlib and elsewhere in Syria,”
the UN undersecretary-general for political affairs, Rosemary DiCarlo, called
on Twitter for all parties to show restraint. “If anyone – incredibly – still needed
a reason to stop the fighting there,” she added, “this is it.”
As unexpected and unlikely as they may seem amidst the apocalyptic pan-
demic, the geopolitical developments in many countries, like Yemen and Af-
ghanistan, seem to be somewhat positive.3 In Yemen, torn apart with years of
conflict and strife, the Saudi Arabia after years of interference is likely to call it
a day with a ceasefire, while in a watershed development in Afghanistan, the
Afghan government and the Talibans are likely to come to dialogues about the
return of peace to Afghanistan, even as the USA and its allies pull out their
troops. Along the India-China border, heightened tensions have been mitigated
with careful deliberation. So politically fractious countries seem to be gathering
some token political consensus, forging some amount of political unity and cre-
ating some semblance of political will, under the mightier common enemy of
the pandemic. The pandemic seems to be plastering the fault lines of a fractious
geopolitical structure.
Violence amidst virus 77
A priori however, it would be rather presumptuous to argue that the terror

of the pandemic would undoubtedly alleviate differences and help reduce con-
flicts. On one hand, confrontation would expose both states and violent non-
state groups to contamination, potentially causing massive loss of human lives.
Moreover, attacks aimed at signalling discontent or making bigger demands are
unlikely to generate media and diplomatic attention at the level it otherwise
would have, sans the pandemic. Hence, there might be a mitigation in the in-
tensity of conflicts. On the other hand, the capacity of the states to retaliate will,
to a large extent, be compromised, since huge financial and human resources
will have to be dedicated to fighting the widespread health and economic effects
of the pandemic. Hence, the belligerent groups may have greater incentive to
launch attacks. However, as this chapter shows, exploration of three of the con-
flicts suggests the factors inhibiting conflict must have gained greater ground so
that overall conflict situations seemed to have improved.
Quite the opposite story emerges however, when we study violence at a more
household level. The enforcement of lockdown measures worldwide to contain
the spread of the deadly virus has resulted in compulsory confinement of men
and women of the same household, irrespective of whether or not they are in
an abusive relationship. This has resulted in a spike of cases of domestic violence
inflicted on women the world over and is proving to be an additional bane on the
psychological, mental, emotional and physical health of the women – a situation
that has come to be reckoned as the “shadow pandemic.”
This chapter is not intended to be a comprehensive account of all major conflicts
in the world but rather somewhat focused on those that seem to have reached criti-
cal milestones, especially triggered by pandemic-like circumstances of the world.4
Besides violence at the “macro” level in some sense, this chapter touches upon
violence at a more “micro” level – the very sensitive and definitely not any less
important matter of domestic violence or violence inflicted on women during the
pandemic situation and the consequent lockdown.5 The rest of the chapter is as fol-
lows: in Section 5.2, we describe in some details three of the conflicts in the world
(conflict in Yemen, the Indo-China standoff, the Afghan War); in Section 5.3,
we briefly discuss the “shadow pandemic”; and Section 5.4 concludes.
5.2 Conflicts in the world

In this section, we focus on three conflicts around the world – the conflict in
Yemen, the Indo-China standoff and the Afghan War.
5.2.1 Conflict in Yemen

Many scholars, specifically in the context of Yemen,6 had apprehended an in-
crease in internal conflict and increased foreign interference during a pandemic
situation. However, that is not what we see in reality.7 Let us look at the conflict
in Yemen closely to see how conflicts have fallen post-pandemic.
5.2.1.1 Background
In spite of being resource-rich, Yemen is one of the poorest countries of the
world.8 More than 80% (2018) of the population lives in poverty and it ranks
165th (nominal, 2018) according to GDP per capita. At the time of writing this
chapter, the number of corona affected in the country is 2,0169 (which is small
compared to the affected in other countries). However, it is engaged in conflict
with a Saudi Arab-led coalition, and the latter is a rich country.10 And the num-
ber of corona affected in the country is slightly more than 3 lakh.11
5.2.1.2 Civil war in Yemen

Yemen has been reeling under conflicts for decades12 and faces one of the worst
humanitarian crises. The civil war has internally displaced and killed several
thousand people.13 The Iran-Saudi rivalry has also intensified over Yemen. Re-
cently, in March, just around the spread of the corona pandemic, Yemen ex-
pressed their desire for a mutually acceptable ceasefire, followed by a ceasefire
declaration from a Saudi-led alliance in April 2020.
5.2.1.3 History
The main conflict in Yemen is between the UN-recognised government of
Mansour Hadi, supported by Saudi Arabia and the United Arab Emirates
(UAE), and the Iranian-backed Houthis. The Houthis belong to the Zaydi sect
of Shias, which constitute nearly 40% of the Yemeni population. The Zaydis
had ruled over northern Yemen for almost a millennium before being over-
thrown in a coup in 1962. For the next three decades, the Zaydis who were
at the top of the social order were marginalised both politically and economi-
cally by the government. Finally Mohammad Badr al-Din Houthi along with
Muhammad Izzan started the Ansar Allah movement which carried out ex-
tensive military campaigns during 2002–09 in the hope of securing greater
political participation.
The Houthis captured capital Sanaa in September 2014 and placed President
Mansour Hadi under house arrest. In January 2015, President Hadi resigned and
escaped to Saudi Arabia, where he pleaded to the international community to
restore his elected government in Yemen. As a result, in March 2015, an alliance
led by Saudi Arabia and UAE and comprising of several other Arab countries,
with logistics and intelligence support from the USA, commenced the military
campaign against the Iranian-backed Houthi militia in Yemen.
Yemen has also become the centre stage of the theatrics of Saudi-Iran rivalry.
The Saudi-led Operation Decisive Storm, often dubbed as Saudi Crown Prince
Mohammad Salman’s war, commenced in 2015 with massive bombing cam-
paigns and naval blockade, and it was probably believed that it could securely
reinstate Mansour Hadi’s government in Sanaa. Instead, Iran has supported the
Houthis and they have put up a strong defence, prolonging the conflict with a
continuous military stalemate in various provinces.14
5.2.1.4 Covid-19 in Yemen

At a time when the world is struggling to fight a pandemic, the focus of the
parties must shift away from fighting one another to ensuring that the popula-
tion will not face even graver risks (Martin Griffiths, the UN special envoy for
Yemen).
In an already war-torn country, bloody with decades of civil war, the advent
of Covid-19 has thrown life and preparations completely haywire. Moreover, the
USA and World Food Programme have recently announced a cut in healthcare
aid and other kinds of aid to Yemen owing to budgetary constraints, which are
likely to affect several Houthi-controlled areas. Also the Donald Trump admin-
istration’s move to halt funding to WHO would severely undermine WHO’s
efforts in Yemen. The UN Secretary-General has called15 for an “immediate
global ceasefire in all corners of the world” and made an appeal “to put armed
conflict on lockdown and focus together on ‘the true fight of human lives’.”
Driven by the seriousness of the crisis, the Saudi Arabia-led coalition (the coa-
lition includes UAE and Oman siding with Saudi Arabia) announced a two-week
ceasefire on April 9. Houthis welcome the move but are somewhat circumspect
and want more guarantee from the UN. Despite the declaration of a “compre-
hensive ceasefire” by Saudi Arabia, Tehran is yet to make an official statement
on the issue. However, given the impact of Covid-19 pandemic in Iran and its
sanctions hit economy, it is unlikely that Tehran would have sufficient resources
to actively stay involved in the Yemeni conflict. Riyadh too is exhausted with
the war and deeply affected by falling oil prices in addition to its internal politi-
cal problems. In such a situation, Riyadh may look for an honourable exit from
Yemen’s war theatre. It can be argued that under present circumstances, both
Iran and Saudi Arabia have sufficient number of reasons to make the ceasefire
work.16 However, since 2015, the Yemeni conflict had been a saga of ceasefires
and their relentless violations. But the malaise of Covid-19 – a common concern
for all in the region – may be a reason for some optimism.
5.2.2 India-China conflict

Here, we look at the India-China standoff, especially in the context of the recent
flash point pertaining to the control of the Galwan valley.17 The neighbouring
armies of India and China are engaged in the standoff in eastern Ladakh. And
they are locked in a standoff, not just for territorial gains on land but enhanced
domination of the northern bank of Pangong Tso, a resource-rich lake. Pangong
Tso or Pangong Lake is a lake in the Himalayas, and extends from India to the
Tibetan Autonomous Region, China, where approximately 60% of the length of
the lake lies within the Tibetan Autonomous Region. The standoff at Ladakh’s
Galwan Valley has escalated in recent weeks due to the infrastructure projects
that India has undertaken in the recent years. India is building a strategic road
through the Galwan Valley – close to China – connecting the region to an air-
strip. China is opposed to any Indian construction in the area.
5.2.2.1 Background
The border, or Line of Actual Control, is not demarcated, and both the countries
differ in their perceptions of the line, leading to regular border “transgressions.”
Often these don’t escalate tensions and a serious standoff like the current one is
uncommon, though not unseen – this is the fourth since 2013. Both countries’
troops have patrolled this region for decades, as the 2,200-mile border has long
been contested, including leading to a war in 1962, popularly called the Sino-
Indian War or the Indo-China War.18
The main cause of the 1962 war was a dispute over the sovereignty of the
widely separated Aksai Chin and Arunachal Pradesh border regions. Aksai Chin,
claimed by India to belong to Ladakh and by China to be part of Xinjiang,
contains an important road link that connects the Chinese regions of Tibet and
Xinjiang. China’s construction of this road was one of the triggers of the conflict.
There had been a series of violent border skirmishes between the two coun-
tries after the 1959 Tibetan uprising, when India granted asylum to the Dalai
Lama. India initiated a defensive Forward Policy from 1960 to hinder Chinese
military patrols and logistics, in which it placed outposts along the border, in-
cluding several north of the McMahon Line, the eastern portion of the Line of
Actual Control proclaimed by Chinese Premier Zhou Enlai in 1959.
Much of the fighting took place in harsh mountain conditions, entailing large-
scale combat at altitudes of over 4,000 metres (14,000 feet). As the Sino-Soviet
split heated up, Moscow made a major effort to support India, especially with the
sale of advanced MiG fighter aircraft. This was the first war between India and
China. Following the end of the war, both sides kept forward armed positions
and a number of small clashes broke out, but no large-scale fighting ensued.
5.2.2.2 Amidst the pandemic

Recently, there has been escalation along the border.19 The border skirmishes,
which started on May 5, 2020 at the Galwan Valley in the high-altitude Ladakh
region in the north, and then at the Naku La mountain pass in northeastern
Sikkim three days later, have led to a military and diplomatic impasse between
the two countries. Thousands of soldiers on both sides are camping along the
undemarcated Line of Actual Control.
Recently, however, diplomatic talks have yielded positive results and we see
disengagement of troops by India and China along the Line of Actual Control,
further corroborating the overall prediction of the model of lowering of conflict
levels post-pandemic.20 Thereafter again, on the night of August 29/30, Chinese
troops violated the consensus arrived at during military and diplomatic engage-
ments during the ongoing standoff in Eastern Ladakh and carried out provoca-
tive military movements to change the status quo.21
Indian authorities (External Affairs Minister S. Jaishankar) emphasised that a
solution for the recent standoff between India and China “has to be found in the
domain of diplomacy.”22 Hence, though tensions are high, it seems that both the
countries have resolved to resort to diplomacy to overcome the recent impasse.
5.2.3. Afghan conflict

Amidst the apocalyptic scenario the world over, it is mildly reassuring that there
is a silver lining across the rugged terrain of Afghanistan, fairly acutely afflicted
by the coronavirus pandemic. While the development is wholly unrelated to
the dreadful illness, the forward movement in geopolitics is unmistakable two
wholly unrelated quantities. The political consensus is critical in terms of gov-
ernance in a chronically fractious country. The release of prisoners will be a
watershed development in Afghanistan, although many Afghans will continue to
have misgivings about the return of the Taliban to the mainstream.
5.2.3.1 Background
Afghanistan War,23 basically a war launched by the USA on the land of Afghan-
istan, was sparked off by the September 11 attacks by the Afghanistan-based
al-Qaeda on the Twin Towers in New York in the USA, and after the Taliban
had refused to hand over the man behind them, Osama bin Laden. The first
phase was mainly targeted at toppling the Taliban, the ultraconservative po-
litical and religious faction that ruled Afghanistan and provided sanctuary for
al-Qaeda. The USA was promptly joined by an international coalition and the
Taliban were quickly removed from power. This phase was brief and lasted for
about two months.
However, the Taliban turned into an insurgent force and continued deadly
attacks, destabilising subsequent Afghan governments. In the second phase, from
2002 until 2008, the USA followed a strategy of defeating the Taliban militarily
and rebuilding core institutions of the Afghan state.
The third phase began in 2008 and accelerated with US President Barack
Obama’s 2009 decision to temporarily increase the US troops in Afghanistan
that saw the number of American soldiers in Afghanistan top 100,000. The larger
force was used for protecting the population from Taliban attacks, on one hand,
and for supporting efforts to reintegrate insurgents into Afghan society, on the
other hand. The surge helped drive the Taliban out of parts of southern Afghan-
istan, but it was never destined to last for years.
The strategy also laid down a timetable for the withdrawal of the foreign
forces from Afghanistan that stipulated that beginning in 2011, security respon-
sibilities would be gradually transferred over to the Afghan military and police.
However, the theoretical approach largely failed to convert to reality and achieve
its aims. Taliban attacks and civilian casualties remained disturbingly high (and
in 2018 the BBC found they were active across 70% of Afghanistan) while many
of the Afghan military and police units taking over security duties appeared to
be ill-prepared to stand up to the Taliban. The US and NATO combat mission
formally ended in December 2014, ending a 13-year Afghanistan War that went
down in history as the longest war ever fought by the USA. There are many
reasons why the war has lasted so long. But they include a combination of fierce
Taliban resistance, the limitations of Afghan forces and governance, and other
countries’ reluctance to keep their troops longer in Afghanistan. There’s no ques-
tion the Taliban have their roots in Pakistan, and that they were able to regroup
there during the US invasion. But Pakistan has denied helping or protecting
them – even as the USA demanded it do more to fight militants.
Nearly 3,500 members of the international coalition forces have died in Af-
ghanistan since the 2001 invasion. The figures for Afghan civilians, militants
and government forces are more difficult to ascertain though in a February 2019
report, the UN said that more than 32,000 civilians had died, while according to
The Watson Institute at Brown University, some 58,000 security personnel and
42,000 opposition combatants have been killed.
5.2.3.2 Towards peace

Since 2011, Qatar has hosted Taliban leaders who have moved there to discuss
peace in Afghanistan. It has been a chequered process but following nine rounds
of US-Taliban talks in Qatar, the two sides seemed close to an agreement. Wash-
ington’s top negotiator announced in September that the USA would withdraw
5,400 troops from Afghanistan within 20 weeks as part of a deal agreed “in prin-
ciple” with Taliban militants.
The USA and the Taliban signed a peace deal on February 29 in the Qatari
capital of Doha,24 which included an agreement for the release of some 5,000
Taliban prisoners in exchange for 1,000 Afghan security forces.25 The prisoner
exchange deal would serve as a precursor to the commencement of intra-Afghan
talks (between the Talibans and the Afghan government) to usher in peace in the
country after nearly 20 years.26
The main provisions of the agreement are as follows:
i Within the first 135 days of the deal the USA will reduce its forces in Afghan-
istan to 8,600, with allies also drawing down their forces proportionately.27
ii The deal also provides for a prisoner swap. Some 5,000 Taliban prisoners and
1,000 Afghan security force prisoners would be exchanged 28 by March 10,
when talks between the Taliban and the Afghan government are due to start.
i ii The USA will also lift sanctions against the Taliban and work with the UN
to lift its separate sanctions against the group.
iv The main commitment the Taliban made to Washington was to sever ties
with al-Qaeda and prevent it and other militant groups from using Afghan-
istan as a base for international attacks.
The Taliban delegation arrived in Kabul on March 31 and held several meet-
ings with Afghan government officials to discuss technicalities for the prisoner
exchange.29 This was the first official delegation of the Taliban in the Afghan
capital in 19 years since the group was ousted from power in 2001 by the USA.
On April 9, 2020, the Afghanistan government released the first batch of 100
Taliban prisoners as a part of a prisoner-swap process expected to pave way for
the intra-Afghan talks.30
According to latest news31 (September 5, 2020), a Taliban delegation has re-
turned to Qatar, paving the way for the start of peace talks with the Afghan
government that are expected to take place in the Gulf state. Washington has
ramped up pressure on Afghans on both sides of the conflict to get started with
their negotiations to decide what a post-war Afghanistan might look like.
The talks come against a backdrop of a diminishing US military presence,
and Donald Trump has signalled that US soldiers will head home whether or
not Afghan parties reach a peace deal. There were 13,000 US forces personnel in
the country when the withdrawal agreement was signed in February, and troop
numbers have already fallen dramatically.
5.2.3.3 International support

The developments in Afghanistan have seen huge international support. UN
Secretary-General Antonio Guterres stressed “the importance of sustaining
the nationwide reduction in violence, for the benefit of all Afghans.” Nato
Secretary-General Jens Stoltenberg: “We went in together in 2001, we are going
to adjust [troop levels] together and when the time is right, we are going to leave
together, but we are only going to leave when conditions are right.” UK Defence
Secretary Ben Wallace: “I welcome this small but important step towards the
chance for Afghans to live in peace, free from terrorism… We remain absolutely
committed to building an Afghanistan that is a strong partner for decades to
come.”
5.2.3.4 Amidst the pandemic

Some observers feel32 that despite the pandemic and the critical situation of the
Afghan authorities to tackle it, and despite the peace deal signed on February 29,
2020, the war in Afghanistan continues.
U.N. Secretary General Antonio Guterres called for a global human-

itarian ceasefire to combat COVID-19, including in Afghanistan.
Secretary General of the Organization of Islamic Cooperation Yousef A.

Al-Othaimeen appealed twice for a ceasefire. American and European dip-
lomats have called on the Taliban to announce a ceasefire. Even President
Ashraf Ghani of Afghanistan called for a ceasefire during Ramadan, a holy
month for Muslims.
In spite of an all-out effort to contain war and concentrate efforts on controlling

the Covid situation, it has not borne much fruit.
As Afghan cities are in lockdown due to the spread of the coronavirus, the
war has shifted into rural areas, where the strategic gains are made or lost. The
Taliban attacks are aimed at district compounds and security outposts, while
the Afghan forces rely on airstrikes and raids to defend themselves.
The first case of coronavirus in Afghanistan33 was confirmed on February
24, 2020, in Herat province. As of July 8, 33,594 cases and 936 deaths have been
confirmed. Nonetheless, only 77,501 people have been tested to date, which is
why the confirmed cases are estimated to be only a tenth of the actual case num-
bers. For example, the government of Herat province estimated that in the week
leading up to June 12, 280 people died in the province of suspected Covid-19
without ever seeing a medical professional.
Even under normal circumstances, the opportunity costs of war are huge, and
they are especially so in an already war-stricken country under the affliction of
a growing pandemic.
War material – rifles, pistols, machine gun cartridges, hand grenades, rock-
ets, shotgun slugs, mortar rounds, tank ammunition, and more – has cost
Afghanistan millions of dollars. Those financial resources could have been
used to purchase protective equipment, face masks, and ventilators to save
lives from COVID-19.
It is not clear how the Taliban are handling the virus, but the Afghan forces have
set up measures to prevent the spread of the virus within the ranks. The NATO
support mission supplies Afghan forces with soap, gloves, eyeglasses and disin-
fectant across the country, helping them protect themselves from the pandemic.
“The Defense Ministry established teams to fight the coronavirus within the
Afghan forces,” said Fawad Aman, spokesperson for the Defense Ministry. “Every
unit and place is disinfected and forces are regularly examined for the coronavirus.”
Colonel Baser, who commands a brigade, said that his troops are observ-
ing the quarantine. Service members avoid unnecessary movement, especially to
public places such as bazaars and marketplaces. Barring an emergency, the forces
stay in their units. Still, there are limit to social distancing amid a war zone.
But for many on the frontlines, Covid-19 is a secondary enemy. “The fatality
rate of Taliban virus is still much higher than the coronavirus,” said one Afghan
solider.
Covid-19 has given the US-Taliban deal and the peace process a rough ride,34
as the country shifts its resources and time to contain the pandemic. According
to latest news, a Taliban delegation has arrived in Qatar to take talks with the
Afghan government forward.
The spread has also put internal and external pressure on the government to
release Taliban prisoners, which was one of the key demands of the Taliban, due
to the possibility of an outbreak in Afghan jails.
Ministry of Interior Affairs has banned large gatherings and has implemented
strict lockdown. The government had also suspended the Persian New Year cel-
ebration, ordered a complete shutdown of educational institutions and sealed
provinces bordering Iran, i.e. Herat, Nimroz and Farah. The government an-
nounced the release of 10,000 prisoners in effort to combat the spread in jails. As
announced, the government has been releasing those who have committed small
crimes and are a less threat to the country.
Overall, it does seem to be the case that peace processes are headed in the right
direction in spite of the catastrophic pandemic raging across the globe.
5.3 Shadow pandemic – violence against women

Rows often break out purely to bad temper, which is becoming chronic.
As the violence at the macro level seems to abate somewhat, it is altogether an op-
posite story for violence at a more subtle and micro level – there has been marked
increase in violence against women during the pandemic.35 With 90 countries in
lockdown, 4 billion people are now confined at homes to prevent the spread of
the virus which is having this unwanted repercussion of what has been termed as
the “shadow pandemic” of growing violence against women.
In countries like Argentina, Canada, France,36 Germany, Spain, the UK,37
and the USA,38 there are increasing reports of domestic violence during the crisis
and greater demand for emergency shelters. Helplines in Singapore39 and Cyprus
have registered an increase in calls by more than 30%, whereas, in Australia, 40%
of frontline workers in a New South Wales survey reported increased requests for
help with violence that was escalating in intensity. In Albania,40 the pandemic
has only intensified gender-based violence. One in two women in Albania has
been subjected to sexual, physical or psychological violence, according to a sur-
vey supported by the Swedish government, UNDP and UN Women. For nearly
half, that violence has come from a partner.41
In India42 too, plight of women has followed a similar trajectory. National
Commission for Women has confirmed a rise in the incidents of domestic vi-
olence since March 25, 2020. According to data,43 during the first four phases
of the Covid-19-related lockdown, Indian women filed more domestic violence
complaints than those recorded in a similar period in the last ten years. Between
March 25 and May 31, 2020, 1,477 complaints of domestic violence were made
by women, which was higher than those received between March and May in
the previous ten years.44 And yet it is only the tip of the iceberg, since about 86%
women who experienced violence never sought help, and 77% of the victims
did not even mention the incidents of violence to anyone. Among the 14.3% of
victims who sought help, only 7% reached out to relevant authorities – the po-
lice, doctors, lawyers or social service organisations. But more than 90% of the
victims sought help only from their immediate family.
Even under normal circumstances, there is huge under-reporting of domestic
and other forms of violence – less than 40% of women who experience violence
seek help of any sort or report the crime. Less than 10% of those women seeking
help go to the police. The current circumstances make reporting even harder,
including limitations on women’s and girls’ access to phones and helplines and
disrupted public services like police, justice and social services. This also com-
promises the care and support that survivors need – like clinical management of
rape, and mental health and psycho-social support. They also fuel impunity for
the perpetrators. In many countries, the law is not on women’s side – one in four
countries has no laws specifically protecting women from domestic violence.
And the trend is understandable – women are locked up with the perpetrators
and routes of either them escaping or outside help coming through are quite
closed under the lockdown scenario, hence making them more vulnerable. They
are even more frightened to report since they are in the same house with the very
people she would be reporting against – hence there is greater fear of escalation
of violence in case their reports come to be known to the perpetrators. Earlier
women would have a window of relief when the man (the perpetrator of vio-
lence) went out for work or when she was away for work herself. Given lockdown
has these avenues, domestic scenarios are much graver.
Even before Covid-19 existed, domestic violence was already one of the
greatest human rights violations. In 2019, 243 million women and girls (aged
15–49) across the world have been subjected to sexual or physical violence by
an intimate partner. A woman is tied to her man in intricately complicated and
discreet networks. Especially in India, she is socially, religiously, emotionally as
well as financially tied to a partner, which makes it very difficult for her to break
away. Even in cases where the woman is earning, the assets are usually held in
the man’s name. There is an enormous amount of social sanctions and stigma
that broken households face – often daughters are unable to return to unwilling
parents, court proceedings are expensive and unduly protracted, and there is the
overarching concern for one’s children and them having the father’s identity.
Hence more often than not, women continue in abusive relationships even under
normal circumstances. And now with the raging pandemic, these circumstances
stand much more worsened at best and completely non-existent at worst.
If not dealt with, this shadow pandemic will also add to the economic impact
of Covid-19. The global cost of violence against women had previously been
estimated at approximately USD 1.5 trillion. That figure can only be rising as
violence increases now, and continues in the aftermath of the pandemic. The
increase in violence against women must be dealt with urgently with measures
embedded in economic support and stimulus packages that meet the gravity
and scale of the challenge and reflect the needs of women who face multiple
forms of discrimination. Some countries like Panama and Peru have enforced
gender-based lockdown days.45 For example in Peru, men can leave their homes
on Mondays, Wednesdays and Fridays with an ID, while women on Tuesdays,
Thursdays and Saturdays. The gender-day scheme is exactly the opposite in Pan-
ama. While on Sundays it is lockdown for everyone. This may lead to some
relaxation as far as gender tensions arising out of confinement in close quarters
are concerned.
5.4 Conclusion
… a fireball in the night of complacency and forgetting: ‘[Rieux] knew
that … the plague bacillus never dies or vanishes entirely… it can re-
main dormant for dozens of years in furniture or clothing, … it waits
patiently in bedrooms, cellars, trunks, handkerchiefs and old papers, and
… perhaps the day will come when, for the instruction or misfortune
of mankind, the plague will rouse its rats and send them to die in some
well-contented city.’
Even as I write this chapter there are developments on many fronts. Intra-Afghan
peace talks have begun from September 1246 and India has joined the inaugu-
ral session.47 In spite of worldwide apprehension of an exacerbation of conflicts
post-pandemic, given lack of monitoring and probable interventions by world
organisation, we see, in many cases, that conflicts have actually fallen.
In Yemen, Saudi Arabia-led coalition of countries, which has been locked in
conflict with the Houthis in Yemen for the last five years, has called for a cease-
fire. In the India-China standoff, after an initial show of hostilities, troops have
started disengaging from the border in a significant de-escalation of conflict.
The Afghans and Talibans are facing each other for the first time and the world
must live in hope and optimism. If there is any remotely positive aspect of a pan-
demic of such proportions, it is probably in the slight amelioration of conflicts
worldwide.
There have been other developments in a positive light as well – there have
been at least two peace deals brokered by Trump in West Asia48 – deals of United
Arab Emirates (UAE) and Bahrain to normalise diplomatic ties with Israel.49
On the domestic front though, there is likely to be a worsening of violence
faced by women, due to confinements imposed by worldwide lockdowns, a nec-
essary measure to contain the spread of the virus. One hopes that NGOs and
governmental organisations can bring enough measures in place to tackle such
an upsurge of gendered violence.
Hence on the distant horizon of the rugged terrain of deaths and diseases,
there seems to be a shimmer of silver lining of infinite human resilience and
endless empathetic patience that will surely help us tide over not just mindless vi-
olence and unconquerable pestilence but also deeper maladies of gender inequal-
ity. In conclusion, let us remember the profound realisation of life that Camus
proclaims through contemplation of his protagonist (Dr Rieux):
All that a man could win in the game of plague and life was knowledge
and memory.
Notes
1 Markel, Howard. Contemplating Pandemics: The Role of Historical Inquiry in Developing
Pandemic-Mitigation Strategies for the Twenty-First Century. Institute of Medicine (US)
Forum on Microbial Threats. Ethical and Legal Considerations in Mitigating Pan-
demic Disease: Workshop Summary. Washington (DC): National Academies Press
(US); 2007. 1, Learning from Pandemics Past. Available from: https://www.ncbi.
nlm.nih.gov/books/NBK54171/.
2 “Will coronavirus slow the world’s conflicts or intensify them?” AFP, United Na-
tions, March 22, 2020, https://www.deccanherald.com/international/will − coro-
navirus − slow − the − worlds − conflicts − or − intensify − them − 816448.html
(Accessed on July 14, 2020).
3 In a related theoretical paper (Violence amidst Virus: A Game-Theoretic Exploration
of Conflict during a Pandemic. (2020) Theoretical Economics Letters, 10, 1292–1306.
doi: 10.4236/tel.2020.106079.), I have developed a simple game-theoretic model by
incorporating possible effects of the pandemic, which actually predicts that for a large
range of parametric restrictions, conflicts are likely to decrease with the onset of a
pandemic.
4 Even in this category it is certainly not exhaustive. It rather humbly submits to deal with
a few important headlines-grabbing conflict outcomes during the pandemic months.
5 The author wishes to thank an anonymous referee for bringing this aspect into the
context of the present essay.
6 Yuksel, Engin, Nancy Ezzeddine, Rena Netjes, Beatrice Noun, Hasim Tekines, and
Erwin Van Veen. (2020) Pandemic or Pandemonium? Covid-19 and Conflict in the Middle
East. Clingendael Institute.
7 The author has an essay “‘Dancing on the Heads of Snakes’: A Glimpse into Yemen,”
available at https://countercurrents.org/2020/12/dancing-on-the-heads-of-snakes-
a-glimpse-into-yemen/, where she explores the history of the country to conclude
that the pandemic has indeed been more of a boon than bane for the nation.
8 It contains oil and gas resources and has productive soil. See Wikipedia webpage
‘Economy of Yemen’ https://en.wikipedia.org/wiki/Economy_of_Yemen (Accessed
on July 6, 2020).
9 See https://www.worldometers.info/coronavirus/country/yemen/ (Accessed on
September 16, 2020).
10 See https://www.worldometers.info/coronavirus/country/yemen/ (Accessed on
11 See https://www.worldometers.info/coronavirus/country/saudi-arabia/ (Accessed
on September 16, 2020).
12 Please refer to the article “Conflict in Yemen amid COVID-19 Pandemic” by Meena
Singh Roy and Prabhat Jawla (April 24, 2020), available at https://idsa.in/idsacomments/
yemen-covid-19-msroy-pjwala-240420 (Accessed on July 6, 2020), for details.
13 According to United Nations High Commissioner for Refugees (UNHCR), “Over

80 per cent have been displaced for more than a year.” In 2019, the death toll in Yemen
had reached 100,000 since the war began five years ago, which includes 12,000 civil-
ian casualties, as per the Armed Conflict Location and Event Data Project (ACLED).
14 The Saudi-Iran conflict has also been dubbed as a Shia-Sunni conflict or proxy war.
Over the last five years, Saudi Arabia has carried out extensive military campaigns
that have only emboldened the Iranian support to Houthis – Iran is now said to facil-
itate the shipment of small arms to include heavy artillery, unmanned aerial vehicles
(UAVs) and even small range surface-to-surface missiles (SSMs).
15 See “COVID-19: UN chief calls for a global ceasefire to focus on ‘the true fight of our
lives’,” UN News, March 23, 2020, https://news.un.org/en/story/2020/03/1059972
(Accessed on April 17, 2020).
16 The cooperation of UAE, a partner of Riyadh in the coalition, would be equally
crucial. The UAE, like Saudi Arabia, too seeks to end its direct military involvement
in the conflict. In March, the UAE had sent medical supplies to Iran amid the coro-
navirus crisis, followed by an official call between the foreign ministers of the two
countries. This signals a positive development that might serve as an opportunity for
Tehran and Riyadh to bridge gaps through Abu Dhabi.
17 Please refer to “Galwan Valley Clash,” June 29, 2020, https://www.drishtiias.com/
daily-updates/daily-news-editorials/galwan-valley-clash (Accessed on July 5, 2020),
and “Pangong Tso,” Wikipedia, https://en.wikipedia.org/wiki/Pangong_Tso (Ac-
cessed on July 5, 2020), for many of the facts in this discussion.
18 See Wikipedia Sino-Indian War, https://en.wikipedia.org/wiki/Sino-Indian_War
(Accessed on September 14, 2020).
19 See Link between Virus, ‘Chinese incursions’: Indian military experts, Cheena Ka-
poor, May 31, 2020 (Accessed on September 14, 2020). This article states “Indian
military experts say there was a link between China’s mishandling of the COVID-19
pandemic and its latest military incursions into the Ladakh region, resulting in a face-
off between the armies. ‘Globally, China is losing its leverage as it is believed to have
caused the pandemic. Industries are looking to move out of China. This is causing
China to divert attention from the COVID-19 situation,’ Lt. Gen. Vinod Bhatia, In-
dia’s former director-general of military operations (DGMO), told Anadolu Agency.”
This article, however, doesn’t dwell on this aspect of the conflict.
20 Refer to https://www.thehindu.com/opinion/editorial/days-of-disengagement-on-
india-china-lac-standoff/article32016057.ece, June 2, 2020, for example.
21 China taking advantage of Covid-19 outbreak, India one such example: US dip-
lomat, September 3, 2020, https://www.hindustantimes.com/india-news/china-
taking-advantage-of-covid-19-outbreak-india-one-such-example-us-diplomat/
story-K4Ni57IRarLYbKgshMsf9O.html (Accessed on September 15, 2020).
22 Ladakh standoff: Space for diplomacy shrinking fast as India, China remain on the edge
of military conflict, Sreemoy Talukdar, September 10, 2020, https://www.firstpost.
com/india/ladakh-standoff-space-for-diplomacy-shrinking-fast-as-india-china-
remain-on-the-edge-of-military-conflict-8802311.html (Accessed on September 15).
23 Much of this discussion is obtained from ‘Afghanistan War’ 2001–2014 written
by Griff Witte, London bureau chief, Washington Post. Available at https://www.
britannica.com/event/Afghanistan-War. See also Afghan conflict: USA and Tali-
ban sign deal to end 18-year war, February 29, 2020, https://www.bbc.com/news/
world-asia-51689443.
24 The deal was signed by US special envoy Zalmay Khalilzad and Taliban political
chief Mullah Abdul Ghani Baradar with US Secretary of State Mike Pompeo as a
witness. In a speech, Mr Pompeo urged the militant group to “keep your promises
to cut ties with al-Qaeda.” Mr Baradar said he hoped Afghanistan could now emerge
from four decades of conflict. “I hope that with the withdrawal of all foreign forces
from Afghanistan the Afghan nation under an Islamic regime will take its relief and
embark on a new prosperous life,” he said.
25 Afghanistan government releases first group of Taliban prisoners, IANS, Kabul, April
9, 2020, https://www.thestatesman.com/world/afghanistan-government-releases-
1st-group-taliban-prisoners-1502875411.html (Accessed on September 6, 2020).
26 Critics of the peace process say they fear the Taliban may be using it to reduce the
US military presence, and will take up arms again when the government in Kabul –
which they have always denounced as puppets – has less international support on the
ground.
27 The move would allow US President Donald Trump to show that he has brought
troops home ahead of the US presidential election in November.
28 According to the provisions of release, 1,500 Taliban prisoners would be released
in batches of 100 every day, and the remaining in batches of 500 every two weeks.
“The National Directorate of Security and the Attorney General’s Office thoroughly
vetted the prisoners, who have taken an oath never to return to the battlefield,” the
statement said, adding that they had received a similar assurance from the Taliban
leadership.
29 Though the actual process was a rather bumpy one – first, the Taliban decided to boy-
cott the talks, terming the meetings with Afghan officials as “fruitless,” and accusing
Kabul and Washington of “deliberately” delaying the prisoners’ release. However, the
Afghan government blamed the Taliban for the delay, claiming the insurgents had
been insisting on the release of 15 top militant commanders who have been involved
in big attacks. The government stressed they first wanted to release those prisoners
who posed a lesser threat to the internal security of the country.
30 Afghanistan releases more Taliban prisoners to pave way for talks. The Guardian.
https://www.theguardian.com/world/2020/aug/14/afghanistan-releases-more-
taliban-prisoners-to-pave-way-for-talks.
31 Aljazeera news Taliban team returns to Doha for intra-Afghan peace talks, September 5,
2020, https://www.aljazeera.com/news/2020/09/taliban-team-returns-doha-intra-
afghan-peace-talks-200905122531188.html.
32 See, for example, Afghanistan: War in the Time of Coronavirus by Ezzatullah
Mehrdad, April 30, 2020, https://thediplomat.com/2020/04/afghanistan-war-in-
the-time-of-coronavirus/ (Accessed on September 5, 2020). And also see, The
coronavirus pandemic hasn’t stopped the war in Afghanistan, by Emran Feroz
and Mohammad Zaman, April 16, 2020, https://www.vox.com/covid-19-
coronavirus-world-international-response/2020/4/16/21220611/coronavirus-
afghanistan-war-taliban-covid-19-cases-deaths (Accessed on September 5, 2020).
33 See Afghanistan’s Covid-19 crisis has been fuelled by armed conflict, July 10, 2020,
https://theconversation.com/afghanistans-covid-19-crisis-has-been-fuelled-by-
armed-conflict-141924 (Accessed on September 5, 2020).
34 See Six Challenges in Afghanistan’s War with Covid-19, by Sukanya Bali, 22 May,
2020, https://diplomatist.com/2020/05/22/six-challenges-in-afghanistans-war-with-
covid-19/ (Accessed on September 5, 2020). However Cry Freedom! The States-
man Desk, New Delhi, April 4, 2020, https://www.thestatesman.com/opinion/
cry-freedom-3-1502873670.html holds somewhat positive view.
35 See Violence against women and girls: the shadow pandemic, Statement by Phumzile
Mlambo-Ngcuka, Executive Director of UN Women, April 6, 2020 (Accessed on
September 7, 2020).
36 Domestic violence cases jump 30% during lockdown in France, by Euronews, last
updated 28/03/2020, https://www.euronews.com/2020/03/28/domestic-violence-
cases-jump-30-during-lockdown-in-france (Accessed on September 9, 2020).
37 Coronavirus: I’m in lockdown with my abuser, by Megha Mohan, Gender and iden-
tity correspondent, March 31, 2020, https://www.bbc.com/news/world-52063755
(Accessed on September 9, 2020).
38 Domestic violence cases escalating quicker in time of Covid-19, by Loi Almeron,
March 27, 2020, https://missionlocal.org/2020/03/for-victims-of-domestic-violence-
sheltering-in-place-can-mean-more-abuse/ (Accessed on September 9, 2020).
39 Domestic violence cases escalating quicker in time of Covid-19, by Loi Almeron,

March 27, 2020, https://missionlocal.org/2020/03/for-victims-of-domestic-violence-
sheltering-in-place-can-mean-more-abuse/ (Accessed on September 9, 2020).
40 Domestic violence cases escalating quicker in time of COVID-19, by Loi Alm-
eron, March 27, 2020, https://missionlocal.org/2020/03/for-victims-of-domestic-
violence-sheltering-in-place-can-mean-more-abuse/ (Accessed on September 9,
2020).
41 In Albania, Kushti writes,
Remzie (not her real name) and her children were forced to sleep in the shop
where she works as a dressmaker “I am happy the owners have allowed me to
sleep here. I cannot go back to my house, as my husband has become more violent
than usual,” she says. “Not a day passed without him beating me and the kids. So,
I left.” Anila (not her real name) said her husband became violent when he lost
his job in a bar “I come home and find him frustrated, shouting, sometimes he
gets violent. I’m not as worried for myself as I am for leaving my children alone
with him all day now that they don’t go to school. But what other choice do I
have?” she says. Most women request anonymity. “My children are there, so is
my mother from time to time, but none of us can stop my husband from releasing
his anger on us. The first week of the quarantine was slightly better, we were all
optimistic things would go back to the way they were in no time. But with each
passing day, no end seems to be in sight and my husband’s mood swings only get
worse. I have nowhere to go.” These stories come from women living in Tirana,
the capital and the largest city of Albania.
Things are likely to be much worse in the rural areas where help is even more scarce.
42 See Lockdown – How it impacts women, by Vageshwari Deswal, April 4, 2020, in
Legally Speaking, India, Times of India, https://timesofindia.indiatimes.com/blogs/
legally-speaking/lockdown-how-it-impacts-women/ (Accessed on September 10,
2020).
43 Domestic violence complaints at a 10-year high during Covid-19 lockdown, by Vig-
nesh Radhakrishnan, Sumant Sen, Naresh Singaravelu, June 22, 2020, https://www.
thehindu.com/data/data-domestic-violence-complaints-at-a-10-year-high-during-
covid-19-lockdown/article31885001.ece (Accessed on September 10, 2020).
4 4 Domestic violence complaints at a 10-year high during Covid-19 lockdown, by Vig-
nesh Radhakrishnan, Sumant Sen, Naresh Singaravelu, June 22, 2020, https://www.
thehindu.com/data/data-domestic-violence-complaints-at-a-10-year-high-during-
covid-19-lockdown/article31885001.ece (Accessed on September 10, 2020).
45 Peru, Panama limit movement by gender in bid to slow the coronavirus, by
Colin Dwyer, April 3, 2020, https://www.npr.org/sections/coronavirus-live-
updates/2020/04/03/826604070/peru-panama-limit-movement-by-gender-in-bid-
to-slow-the-coronavirus (Accessed on September 10, 2020).
46 Warring Afghans meet to find peace after decades of war, September 12, 2020,
https://www.hindustantimes.com/world-news/warring-afghans-meet-to-f ind-
peace-after-decades-of-war/story-XkcWK7v1cqiQkuQ6xeWkEJ.html (Accessed on
47 India joins inaugural session of intra-Afghan peace talks, https://timesofindia.
indiatimes.com/india/india-joins-inaugural-session-of-intra-afghan-peace-talks-
in-doha/articleshow/78080742.cms (Accessed on September 13, 2020).
48 India joins inaugural session of intra-Afghan peace talks, https://timesofindia.
indiatimes.com/india/india-joins-inaugural-session-of-intra-afghan-peace-talks-
in-doha/articleshow/78080742.cms (Accessed on September 13, 2020).
49 According to news, they will exchange embassies and ambassadors, begin direct
flights between their countries and launch cooperation initiatives across a broad range
of sectors, including health, business, technology, education, security and agriculture.
6
BROTHERS IN ARMS? COVID-19 AND
HINDU-MUSLIM CONFLICT IN INDIA1
Anirban Mitra and Arnab Mukherji
6.1 Introduction
Ethnic violence in its multifarious manifestations – religious, racial, linguistic,
etc. – is a veritable menace which threatens humanity all over the globe. Such
events have only intensified over the last century (see, e.g., Miguel, Satyanath
and Sergenti (2004)). India is no stranger to ethnic violence by any means. Hor-
rific incidents of religious and caste-based violence have dotted the length and
breadth of the nation over several decades, if not centuries. Recurrent episodes of
Hindu-Muslim conflict in India (going back to the Partition of the Indian sub-
continent in 1947 and earlier) have continued through the second half of the
twentieth century, accounting for over 7,000 deaths over 1950–2000, and many
more at the time of Partition. Indeed, twenty-first-century India is just more of
the same, and in its pervading sense of menace and repression, possibly much
more.
Researchers from various disciplines have approached the issue of Hindu-
Muslim violence from different angles. The view espoused by Huntington
and others, that it is all about “a clash of civilisations”, however disheartening,
cannot be wished away in light of the unabashed, untrammelled brutality
that has been on display in some of the violent encounters. However, that is
surely not the complete explanation. Wilkinson (2004) has explored how the
degree of political competition in state-level elections can affect the extent
to which (religious, ethnic) minorities become important electorally; that,
in turn, can and does influence the attitude of state governments towards
riot-containment.
DOI: 10.4324/9781003220145-7
Brothers in arms? 93
Mitra and Ray (2014) follow the line espoused in Engineer (1984, 1987 and
1994) and others (see, e.g., Upadhyaya (1992), Rajgopal (1987), Khan (1992),
Bagchi (1990) and Das (2000)) by highlighting an economic component to
Hindu-Muslim conflict. They model inter-group conflict driven by economic
changes within groups and make the following prediction: if group incomes
are low, increasing group incomes raises violence against that group, and lowers
violence generated by it. Using data on Hindu-Muslim violence from 1979 to
2000, they show that regional Hindu-Muslim violence rises in response to an
increase in the regional average of Muslim incomes. The opposite is true when
regional Hindu incomes rise. These empirical findings are of interest in their
own right, but one can go further by using their theory as a device for the in-
terpretation of the empirical patterns. In their work, they suggest that Hindu
groups have largely been the aggressors in Hindu-Muslim violence in India, or
at least in Hindu-Muslim violence driven by instrumental, specifically economic
considerations.
Varshney (2003) has underlined the role of civic engagement in a multi-
cultural society in influencing the pattern of such inter-group violence. In a
related vein, Jha (2013) provides evidence that the degree to which medieval
Hindus and Muslims could provide complementary, non-replicable services and
a mechanism to share the gains from exchange has resulted in a sustained legacy
of ethnic tolerance in some South Asian towns. Specifically, he finds that me-
dieval ports, despite being more ethnically mixed, were five times less prone to
Hindu-Muslim riots between 1850 and 1950, two centuries after Europeans dis-
rupted Muslim overseas trade dominance, and remained half as prone between
1950 and 1995.
The effects of such sectarian violence are not easy to capture – the loss of lives
and livelihoods are a gross underestimate of the overall deleterious impact on so-
ciety. The erosion of trust between members of the different communities serves
to undermine the social capital in the affected areas – this, in turn, depresses both
economic activity and the morale of society.
Now consider the coronavirus pandemic and let us examine some of its salient
aspects in the Indian context. In India, the pandemic’s spread and impact have
been considerably varied across the different states. To elaborate, the distribution
of the disease load has been quite uneven both across states and across rural and
urban areas. On 10 May, the eight most affected Indian states – Delhi, Gujarat,
Madhya Pradesh, Maharashtra, Rajasthan, Tamil Nadu, Uttar Pradesh and West
Bengal – accounted for 90% of the country’s total disease load, while the eight
least affected states (located predominantly in the north-eastern part of the coun-
try) contained 1% of India’s disease load. Moreover, on 10 June, the three metro
cities in India – namely, Chennai, Delhi and Mumbai – harboured more than
50% of the country’s active disease load (see Jalan and Sen (2020)).
94 Anirban Mitra and Arnab Mukherji
The spread of the infection in society depends upon the extent and na-
ture of human engagement and adherence to protocols (wearing appropriate
face coverings, social distancing, regular washing/sanitising of the hands, etc.).
Also, for the successful containment of the virus, there has to be trust between
private individuals and that between the citizens and the public bodies (see,
e.g., Heller (2020), Jalan and Sen (2020), Tharoor (2020) on Kerala’s experi-
ence). Without the trust of the citizens on the public authorities, policies like
lockdowns (howsoever localised) cannot bear fruit. Relatedly, private indi-
viduals have to repose faith in one another as the “externalities” generated
by the breach of the safety protocols are quite sizeable, given the highly con-
tagious nature of the virus. Thus, a certain amount of coordinated action –
predicated by trust – is required for the successful combatting of the pandemic.
This is more likely to arise organically in areas which have high social capital.
Therefore, populations routinely scarred by communal violence are unlikely to
evince such coordination.
However, there is another aspect to the dynamic of religious conflict which
suggests a contrarian view. Observe that the degree of physical engagement
across the different religious communities in communally charged areas is likely
to be low. Moreover, the flow of information and the level of cooperation within
each religious group are likely to be higher in riot-prone areas; after all the sense
of (religious) identification is heightened in such places. Given that residential
segregation along religious lines is fairly high (see, e.g., Klasnja and Novta (2016))
within cities, towns and villages in most of the country, the heightened cooper-
ation within each community may serve to facilitate measures in curtailing the
spread of the virus. This implies that the rate and impact of the infection would
be lower in such riot-prone areas.
The preceding discussion clearly suggests that the net effect may well depend
upon the relative strengths of these opposing forces. This observation motivates
our empirical exercise. We collate data on various Covid-related outcomes –
specifically, numbers infected, deceased, number of active cases and recovered –
at the district level and match that to the district’s past levels of Hindu-Muslim
conflict (between 2014 and 2018, both years inclusive). Using a rich set of control
variables, we examine the association between the different Covid-related out-
comes and the past conflict measures. We uncover a robust correlation between
these two sets of variables. Specifically, we find that higher levels of Hindu-
Muslim conflict (2014–18) are associated with a lower number of Covid-19 in-
fections, deaths and active cases. This pattern is fairly consistent and holds across
a wide range of specifications.
We follow Mitra and Ray (2014, 2019) in guiding our choice of control var-
iables. As Mitra and Ray (2014, 2019) document the importance of various eco-
nomic variables like Muslim and Hindu per capita expenditures (or their ratio) in
predicting the pattern of Hindu-Muslim conflict, we duly construct such meas-
ures and employ them as controls since these economic factors could also have an
independent effect on the Covid-related outcomes.
We wish to draw attention to one particular control variable – namely, the

share of Lok Sabha seats won by the BJP in the district in the 2014 national elec-
tions. The idea of measuring BJP presence (details below) is to create an indicator
for the pro-Hindutva political forces operating in the region. It should be noted
that the BJP is not the sole political player in terms of the ideology and rhetoric
relating to the establishment of a “Hindu-rashtra” (roughly translates as “nation
of the Hindus”). But as a first pass, focusing on the BJP does seem reasonable.
We briefly discuss two recent studies whose findings further justify our approach.
Nellis, Weaver and Rosenzweig (2016) analyse data between 1962 and 2000
to show that the election of Congress legislators in close elections curtailed riot
incidents by about 10% over this period. They focus on assembly constituencies
that faced close contests and then compare differences in rioting in constituencies
where the Congress won to those constituencies where the Congress lost elec-
tions to arrive at their findings. They also argue that rioting polarises voters and
harms the electoral prospects of the Congress party in ensuing elections, while it
has the opposite effect on the fortunes of its principal rival, the BJP, which gains
vote share following such polarisation. Iyer and Shrivastava (2018) analyse the ef-
fect of Hindu-Muslim riots on state election results. They combine data on geo-
coded riots with data on state elections and on demographics and public goods
provision for the period 1981 through 2001. To safeguard against the possible
endogeneity of riot occurrence to various ambient conditions, they introduce a
new instrument which draws upon the random variation in the day of the week
that Hindu festivals fall on, as set by a lunar calendar. The probability of a riot
increases if a Hindu festival falls on a Friday, the holy day for Muslims. In this
manner, they are able to isolate the causal effect of riots on electoral results. They
find that riots occurring in the year preceding an election increase the vote share
of the BJP by a minimum of 5 percentage points.
In the next section, we discuss the data and the empirical strategy and then
present our findings.
6.2 Empirical analysis

We begin by describing our various datasets.
6.2.1 Data
We utilise multiple Covid-related outcomes in our analysis. Covid19India.org
(C19I) makes available its data for further research and development through api.
Covid19india.org. C19I is a volunteer-driven and crowdsourced repository that
uses a number of publicly available data sources to cleanly archive raw counts of
total cases, new cases, mortality, testing patterns and other publicly available data
with time and geographic indicators (both state and district). They also report
a range of statistics such as the case fatality ratio, growth rate, etc. based on the
publicly available data. C19I constructs its raw counts from data shared by state
press bulletins released by different state governments, official Chief Minister and
Health Minister social media handles, Press Bureau of India, Press Trust of India
and ANI reports; they also reflect updates in their data based on official correc-
tions as and when they are documented. The data reported by the Ministry of
Health and Family Welfare, Government of India and C19I are broadly similar,
particularly for archival data. C19I’s data series provides more recent daily up-
dates countrywide due to its active volunteer network.2
We construct four distinct variables using the above dataset: the numbers in-
fected, deceased, active cases and the recovered cases as of 24 October 2020. All
of these are constructed at the district level.
Next, we turn to our conflict data. Like most other studies involving the topic
of Hindu-Muslim conflict in India, we turn to the Varshney–Wilkinson (2004)
dataset on Hindu-Muslim violence. The baseline Varshney–Wilkinson dataset
runs from 1950 to 1995. We use an extension of it for our analysis. Specifically,
we use data on conflict between 2014 and 2018. The dataset summarises re-
ports from The Times of India, a leading national newspaper, on Hindu-Muslim
conflicts in India. This dataset has information on deaths, injuries and arrests.
For every report of Hindu-Muslim violence, the dataset provides the date of
incidence of the riot, the name of the city/town/village, the district and state,
its duration, the number of people killed, injured and arrested, and the reported
proximate cause of the riot.
We use three different count measures from the dataset: the number of people
killed or injured (“casualties”), the number of people killed or the number of riot
outbreaks over the period. In all cases, we take aggregates over a five-year period
(2014–18) in each district. This is in the spirit of Mitra and Ray (2014, 2019).
The 68th round of the NSSO consumption expenditure survey which covers
101,662 households has been used in the construction of several key variables.
The survey was conducted over the period of a year, starting on July 2011 and
concluding in June 2012. Based on the data in this survey, we constructed several
district-level variables. Specifically, we created the average per capita expenditure
at the district level, the population measure, the literacy rate (also, the primary
education completion rate), the proportion of the district which is urban. In
addition, we created variables which have been shown to be strongly associated
with Hindu-Muslim violence – namely, the ratio of Muslim-to-Hindu per capita
expenditure, the proportion of Muslims and the degree of religious polarisation.
As mentioned earlier, we also control for the presence of the BJP in the district
in some of our regression models. Specifically, we use the share of Lok Sabha
seats won by the BJP in the district in the 2014 national elections. Data on the
Lok Sabha election outcomes are freely available on the website of the Election
Commission of India. The main idea is that BJP’s presence in the district may
affect Hindu-Muslim conflict (in fact, Mitra and Ray (2019)) present evidence of
this) and given that the central government is actually led by the BJP, the tackling
of the pandemic in the district. Given that the Lok Sabha is a national body and
given that the BJP is a national party with a distinct ideology, it makes sense to
see how the BJP MPs affect the climate for riots. It is of course possible to use
other measures – for example, the state assembly shares – but then there are two
distinct issues which one needs to account for:
i the party’s ideology/stance/policy could vary from state to state especially

given the caste and religious composition of the electorate. In other words,
a BJP MP and a BJP MLA could mean different things for riots and the war
on the Covid-19, and
ii state-level competition, by itself, can have an independent effect on riots
(à la Wilkinson (2004)). So, when using BJP’s assembly seat share, one could
be picking up some state-level competition effects. It might be difficult to
isolate a pure “BJP effect” from the state-level competition effect.
Therefore, keeping these considerations in mind, we proceed with this rather

conservative measure of BJP presence, namely, the share of Lok Sabha seats won
by the party in the constituencies forming the district.
6.2.2 Empirical specification

We are interested in the effects of prior levels of Hindu-Muslim violence in a
district on various Covid-19 health outcomes in the district. This motivates the
baseline OLS specification that we use:
Covid d = β0 + β1Conflict d + X d′γ + errord
where d indexes district and X is the vector of control variables at the district level.
We utilise different measures for our dependent variables, namely, the Covid-
related outcomes. Specifically, we use the numbers infected, deceased, active
cases and the recovered cases at the district level as of 24 October 2020.
Among the important variables on the right-hand side are, of course, the vari-
ous measures of Hindu-Muslim violence – casualties, killed and outbreak. We will
also pay special attention to the ratio of Muslim-to-Hindu per capita expenditures
and the district’s share of the BJP in the Lok Sabha: both have the potential to
affect Hindu-Muslim violence and the Covid-19 outcomes. Apart from these core
variables, population and some measure of Muslim presence are always included
as controls in every specification. Muslim “presence” is measured in two ways: we
use either the share of Muslim households in the district, or a measure of Hindu-
Muslim polarisation along the lines proposed by Esteban and Ray (1994) and
Montalvo and Reynal-Querol (2005).3 To be sure, in all the regressions we either
control for Muslim percentage or religious polarisation but never both simulta-
neously. The correlation between these two variables is very high (about 0.93).4
In some of the specifications, we include state dummies as controls. This is
to account for any state-specific factors which could influence the pandemic
outcomes. One factor which easily springs to mind is the rate of testing. We do
not have reliable data on the rate of testing across the various districts. However,
given that the state governments directed the core policy of testing in their re-
spective states, one hopes that the state fixed effects would serve to capture this
aspect in the regression models. In every regression specification, we cluster the
standard errors by state.
6.2.3 Results
We first report the regression results where the outcome is the number of con-
firmed cases (district wise) and where the main independent variable of interest
is Casualties.
Table 6.1 contains a collection of such regressions. In column 1, the specifica-
tion is very parsimonious – apart from the population measure of the district, no
other control variable is employed.
The coefficient on Casualties in this column is negative and statistically signif-
icant. In column 2, we introduce three additional controls: the literacy rate, the
extent of urbanisation and the average per capita expenditure (entered logarith-
mically). While the literacy rate exhibits no marked correlation with the number
TABLE 6.1 The effect of Hindu-Muslim conflict (Casualties) on Covid-19 confirmed cases: OLS
district-level regressions
Confirmed Cases [1] [2] [3] [4] [5]
Casualties −137.166** −169.052*** −147.878** −129.025** −118.090*

(51.625) (58.724) (51.747) (56.247) (58.156)
Population (log) 20,461.834*** 17,124.557*** 17,666.582*** 16,890.215*** 13,701.651**
(5,765.677) (4,562.823) (4,573.008) (4,737.012) (5,009.582)
Literacy 45.578 18.117 13.072 −47.651
(85.720) (85.978) (79.130) (47.361)
Urbanisation 28,172.584** 28,304.410** 30,770.580*** 47,156.375***
(10,369.395) (10,023.541) (9,013.279) (9,669.368)
Av. per capita exp. (log) 7,835.796* 8,043.071* 7,080.506 9,205.624
(4,050.989) (4,075.682) (4,126.182) (5,422.244)
Muslim/Hindu exp. ratio 1,313.919 1,384.984 671.702
(1,045.903) (1,145.038) (990.229)
Muslim % −162.302 −154.226 −84.659
(109.904) (109.696) (78.400)
BJP LS seat share −4,056.808 −1,861.770
(2,973.206) (1,895.176)
State dummies N N N N Y
Observations 457 457 457 457 457
Adjusted R 2 0.251 0.330 0.331 0.334 0.408
Sources and Notes: Extension of the Varshney–Wilkinson dataset on religious riots (2014–18), National Sample Survey
68th round, Covid19India.org (C19I) for Covid-19 data and the Election Commission of India for Lok Sabha out-
comes data. The dependent variable is the total number of confirmed Covid-19 cases at the district level. Casualties
(killed + injured) is the district-level aggregate over the five-year period, 2014–18. Robust standard errors clustered
by state in parentheses. *significant at 10%, **significant at 5%, ***significant at 1%.
of confirmed cases, both the urbanisation and the average per capita expenditure
variables display positive and statistically significant coefficients. This is hardly
surprising as most of the Covid-19 cases reported have been largely in urban cen-
tres which also tend to be richer than their rural counterparts. The coefficient on
Casualties in this column too is negative and statistically significant.
In column 3 of Table 6.1, we incorporate two more control variables which
potentially affect both Hindu-Muslim conflict and Covid-19 outcomes. These
are the Muslim-to-Hindu per capita expenditure ratio and the percentage of
Muslim households in the district. Neither of these two variables seem to have
any bearing upon the outcome variable. The coefficient on Casualties in this
column remains negative and statistically significant. In column 4 of this table,
we include the key political variable of our analysis – the district’s share of the
BJP in the Lok Sabha. Mitra and Ray (2019) find this variable to be salient in
determining the extent of Hindu-Muslim violence in the recent years. In terms
of confirmed cases of Covid-19 however, this measure of BJP presence does not
seem to have any statistically significant effect. The coefficient on Casualties in
this column too is negative and statistically significant.
In the last column of this table, we add the state fixed effects along with all
other variables which have been used in the preceding regressions. The inclusion
of the state dummies does not seem to change the basic pattern very much. The
coefficient on Casualties in this column remains negative and statistically signif-
icant. Moreover, the size of the coefficient does not vary excessively across these
different specifications. This is suggestive of a stable pattern between conflict (as
captured by Casualties) and Covid-related outcomes (as captured by the number
of confirmed cases).
When one uses either Killed or Outbreak as the measure of religious conflict –
in place of Casualties – the results are substantially unaltered. The negative as-
sociation between the measure of conflict and the number of confirmed cases
is exhibited here too. We omit presenting these results in the interest of space.5
Next, we use the number of deceased (due to Covid-19) as the dependent
variable. Table 6.2 contains a collection of such regressions. In columns 1–3, the
measure of Hindu-Muslim violence is Casualties. In column 4, we use Killed
in place of Casualties whereas Outbreak is used in column 5 as the indicator of
Hindu-Muslim violence.
In column 1 of Table 6.2, the coefficient on Casualties is negative and sta-
tistically significant – this suggests an increase in Casualties results in a fall
in the number of Covid-19 fatalities. The other control variables display the
same pattern as with Confirmed cases in Table 6.1; specifically, the urban-
isation variable is positive and statistically significant while the Muslim-to-
Hindu per capita expenditure ratio and the percentage of Muslim households
are not so. In column 2, we add the BJP share variable. This leaves the result
unaltered; the BJP variable appears to be uncorrelated with Covid-19 fatalities
while Hindu-Muslim casualties seem to have a negative relation. In column 3,
we incorporate the state dummies. This does not change either the sign or the
TABLE 6.2 The effect of Hindu-Muslim conflict (Casualties, Killed and Outbreak) on Covid-19
deaths: OLS district-level regressions
Deceased [1] [2] [3] [4] [5]
Casualties −2.379** −2.273** −2.497*

(0.951) (1.009) (1.277)
Killed −66.765***
(21.140)
Outbreak −22.100*
(12.492)
Population (log) 276.191** 271.817** 233.142** 226.170** 235.055**
(108.391) (110.630) (106.580) (102.705) (106.826)
Literacy 1.351 1.323 −0.920 −0.644 −0.641
(1.765) (1.786) (0.990) (0.971) (0.978)
Urbanisation 738.280*** 752.173*** 997.516*** 975.989*** 1,025.378***
(207.841) (200.776) (219.161) (215.011) (227.999)
Av. per capita exp. (log) 34.953 29.531 96.822 119.605 96.293
(57.239) (60.291) (88.261) (94.178) (85.221)
Muslim/Hindu exp. ratio 28.929 29.330 13.850 12.140 13.112
(23.288) (23.902) (21.700) (20.946) (21.919)
Muslim % −2.675 −2.630 −1.534 −1.520 −1.659
(1.908) (1.969) (1.463) (1.565) (1.481)
BJP LS seat share −22.853 −49.694 −52.667 −56.235
(40.869) (41.108) (41.682) (42.572)
State dummies N N Y Y Y
Observations 457 457 457 457 457

Adjusted R 2 0.273 0.272 0.383 0.384 0.386
Sources and Notes: Extension of the Varshney–Wilkinson dataset on religious riots (2014–18), National Sample
Survey 68th round, Covid19India.org (C19I) for Covid-19 data and the Election Commission of India for Lok
Sabha outcomes data. The dependent variable is the total number of reported Covid-19 deaths at the district
level. Casualties (killed + injured) is the district-level aggregate over the five-year period, 2014–18, Killed and
Outbreak are defined analogously. Robust standard errors clustered by state in parentheses. *significant at 10%,
**significant at 5%, ***significant at 1%.
magnitude (largely) of the coefficient on Casualties – it continues to be negative

and statistically significant.
In column 4, we replicate the regression specification of the one in the preced-
ing column with one single change – Killed replaces Casualties. The result is by
and large the same; the coefficient on Killed is negative and statistically signifi-
cant much like that on Casualties.6 In column 5, we utilise Outbreak in place of
Casualties or Killed. The rest of the control variables are kept the same (includ-
ing the use of state dummies). The result mirrors that for the case of Killed or
Casualties – the coefficient on Outbreak is negative and statistically significant
while that on urbanisation is positive and statistically significant.
On the whole, the results in Table 6.2 affirm the core message from the re-
sults in Table 6.1. There is a consistent pattern between a district’s past levels of
Hindu-Muslim violence and its record in terms of Covid-19 confirmed cases or

fatalities. Our results clearly suggest a negative association between these two
sets of variables.
We also examine the relation between our three indicators of Hindu-Muslim
conflict and the number of Covid-19 Active cases. The patterns are essentially
the same with one notable exception. The statistical significance of the coeffi-
cient on the urbanisation variable is much diminished.7
In fact, for all the three measures of Hindu-Muslim conflict, the coefficient
on the urbanisation variable is highly statistically significant only when the state
dummies are included in the regression specification.
We also have data on the number of recoveries from Covid-19 at the district
level. Using this information, we construct a variable which we call “Recovery
rate” in the following manner:
Recovery rate = Number of Recoveries/Number of confirmed cases.
We examine the effect of the three indicators of Hindu-Muslim conflict on
this variable. Even though we have observed the stable pattern of a negative
association between the other Covid-19 outcomes – namely, confirmed cases,
numbers deceased and active cases – there is no compelling reason to expect that
conditional on being infected, the recovery rate should be, ceteris paribus, any
different in districts prone to violence between Hindus and Muslims from those
districts where communal harmony abides. After all, the recovery rate depends
almost entirely upon physiological factors of those affected and not upon the
communal activities of those around the affected. Hence, one would not expect
to observe any relation of this variable with the area’s track record on communal
violence.
Our regression results confirm the above conjecture. The coefficients on the
three indicators of Hindu-Muslim conflict evince no convincing pattern what-
soever in the various specifications. In most of the cases, the coefficients on the
conflict indicators are not statistically significant.8 This is in marked contrast
to the pattern observed with the other Covid-19 outcomes, where the negative
association leaves us with much food for thought.
6.3 Concluding remarks

We have tried to forge a link between the performance of a typical Indian district
in terms of the severity of the outbreak of Covid-19 and contrast this with its
historical proclivities towards Hindu-Muslim violence. To gain a clear under-
standing of the theoretical underpinnings of such a potential link, we delved
into an examination of (i) the key factors which affect the dynamics of Hindu-
Muslim conflict and (ii) those factors which are impacted by the scourge of such
communal animosity. In particular, we discussed the idea of social capital being
both a determinant of inter-religious battles and also something which is trans-
formed (in myriad ways) by the fires of communal vengeance. As the coronavirus
pandemic continues to rage on, our knowledge of the factors that stoke it have
steadily increased. In particular, it is difficult to deny that the ambient level of

social capital plays a role in curtailing its spread. Thus, the social capital in an
area is a crucial factor for both the endurance of communal harmony and the
containment of the coronavirus pandemic. Hence, we focus our attention on this
critical link.
While Varshney (2003) puts forward the idea of how civic engagement may
prevent communal fault lines from emerging, we additionally consider how the
disruption of communal harmony may mould social capital in a certain way. To
elaborate, areas with higher social capital – and hence characterised by greater
civic engagement – are likely to experience fewer incidents of Hindu-Muslim
violence (à la Varshney (2003)). Given the nature of the coronavirus pandemic,
a successful programme of its curtailment requires mutual trust and cooperation
across and between various communities and also between the citizenry and the
public bodies. Therefore, higher social capital would facilitate the suppression of
the coronavirus pandemic. Combining these strands, one could make the argu-
ment that lower communal violence is associated with higher social capital and
thereby would lead to lower Covid-19 infections, deaths and the like.
However, there is another aspect to this religious conflict and social capi-
tal relationship which may be relevant for a country like India where residen-
tial segregation along religious lines is fairly pervasive. This relates to the point
made above about how the disruption of communal harmony may mould so-
cial capital in a particular manner. Areas which experience recurrent incidents
of Hindu-Muslim violence tend to enhance cooperation within each religious
community – this derives from a sense of identification with members of one’s
community along with a sense of alienation from the other community. Thus,
such areas may actually have greater information flows and cooperation within
Hindus and separately within Muslims. This may result in more coordinated
action – by each religious group – due to this “fractured” growth of social cap-
ital (within each opposing religious community). In addition, other factors may
further reinforce this – e.g., in conflict-prone areas one would naturally expect
more social distancing and limited social interaction certainly between commu-
nities and also within communities (it may not be safe to step out and there may
be limited community spaces for interaction). Thus, the severity of the coronavi-
rus pandemic may actually be lower in (religious) riot-prone areas.
In sum, the logical implications of social capital are ambiguous and in fact
yield opposing predictions in terms of the association between an area’s history
of Hindu-Muslim violence and its current performance in terms of Covid-19
outcomes. This observation makes it incumbent upon one to take up the empir-
ical investigation seriously to evaluate the relative strength of the two opposing
forces. In fact, this is precisely what we have done here.
We gathered data on various Covid-related outcomes – specifically, numbers
infected, deceased, number of active cases and recovered – at the district level and
then matched that to the district’s past levels of Hindu-Muslim conflict (between
2014 and 2018, hence a five-year period predating the outbreak of Covid-19).
Using a rich set of control variables, we examined the association between the
different Covid-related outcomes and the past conflict measures. Our empirical
investigations unearth a robust correlation between these two sets of variables.
Specifically, we find that higher levels of Hindu-Muslim conflict (2014–18) are
associated with a lower number of Covid-19 infections, deaths and active cases.
This pattern is fairly consistent and holds across a wide range of specifications.
On the whole, it appears that the second channel – the strengthening of so-
cial capital within each opposing community as a result of higher incidence of
Hindu-Muslim violence – seems to win out over the first one. While that may
indeed be the case, we only offer a cautious endorsement of this theory. Our
reasons for doing so are outlined below.
While our empirical analysis draws upon data for over 450 districts in India
spread over 20 major states, we cannot claim that the effects that we uncover are
strictly causal. Admittedly, we ensured that the independent variables were suf-
ficiently lagged to rule out any reverse causation issues. Nevertheless, given that
Hindu-Muslim conflict does not occur randomly across the various districts in
our sample, it is difficult to argue that the rich set of control variables we employ
is sufficient to rule out any omitted variables bias concerns. Hence, the patterns
while consistent and robust are to be interpreted as correlations and not proof of
causation.
Even if one overlooks the correlation versus causation issue, there is another
conceptual issue to confront. One ought to recognise that the channel of social
capital need not be the only one connecting Hindu-Muslim violence to the
various Covid-19 outcomes. There could be other factors which are related to
both communal violence and the spread of the pandemic and which have not
been covered by our set of control variables. While no such obvious confounder
springs to mind, one may not rule out this possibility.
In spite of the reservations listed above, our exercise does make a valuable
contribution to our understanding of the spread of the pandemic and its rela-
tion to Hindu-Muslim violence. This work, to the best of our knowledge, is
the first one to formally investigate such a link with detailed microeconomic
data. One needs to build on this exploratory empirical exercise to both sharpen
the empirical findings and also pinpoint the mechanism(s) behind the empirical
relationship. This would not only further our understanding of the dynamics of
Hindu-Muslim violence but also help devise appropriate policy measures which
can harness the spread of contagious (and pernicious) pandemics like the present
one.
Notes
1 We are grateful to Vrinda Anand and Debraj Ray for generously sharing their data
on Hindu-Muslim conflict (2014–2018) and on the BJP Lok Sabha outcomes (from
the 2014 Indian national elections). We also wish to thank Meenu Jothsna for her able
research assistance. All remaining errors are solely ours.
2 More about this repository is available at: https://www.covid19india.org/about.
3 The degree of religious polarisation for a region is defined by 4 ∑s 2

j (1 − s j ) for
j = H , M where H denote Hindus, M denote Muslims and s j denotes the population
share of j in the region.
4 In some areas, there are other dominant religious groups (like Sikhs in Punjab), so
that Muslim percentage and Hindu-Muslim polarisation measure different things.
But these cases are exceptions rather than the rule.
5 These results are available from the authors upon request.
6 To be sure, the magnitude is much larger for Killed, and the coefficient is significant
at the 1% level, but we refrain from speculating on the causes behind this. The core
message we wish to emphasise is that both measures (Killed and Casualties) yield
qualitatively similar results.
7 We omit presenting these results in the interest of space. These results are, however,
available from the authors upon request.
8 Even when they are, they are only marginally so and very dependent on the inclu-
sion/exclusion of certain control variables.
References
Bagchi, A. (1990), “Predatory Commercialization and Communalism in India,” in S.
Gopal (ed), Anatomy of a Confrontation, New Delhi: Penguin, 193–218.
Das, S. (2000), “The 1992 Calcutta Riots in Historical Continuum: A Relapse into
‘Communal Fury’,” Modern Asian Studies, 34, 301.
Engineer, A. (1984), “The Causes of Communal Riots in the Post-Partition Period in
India,” in A. Engineer (ed.), Communal Riots in Post-Independence India, Hyderabad:
Sangam Books, 33–41.
Engineer, A. (1987), “Meerut: the Nation’s Shame,” Economic and Political Weekly, 22, June
20, 969–973.
Engineer, A. (1994), “Communal Violence in Kanpur,” Economic and Political Weekly,
29(9), February 26, 473–474.
Esteban, J., and D. Ray. 1994. “On the Measurement of Polarization.” Econometrica 62,
819–851.
Heller, P. (2020) “A Virus, Social Democracy, and Dividends for Kerala,” The Hindu, 18
April.
Iyer, S., and A. Shrivastava (2018), “Religious Riots and Electoral Politics in India”, Jour-
nal of Development Economics, 131, March, pp. 104–122.
Jalan, J., and A. Sen (2020). “Containing a Pandemic with Public Actions & Public Trust:
The Kerala Story”, forthcoming, Indian Economic Review.
Jha, S. (2013), “Trade, Institutions and Ethnic Tolerance: Evidence from India,” American
Political Science Review, 107(4), November 2013.
Khan, D. (1992), “Meerut Riots: An Analysis,” in P. Kumar (ed.), Towards Understanding
Communalism, Chandigarh: Centre for Research in Rural and Industrial Develop-
ment, 465.
Klasnja, M., and N. Novta (2016). “Segregation, Polarization, and Ethnic Conflict”,
Journal of Conflict Resolution, 60(5), 927–955.
Miguel, E., Satyanath, S., and E. Sergenti (2004), “Economic Shocks and Civil Conflict:
An Instrumental Variables Approach,” Journal of Political Economy, 112, 725–753.
Mitra, A., and D. Ray (2014), “Implications of an Economic Theory of Conflict:
Hindu-Muslim Violence in India,” Journal of Political Economy, 122(4), 719–765.
Mitra, A., and D. Ray (2019), “Hindu-Muslim Violence in India: A Postscript from the
21st Century,” in J.-P. Carvalho, S. Iyer and J. Rubin (eds.), Advances in the Economics
of Religion. Volume 158, International Economic Association Series, Palgrave Macmillan,
229–248.
Montalvo, J. and M. Reynal-Querol (2005), “Ethnic Polarization, Potential Conflict,
and Civil Wars”. American Economic Review 95, 796–813.
Nellis, G., Weaver, S., and S. C. Rosenzweig. (2016), “Do Parties Matter for Ethnic
Violence? Evidence from India”. Quarterly Journal of Political Science, 11(3), 249–277.
Rajgopal, P. (1987), Communal Riots in India. New Delhi: Uppal Publishing House/Cen-
tre for Policy Research.
Tharoor, S. (2020) “The Kerala Model,” Project Syndicate, 11 May.
Upadhyaya, A. (1992), “Recent Trends in Communal Violence: A Case Study of Vara-
nasi,” Proceedings of the International Conference on Conflict and Change, Ulster/United
Nations University, Portrush, Northern Ireland.
Varshney, A. (2003), Ethnic Conflict and Civil Life: Hindus and Muslims in India. New Ha-
ven, CT: Yale University Press.
Varshney, A., and S. Wilkinson (2004), “Dataset on Hindu–Muslim Violence in India,
Version 2.” October 8, 2004.
Wilkinson, S. (2004), Votes and Violence: Electoral Competition and Ethnic Riots in India,
Cambridge: Cambridge University Press.
PART II
Macroeconomic issues
and environment
7
THE LONG SHADOW OF EPIDEMICS1
Shankha Chakraborty and Mausumi Das
7.1 Introduction
As of 28 February 2021, more than a year since the outbreak of the pandemic in
China, India has reported more than 11 million cumulative cases of Covid-19
and more than 157,000 deaths (Dong et al., 2020). With 17.7% of the world’s
population, that tally accounts for 9.8% of global cases and 6.2% of global deaths
from the pandemic. Aggressive policy measures have slowed the progress of the
disease as Figure 7.1 shows, though India’s peak has been significantly higher
than China and Brazil, two other large developing countries.
Relative to population, India’s 802.80 confirmed cases (11.37 deaths) per
100,000 is significantly better than Brazil’s 4,888.26 (118.32) and the USA’s
8,490 (152.46), but worse than China’s 6.92 (0.33).
The economic cost of the pandemic has been surprisingly high for India. The
National Statistical Office estimates that during April–June 2020, aggregate out-
put shrunk by almost a quarter compared to the same period in 2019 (Economist,
2020) and the annual growth is expected to drop to −3.2% from 4.2% last year
(World Bank, 2020). It has been argued that this high economic cost is due to
India’s aggressive lockdown strategies. A recent study finds that policy-induced
indirect causes have been more important than the direct causes – mortality,
morbidity, stressed healthcare system – of Covid-19, and the economic risk is
higher in South Asia and sub-Saharan Africa (Noy et al., 2020).
Yet there is room to question whether direct or indirect costs of the pandemic
are more relevant. Sheridan et al. (2020) find that in Scandinavian countries
where consumer spending fell sharply, lockdown measures were responsible for
only a small fraction of it.2 Similarly, Cantore et al. (2020) report that the aver-
age loss in industrial production in April 2020, relative to December 2019, was
similar across developing and developed countries: 22% in lower middle-income
DOI: 10.4324/9781003220145-9
110 Shankha Chakraborty and Mausumi Das
Daily
FIGURE 7.1 new cases of Covid-19: Brazil, China, India and the USA, from 22
January 2020 to 20 February 2021. Seven-day moving average.
Data: Johns Hopkins Coronavirus Resource Center (Dong et al., 2020).
Life Expectancy at Birth, 1950-2015 Life Expectancy at Age 45, 1950-2015

80 40
65 35
LE45
LE0
50 30
35 25
20 20
1950-1955 1960-1965 1970-1975 1980-1985 1990-1995 2000-2005 2010-2015 1950-1955 1960-1965 1970-1975 1980-1985 1990-1995 2000-2005 2010-2015
Year Year
India Brazil China US India Brazil China US
FIGURE 7.2 Life expectancies in Brazil, China, India and the USA.
Data: United Nations World Population Prospect, 2019.
countries, 24% in upper middle-income countries and 18% in high-income

countries. Some of the poorest regions, such as sub-Saharan Africa, have not
economically suffered much despite being affected by Covid-19 while other
countries are suffering even without a significant Covid-19 burden.
In light of this mixed evidence, we think it is important to step back and first
understand how and through which channels Covid-19 is expected to affect the
economy. We propose a simple dynamic model, inspired by Chakraborty and
Das (2019), to highlight some of these channels. The model is situated in the
broader context of developing country health problems, particularly mortality.
The developing world has made remarkable strides in saving lives since the 1950s
(Figure 7.2). India’s life expectancy at birth, for example, rose from a meagre
36.62 in 1950–55 to 67.57 by 2010–15, reducing a gap of 32.09 years relative to
the USA down to 11.13 years. Gains in other large developing countries such as
Brazil and China have been higher still.
The long shadow of epidemics 111
A large portion of these gains have come from averting child deaths. Less pro-
gress has been made towards averting premature deaths in working age and old age.
Life expectancy at age 45 in India, for example, has gone up from 20.17 to 29.73:
here a gap of 8.83 relative to the USA has gone down proportionately less, to 6.4.
India’s life expectancy at age 65 (not shown above) has gone from 9.8 to 14.36 over
this time period: a gap of 4.4 years relative to the USA has now increased to 5.04.
One factor that accounts for this slower catchup in adult life-expectancy is
a previously high infectious disease burden. Developing countries such as India
have a high incidence of circulatory, cardio-vascular and degenerative diseases
and mortality among its middle-aged population that is disproportionate to their
age structures. 35% of deaths among 35–64 year olds in India are due to cardio-
vascular diseases compared to 12% in the USA and 22% in China (Leeder et al.,
2003). A large body of literature has now emerged that connects this heightened
mortality risk to early-life exposure to infections. Health shocks such as epi-
demics in early life can have lifelong morbidity and mortality consequences (see
Aksan and Chakraborty, 2014, and Section 7.3 for some of the evidence) and
this connection informs our understanding of the short- and long-run effects of
Covid-19. Working in India’s favour are its younger demographic structure – the
median age is 28 – and low reported case fatalities from Covid-19. However,
exposure to the virus alone – one survey puts seroprevalence in Delhi at more
than 50% (Elegant, 2021) – and the large literature on the long-term effects of in-
fectious disease exposure suggest the effects may drag on for several generations.
In the model presented below, developing and advanced economies are pri-
marily distinguished by their adult mortality risk. We show that higher mortal-
ity induces agents to invest more in tangible assets (physical capital) instead of
intangible human capital. In developing countries, this provides some degree of
insurance against mortality shocks, including pandemic-related mortality effect.
At the same time, physical assets are typically more productive in informal and
manufacturing activities that are less flexible in adapting to a pandemic shock
either because they require repeated physical involvement with complementary
inputs and the marketplace or because remote work is less viable. This inflexibil-
ity amplifies the morbidity effect of the pandemic. Ex ante it is unclear whether
low-income, middle-income or high-income countries would pay a higher direct
cost from an epidemic shock; the net effect depends on other structural factors.
Section 7.2 below outlines the basic endogenous growth model with adult
mortality risk. Section 7.3 then studies the effect of higher mortality and morbid-
ity from an epidemic shock for current and future generations and how structural
factors – correlation of the shock with health infrastructure and sectoral compo-
sition of output – shape the economic costs.
7.2 Model
We consider an overlapping-generations (OLG) economy with a constant cohort
population of measure 1. People live for two periods, adulthood and middle
age, with survival from youth to middle age given by the exogenous probability
p ∈ (0,1). A young agent starts her economic life with an inherited wealth Wt,
received either as intended bequest (if parent is alive) or as accidental bequest (if
parent died prematurely). She does not work or consume during youth and in-
vests her entire inherited wealth in acquiring productive assets (capital and/or ed-
ucation) to be employed in the next period. Investing e units of goods generates
as many units of human capital/education in the next period through schooling.
Likewise, investing s units of goods generates as many units of physical capital in
the next period through savings.
Each agent is endowed with one unit of raw labour in the second period of her
life. In addition, she may acquire some education (e) and some physical capital
(s) depending on her schooling and savings decisions in the previous period. We
assume that there is no constraint on time and an agent can be engaged simulta-
neously in multiple occupations. We shall focus on two specific occupations: one
that combines physical capital with raw labour, and another that entails the use
of human capital alone.3 The first occupation generates a gross income Rs, s de-
noting the capital per unit of raw labour, while the second occupation generates
an income we, where w > R > 1.
Agents have preference over only second period consumption. An agent who
has survived to the middle age and in possession of one unit of raw labour, s
units of physical capital and e amount of education earns a second period income
we + Rs. She consumes part of her second period income, and part she allocates as
bequest to her single offspring. In case the agent dies before making it to the mid-
dle age, her entire human capital is wasted while the physical capital investment
is passed on to the child. The latter constitutes an unintended, or accidental,
bequest. The agent values both types of bequest; in other words, she values the
ability to pass something on to her child either way.
The agent’s decision problem is
max EU t = pc tσ+1 +  pbtσ+1 + (1 − p )( Rst )σ  , σ ∈ (0,1)

(7.1)
subject to
et + st = Wt (7.2)
bt+1 + c t+1 = wet + Rst (7.3)
et ≥ 0, st ≥ 0, bt+1 ≥ 0 (7.4)
where
 b , with probability p
t
Wt = 
 Rst −1, otherwise
and the last constraint follows from the no-borrowing assumption. More
succinctly
max EU t = p [w(Wt − st ) + Rst − bt +1 ]σ + pbtσ+1 + (1 − p )( Rst )σ .

{st ≥0, bt +1 ≥0}
This leads to the two first order conditions for st and bt+1:
∂ EU t
= − pσ (wet + Rst − bt +1 )σ −1 (w − R ) + (1 − p )σ ( Rst )σ −1 R ≤ 0, st ≥ 0
∂ st
∂ EU t σ −1
= − pσ (wet + Rst − bt +1 ) + pσbtσ+1−1 ≤ 0, bt +1 ≥ 0
∂ et
with their associated complementary slackness conditions. As long as both e and
b are positive, these imply
w
st = Wt (7.5)
w − R + φ( p)
 w 
et = 1 − Wt (7.6)
 w − R + φ ( p ) 
1
bt+1 = [wet + Rst ] (7.7)
2
where
 p(w − R ) 1/(1−σ )
φ ( p ) ≡ 2R  
 (1 − p )R 
is an increasing function of p. Now observe that et ≥ 0 as long as st ≤ Wt . This
condition is satisfied iff
w R
≤ 1⇔ p ≥ ≡ pˆ .
w − R + φ( p) R + 21−σ (w − R )
If this is satisfied, the ratio of human-to-physical capital for the agent is
ht+1 w − R + φ ( p ) R
= − 1 = λ( p ) (7.8)
kt+1 w w
 p / pˆ 1/(1−σ )
where λ( p ) ≡   − 1 is increasing in p: for increases in p above
 (1 − p ) / (1 − pˆ ) 
p̂, the agent invests disproportionately more in human capital. Intergenerational
wealth dynamics for p ≥ p̂ follows:
  w + wλ( p )  RWt
   , with prob. p
  w + Rλ( p )  2
Wt+1 = 
  w 
   RWt , otherwise.
  w + Rλ( p ) 
What if p < p̂ ? Here the agent is at a constrained optimum with the entire inher-
itance invested in capital:
st = Wt (7.9)
et = 0 (7.10)
RWt
bt+1 = (7.11)
2
and wealth evolves according to
 RW / 2, with prob. p
t
Wt +1 = 
 RW t, otherwise.
7.2.1 Production
A single final commodity is produced in two sectors: one sector uses capital and
unskilled/raw labour, the other uses only skilled labour, in line with the occupa-
tion choices people have. Technology in the physical capital-dependent sector is
represented by the following production function:
YtK = At K tδ L1t−δ , 0 < δ < 1, (7.12)
where K t and Lt represent aggregate physical capital and raw labour respectively.
The total factor productivity of this sector, At , depends on the economy’s aver-
age capital stock kt ≡ K t / N t (i.e., capital stock per unit of cohort population N t )
through knowledge spillovers such that
At = ρktβ , 0 < β < 1.

Under the specific assumption that δ + β = 1, and noting that the total cohort
population and aggregate raw labour are identically measured by N t = Lt = 1,4
the production function in the capital-dependent sector reduces to a simple AK
form:
YtK = ρ K t .
Technology in the human capital-dependent sector, on the other hand, is linear
in human capital:
YtH = ω H t . (7.13)
Aggregate output in the economy is the weighted average of outputs from the
two sectors
Yt = (1 − αt )( ρ K t ) + αt (ω H t ); ω , ρ > 0, (7.14)
where ρ and ω are the constant marginal products of capital and human capital
respectively, and αt ∈ (0,1) is the relative importance of human capital-dependent
activities in the economy. Assume for now that α is constant. The wage rate and
net interest rate are constant too, given respectively by
w = αω , (7.15)
r = (1 − α )ρ. (7.16)
We further assume that
α(ω + ρ ) > 1 + ρ. (A1)
Assumption (A1) ensures that w > R ≡ 1 + r for all t .
7.2.2 Balanced growth path

The dynamic equations for wealth accumulation allow us to characterize the
balanced growth path for this economy. Since factor returns are constant, for a
given p, the portfolio choice of assets, consumption and bequest decisions of a
household depend proportionally on inheritance Wt. Hence, the following equa-
tions describe the balanced growth path for the economy. For all p ∈ (0, pˆ ),
 b
Wt +1  R / 2 − 1 ≡ gW , with prob. p
gW ≡ −1= 
Wt a
 R − 1 ≡ gW , otherwise
and for p ∈  pˆ ,1) ,
 
  1 + λ( p)  R − 1 ≡ g b , with prob. p
  1 + R λ( p )  2 W
W w
gW ≡ t +1 − 1 = 
Wt  1 
a
 R  R − 1 ≡ gW , otherwise.
 
 1 + w λ ( p ) 

Consider now how these growth rates respond to p. Observe that for p ∈ (0, pˆ ,
b a
the growth rates gW and gW are independent of p. Moreover, for this range of p
b > a
values, gW < gW . For p ∈  p̂,1) , gW
b a
gW according to
<
1 + λ( p ) > >
1 ⇔ λ( p ) 1.
2 < <
Since λ( pˆ ) = 0, lim p→1 λ( p ) = ∞ and λ( p ) is a monotonically increasing function

of p, by the Mean Value theorem there exists a p ∈  p̂,1) such that λ( p ) = 1 and
gW
gWa (w.p. 1 − p)
R
R gWb (w.p. p)
2
p
0 p p̄ 1
FIGURE 7.3 Balanced growth rate.
b > a >
p = R / w . This means for p ∈  p̂,1) , gW gW according to p p . Further note
< <
a a
that for this range of p values, gW is decreasing while gW is increasing in p.
b
Based on these properties, Figure 7.3 illustrates the balanced growth rates gW
a
and gW for different values of p.
Applying the law of large numbers to the continuum of households in this
b a
economy, the growth rate is the weighted average of gW and gW ,
b a
gW = pgW ( p ) + (1 − p ) gW ( p)
where the economy is either in the p ∈ (0, pˆ ) regime or the p ∈  p̂,1) regime.
Observe that as we move from low p to high p, the average growth rate is
b a
affected not only because gW and gW are functions of p but also because the
a b
relative weight shifts from gW to gW . Indeed, for all p ≤ p , an improvement
in longevity is necessarily accompanied by a fall in the average growth rate of
the economy. Only a sharp rise in life expectancy (such that p moves above
p) provides a definitive positive boost to the economy’s growth rate. Such non-
monotonic relationship between life expectancy and per capita growth rate has
been reported in the literature, but existing explanations typically point towards
demographic transitions that accompany mortality reductions (e.g., Cervellati
and Sunde, 2011). The alternative explanation here is based on a mechanism
of self-insurance. In countries exhibiting high mortality, households self-insure
themselves by investing in assets that are less susceptible to mortality shocks. The
result provides a new interpretation of the existing empirical relationship.
There is another interesting implication of this result from cross-country
perspective. Note that p – the cut-off point at which the average growth rate
responds positively to an increase in longevity – depends on the economy’s wage-
rental ratio R / w . Thus, economies where the return to human capital is much
higher relative to that of the physical capital (due to differences in technology for
example) will experience this growth turn-around at a much lower level of mor-
tality reduction. In other words, in advanced countries, where R / w is relatively
low, one is likely to find a positive relationship between longevity improvement
and growth, while in less developing countries, where R / w is relatively high,
one would find the relationship between longevity improvement and growth to
be negative.
7.3 Short- and long-run effects of pandemic

A pandemic can potentially affect the macroeconomy through three distinct
channels. First, as morbidity goes up, there is a direct and immediate impact on
labour supply, reducing the effective stock of both unskilled labour and skilled
labour (human capital). Second, to the extent the pandemic leads to a tempo-
rary increase in late-life mortality, wealth dynamics and economic growth are
affected. Third, if the pandemic has a longer term effect on mortality (a fall in p
which persists beyond the current generational time), people would adjust their
investment portfolios from human capital to physical capital.
Yet, these effects do not operate in isolation. Their interaction with various
preventive measures undertaken to combat the pandemic (e.g., a lockdown pol-
icy) may further attenuate or accentuate these effects depending on pre-existing
economic and health conditions. Moreover, their impact would not be symmet-
ric across labour categories, sectors or economies where underlying economic
and health infrastructure differ.
In the discussion that follows, we use the endogenous growth framework
developed in the previous section to analyse the morbidity and mortality effects
of a Covid-19 like health shock. We first consider a benchmark case where rich
and poor countries are assumed to be similar in every respect except in terms of
the underlying mortality risks. We also ignore the role of any preventive policy
measure undertaken by the government and/or endogenous adjustment of job
location by the population in the face of the pandemic, which imply that the
Covid-19 health shock would affect the entire labour force symmetrically. This
benchmark case – albeit artificial – allows us to focus on the pure mortality/
morbidity effects of the pandemic. We then extend the analysis to more realistic
cases where these effects interact with various structural parameters that differ
across economies owing to differences in technology, production organization
and composition of the labour force.
7.3.1 Symmetric impact across countries

In the absence of any policy or conscious attempt at changing the work environ-
ment, the immediate impact of a pandemic would be on the working population
in terms of both morbidity and mortality. The first would affect aggregate out-
put in the short run, while the second would cause a deviation in the balanced
growth path.
7.3.1.1 Effect on aggregate output

To analyse the macroeconomic effects of the first, let us revisit the sectoral
production functions. Recall that the physical capital-dependent sector com-
bines physical capital and raw labour to produce an output according to (7.12).
Pandemic-related morbidity would reduce the effective labour supply Lt , but not
affect the average stock of capital kt . Indeed, the total cohort population is still
represented by N t = 1, but now Lt ≠ N t . Suppose the pandemic reduces the work
capacity of raw labour by a proportion (1 − γ) ∈ (0,1). Then Lt = γ N t and produc-
tion in the K -intensive sector becomes YtK = ρ( K t / N t )β K tδ (γ N t )1−δ = ργ 1−δ K t .
In the human capital-intensive sector, the pandemic is likely to reduce the pro-
ductivity of human capital as well. Let us assume that the productivity of human
capital falls by a proportion (1 − µ ). Hence, this sector’s output is now determined
by YtH = ωµ H t. Post-pandemic aggregate output in the economy is therefore
( )
Yt = (1 − α ) ργ 1−δ K t + α(ωµ H t ) (7.18)
and wage and net interest rates

w = αωµ , (7.19)
r = (1 − α )ργ 1−δ , (7.20)

respectively.
The pandemic predictably lowers aggregate output Yt . However, for the same
percentage decline in the efficiency units of raw labour and human capital (in
other words, for γ = µ ), the fall in output is greater in the H -intensive sector than
in the K -intensive sector.5 The intuition for this is simple. In the K -intensive
sector, the production technology is CRS in both its inputs – physical capital
and raw labour. Since the pandemic impacts only one of these inputs, its ef-
fect on the sectoral output gets dampened. In the H -intensive sector on the
other hand, the production technology is CRS in its single input – human cap-
ital. Hence, any pandemic-induced reduction in H reduces the sectoral output
equi-proportionately.
Second, the magnitude of impact on aggregate output in the short run depends
on α . The higher is α , the greater the importance of the human capital-intensive
sector in aggregate production and the greater the drop in output in the short
run. This effect will presumably be more pronounced in developed countries
which rely more heavily on human capital-intensive sectors. Concentration of

production in the human capital-intensive sector may well be an endogenous
outcome of higher longevity in the developed economies – a possibility we touch
upon towards the end of this section.
7.3.1.2 Effect on the growth rate

The pandemic affects not only aggregate output in the short run but also the
growth rate. There are two independent mechanisms at work here. One works
by lowering the factor returns – as derived in (7.19) and (7.20) – which directly
b a b
affects gW and gW . The other operates through a fall in the relative weight of gW
a
vis-à-vis gW as the mortality rate temporarily rises due to the pandemic.
b
We first analyse the direct effect of the pandemic on the growth rates gW and
a
gW working through the factor returns. As we have already seen, under the pan-
demic, the wage rate falls from w to w , while the net interest rate falls from r to
r .6 If this fall is temporary, affecting only the current working population (older
generation), then it will simply reduce the amount of bequeathed resources (both
intended and accidental) or wealth for one period, without inducing any change
in pattern of investment (e.g., moving away from one form of capital to another)
by the current young generation. If, on the other hand, the productivity loss as-
sociated with the pandemic is expected to persist, then it will alter the investment
decisions of the current young, which in turn will have a more lingering effect
on the balanced growth rates.
In the context of the Covid-19 pandemic, the last effect is arguably weak: at
the moment, it seems unlikely that the pandemic will continue unabated long
enough to cause a permanent reduction in productivity. In that scenario, the
b a
direct impact on the balanced growth rates will be minimal: gW and gW will dip
somewhat in period t as less wealth is passed on to the next generation, only to
go back to their previous levels in the subsequent periods. To see this point more
clearly, let us go back to the dynamic equation that characterizes the intergener-
ational evolution of wealth. Recall that the inherited wealth level of the younger
generation in period t is:
 1
 bt = (wt et −1 + Rt st −1 ) , with prob. p
Wt =  2
 Rs , otherwise.
 t t −1
The wealth level they will pass on to their progeny is given by the following
equations. For all p ∈ (0, pˆ ,
 R W
t +1 t
 , with prob. p
Wt+1 =  2
 R W, otherwise
 t+1 t
and for all p ∈ ( pˆ ,1),
  wt+1 + wt+1λ( p )  Rt+1

   W, with prob. p
  wt+1 + Rt+1λ( p )  2
Wt+1 = 
  wt+1 
   Rt+1Wt. otherwise
  wt+1 + Rt+1λ( p ) 
 p   αω − (1 − α )ρ − 1 1/(1−σ )
where λ( p ) ≡ 2    − 1.
 1 − p   1 + (1 − α )ρ 
We have used time-specific subscripts for factor incomes here to clearly demar-
cate the current w and R (which are impacted by the pandemic) from the next
periods w and R (which are unaffected). Indeed, the relevant and values for the
two time periods are as follows:
wt = w = αωγ ,
Rt = 1 + r = 1 + (1 − α )ργ 1−δ
while
wt+1 = w = αω ,
Rt+1 = 1 + r = 1 + (1 − α )ρ.
In terms of the dynamic equations, the pandemic therefore only affects Wt, leav-
ing other terms appearing on the right-hand side of the equations unchanged.
More specifically, the pandemic reduces the wealth level of the current genera-
tion as follows:
 1
t −1 { } t−1
 αωγ e + 1 + (1 − α )ργ 1−δ s  , with prob. p
2 
Wt = 
 {1 + (1 − α )ργ 1−δ } s , otherwise.
 t −1
To be sure, the reduction in Wt will not be the same for all p. For all p > p̂ , where
agents earn some skilled wage income in addition to the interest income, the
magnitude of the fall will be higher (since γ impacts w proportionately more than
R). This will result in a temporary dip in next period’s growth rate, the dip being
more pronounced at higher values of p.
Now consider the effect of mortality. If the fall in p is a temporary shock to
the currently old, short-run growth is affected through a temporary fall in the
b a
relative weight of gW vis-à-vis gW . Once again the magnitude of this effect will
differ across p. Suppose, for example, longevity temporarily falls to pτ < p . For
b a
all p < p , gW is lower than gW , which shifts the relative weight from a lower
growth value to a higher growth value; this would tend to soften the growth
impact of increased mortality. In contrast, the mortality effect of the pandemic
growth rate
b
gW
gW
gW
a
gW
FIGURE 7.4 Growth effect of mortality shock.
will be more pronounced in the p > p regime as in Figure 7.4 illustrated for
τ = 0.9 .7 Here, a temporary fall in longevity shifts the relative weight from a
higher growth value to a lower growth value gW and therefore would be accom-
τ
panied by a greater fall in the average growth rate, to gW .
What if the morbidity and mortality effects of the Covid-19 pandemic are
not temporary? Given advances in medicine and herd immunity, it is natural to
think the pandemic itself will wane before long and, however fatal the disease is
to the aged, it will be a thing of the past. Much evidence has accumulated, how-
ever, that short-duration epidemics in particular, and longer-duration early-life
exposure to infectious diseases more generally have morbidity consequences that
affect late-life health of current adults and children (Barker, 1994). For example,
some studies show that cohorts that enjoyed lower infectious disease mortality in
childhood also enjoyed lower cardio-vascular diseases and mortality later in life
(Crimmins and Finch, 2006). Data from Taiwan show elevated cardio-vascular
and mortality hazard in old age from higher malaria exposure early in life (Chang
et al., 2014). Conversely, about half of the longevity gains among 50- to 64-year-
old Union Army veterans in the USA may have been due to lower incidence of
infectious diseases in childhood (Costa, 2003). More well known in the econom-
ics literature is evidence on the foetal origins hypothesis. Males in utero who were
exposed to the 1981 Spanish flu epidemic had a 23% higher rate of heart disease
in their 60–80s in the USA (Mazumder et al., 2010) and higher cancer and heart
disease mortality after age 54 in Sweden (Helgertz and Bengtsson, 2019). Some
of these effects are picked up by aggregate statistics we see for India in Figure 7.2:
India’s past high infectious disease burden continues to depress survival gains for
cohorts 45 years and older.
It stands to reason that if the pandemic were to affect large swathes of the
Indian population, we can expect lingering effects on the growth trajectory.
If people are aware of the links between their future longevity and current expo-
sure to Covid-19, their investment should shift more towards physical assets like
capital. In that case, the growth effects of lower p could be substantive, especially
if the difference between the marginal products of the two forms of capital is
sizable.
Summing up, we find that when the productivity impact of the pandemic
is exogenous and symmetric across countries, poorer economies which already
exhibit a high (pre-pandemic) mortality rate are likely to have an advantage
vis-à-vis their richer counterparts. Since production in these economies depends
more on tangible inputs like land and physical capital than human capital, the
effect of the pandemic would be muted: loss of lives would not be accompanied
by a substantial loss of aggregate income or wealth. Hence, the adverse impact on
growth would be less pronounced. Needless to say, the growth impact would be
more significant if Covid-19 turns out to have long-term health effects that get
amplified through investment choices across generations.8
7.3.2 Asymmetric effect across countries

In the morbidity and mortality effects discussed above, the only factor that dif-
ferentiated countries was the mortality rate. We showed how, ceteris paribus,
countries with higher mortality (lower p) can have greater resilience against the
pandemic because of their greater reliance on capital-intensive production.
Poorer countries differ in other structural aspects too. In particular, poorer
countries have worse health infrastructure that may, controlling for age compo-
sition,9 raise mortality from pandemics. That, plus their higher population den-
sities, may necessitate stricter physical distancing policies that would compound
the productivity loss from the pandemic itself. Moreover, being dependent on
secondary-sector activities, they also have less scope to re-organize production.
Both these would counteract the advantages highlighted in the benchmark case.
The impact of these counterbalancing forces gets exacerbated due to a lower
share of the human capital in the aggregate production process. We consider
these possibilities here.
7.3.2.1 Human capital and production organization

In the benchmark case, we assumed that the morbidity impact of the pandemic
is symmetric for skilled and unskilled labour ( µ = γ ). Yet one crucial feature of
skilled labour (and the human capital-intensive sector) is its flexibility in terms
of production organization. While machine-based production requires physical
proximity to machines, a production process based on human capital can be
carried out (more easily) from a remote location. This implies that at the onset
of the pandemic, the skilled labour force would easily adjust their location and
work from home, while the same option would not be available to the unskilled
workers. Hence, skilled workers are less likely to be infected by the disease than
the unskilled worker. Once we adjust for this organizational flexibility in the
H -intensive sector, the associated productivity loss in this sector would actually
be lower. In other words, µ > γ .
The relative impact of the pandemic on the sectoral outputs is now ambigu-
ous. While the sector-specific outputs are still given by
YtK = ργ 1−δ K t
and
YtH = ωµ H t ,
the productivity losses across the two sectors are no longer directly comparable:
both γ 1−δ and µ are now greater than γ . However, flexibility in terms of remote
working means that the loss of productivity in the H -intensive sector would be
minimal. Hence, it seems reasonable to assume that 0 < γ < γ 1−δ < µ < 1. As be-
fore, post-pandemic aggregate output and the corresponding factor returns are
given respectively by
( )
Yt = (1 − α ) ργ 1−δ K t + α(ωµ H t ),
w = αωµ ,
r = (1 − α )ργ 1−δ .
In contrast to the previous case, it is the K -intensive sector that is hit harder by
the pandemic now. The intensity of the impact will again depend on α . The
higher is α , the lower the drop in aggregate output in the short run. Developed
countries where production is concentrated in the human capital-intensive sec-
tors will therefore be less affected by the pandemic now. A similar conclusion
follows when we consider the effect on the growth rate. For illustrative purposes,
focus on the p ∈  p̂,1) regime where human capital accumulation is positive and
post-pandemic wealth dynamics is determined by
   [1 + (1 − α )ρ ]
 αω + αωλ( p )
  Wt with prob. p
  αω + [1 + (1 − α)ρ]λ( p )  2
Wt+1 = 
  αω 
   [1 + (1 − α )ρ ]Wt otherwise.
  αω + [1 + (1 − α )ρ ]λ( p ) 
Recall that the pandemic-specific parameters µ and γ affect wealth dynamics

only through Wt where
 1
t −1 { } t−1
 αωµ e + 1 + (1 − α )ργ 1−δ s  , with prob. p
2 
Wt = 
 {1 + (1 − α )ργ 1−δ } s , otherwise.
 t −1
Since µ > γ 1−δ , the pandemic-related fall is less in the H -sector. Working popu-
lation in higher p countries will have higher share of their income coming from
human capital; hence, the fall in current income and the concomitant fall in
wealth will be less in these countries. Growth dips down less.
7.3.2.2 Mortality, productivity loss and sectoral composition of output

In our analyses so far, we have taken the sectoral share of output (α ) and the
productivity impacts of the pandemic ( µ and γ ) as exogenous, although we have
indicated that they differ across high vis-à-vis low-mortality countries. In this
section, we make these functional relationships explicit.
The correlation between the mortality rate and the productivity loss from
the pandemic is apparent. In poorer economies, public health facilities are often
dysfunctional, which is indeed the predominant cause of high mortality in these
countries. Restricted availability of public health facilities also imply that people
affected by the disease (especially in milder form) would often go untreated,
thereby increasing its infectivity. Hence, morbidity would go up, lowering pro-
ductivity levels in either sector. This indicates that µ and γ are both endogenous
and increasing in p.
Recall from our previous discussion that µ and γ affect the growth rate in
the short run by reducing the current wealth level Wt. Let the joint effect of µ
and γ on Wt be captured by θ ( p ) ∈ (0,1) such that θ ′( p ) > 0 . The corresponding
post-pandemic growth rates in the p ∈ ( pˆ ,1) regime would then be given by
   1 + (1 − α )ρ 
 αω + αωλ( p ) b
  θ ( p ) − 1 ≡ gW , with prob. p
  αω + [1 + (1 − α ) ρ ] λ ( p )  2 
W
gW ≡ t +1 − 1 = 
Wt   αω 
   [1 + (1 − α )ρ ] θ ( p ) − 1 ≡ gW
a
, otherwise.
  αω + [1 + (1 − α )ρ ] λ( p ) 
For expositional convenience, suppose θ ( p ) = pε, ε > 0 . As we now move from

b a
low p to high p, the growth rates gW and gW are affected not only through
investment choices but also through θ ( p ), the latter capturing the effect of the
pandemic. These growth rates and the corresponding average growth rate for
various values of p are depicted in Figure 7.5.
Comparing the growth rate for a positive ε case with one where there is no
productivity effect of the pandemic (ε = 0 ), we conclude that when the produc-
tivity loss due to pandemic is correlated with the country-specific mortality rate,
the growth impact on poorer/high mortality countries will be higher.
As a final exercise, we entertain the possibility that sectoral composition of
aggregate output differs systematically between rich and poor countries. More
specifically, richer countries have lower mortality rates and, consequently, have
higher α . This correlation could be the outcome of a skill bias in technological
growth rate
g̃ b
W
g W (ϵ=0)
g W (ϵ=0.3)
g̃ a
W
FIGURE 7.5 Asymmetric effect of morbidity shock.
change whereby skill formation increases the productivity of skilled labour more
than other production inputs. Our model can deliver such an outcome with an
additional assumption.
Suppose we restrict to p ≥ p̂ . Then we know for each household the ratio
of human to physical capital investment is given by (7.8). Since households are
identical and of unit measure, we have H t / K t = Rλ( p ) / w for all t since factor
prices are constant. Now suppose the world technology frontier entails a certain
αFt while for any country, its relevant α is a function of how much of the frontier
technology it absorbs. The absorption rate is an increasing function of its H / K
ratio á la Nelson and Phelps (1966):
αt = αFt ϕ ( H t / K t ) ≤ αFt
where ϕ ′ > 0, ϕ (0) = 0, ϕ (∞) = 1. A candidate function is ϕ ( x ) = x / (1 + x ). In a

balanced growth path, if αFt is constant we have
αF
α= ,
1 + w / {Rλ( p )}
an increasing function of p via λ( p ).

A full-fledged analysis of this relationship and its implication for the growth
rates would entail solving a fixed-point problem whereby the factor returns
(w and R) depend on α and the value of α in turn depends on w / R . Instead of
pursuing this line, for brevity consider the effect of two exogenous α values: one
relevant for the rich country, another for the poor. In Figure 7.6, R corresponds
to α = 0.75 and P to α = 0.5: the richer country is assumed to be at higher values
of p and α .10
Our specific interest here lies in exploring the impact of a pandemic on P (low
p -low α ) vis-à-vis R (high p -high α ). For this purpose, we interact α with the
growth rate
gR
W (ϵ=0)
gR
W (ϵ=0.3)
gP
W (ϵ=0)
gP
W (ϵ=0.3)
FIGURE 7.6 Morbidity effects when sectoral output composition differs.
pandemic-related morbidity effect θ ( p ). Comparing the growth rates with mor-

bidity effect (ε = 0.3) and without (ε = 0.3), we find that not only is the average
growth rate much lower in P compared to R, but the decline in growth due to
the morbidity effect is also sharper. From this, we can conclude that the adverse
growth impact of the pandemic gets compounded in the poorer countries due to
lower share of human capital in the production process.
The main take-away from the discussion in this section is that while poorer
countries have greater resilience to a pandemic due to their inbuilt self-insurance
mechanism in terms of greater reliance on physical capital, the same also under-
mines their ability to reorganize production by switching to remote work. More-
over, their low share of human capital impedes adoption of suitable technologies/
occupations that facilitates working from home, which further slows their recovery.
7.4 Conclusion
Conventional wisdom suggests that poorer countries will be naturally more sus-
ceptible to global health shocks like the Covid-19 pandemic. This chapter ex-
plains why that may not be necessarily true. Poor countries with high mortality
also invest more in physical capital which provides some insurance against such
health shocks. In contrast to India’s projected growth contraction, neighbouring
Bangladesh for example is expected to grow at 1% next year compared to 1.6%
last year (World Bank, 2020). Yet, effectiveness of this insurance mechanism is
limited, especially in the face of structural bottlenecks like weak health infra-
structure and a large population of unskilled labour force, not adaptable to new
forms of production organization. A comparison between India and Bangladesh
is truly illustrative here. Over the last two decades, Bangladesh has achieved
remarkable improvements in many social development indicators, notably in

health and education, while India has by and large ignored its social sector de-
spite decades of high growth. The mechanisms highlighted in this chapter point
to a robust social sector as a complementary input for sustained growth because
of its ability to absorb unexpected macroeconomic shocks such as Covid-19.
Notes
1 Thanks to Saurabh Gupta for expert research assistance and an anonymous reviewer
for helpful feedback. Mausumi Das acknowledges financial assistance from the Bill
and Melinda Gates Foundation.
2 For example, Sweden pursued a hands-off, herd immunity-based strategy while Nor-
way went for a comprehensive lockdown. In the second quarter of 2020, Swedish
GDP fell by 8.3% while Norway’s fell by 5.1%. Sweden paid an additional cost: its
Covid-19 death rates per 100,000 were 11 times that of Norway’s (Barry, 2020).
3 To fix ideas, one could think of the two occupations as running a printing press
(which requires running a machine using physical labour), and simultaneously writ-
ing a book (which requires education and knowledge).
4 If we take into account cohort-level mortality, then the measure of working popula-
tion and amount of raw labour will be given by N t = Lt = p . However, since the ex-
ponent of the L term in the aggregate production function reduces to zero, it makes
no difference to the results.
5 Note that YtK decreases by a proportion 1 − γ 1−δ < 1 − γ , while YtH decreases by a pro-
portion 1 − µ = 1 − γ .
6 We continue to assume that γ = µ , i.e., impact of the pandemic on the two types of
labour (unskilled and skilled) is symmetric.
7 This and subsequent figures use the numerical values σ = 0.5, α = 0.5, δ = 0.3, ω = 25, ρ = 3
0.5, δ = 0.3, ω = 25, ρ = 3 unless noted otherwise.
8 See Michl and Tavini (2020) for a model where a temporary Covid-19 recession
inflicts permanent damage by lowering the rate of technological change. In contrast,
long-term effects in our model operate through behavioural responses in generational
time.
9 Philip et al. (2020) attribute India’s low case fatality from Covid-19 – 2.2–3% versus
about 4% for the world – to the younger age groups that have been stricken by the
disease so far. Other factors such as under-reporting, viral load and prior immunity
may be also at play (Mukherjee, 2021).
10 Observe that the same decrease in p in each country has different effects. In P , it can
even raise the growth rate while in R, growth falls for sure. This reaffirms the results
derived in Section 7.2, where we have shown that the effect of mortality reduction
on the average growth rate is non-monotonic due to a mechanism of self-insurance
operative in countries with high mortality.
References
Aksan, A-M. and S. Chakraborty (2014), “Mortality versus Morbidity in the Demo-
graphic Transition,” European Economic Review, 70: 470–492.
Barker, D.J.P. (1994), Mothers, Babies, and Disease in Later Life, London: British Medical
Journal Publishing Group.
Barry, J.M. (2020), “What Fans of ‘Herd Immunity’ Don’t Tell You,” New York Times,
Oct. 19, 2020.
Cantore, N., F. Hartwich, A. Lavopa, K. Haverkamp, A. Laplane, and N. Rodousakis

(2020), “Coronavirus: The Economic Impact,” Policy Brief, UNIDO, 10 July 2020.
Cervellati, M. and U. Sunde (2011), “Disease and Development: The Role of Life Expec-
tancy Reconsidered,” Economics Letters, 113(3): 269–272.
Chakraborty, S. and M. Das (2019), “Death and Capital: Investment with and without
Annuities”, Macroeconomic Dynamics, forthcoming.
Chang, S., B. Fleisher, S. Kim, and S. Liu (2014), “Long-term Effects of Early Childhood
Malaria Exposure on Education and Health: Evidence from Colonial Taiwan”, Eco-
nomic Development and Cultural Change, 62(3): 519–536.
Costa, D.L. (2003), “Understanding Mid-Life and Older Age Mortality Declines: Evi-
dence from Union Army Veterans,” Journal of Econometrics, 112(1): 175–192.
Crimmins, E.M. and C.E. Finch (2006), “Infection, Inflammation, Height, and Longev-
ity”, Proceedings of the National Academy of Sciences, 103: 498–503.
Dong, E., H. Du and L. Gardner (2020), “An Interactive Web-based Dashboard to Track
COVID-19 in Real Time”, The Lancet Infection Diseases, 20(5): P533–P534, May 01,
2020. DOI: https://doi.org/10.1016/S1473-3099(20)30120-1
Elegant, N.X. (2021), “Survey Finds 56% of Delhi Has COVID Antibodies, Adding to
India’s Declining Cases Mystery,” Fortune, Feb. 2. https://fortune.com/2021/02/03/
delhi-covid-antibodies-india-declining-cases-herd-immunity/
The Economist (2020), “India’s Economy Shrinks by a Quarter as Covid-19 Gathers Pace,”
Sept 3. https://www.economist.com/asia/2020/09/03/indias-economy-shrinks-by-a-
quarter-as-covid-19-gathers-pace
Helgertz, J. and T. Bengtsson (2019), “The Long-Lasting Influenza: The Impact of Fetal
Stress during the 1918 Influenza Pandemic on Socioeconomic Attainment and Health
in Sweden, 1968–2012,” Demography, 56: 389–1425.
Leeder, S., S. Raymond and H. Greenberg (2003), “A Race Against Time: The Chal-
lenge of Cardiovascular Disease in Developing Economies,” Center for Global Health
and Economic Development, Columbia University and University of Sydney.
Mazumder, B., D. Almond, K. Parka, E. M. Crimmins and C. E. Finch (2010), “Lin-
gering Prenatal Effects of the 1918 Influenza Pandemic on Cardiovascular Disease”,
Journal of Developmental Origins of Health and Disease, 1(1): 26–34.
Michl, T. and D. Tavani (2020), “Path Dependence and Pandemics in a Classical Growth
Model,” PERI working paper, University of Massachusetts – Amherst.
Mukherjee, S. (2021), “Why Does the Pandemic Seem to Be Hitting Some Countries
Harder Than Others?” The New Yorker, March 1 2021 issue.
Nelson, R., and E. Phelps (1966), “Investment in Humans, Technological Diffusion, and
Economic Growth,” American Economic Review, 56(1/2): 69–75.
Noy, I., N. Doan, B. Ferrarini and D. Park (2020), “The Economic Risk of COVID-19
in Developing Countries: Where Is It Highest?” in S. Djankov and U. Panizza (eds.)
COVID-19 in Developing Economies, CEPR.
Philip, M., D. Ray and S. Subramanian (2020), “Decoding India’s Low Covid-19 Case
Fatality rate,” NBER Working Paper No. 27696.
Sheridan, A., A.L. Andersen, E.T. Hansen, Niels Johannesen (2020), “Social Distancing
Laws Cause Only Small Losses of Economic Activity during the COVID-19 Pan-
demic in Scandinavia,” Proceedings of the National Academy of Sciences, 117(34): 20468–
20473. DOI: 10.1073/pnas.2010068117
World Bank (2020), World Economic Prospects, June 2020, Washington, DC.
8
THE MACROECONOMIC IMPACT OF
COVID-19 ON THE INDIAN ECONOMY
An exploratory framework
Ajitava Raychaudhuri
8.1 Introduction
The discussion surrounding Covid-19 has been centred on two major
d imensions – the impact on life and the subsequent displacement in liveli-
hood. The second issue is in the natural domain of the economists although
the first dimension has also attracted discussion on health economics factors.
Although a number of articles have been written on the economic impact of
the pandemic, most of them have focused on the fiscal and monetary stimu-
lus packages undertaken by governments across the continents (Dasgupta and
Rajeev, 2020; Forlano and Wolf, 2020) and one of the papers has highlighted
the role of containment policy in saving lives although it increases the recession-
ary trends (Eichenbaum, Rebelo and Trabandt, 2020). This naturally touches
on macroeconomic dimensions of such packages, although the discussions are
somewhat based on standard notions of policies needed to counter short-term
adverse economic shocks. Another series of articles have tried to simulate the
effects of Covid-19 on changing the trajectory of growth of nations (McKibbin
and Fernando, 2020; Mishra, 2020; World Bank, 2020). The present chapter
is an effort to capture and extend some of these arguments in terms of simple
macroeconomic models, both static and dynamic, so as to address the short-run
and long-run issues. The short-run issues are more in the nature of demand-
and supply-side managements while the long-run issues are in terms of growth
dynamics. Section 8.2 gives an overview of the Indian economy post Covid-19.
Section 8.3 and its sub-sections discuss the short-run issues and Section 8.4 has
highlighted the long-run issues. The supply side shock is dealt in Section 8.5,
while Section 8.6 has the concluding remarks.
DOI: 10.4324/9781003220145-10
130 Ajitava Raychaudhuri
8.2 State of the Indian economy post Covid-19

The short-run problem post Covid-19 is a classic example of short-run man-
agement of demand and supply to ensure output, employment, prices, interest
rate and exchange rate. Indian economy already faced an acute problem of un-
intended accumulation of inventory in several sectors, especially consumer du-
rables, and shortage of demand in some related services (like transportation and
travel). The real GDP at market prices with 2011–12 as the base year grew at a
rate of 4.2% in 2019–20 and the nominal GDP grew at a rate of 7.2% compared
to corresponding growth rates of 6.1% and 11.1% in 2018–19 (Provisional esti-
mate, GOI, June 2020). The budget of 2020–21 had projected a rate of growth
of nominal GDP of 10% (Budget 2020–21, GOI, 2020). The main reason for bad
performance in 2019–20 was a meagre growth of real GDP in the last quarter
of 2019–20 at a rate of 3.1% partly due to Covid but more due to an ongoing
trend of slowing down of growth in the economy (Monthly Economic Reports,
GOI, March–April 2020). The projected growth in 2020–21 is directly affected
by Covid and the first quarter growth in the current fiscal year has a significant
negative impact on the GDP. The estimates of this impact vary: the National Sta-
tistical Office (NSO) of the Government of India calculated the decline in first
quarter of real GDP (on a year-to-year basis) as 23.9%, which is one of the high-
est among the major economies (NSO (2020c), MOSPI, Government of India,
August 31, 2020), the World Bank estimated a decline of GDP for the financial
year 2020–21 of 3.2% (Global Economic Prospects, World Bank, June 2020), the
SBI estimated a decline of 16.5% in the first quarter of real GDP (SBI Ecowrap,
August 17, 2020), the IMF projected a 6.4% decline of real GDP for 2020–21
(World Economic Outlook, IMF, June 2020) while Fitch Rating Agency has
projected a 10.3% fall in GDP in 2020–21 – in September 2020 downgrading its
earlier estimate of a fall of 5% made in June 2020 (Economic Times, September
9, 2020). However, according to the Economic Survey of India for 2020–21, the
gross national income in constant 2011–12 prices grew by 4.4% in 2019–20 (pro-
visional estimate) and will fall by −7.9% in 2020–21 (advanced estimate). As per
the same document, the expected growth of real GDP in 2021–22 is 11%. With
World Bank projecting a 5.2% decline in world GDP in 2020 (Global Economic
Prospects, World Bank, June 2020), the rest of the world also provides no great
hope for the external sector pulling up the Indian economy in 2020–21 (Mishra,
2020).
Now the decline in the GDP growth in the short run may be due to falling
aggregate demand (AD) or falling aggregate supply (AS) or both. If one looks
at the components of AD, the most significant slowdown in 2019–20 in current
prices (with 2011–12 base), as revealed by the data published by the National Sta-
tistical Organisation (NSO, 2020a) in its National Accounts Statistics (NAS) of
2020, is in private final consumption expenditure (growth of 8.96% in 2019–20
compared to 11.52% in 2018–19). Similarly, some of the declines are in gross fixed
capital formation (−0.39% change in 2019–20 compared to 14.96% in 2018–19),
The macroeconomic impact on India 131
exports (−0.55% change compared to 17.27% in the same periods) and imports
(−3.79% change compared to 19.11% in the same periods). The government final
consumption has risen in percentage terms by 16.27% in 2019–20 compared to
14.48% in 2018–19. Most worrisome is that on a quarterly basis, the data reveal
a decline in private final consumption expenditure of 26.68% in the first quarter
of 2020–21 (NSO (2020c), MOSPI, Government of India, August 31, 2020),
although the Economic Survey of 2020–21 mentions the final decline in pri-
vate consumption to be much less for 2020–21 at −0.6% (Ministry of Finance,
GOI, Economic Survey 2020–21 Volume II, p.43). Clearly, the decline in private
consumption expenditure is a major factor since it is about 60% of GDP. As the
Monetary Policy Committee of RBI opined, consumer confidence has no sign
of improvement (RBI Bulletin, August 2020, p.10). One needs some deep in-
trospection about the falling growth in the private final consumer expenditure.
Further, the decline of imports was faster than the decline in exports even in
2019–20, which shows the declining demand for imported intermediate and final
goods in the economy. As per Economic Survey of 2020–21, in 2020–21 fiscal,
India recorded a current account surplus of 3.1% of GDP. In fact in the fiscal
year 2020–21, during April–December 2020, equities witnessed inflow of FII at
USD 30.0 billion, which is five times its previous year value – India was the only
country among emerging markets to receive equity FII inflows in 2020. This
shows that this capital inflow could provide much needed resources for India to
raise capital expenditure which can both boost up demand through multiplier
and boost up supply in the long run through a rise in supply.
Further, the non-food credit going to the corporate sector also shows a slow
growth in 2019–20 of 6.1% compared to 13.4% in 2018–19 (RBI Bulletin, January
and August 2020), which shows the reluctance of taking credit on the part of risk-
averse investors and hesitancy of extending credit on the part of over-cautious
banks. The situation worsened in the first quarter of 2020–21. Thus, private in-
vestment also slowed down considerably. So unambiguously falling AD appears
to be one of the root causes for the short-run demand-supply gap (Mishra, 2020,
p. 4). Even the government tax incentives announced sometime in January 2020
through major cuts in corporation tax worth C1.45 lakh crores, C50,000 duty re-
fund scheme for exporters along with sops for telecom and housing sectors could
not boost investors’ sentiment. As per budget speech of the Finance Minister of
India (Ministry of Finance, Government of India, 2021, p. 1) on February 1,
2021, the fiscal stimulus immediately after lockdown of the Indian economy in
March 2020 was C2.76 crores (about 1.2% of the GDP). Later on, a combina-
tion of fiscal and monetary policies to boost investment and consumption mainly
through cheaper guaranteed credit under Atmanirbhar Bharat package in three
tranches was C27.1 lakh crores amounting to 13% of GDP. But the direct stimulus
through transfer of money to beneficiaries was much less – economic survey of
Government of India for 2020–21 puts it around C69,000 crores (Economic Sur-
vey 2020–21, Volume 1, pp. 33–34). This aspect forms one of the cornerstones of
the theoretical conjectures discussed later in the chapter.
The shock on the supply side came from massive reverse migration of la-
bour force witnessed in the economy after lockdown was announced in March
2020. This suddenly created a void in the labour market which possibly increased
transaction costs for firms when unlocking of the economy slowly started from
May onwards (Dasgupta and Rajeev, 2020). In fact as per Census 2011 about
41 million workers moved from rural to urban areas (Singh et al., 2020) and
in the month of May 2020 alone, taking special trains, about 55 lakh workers
returned to their home states, facing tremendous uncertainty to their life and
livelihood (Singh et al. 2020, p. 26). The other type of supply shock came from
the intermediate inputs like parts and components due to breakdown of supply
chain especially from China which controls around 22% of global industry out-
put (OECD, March 2020). OECD particularly mentions disruption in supply
of intermediate inputs in computers, electronics, pharmaceuticals and transport
equipment (OECD, March 2020, p. 5). So this clearly affects the AS adversely as
a temporary negative external shock.
In any event, such simultaneous adverse demand and supply shocks not only
affect GDP in the short run but also affect employment. The Periodic Labour
Force Survey (PLFS from NSO (2020b)) annual report of 2018–19 shows the
percentage of informal workers in India stood at 68.4% while those who are in
regular wage/salaried employment, around 67.8% of them had no written con-
tract and 55.9% had no social security benefit (PLFS, 2018–19, Section 3.4). This
clearly shows the vulnerability and uncertainty in the life of majority of workers
in India. This plays a major role in the current situation of post Covid-19 as will
be explained in our theoretical framework later.
In this context, one needs to mention the importance of medium, small and
micro enterprises (MSME) in India. As the National Sample Survey of 73rd
round had revealed, the MSME sector generated 110.98 million employment in
2015–16 which is roughly 40% of total workforce participation in the economy
(Annual Report 2018–19, Ministry of MSME, GOI, 2020). The MSME sector
comprise of both manufacturing and services, but one must not forget that the
manufacturing sector also generates employment for many services which de-
pend on the manufacturing sector, for example, transport, logistics, retail trade,
finance, accounting, repair and maintenance, etc. In the budget speech in Febru-
ary 2021, the Finance Minister of India reiterates the productivity-linked incen-
tive scheme (PLIS) for ten important industries for the next five years to the tune
of C1.97 lakh crores which will help the MSME sector to get more integrated
to the global value chain (Budget speech 2021, Government of India, p. 7). The
government has also raised the upper limit of the MSME sector in terms of turn-
over so that they may attain the desired efficiency level for this purpose.
An equally disturbing trend observed in the Indian economy in the last few
years is the rising income inequality as well as wealth inequality. Since income
inequality could not be calculated in the early years for the lack of data from
income tax authorities as well as very limited coverage of income tax, inequality
in India is historically calculated from household consumer expenditure surveys
carried out by National Sample Surveys. The data show unambiguously a trend
towards rising inequality in personal income and expenditure in India. As sum-
marized by one author, consumption gini coefficient had increased from 0.3
in 1993–94 to 0.359 in 2011–12. Similarly, income gini had risen from 0.548
in 2004–05 to 0.553 in 2011–12 (Dev, January 2018, p. 10). As per the World
Inequality Report (WIR), “During the 2000s (In India from 2000–2014, italics
mine), the annual real income growth of the top 1% was close to 8.5%, followed
by the top 10% at around 7% and the bottom 50% at less than 2.5% (WIR, 2018,
p. 128)”. This trend is arguably accelerated after the appearance of Covid-19 pan-
demic. So one needs to factor in this rising unequal distribution in a theoretical
framework to judge the efficacy of fiscal stimulus to tackle Covid-19 situation.
8.3 A
theoretical framework for short-run
demand-supply management
The above description clearly points out to several areas where institutional in-
terventions may be productive to manage short-run demand-supply imbalance.
This may follow a typical Keynesian prescription of “pump priming” since it has
got some similarity with the Great Depression of 1929, but there is an important
difference. This time a health pandemic with a great deal of uncertainty is an
additional dimension. So clearly the degree of uncertainty for people is signif-
icantly higher today than 1930s. So this needs to be explicitly modelled in an
otherwise Keynesian macro framework. In addition, the open economy implica-
tions should be incorporated to account for the global trade and foreign invest-
ment downturns. As pointed out earlier, the plausible rising income inequality
consequent upon Covid-19 may also play an important role in understanding the
efficiency of typical fiscal stimulus in the present scenario.
The following theoretical model is typically Keynesian with some modifica-
tions to incorporate uncertainty in household aggregate consumption expend-
iture and an undesirable shift of income distribution against the low-income
segment of the population. In addition, investors’ sentiment which reflects the
uncertain market prospects may be paid attention too. The first part essentially
deals with the AD in an open economy and how this can be managed through
fiscal and monetary actions. Next, the AS shocks are introduced in the light of
reverse migration of labour and supply chain disruptions. Finally, some idea is
formed in terms of AD and AS interactions and the resultant scope of policy
interventions.
8.3.1 The traditional standard model of fiscal stimulus

and its modifications
Since the present institutional stimulus packages are basically Keynesian in spirit,
let us first state the traditional model of AD in an open economy as follows. All
functions are taken as first-order linear approximations.
8.3.1.1 Product market equilibrium and the IS curve

Consumption function: C = a1 + b1(Y − T ), a1 > 0,0 < b1 < 1 (8.1)
_
Investment function: I = I + αY − β i, α > 0, β > 0 (8.2)
Net exports: NX = a2 + b2Y * − b3Y + b4 e (8.3)
Government expenditure: G = G0 (8.4)
where C = aggregate consumption, Y = domestic GDP, T = net tax, e = ex-

change rate = rupee/dollar, I = autonomous investment, i = real interest rate,
Y* = foreign GDP and G = government expenditure.
Thus, the IS curve will have a negative slope provided (b1 − b3 + α) < 1, which
is required for a positive output level.
8.3.1.2 Money market equilibrium and the LM curve

Md
Money demand function: = kY + δ1 − δ2i, δ1 > 0, δ2 > 0 (8.5)
P
Money supply function: M S = M 0 (8.6)
Price level: P = P0 (8.7)
where M 0 and P 0 are given.

The LM curve has an unambiguous positive slope.
8.3.1.3 Balance of payment equilibrium and the BP curve

Capital account and Net Foreign Capital Inflow:
NFI = γ1 + γ 2 (i − i * ), γ1 > 0, γ 2 > 0 (8.8)
where i* is the world interest rate.

Given NX function as stated earlier, the BP curve is found from the condition
that NFI + NX =0 (8.9). The BP curve has a positive slope unambiguously.
Now taking the above as a system which solves for Y, i and e from the three
equations of IS, LM and BP, the final stability requires a weaker stability con-
dition namely, b1 + α < 1, which ensures a negative slope of the IS curve and a
positivity of output level too. It can also be checked that if the above property is
satisfied then an expansion of government expenditure will lead to more expan-
sion when the BP curve is steeper than the LM curve, which is well known in
the standard open economy macroeconomics.
8.3.2 Modified standard model

Here, we introduce two important modifications to the existing basic model, namely:
(a) Distributional issue and

(b) Uncertainty
Let there be two groups of income earners, with bi = MPC of group i and Yi =
GDP accruing to group i, i = 1,2. If we assume Y1 > Y2, b2 > b1 (low-income
group has higher MPC), then Y = Y1 + Y2 = GDP. Also define the income
Y Y
shares as ω1 = 1 and ω2 = 2 . Also, we make a simplifying assumption that
Y Y
low-income group gets a positive net lump-sum subsidy S and the high-income
group pays a net lump-sum tax T.
Under uncertainty, households do not have knowledge about their income
with complete certainty. So instead of actual income, they take their consump-
tion decision based on expected income. Thus, we may write Expected GDP =
E(Y1 ) E(Y2 )
E(Y ) = E(Y1) + E(Y2). Thus, we retain ω1 = , ω2 =
E(Y ) E(Y )
Now suppose E(Y ) = p(Y + ε ) + (1 − p )(Y − ε ), p ∈ (0,1) where p is the probabil-
ity of having high income. Thus, if one takes total differential of this expected
income, then we have
dE(Y ) = dY + dp(Y + ε − 1 + ε ) = dY + 2εdp (8.9)
So under uncertainty, the change in actual income will be more for any corre-
sponding change in expected income if the probability of having a good state
worsens, that is dp < 0.
Now suppose that the MPCs of the two groups are also functions of the
probability of having high or low income. Thus, 1 > b1 = b1(p) > 0 and 1 >
b2 = b2(p) > 0, 0 < b1′ < b2′ < 1. The system of equations now looks as follows.
8.3.2.1 Product market equilibrium

E(Y ) = a1 + b1( p )(ω1E(Y ) − T ) + a2 + b2 ( p )(ω2 E(Y ) + S ) + I + α E(Y )
− β i + a3 + b3Y * − b4 E(Y ) + b5e + G (8.10)
where ai = constant terms, i = 1, 2 ,3, and the other terms are similar to the
standard model. Here, it is assumed that only households are affected by this
uncertain income while they decide how much to consume, but for simplicity
import propensity and investment propensity are assumed to be constant.
8.3.2.2 Money market equilibrium

M
= kE(Y ) + δ1 − δ2i (8.11)
P
P = P0 (8.12)
8.3.2.3 Balance of payment equilibrium

NFI = Net Factor Inflow = γ1 + γ 2i − γ 2i *
Net Exports = NX = a3 + b3Y * − b4 E(Y ) + b5e
BOP Equilibrium: γ1 + γ 2i − γ 2i * = −a3 − b3Y * + b4 E(Y ) − b5e (8.13)
Suppose now the government increases G and S simultaneously to tackle reces-

sion due to pandemic like Covid-19. (this is something done in most of the coun-
tries including India). At the same time p falls due to an increase in uncertainty
about achieving the higher value of GDP. This naturally leads to a fall in E(Y )
also. Given the arguments in the previous section, we assume this leads to a rise
in the share of high-income group ω1 and a corresponding decrease in ω2. Thus,
the aspects of uncertainty and changing distribution of income are incorporated
in this modified Keynesian standard model.
Now, as mentioned above, suppose G, S, p and ωi change simultaneously.
In the absence of specific values of change for the uncertainty and distribu-
tional variables, we derive certain conditions for different types of changes
in income to take place. Simple calculations produce the following compar-
Zδ b
ative static result (details are in the mathematical appendix): dE(Y ) = 2 5 ,
∆
where the denominator is positive if 1 > b1ω1 + b2ω2 + α. In the numerator
Z = dG + b2dS + b2′ S + dp{E(Y0 )(b1′ω1 + b2′ ω2 )} + E(Y0 )dω1(b1 − b2 ) − b1′T , where
E(Y0 ) is the initial expected equilibrium GDP.
In addition, clearly from (8.9) dYu = dE(Y ) − 2εdp , where the “u” subscript
after Y indicates GDP under uncertainty to distinguish it from the stand-
ard case which is a normal situation. If there is no uncertainty and no sepa-
rate MPCs between the two groups, the normal case will have the following:
b1 = b2 = b = constant, dp = 0, p = 1, ε = 0; hence, b1′ = b2′ = 0. Therefore, the change
in income under a simultaneous change in G and S will lead to the following
(dG + bdS )δ2b5
change in income: dYN = , where the denominator is the same as
∆
in the standard model above. Clearly, given a worsening of the probability of
having a good state of the economy, dp < 0, the change in expected income can
itself be negative despite a fiscal stimulus. In other words, the numerator of dE(Y )
will be negative if Z in the numerator of the above expression for dE(Y ) turns
out to be negative. The expression for Z written above will be negative if fiscal
stimulus in the form of (dG + b2dS + b2′ dS ) is small but changes in uncertainty and
distribution of income against lower income group (captured by the rest of the
terms in the expression for Z written above) are significantly large. In addition,
we note that the difference between income change in the normal case and under
certainty is given by
dYN − dYU = −b2′ S + b1′T − dp{E(Y )(b1′ω1 + b2′ ω2 )} − E(Y )dω1(b1 − b2 ) + 2εdp
( < 0) ( > 0) ( > 0) ( > 0) ( < 0)
So it is clear that the change in Y in the presence of significant changes in un-
certainty and changing income distribution will be much less than the standard
case for the same amount of fiscal stimulus in the two cases. Thus, the amount of
fiscal stimulus needed may be several times more in the former case if one wants
to get a similar multiplier effect on income expansion. So the perceived change
in demand may well fall short of actual if one fails to take account of uncertainty
and changing income distribution in the wake of Covid-19 pandemic. Thus, the
existing fiscal stimulus of about 1% of GDP may not be very effective.
The above calculation will show more adverse result if one adds two more
realities – one is the fact that in the presence of worldwide pessimism, invest-
ment sentiment captured by I may well become significantly adverse (Fornaro
and Wolf, 2020). The other is the strong belief in Indian policy perspective that
monetary policy through expansion of liquidity is the real antidote to tackle the
recessionary tendency. The multiplier value will then change as follows:
(Z + dI )δ2b5 + b5 ( β + γ 2 )dM
dE(Y ) =
∆
So one part of the numerator has a negative effect since dI < 0 and the other part
has a positive effect through depressing the interest rate via monetary expansion
since dM > 0. This will cause a change in the sign and magnitude of dE(Y ) pro-
vided the change in money supply is very large and the response coefficients to
interest rate of domestic and foreign investment, namely β and γ2 respectively,
are strong enough. The Government in India has assumed these to be true and
put more faith in the monetary expansion to tackle the recession aftermath the
Covid-19 pandemic. Future data will show the efficiency of such policies.
8.3.3 The current account surplus puzzle

The current account surplus observed for the first quarter of 2020–21 fiscal year
for India has attracted some attention. A look at the data clearly shows the sur-
plus, somewhat contrary to the general trend for last 20 years and more, has
occurred because the exports of goods and services have fallen faster than their
imports, leading to such a temporary current account surplus. So this is a peculiar
result – the textbook analysis discusses this in terms of a rise in exports faster than
imports and the consequent impact on output, interest rate and exchange rate.
Thus, the appearance of current account surplus in a contractionary economy
along with expansionary fiscal and monetary policies has to be incorporated
in a standard IS-LM-BP framework (introduced earlier in Section 8.3.1) with
necessary modifications for uncertainty and changing income distribution. The
following diagrammatic representation will be useful to discuss this.
i IS0/ IS0 BP0 LM0

BP0//
A LM0/
/
A
A//
A///
BP0/// IS0/// IS0//
FIGURE 8.1 IS-LM-BP adjustments post Covid.

The initial position is at A. We assume the BP curve is steeper, implying cap-

ital inflow in the country requires higher interest rate than what the domestic
investors need. The Covid simply pushes the domestic (internal) equilibrium
to A /. At A / net exports are assumed higher, a case which happened for India.
The reason is that Y and Y* are both lower, but imports fall faster than exports.
This should theoretically lead to current account surplus and an appreciation in
the exchange rate leading to a rightward shift of the BP curve. Thus, A/ is not
a sustainable equilibrium. However, fiscal and monetary expansionary policies
are aimed at pushing the AD upwards. In the diagram, this may shift the LM
curve rightwards but the impact on the IS curve is not foregone because of the
increased uncertainty and changing income distribution. So two alternative sce-
narios are possible as given by the points A // (an increase in output exceeding the
original outcome before Covid) and A /// (a decrease in output compared to the
original outcome). Which of these will be true depends on the strength of fiscal
and monetary policy in the background of increased uncertainty and changing
income distribution post Covid as outlined in the earlier section discussing the
modified standard model in Section 8.3.2 above.
8.4 Aggregate supply shocks

The above discussions all deal with AD shocks. This does not mean that there
is no shock on the supply side. Two types of shocks have been noted in various
discussions. One is the return of migrant labour to their home states as discussed
in Section 1.1 above. This may create higher transaction costs as noted above.
The other is the supply-side disturbances created by breakdown of supply chains,
mainly on account of disruptions initially in China and restrictions on flights,
shipping and travels (Barua, 2020, section 5.1). One can portray this in a heuristic
manner with the help of simple AS and AD diagrams.
Using the standard macro theory, assuming away the full employment con-
straint since it is not binding in the case of a country like India, the AS curve
is represented by Y = F(P), F/(P) > 0. Also, the production function is well be-
haved and regular and given as Y = f(L, K, M), where L = labour, K = capital
and M = intermediate inputs. To make the discussion focused suppose all the
intermediate inputs are imported. Due to Covid-19, the availability of labour is
certainly capped at a level L1 less than the existing employment at the pre-Covid
output level given by L 0 where L1 < L 0. Similarly suppose the intermediate
inputs are constrained at the level M1 < M 0. So the AS function shifts leftwards
and the maximum output gets constrained at an output level, say Y1. The point
is whether this becomes a binding constraint. In the absence of more elaborate
data, one cannot say definitively, but let us show the binding case to get the feel
of a difficult scenario post Covid. Figure 8.2 clarifies the point.
Figure 8.2 shows no change in the technology or marginal productivity of
variable factor labour, but the sudden non-availability of labour or intermediate
inputs puts an upper limit to the output in the short run after Covid-19. As a
P AS/
AD
AS// AS
/
AD AD //
P0
P0// P0/
Y0
Y0/ Y1 Y0// Y
FIGURE 8.2 Post-Covid shift in the AS curve along with AD curve.
result effectively the AS curve becomes vertical at output level Y1. This does not
pose immediate problem since the AD curve shifted more after Covid to AD/,
leading to an equilibrium output Y0/ less than the original equilibrium value Y0.
However, the fiscal and monetary expansion measures may or may not be able
to push the AD curve rightwards. Taking an optimistic view, a shift of the AD
curve to the AD//curve may face the constrained AS curve as binding and output
may be capped at Y1. But a relaxation of labour and intermediate input constraint
may shift the AS curve rightwards to AS// and lead to a new equilibrium at Y0//.
It all depends on relative shifts of AD and AS curves and we have drawn the
most optimistic situation. Just to complete the story, initially the overall price
level will show a declining trend to P 0/ although there may be sectoral price
rise especially of food items. The National Statistical Office (NSO) of India has
just published wholesale price index (WPI) data which show exactly this – food
articles have a price rise of 2.5% in the first quarter of 2020–21 while the overall
index has fallen by 2.3% (NSO, MOSPI, Government of India, August 31, 2020).
Once the boosting measures raise AD, the prices will tend to rise to P 0//.
8.5 The long-run growth after Covid-19

The long-run growth depends on the amount of domestic savings and foreign
capital inflow. It is pertinent to mention that the gross domestic savings in In-
dia showed a declining trend in the decade of 2010–20, falling as a percentage
of GDP from 34.68% in 2011–12 to 30.16% in 2018–19 in current prices (with
2011–12 as base) (National Account Statistics, NSO, MOSPI, Government of
India, 2020b). The gross capital formation as percentage of GDP in the same
period was 39% and 32.2% respectively. So the foreign capital inflow from the
rest of the world has also fallen as a percentage of GDP in recent periods. The
outbreak of Covid-19 pandemic has reduced the household consumer expendi-
ture in the first quarter of 2020–21, thereby perhaps increasing some amount of
precautionary savings. However, the fiscal injections in the Western countries
may have gotten their way in some form of foreign capital inflow in India so that
the inflow of capital from foreign countries have recovered well in last couple of
months. However, the falling domestic savings ratio with a projected increase in
government deficit in India may not augur well for investment. Thus, the growth
prospect may well turn out to be somewhat pessimistic.
Following the introduction of distributional issues in our previous theoretical
models in Section 8.3, let us incorporate this in a standard Solow-type one-sector
growth model. This will give some idea about the implication of changing in-
come distribution on savings rate and the role of different components of gross
capital formation in boosting up growth in India. For the time being, let us
ignore the imported intermediate inputs in the production function mentioned
in Section 2.4.
The aggregate production function in the economy is taken as Y = F(K, L)
Y K 
having all the regularity properties including CRS, so that = y = f   = f (k ).
L L 
Let us write the savings function as follows:
S = s2W + s1R = s2 ( f (k ) − kf ′(k )) + s1 f ′(k )k, where s1,s2 are savings propensities
of workers having wage income W and capitalists having profit income R from
standard neo-classical theory of distribution under CRS production function.
Now if we assume that wage earners form the low-income group in the sense
that they have lower average income, then following the assumptions made about
the MPCs we must have s 2 < s1. Assume that gross capital formation or gross
investment has three components: I = I P + IG + I F , where the three types of in-
vestments stand for private, government and foreign respectively (MOSPI, Gov-
ernment of India, 2009). The fundamental equation of balance in the economy
is given by
S = National Savings = I P + IG + I F + CA (8.14)
where CA = current account balance. Since Foreign Investment = capital inflow

from abroad = capital account balance = KA = −CA = IF. Naturally, if CA < 0,
KA > 0, thus IF > 0.
So following Solow-type one-sector growth model with incorporation of dif-
ferential propensity to save we have Id = domestic investment = IP + IG which
depends on the savings in the economy. The foreign investment may be taken as
largely exogenous to the economy. Thus, the Solow growth equation needs to
be modified to understand the savings investment balance also. Further, overall
aggregate capital stock is given by K = Kd (Domestic)+ Kf (Foreign). Thus, invest-
. . .
. K K K Kf Kf
ment appears as K = I , so that= gK = d . d + = g K d α1 + g K f α 2 .
K Kd K K f K
We assume, as mentioned earlier, g K f α 2 = constant = A, so g K = g K d α1 + A.
.
K k
Now, define k = , gk = = g K − n, where n = growth rate of labour. From the
L k
above, it is obvious that
.
k = α1 g K d k + Ak − nk (8.15)
Using the fundamental balance relation in the economy as expressed in equa-

tion (8.14), we assume that domestic saving finances the accumulation of do-
mestic capital since the foreign capital inflow is assumed to be exogenous to
the economy. Thus, we have the following after using the value of S derived
above:
.
K d = S = s2 [ f (k ) − kf ′(k )]L + s1 f ′(k )K which implies that the growth rate of
domestic capital is given as
.
Kd 1 1
= g K d = s2 [...] + s1 f ′ (8.16)
Kd α1k α1
Using the definition of the production function, we have gY = growth rate of

output which can be shown as
f′ . f′
gY = n + k = n + α1 g K d k + ( A − n )k  (8.17)
f f
(From equation (8.15) above after simplification, we obtain
f′
gY = n + [ s2 f + ( s1 − s2 )kf ′ + Ak − nk ] (8.18)
f
From this, differentiating gY with respect to k we obtain the following expression
(see mathematical appendix):
∂ gY  f ′′k  f′ f′
=  gk + [( s1 − s2 ) f ′ − n ][1 − sh( K )] + A[1 − sh( K )] (8.19)
∂k  f  f f
where Sh(K) = share of capital input in the total production which is a fraction.
Clearly in equation (8.18), the first two terms are negative but the last term is
positive. So to ensure a positive growth in output with increase in capital per
head, the foreign investment plays a very important role since a large value of A
can overturn the negative impact of diminishing marginal productivity of capital
in the economy, reducing the profit share in the economy and affecting the sav-
ings rate, accumulation of capital and growth in the economy. One very poten-
tially productive avenue of foreign investment may be in the creation of global
value chain by the large MNCs consequent upon the trend of diversification
of GVC being witnessed post Covid-19. This will also open up possibilities of
sustainable growth for the ailing MSME sector in India which produces around
40% of manufacturing output in India and provides employment to about 40%
jobseekers across the manufacturing and services sectors (Annual Report, Min-
istry of MSME, Government of India, 2019). In fact, the foreign investment can
boost the infrastructure investment as well. This solves the problem of crowding
out of private investment in the face of faltering domestic investment of the
economy (National Accounts Statistics, NSO, MOPI, Government of India, Au-
gust 2020). Boosting of infrastructure spending is an appropriate strategy at the
moment which will create job and boost up demand for non-food sector as well.
The apprehension of non-food price rise may not be that dominating given the
possibility of slow movement of AD coupled with some improvement of capacity
utilization of industrial sector.
8.6 Conclusions
The advent of Covid-19 has brought unforeseen hardship to both life and
livelihood all across the world. All forecasts predict fall in world output in the
year 2020–21 (World Bank, 2020; IMF, 2020), amidst the ominous result of
one of the highest fall in the GDP of India in the first quarter of 2020. The
most common policy response to such catastrophe is expansionary fiscal pol-
icy risking a rising fiscal deficit along with higher debt to GDP ratios. India
has taken such fiscal measures as well but with greater emphasis on monetary
policy. The present chapter highlights two issues commonly missed in the
discussion – this is the rising uncertainty due to the Covid-19 pandemic and
probably change in the distribution of income against the group having higher
marginal propensity to consume. As a result, the perceived multiplier effect of
the fiscal boost in India may be much less justifying the call for a much larger
fiscal package. The monetary policy may also not be very effective in the face
of uncertainty in the market as well as weak response to a falling interest rate
by the domestic investors in such circumstances. Similarly, the supply side may
face bottlenecks in terms of non-availability of migrant labour force and dis-
ruptions in the availability of intermediate inputs due to broken supply chains.
In the short run, efforts must be directed to ease such supply-side constraints
as far as feasible.
The long-run growth is also a major point of concern. Once the income
distribution is incorporated, a change in this distribution in favour of the profit
earners having a higher marginal propensity to save may not help overcome the
effect of falling marginal productivity of capital with a rising capital per head.
But extending the Solow one-sector model to incorporate the fundamental mac-
roeconomic balance equation of saving and investment in an open economy, one
can see the importance of foreign investment in the process. The huge fiscal stim-
ulus in the USA, the UK, the European Union and Japan may generate surplus
funds which may find their way to the Indian investment market. In fact, OECD
investment statistics shows that the first quarter of 2020 shows a 32% fall in FDI
inflow over the last quarter of 2019 for the non-OECD G20 countries, whereas
in that group India has shown a 15% growth in the same period (OECD, July
2020). This is reiterated in the economic survey of India for 2020–21 (Economic
Survey, 2020–21, Volume II, p.31). There are two important consequences to
this – one is the possibility of large companies relocating some of their supply
chain activities to India strengthening the MSME sector in future. The other is
the opportunity to boost up the infrastructure investment in the economy which
may generate the much needed employment and income among the vast majority
of the workers in the economy.
References
Barua, S. (2020), “Understanding Coronanomics: The Economic Implications of the
Coronavirus (COVID-19) Pandemic.” April 1. Available at https://papers.ssrn.com/
sol3/papers.cfm?abstract_id= 3566477, accessed on June 2, 2020.
Dasgupta, D and M. Rajeev (2020). “Paradox of a Supply Constrained Keynesian Equi-
librium: The Covid-19 case”, Economic and Political Weekly, 55(22), May 30. Available
at http://www.epw.in.jadavpur.remotexs.in/journal/2020/22/commentary/paradox-
supply-constrained-keynesian-equilibrium.html, accessed on July 2, 2020.
Dev, S. M (2018). Inequality, Employment and Public Policy. Working paper WP-2018-003,
Mumbai: IGIDR. Available at http://www.igidr.ac.in/pdf/publication/WP-2018-
003.pdf, accessed on April 15, 2020.
Economic Times (2020). “Indian Economy to Contract 10.5 pc in FY21”. September 9.
Available at https://economictimes.indiatimes.com/news/economy/indicators/fitch-
projects-indian-economy-to-contract-10-5-per-cent-in-current-fiscal/printarticle/
77991273.cms
Eichenbaum, M, S. Rebelo and M. Trabandt (2020). The Macroeconomics of Epidemics.
Working Paper 26882. Available at http://www.nber.org/papers/w26882, accessed
on July 6, 2020
Fornaro, L and M. Wolf (2020). “Coronavirus and Macroeconomic Policy”. March,
Available at https://voxeu.org/article/coronavirus-and-macroeconomic-policy, ac-
cessed on July 6, 2020.
International Monetary Fund (IMF) (2020). World Economic Outlook, June. Washington,
D.C.
Ministry of Finance, Government of India (2021). Budget Speech 2021–22. Available at
https://www.indiabudget.gov.in/doc/Budget_Speech.pdf, accessed on February 7,
2021.
Ministry of Finance, Government of India (2021). Economic Survey 2020–21. Avail-
able at https://www.indiabudget.gov.in/economicsurvey/, accessed on February
7, 2021.
Ministry of Finance, Department of Economic Affairs, Government of India (2020).
Monthly Economic Report, March and April 2020. Available at https://dea.gov.in/sites/
default/files/March%20and%20April%202020_0.pdf, accessed on June 20, 2020.
Ministry of Micro, Small and Medium Enterprises, Government of India (2019). Annual
Report 2018–19. Available at https://msme.gov.in/sites/default/files/Annualrprt.pdf,
accessed on August 10, 2020.
Ministry of Statistics and Programme Implementation (MOSPI), Government of India
(2009). Report of the High Level Committee on Estimation of Savings and Investment. https://
www.rbi.org.in/scripts/PublicationReportDetails.aspx?FromDate=3/18/2009&
SECID=21&SUBSECID=0, accessed on September 10, 2020
National Statistical Office (NSO). MOSPI, Government of India (2020a). National
Accounts Statistics. Available at http://mospi.nic.in/publication/national-accounts-
statistics-2020, accessed on September 8, 2020.
National Statistics Office (NSO), MOSPI, Government of India (2020b). Periodic La-
bour Force Survey (PLFS), July 2018-August 2019. Available at https://pib.gov.in/Press
ReleaseIframePage.aspx?PRID=1629366, accessed on August 20, 2020.
NSO, MOSPI, Government of India (2020c). Press Note on Gross Domestic Product for the
first Quarter (April-June 2020–21), August 31. Available at http://www.mospi.gov.in/
sites/default/files/press_release/PRESS_NOTE-Q1_2020-21.pdf, accessed on Sep-
tember 5, 2020.
McKibbin, W and R. Fernando (2020). “The Global Macroeconomic Impacts of

Covid-19: Seven Scenarios”. Available at https://www.brookings.edu/wp-content/
uploads/2020/03/20200302_COVID19.pdf, accessed on September 3, 2020.
Mishra, P. (2020). “Whither India’s Economy Post Covid-19?”, Paper prepared for the
NCAER India Policy Forum 2020.
Organisation for Economic Cooperation and Development (OECD) (2020). Coronavi-
rus: The World Economy at Risk. March. Available at https://www.oecd.org/berlin/
publikationen/Interim-Economic-Assessment-2-March-2020.pdf, accessed on
August 7, 2020.
OECD (2020). Updated Excel Tables for FDI in figures. July 2020. Available at http://www.
oecd.org/investment/statistics.htm, accessed on August 3, 2020.
Reserve Bank of India (2020). RBI Bulletin. August. Mumbai.
Singh, S.K., V. Patel, A. Chaudhary and N. Mishra (2020). “Reverse Migration of La-
bourers amidst Covid-19”, Economic and Political Weekly, LV(32 & 33), August 8, 25–29.
State Bank of India (2020). Ecowrap Issue No 34, FY 21, August 17. Mumbai. Available at
https://www.sbi.co.in/documents/13958/3312806/170820-SBI+Ecowrap+-+The+
Good%3B+The+Bad%3B+The+Ugly.pdf/38684abb-d48c-6f 80-45ad-bd1351c
51c05?t=1597644350762, accessed on August 28, 2020.
World Bank (2020). Global Economic Prospects, June. Washington, DC.
World Inequality Lab (2018). World Inequality Report. Available at https://wir2018.wid.
world, accessed on July 20, 2020.
MATHEMATICAL APPENDIX
A. The result for dY for simultaneous change in G, S with change
in weights and uncertainty
Using equations (8.10)–(8.13), the system can be written as follows:
 1 − b ( p )ω − b ( p )ω − α + bY β −b5  E(Y )
1 1 2 2
  
 k −δ2 0  i 
 b −γ − b  e 
 4 2 5 
 _ 
 A + I + b3Y * + G − b1( p )T + b2 ( p )S 
 _ 
 M 
=  δ1 + 
 p 
 a + γ + b Y * − γ i* 
 3 1 3 2 
 
where A = (a1 + a2 + a3 ).
∆ in the text refers to the determinant value of the above coefficient matrix.
Now, taking total differential on both sides and assuming dG > 0, dS > 0, dω1 > 0, dp <
dG > 0, dS > 0, dω1 > 0, dp < 0 and replacing dE(Y ) = dY + 2dpε we obtain the expression of
dE(Y ) written in the text. The condition for negativity of Z in the text is given as
dG + b2dS + b2′ S < dp{E(Y0 )(b1′ω1 + b2′ ω2 )} + E(Y0 )dω1(b1 − b2 ) − b1′T where E(Y0)
is the original expected GDP level. This is unlikely if the changes in G and S are
large enough, leading to a positive multiplier effect. In the case of small changes in
G and S, the multiplier value will be really low if not negative. Now writing Y N
as GDP in normal year and Y U as the GDP in the year of uncertainly like Covid,
dYN > dYU if b2′ S − 2dpε > b1′T − dp{E(Y0 )(b1′ω1 + b2′ ω2 )} − E(Y0 )dω1(b1 − b2 ). We
assume the change in the share of the rich group as well as the change in uncer-
tainty is not that large which will reverse the above inequality sign. This also
ensures a positive multiplier effect for simultaneous fiscal stimulus in terms of a
change in G and S.
B. Long-run growth derivation of equation (8.18)

Derivation of equation (8.18):
f′ . f′
gY = n + k = n + [α1 g K d k + Ak − nk ]. Now using equation (8.16)
f f
in the text, we obtain α1 g K d k = s2 [ f (k ) − kf ′(k )] + s1 f ′(k )k. Therefore,
f′
gY = n + [ s2 f (k ) + kf ′(k )( s1 − s2 ) + ( A − n )k ]. The above is equation (8.17) in the
f
text. Using this equation, we differentiate both sides by k and obtain: ∂ gY / ∂k =
 f {kf ′ f ′′ + ( f ′)2} − k( f ′)2 f ′ ( A − n ) f ( f ′ + kf ′′) − ( A − n )k( f ′)2 
s2 f ′′ + ( s1 − s2 ) + 
 f2 f2 
= 
 k  k {
 f ′′k  s2 f
}
+ ( s1 − s2 ) f ′ + ( A − n ) +
( A − n) f ′
f
[1 − sh(k )] + ( s1 − s2 )
( f ′)2
f
[1 − sh(k )],
kf ′
where sh(k ) = which is equation (8.18) in the text by inserting the value of gk
f
using equations (8.15) and (8.16).
9
DEVELOPING ECONOMIC HERD
IMMUNITY IN THE FACE OF A
PANDEMIC1
Parikshit Ghosh
9.1 Introduction
It has become conventional wisdom that Covid-19 has placed us between the
Scylla and Charybdis of epidemic explosion and economic collapse. There is a
tough trade-off (Barro et al., 2020) between protecting our health and saving the
economy, goes this view. In a lot of ways, this perspective is flawed.
In the short run, measures to suppress the spread of infection dampen eco-
nomic activity. This suggests a positive contemporaneous correlation between
GDP and case load. However, a high rate of infection alters consumer behaviour
in ways that have a negative impact on the economy in the medium to long run.
People withdraw from travelling, shopping and eating out. The lagged correla-
tion between case load and GDP is, therefore, negative. The dynamic interplay
between the infection rate and economic activity is not unlike that in predator-
prey models (Lotka, 1910; Volterra, 1926). As a consequence, protecting the
economy involves an intertemporal choice—to some extent, incurring short-term
losses prevents greater long-term damage.
Economic resilience in the face of a pandemic cannot be thought of in purely
aggregative terms, either. Issues of incentives and equity are important as well.
Covid-19 has delivered two distinct shocks to the economy—a macroeconomic
demand and supply shock as well as a shock to the income distribution. Job and
income losses have been concentrated among certain sectors and groups, through
no fault of their own. Putting in place social safety nets and redistributive pol-
icies to counter this income distribution shock is important for its own sake. I
will make the case that egalitarian policies also have instrumental value—they are
likely to curb total infections as well as GDP loss in the long run.
The role of incentives in curbing economic activity during a pandemic has re-
ceived little attention. Policy responses have relied almost exclusively on command
DOI: 10.4324/9781003220145-11
Developing economic herd immunity 147
and control measures, such as lockdowns, travel bans, prohibition on large gather-
ings, mask mandates and social distancing guidelines. To the extent policymakers
have tried to apply these restrictions differentially across regions and sectors (e.g.,
by demarcating containment zones in hot spot areas), it has been done through
bureaucratic judgement rather than economic incentives and self-selection. Econ-
omists often overemphasize prices and incentives, ignoring practical difficulties
in implementation. Therefore, the scant attention paid to how tax-and-transfer
policies can be used to curb economic activity selectively and efficiently seems
surprising. This chapter is also an attempt to bridge that gap.
During a pandemic, economic activity, which requires people mingling in
close proximity, creates a classic negative externality. The socially optimum level
of activity will be less than what agents will choose out of self-interest. However,
economic contraction should not happen uniformly—activities which produce
relatively less economic value should contract disproportionately. The problem
that should interest economists is how to allocate cutbacks in production.2 If an
omniscient social planner imposed firm-specific caps on productive activity
without compensatory transfers, the outcome would be extremely regressive.
Street hawkers and small industries may be shut down, while the IT and financial
sectors may be allowed to operate with very little reduction in output in per-
centage terms. The problem with implementing such a plan, other than qualms
of social justice, is that productivity levels and especially productivity shocks are
unobservable to the planner, by and large. One has an intuitive sense that the
hospitality sector has been hit harder than online retail, but it is impossible to
precisely measure productivity across all enterprises and calibrate the restrictions
accordingly.
A Pigouvian tax on all economic activity can be a potential solution to the
problem. The disease externality caused by people going about their business is
plausibly independent of the economic value they are creating. The government
can impose a pandemic tax on production inputs (especially human inputs) to
encourage contraction. Facing the same tax, more productive firms will contract
less, and less productive ones will shrink disproportionately. If the tax is kept
revenue neutral, then its proceeds will be returned to households and businesses
as lump sum tax credits. It could finance a universal basic income for the duration
of the intervention. Note that such a policy is necessarily progressive—richer and
more productive agents engage in greater economic activity and contribute more
in taxes, while everyone gets the same tax credit.
Even a Pigouvian tax scheme is difficult to enforce in practice. Inputs like
work hours and size of the work force are hard to measure. However, the fruits
of economic activity, namely income and profits, are routinely assessed by tax
authorities in all nations. Can the allocative function of a universal3 Pigouvian
tax be proxied through an appropriately amended income tax schedule? In a later
section, using techniques from the optimal taxation literature pioneered by Mir-
rlees (1971), I will show that the answer is in the affirmative. However, to satisfy
the demands of incentive compatibility, an optimal income tax must be even
148 Parikshit Ghosh
more redistributive than an optimal Pigouvian tax levied on inputs. Essentially,

in an income tax regime, agents have an incentive to lie upwards, i.e., make
choices that indicate they are more productive than they truly are. To discourage
this, the marginal tax rate should climb more steeply (or fall more slowly) than in
a Pigouvian regime, leading to greater redistribution.
Lags in income assessment pose a problem for this approach. Agents who suf-
fer large losses will eventually be compensated but it may take months. In most
countries, income assessment is an annual exercise. If credit markets worked
perfectly, agents whose incomes have collapsed due to the Covid shock could
borrow against their future tax credits to finance their immediate consumption
needs. This is, however, far from reality. The government has to tackle not only
the task of redistribution but also the immediate insolvency problems faced by
households and businesses. This has been widely recognized; as a consequence,
governments across the world have rushed to put together financial rescue pack-
ages (comprised of grants as well as loans) from the early days of the outbreak.
The quantum of assistance has, however, been held in check by concerns about
ballooning budget deficits and soaring public debt.4
This financial pusillanimity and dithering has a lot to do with how the prob-
lem has been framed. While it has been gingerly accepted that the most un-
fortunate citizens (e.g., those employed in the hospitality and travel industries)
deserve income support since the pandemic is an act of God, it has scarcely been
acknowledged that the relatively more fortunate among us (e.g., those in gov-
ernment service or education, with secure tenures and paycheques) have a corre-
sponding obligation to provide that support during these troubled times. There
is, after all, no manna from heaven. In other words, a basic income for everyone
early in the pandemic should be financed by a special pandemic tax or CESS
collected later from households and businesses shown to have suffered little or
no income loss. To make sure the tax has the desired incentive effects, it should
have been announced along with the financial rescue packages, with a clear time
frame regarding its application. The strength of the fiscal support provided by
the state has been compromised by the lack of a clear revenue plan to match the
sudden spending obligation.5
The Covid-19 crisis has illustrated another shortcoming of our existing model
of public finance. Individuals have precautionary savings, universities have en-
dowments, corporations carry cash reserves and countries hold buffer stocks of
food grains. In other words, everyone saves for a rainy day, but governments live
from hand to mouth, relying on flows like tax revenues and borrowing from the
bond market to finance their needs. Ghosh and Ray (2021, forthcoming) have
advocated that all nations build up a sovereign wealth fund—a professionally
managed financial portfolio—of the sort currently owned by a handful of states
rich in natural resources (Norway, Saudi Arabia and Alaska are prime examples).
Such a fund can reduce structural inequality through welfare spending or a basic
income to all citizens, but it can also act as a contingency fund, giving govern-
ments the financial muscle to fight against a pandemic or recession.
Infectious diseases put us all in the same boat. The selfish rich may be indiffer-
ent to the health of the poor in general, to the prevalence of diabetes, heart dis-
ease, cancer and malnutrition, but vaccination of the poor is also in the interest of
the affluent. One can escape poverty, squalor and crime behind the high walls of
gated communities, but not so much deadly viruses. As epidemiology textbooks
tell us, what ultimately protects us is herd immunity. A similar inter-dependence
should also be acknowledged in the economic sphere. Regulating social and
economic activity during a pandemic is in the public interest, but unless all social
classes have their basic needs met and livelihoods protected, their incentive to
cooperate in that venture will be weak. Outside of non-democratic nations with
high state capacity, such as China, there are limits to achieving public health out-
comes through coercion alone. The migrant crisis precipitated by India’s sudden
and draconian lockdown is a case in point. A pandemic calls for egalitarian pol-
icies that are necessary to create both biological and economic herd immunity—the
development of collective resilience by sharing the pain and spreading the gains.
The rest of this chapter is organized as follows. In Section 9.2, I present some
empirical evidence in support of the main premise that there is enormous heter-
ogeneity in how different households and firms have experienced the Covid-19
shock. Section 9.3 analyses a theoretical model to derive conclusions about opti-
mal tax-transfer policies. Section 9.4 discusses experiences and lessons specific to
developing countries like India. Section 9.5 concludes.
9.2 The economic impact of Covid-19

Economists are trained to give aggregates (like GDP, inflation, gains from trade,
etc.) first-order importance. Let us bake as big a pie as we can, and then we can
haggle over who gets to eat how much. This instinct has produced headline-
grabbing reports on the drop in GDP experienced by major economies since the
beginning of the pandemic. It has also generated a range of ideas about how to
keep the engine of production running at least on a lower gear—by admitting
back workers who have developed immunity (Dewatripont et al., 2020) or who
are at low mortality risk (Ray and Subramanian, 2020), by alternating between
lockdowns and business as usual (Ferguson et al., 2020), and so on.
Even before the virus struck, the world’s richer economies were wasting nearly
a third of their potential GDP. People didn’t call it an economic catastrophe, they
called it weekends and holidays! Indeed, the five-day work week is a very recent
phenomenon, still confined to the more prosperous parts of the world. If people
are willing to give up such a volume of goods and services for leisure, is it incon-
sistent to make similar material sacrifice for health?
A lot of production and consumption also allows for intertemporal substitu-
tion. The output loss from a two-month lockdown (roughly how long it took
China to bring the new case count down to nearly zero) can be entirely made
up if workers accept a shortened one-day weekend for the rest of the year. There
will be some complications, of course. One can move around when cars and
150 Parikshit Ghosh
computers are made and purchased, but if I have skipped four haircuts while I
was avoiding contact with people, in the end I am going to have one big hair-
cut, and the barber shop will never recover three of the four haircuts it lost to
Covid-19. Some consumption is time sensitive. However, straining to jumpstart
productive activity with a reduced workforce has its own complications. Workers
work in teams, not solo. Imagine running a restaurant when the waiter has devel-
oped immunity but the chef is in quarantine and most patrons are still interested
in takeout.
The Covid-19 shock has been compared to the World Wars and the Great
Depression in magnitude. Unlike wars, there is virtually no destruction of the
economy’s productive capacity brought about by the virus. Studies (Barro et al.,
2020) show that although the First World War and the 1918 Spanish flu pandemic
had roughly similar effects on GDP in the immediate aftermath, the effect of the
flu was much less lasting and probably partly recovered by subsequent increases in
production. The Depression is a different story due to its sheer duration.
However, extremely disruptive events, like a pandemic, war or famine, tend
to have heterogeneous impact on various occupations and cause massive shifts in
the income distribution. This distributive effect is of first-order importance that
public policy must address, looking beyond aggregate shortfalls. One illustration
of this principle comes from Amartya Sen’s (1981) study of the Bengal famine
of 1943 that killed nearly a million people. As Sen documents, grain production
was, if anything, marginally higher in the famine year compared to the preced-
ing one. Shortages appeared in civilian supplies partly due to the diversion of
grains to support the war effort, and partly due to speculative hoarding. How-
ever, this triggered massive shifts in consumption patterns, movements of relative
prices and heterogeneous impact on various occupations. Fishmongers, milk-
men, labourers and barbers saw a massive erosion of their purchasing power.6 It
is the concentration of losses in these particular groups that caused much of the
starvation and death.
A full picture of the heterogeneous economic impact of Covid-19 is yet to
emerge, though the inequality in its epidemiological impact is well documented
(Abedi et al., 2020; Yancy, 2020). Income distribution data come from house-
hold surveys; their frequency, sample size and publication lag make it difficult to
obtain granular data even a year since the outbreak. Nevertheless, a handful of
studies have emerged, documenting the widely disparate impact of Covid-19.
Using credit card, payroll and job postings data from the private sector, Chetty
et al. (2020) have shown how the economic impact in the USA has varied widely
across sectors, social classes and geographies.7 One salient insight from this study
is that although less educated, poorer workers and small businesses have, on av-
erage, suffered larger and more persistent income erosion, past income is far from
a perfect predictor of these losses. Factors like geography and sector of employ-
ment have also played a major role. For example, poorer workers living in richer
ZIP codes and working in occupations that rely on greater human contact (such
as salons and restaurants) have suffered more than their counterparts elsewhere.
Nevertheless, the relief cheques included in the two financial stimulus packages
passed by the US Congress condition payments on past (i.e., pre-pandemic) in-
comes and do not impose new taxes on the relatively fortunate. The economic
impact and policy response in other rich economies follow a similar pattern
(Adams-Prassl et al., 2020).
While heterogeneity is difficult to estimate at the household level, stock mar-
ket data, which is available at a granular level in real time, can give us further
insights into how divergent fortunes have been at the firm and sectoral level.
Changes in the stock prices of various companies since the outbreak of the dis-
ease reflect the market’s expectations regarding their altered revenue and profit
flows, as well as the possible impact on the wage and employment status of their
employees. Figure 9.1 shows the change in the share prices of select companies
between January 15 and April 14. Clearly, there is an enormous divergence in
these movements—Amazon shares rose by more than 20%, while shares of Delta
Airlines and Royal Carribean Cruise Lines dropped by more than 60%. Overall,
the travel, hotel, auto and banking industries took large hits, tech and pharma-
ceutical giants did better than average, and the online retail business actually
prospered.
To get a more long-term and systematic estimate of this heterogeneity, I chose
a calendar period of one year (March 1 to March 1), and compared the cross-
sectional volatility of stock prices in the year of the pandemic to the years in the
previous decade. Specifically, I looked at the annual percentage change in the
stock price of companies that comprise the index and computed its standard de-
viation. Figure 9.2 reports the results for companies that are included in the BSE
Sensex index, while Figure 9.3 shows results for the S&P100 index.
FIGURE 9.1 ercentage change in the prices of select stocks, January 1–April 14,
P
2020.
152 Parikshit Ghosh
50
45
40
35
30
25
20
15
10
5
0
2020-21 2019-20 2018-19 2017-18 2016-17
2015-16 2014-15 2013-14 2012-13 2011-12
FIGURE 9.2 S tandard deviation of annual percentage changes in stock prices for BSE
Sensex companies, 2011–21.
45
40
35
30
25
20
15
10
0
2020-21 2019-20 2018-19 2017-18 2016-17
2015-16 2014-15 2013-14 2012-13 2011-12
FIGURE 9.3 S tandard deviation of annual percentage changes in stock prices for
S&P100 companies grouped by industry.
Clearly, cross-sectional volatility in stock prices has been strikingly higher in

2020–21 compared to the previous decade. For the Sensex, the volatility measure
in 2020–21 is 45.95 compared to an average of 23.98 for the rest of the decade.
For the S&P100, a decadal average of 15.35 has jumped to 39.31 in the last
year. These figures suggest that the earnings, and, consequently, valuations of
firms have diverged in a historically unprecedented manner since the arrival of
Covid-19.
9.3 A model of optimal taxation [model]

As I have argued above, the pandemic has introduced two salient features into
the economic environment. First, economic agents have received negative pro-
ductivity shocks that are heterogeneous and private information. Second, as long
as the virus is around, economic activity generates a negative externality that is
independent of an agent’s productivity. An optimal tax policy should correct for
the negative externality by imposing a tax on economic activity. At the same
time, it should discriminate between agents at different levels of productivity,
suppressing less productive activity to a greater degree than more productive ac-
tivity. I now present a simple model of optimal taxation borrowing the Mirrlees
(1971) framework to study these issues. The aim is to gain some insights into
normative tax policy under these circumstances.
There is a continuum of agents of measure 1. Each agent has a production
function given by y = θ x , where y is the agent’s (pre-tax) income, θ is his pro-
ductivity and x is his level of economic activity (the number of hours devoted
to production, say). Let t be the tax liability of the agent so that his net income
is θ x − t . Economic activity has a private utility cost8 given by the cost function
θ
1 2
2
∫
x . In addition, each agent suffers a utility cost cx, where x = x (θ ) f (θ ) dθ
θ
is the aggregate level of economic activity in the economy. The idea captured
here is that more economic activity makes the virus spread faster, which imposes
a cost on everyone in terms of lost production, medical expenses, etc. For sim-
plicity, I have taken this cost to be linear in the aggregate amount of economic
activity. However, since each individual agent is atomistic, the choice of his own
x does not affect x , and hence the cost arising from the disease burden is a social
externality, not a private cost. To guarantee interior solutions and strictly positive
activity levels, I will impose the condition θ > c .
Let u (.) be the agent’s utility function of income, where u (.) is strictly in-
creasing and concave. I assume that the benefits and costs described above are
additively separable. The agent’s net utility is then given by
1
v = u (θx − t ) − x 2 − cx (9.1)
2
Here, θ is a variable across agents and private information, i.e., the agents them-
selves observe the realization of their own θ but the planner does not. Let F (θ )
be the distribution of θ in the population, with density function f (θ ) and sup-
port θ , θ . Suppose private choices of economic activity and the tax liability of
agents (both dependent on their productivity) are described by twice continu-
ously differentiable functions x (θ ) and t (θ ) respectively. We consider a utilitar-
ian social planner whose objective is to maximize the social welfare function:
θ θ
W= ∫  1 2
u (θ x (θ ) − t (θ )) − 2 x(θ )  f (θ ) dθ − c ∫ x (θ ) f (θ ) dθ (9.2)
θ θ
154 Parikshit Ghosh
We will consider tax policies that redistribute tax receipts back to the agents
as lump sum transfers, subject to meeting some fixed revenue requirement R ≥ 0
for the government.
To throw the points I wish to make into sharp relief, I will assume R = 0 (so
the tax policy is revenue neutral). Furthermore, I will assume u (c ) = c , i.e., the
incentive effects of taxes do not involve any income effect. The two assumptions
together also imply that the social planner has no intrinsic motive for raising
revenue or redistributing income. Under these assumptions, the only purpose for
taxation is correcting the externality generated by the pandemic. As we will see,
the optimal tax policy will nevertheless have a strong egalitarian flavour, which
is a by-product of pursuing this goal.
I will successively investigate three scenarios: (a) laissez-faire (the planner does
not interfere in private choices), (b) Pigouvian taxes (the shock θ is private infor-
mation but economic activity x is observable and the planner taxes it to correct
the externality), (c) Mirrlees income taxes (income y is observable but neither
θ nor x is observable, and the planner chooses an optimal income tax schedule).
9.3.1 Laissez-faire
Each agent chooses a level of activity given by
1 2
x0 (θ ) = arg max θ x − x − cx = θ
x 2
where x is constant from an agent’s perspective but not in the aggregate. Agents’
incomes and utilities are then given by
y0 (θ ) = θ 2 − cEθ (9.3)
1
u0 (θ ) = θ 2 − cEθ (9.4)
2
where E denotes the mathematical expectation operator.
9.3.2 Pigouvian taxes

If activity level x is observable to the planner, it is straightforward that the exter-
nality can be corrected by imposing a marginal tax rate of c on economic activity.
A revenue-neutral tax policy returns the collected taxes by paying a lump sum T1
to all agents. Under such a policy, each agent’s choice of activity level is given by
1 2
x1 (θ ) = arg max θ x − x − cx − cx + T1 = θ − c
x 2
The tax refund T1 can be calculated from the revenue neutrality condition:
θ θ
∫ t ( θ ) . f ( θ ) dθ = ∫ c (θ − c ) f (θ) dθ − T = 0
1
θ θ
which gives us
T1 = cEθ − c 2
As a result, individual incomes and utilities are given by the following expressions:
y1 (θ ) = (θ − c )2 (9.5)
1
u1 (θ ) = (θ − c )2 (9.6)
2
We are now in a position to do a vector comparison of welfare under the two
regimes, laissez-faire and Pigouvian taxes.
Proposition 1
Compared to laissez-faire, a Pigouvian tax regime is progressive, i.e., income and
utility gains of agents are decreasing in θ .
The proof is straightforward. Simply compute the differences y1 (θ ) − y0 (θ )

and u1 (θ ) − u0 (θ ) from the expressions derived above and check that they are
decreasing functions of θ . The intuition is as follows. Under a Pigouvian regime,
all agents pay for the damages they impose on others through productive activ-
ity, and other agents are compensated for their loss via the lump sum payout T1.
Therefore, the net loss due to externalities under this regime is zero for everyone,
as opposed to some positive uncompensated damage under laissez-faire. This gain
from switching to a Pigouvian tax regime is independent of an agent’s productiv-
ity. What remains is the profit and income loss on account of the tax. This is ob-
viously higher for more productive agents, which explains the result stated above.
Therefore, a revenue-neutral Pigouvian tax not only increases aggregate sur-
plus, but also concentrates gains at the lower end. In other words, one of its
by-products is redistribution of income and utility. This serves as a benchmark
for what follows. If the social planner can only observe income y, not the level
of economic activity x, it is still possible to implement the first-best allocation
through an appropriately chosen income tax schedule. However, I will show that
this necessitates even more redistribution than under Pigouvian taxes.
9.3.3 Mirrlees income taxation

Suppose x is not observable but y is. Can the social planner devise an income tax
schedule that will implement efficient levels of economic activity for all agents
(i.e., θ − c , as under Pigouvian taxes)? The answer to this question turns out to
be affirmative. To analyse the problem, I adopt the approach taken by Mirrlees
(1971), and look at truth-telling direct mechanisms. The social planner asks each
agent to report his productivity. Let θˆ be reported productivity. The mechanism
() ()
specifies a permissible (pre-tax) income y θˆ and a tax liability t θˆ for each θˆ.
156 Parikshit Ghosh
A mechanism has the truth-telling property if agents have the incentive to re-
port their productivity truthfully, i.e., θˆ = θ for all θ . By the revelation principle
(Myerson, 1982), for any mechanism that implements a certain allocation, there
is an outcome-equivalent truth-telling direct mechanism. If there exists such a
mechanism that can implement levels of economic activity achievable by Pig-
() ( )
ouvian taxes, then y θˆ = θˆ θˆ − c . The task is to find a function t (.) that will
achieve this end.
Before tackling that problem, it may be useful to attempt an approach that
will not succeed but will yield some intuition about the nature of the t (.) func-
tion that will bring about the desired allocation. What if the social planner sim-
ply asks for the Pigouvian tax for each level of reported productivity, θˆ, i.e., set
()
t θˆ = cθˆ − T1? The agent then solves
( ) 
2
 ˆ  2
1y θ 1  θˆ   ˆ
() () ( )
2
ˆ
max y θ −  ˆ   
θˆ 2 θ  − t θ ≡ 1 − 2  θ   θ − c
     
The solution to this problem gives us the agent’s best response to such a scheme:9
 c 2 c
θˆ = θ 2 +   + > θ for all c > 0
4  4
This shows that under a naive scheme where an agent’s tax liability is set equal
to what it would have been in a Pigouvian world, every agent will have an in-
centive to over-report his productivity. This suggests that if there is any hope for
implementing the first best in this environment, the tax liability has to increase
more steeply with productivity (and hence, income) than in a Pigouvian world,
i.e., it must be more egalitarian.
Proposition 2
() ( ) ()
There exists a direct mechanism y 2 θˆ = θˆ θˆ − c , t2 θˆ , and an implied income
tax schedule τ ( y ) , which implements first-best activity levels, θ − c . This Mir-
rlees mechanism is more egalitarian than the Pigouvian mechanism in that it
increases the burden on agents at higher levels of productivity and decreases the
2 () ()
burden at lower productivity levels, i.e., t θˆ − t θˆ is increasing in θˆ.
1
Faced with any mechanism y 2 (.) , t2 (.), the agent solves
( ) 
2
 ˆ
1  y2 θ
θˆ
()
max y 2 θˆ − 
2 θ
ˆ
 − t2 θ()
 
Using the first-order condition of this optimization problem, and imposing the
() ( )
desired income generation target y θˆ = θˆ θˆ − c as well as the truth-telling
2
condition θˆ = θ , we arrive at the following characterization of the tax liability

function t (.):
 c
t2′ (θ ) =  2 −  c > c since θ > c
 θ
Solving the differential equation gives us
t2 (θ ) = 2cθ − c 2 ln θ − T2 (9.7)
where T2 is the lump sum tax refund paid to the agents. Revenue neutrality
requires
θ
∫t 2 (θ ) f (θ ) dθ = 0
θ
which identifies the value of the lump sum refund:
T2 = c 2 Elnθ − 2cEθ
Deriving an income tax schedule, τ ( y ) , that is equivalent to this direct mecha-

nism is straightforward but not particularly instructive. Recall the following re-
lationship between income and productivity: y = θ (θ − c ). Inverting this function
and ignoring the negative root gives us
c + c 2 + 4y
θ=
2
Replacing this in (9.7), we obtain τ ( y ) .
The central message of this simple model runs somewhat counter to what
often emerges from the optimal taxation literature—there is no conflict between
equity and efficiency when correcting a disease externality is the goal of taxation.
Under such circumstances, allocative efficiency warrants redistribution. Several
related points deserve mention here.
First, the temporal structure of these tax-transfer policies need to be care-
fully designed. Income assessment by tax authorities takes time but due to credit
constraints, those suffering large losses need immediate relief. Therefore, the
tax credit needs to be paid upfront as a basic income for all citizens and tax dues
collected later. Governments must still be prepared to run temporary deficits,
perhaps over several months.
Second, by assuming a utility function linear in income, I abstracted away from
both income effects and risk aversion. Taking those factors into consideration will
reinforce the point that a basic income during a pandemic serves efficiency goals.
If income effects are strong, agents hit with large negative productivity shocks
158 Parikshit Ghosh
will increase economic activity to make up for low returns. The only way to dis-
suade them is through fiscal support. With concave utility of money, a dynamic
extension of the model could be fruitful. Risk averse agents would ideally like to
have income insurance against a Covid-like shock (and not knowing the exact
nature of the shock in advance puts them behind a Rawlsian veil of ignorance).
The need for insurance is unlikely to be fully met by precautionary savings or
the private insurance market (it is an aggregate, not idiosyncratic shock). There
should be good justification for social insurance, which means governments need
to build up a contingency fund by levying taxes even in normal times.10
Third, the pandemic tax being an incremental and temporary levy should per-
haps choose as its base not income but change in income during the pandemic year.
Tax authorities have some idea about baseline productivity of an agent or enter-
prise from past years’ income tax filings. Taxes and tax adjustments on that base is
a separate exercise. The ahistoric and static analysis presented here is an oversim-
plification. All of these considerations can be accommodated in the framework I
have sketched out. I have kept the model simple in order to convey the main idea.
9.4 Developing countries

The problem with implementing a well-targeted tax-transfer scheme in poorer
countries is the lack of state capacity for means testing. Most people are outside
the tax network and the state is unable to assess their household incomes with
any degree of accuracy. In India, for example, only about 7% of working-age
adults file income tax returns. Nevertheless, a crude version can perhaps be im-
plemented. Although income losses under the Covid-19 shock are not perfectly
correlated with income, in a developing country, the correlation is likely to be
high. The poor disproportionately work as daily labourers and self-employed
persons, and are likely to suffer most when the economy contracts. Therefore,
increased funding towards welfare programmes that are targeted at the poor in
normal times may do a good enough job. The problem is, there is typically a
bewildering variety of such programmes, so which should be chosen as a vehicle
for redistribution becomes a vexing question. The choice of delivery mechanism
may have significant effects on inclusion and exclusion errors.
The Indian government’s fiscal relief package chose a kitchen sink approach,
putting small amounts of funds injection into a multitude of schemes. The two
largest schemes—the Mahatma Gandhi National Rural Employment Guarantee
Act (MNREGA) and the Public Distribution System (PDS) for subsidized food
grains—were both roped in for this purpose. The former got a cash infusion of
Rs 40,000 crores, while the latter was funded to raise food rations from 5 kg to
10 kg per head. The government also put little dollops of cash into Jan Dhan
Yojana ( JDY) bank accounts for women, public pension schemes, credit lines
for self-help groups (SHGs), etc. To some extent, this approach has a portfolio
diversification advantage—it increases the probability that a poor family gets as-
sistance in some form or the other. However, especially given the relatively small
size of the actual fiscal stimulus (around 1% of GDP by most estimates, as opposed
to about 9% of GDP in the form of liquidity infusion—loans, interest deferrals,
guarantees, etc.), this strategy spread it out too thin.
The logistical problems that hobbled any serious attempts at cash transfer fol-
lowing the draconian lockdown in India demonstrate that the much touted fi-
nancial inclusion programme under the JAM trinity still has miles to go. The
need of the hour was a comprehensive database of uniquely identified non-rich
households with Aadhaar-linked bank accounts where money could be trans-
ferred without duplication or exclusion. Even in this less than ideal situation,
perhaps an opportunity was missed. It has been suggested (Borah et al., 2020;
Ghosh, 2020) that of all the myriad welfare schemes, the PDS, with its list of
unique ration card holders, which covers roughly the poorest two-thirds of the
population, should have been chosen as the conduit for cash transfers in addition
to in-kind assistance. A detailed analysis is beyond the scope of this chapter, but
it has been estimated that transferring enough cash to guarantee that PDS bene-
ficiaries remained above the poverty line for six months would have cost around
3% of GDP. This was money well worth spending.
Another problem in India is that much of production takes place through very
small firms—the micro, small and medium enterprises (MSME). This sector
accounts for 32% of GDP and employs 110 million people, according to govern-
ment data. These tiny firms do not have deep pockets and could be driven to
bankruptcy or closure if revenues stopped flowing even for a couple of months.
In 2017, however, the Indian government implemented a comprehensive national
value-added tax, the Goods and Services Tax (GST), which requires firms to re-
port monthly sales and costs. This high-frequency granular data can be used to
target tax credit at firms that actually suffer significant losses during lockdowns,
with upfront access to credit to keep them afloat before that help arrives.11
It is perhaps too late to rectify any of this but lessons learnt in hindsight are
still valuable for future crises. It is important to remember that these are not
permanent obligations of governments, like entitlement programs, but one-off
spending. That is an obvious point, but I suspect governments’ reflexes may not
always be well honed on that distinction. After all, their political accountability
extends over a short horizon, so they are loath to frontload costs. As citizens, our
interests are not quite the same.
9.5 Conclusion
When people express concern about ‘the economy’ in a public health crisis, I sus-
pect many are really worried about the share of the vulnerable population. They are
not losing sleep over empty airports but the livelihoods of airline employees. Nev-
ertheless, framing of a problem has an effect on what kind of solutions we look for.
Alarmist rhetoric about falling GDP and nose-diving stock markets implicitly di-
rects our attention towards efforts that will keep the pie from shrinking rather than
policies that will divide it better. We should not succumb to this framing effect.
160 Parikshit Ghosh
I have argued that one does not have to be a bleeding heart to appreciate the
critical importance of social safety nets and redistribution during a pandemic. A
compassionate society demands such measures but there are other reasons too.
Unless strong income protection is given to all citizens, it leaves a private need
and incentive to go out and participate in economic activity. The state lacks the
information and capacity to allocate economic contraction efficiently across sec-
tors, firms and individuals who have experienced widely different productivity
shocks. It can be done indirectly through taxes and transfers, but that requires
political will and fiscal resolve.
Notes
1 I have benefitted from conversations with Maitreesh Ghatak, Bhramar Mukherjee
and Debraj Ray. Vaibhav Ojha provided excellent research assistance. All errors are
my own. Research support from the Bill and Melinda Gates Foundation is gratefully
acknowledged.
2 Some of the desirable pattern of contraction or expansion will arise from uncor-
rected private incentives. For example, cruise lines are likely to see a sharp decrease
in business, while profit-oriented hospitals will expand ICU capacity in response to
demand. The production of masks and sanitizers saw a sharp spike soon after the out-
break, leading even to problems of quality control. However, in the presence of the
externality, the allocation of resources will typically not be first best.
3 A Pigouvian tax is usually targeted at polluting industries, and not applied across the
board. Its purpose is to change not the aggregate level but the composition of produc-
tion and consumption away from dirtier goods. For this reason, its effects cannot be
mimicked by a general income or sales tax under normal circumstances. However,
the externality generated by the pandemic is not specific to certain sectors like the
chemical or coal industry. It is this universal nature of the externality that makes it
relevant to explore the connection between taxing inputs and taxing income.
4 A few months into the pandemic, the financial press in India (for example, Roy,
2020) was abuzz that a generous fiscal stimulus will attract a downgrading of India’s
country rating, increasing the cost of borrowing in international capital markets. Set
aside the issue that credibility of rating agencies themselves should have been down-
graded after the 2008 financial crisis. Surely, if the government announced additional
(temporary) taxes to finance the fiscal expansion, the concerns of deficit hawks could
have been addressed.
5 The pandemic has perhaps also triggered a Keynesian recession which may outlast the
disease. Announcing new taxes during a recession is not the usual prescription—fiscal
stimulus is supposed to be deficit financed. However, postponing taxation beats the
purpose of both incentivising efficient contraction of economic activity and ensuring
distributive justice. During the pandemic, the relatively less fortunate should be sup-
ported by the relatively more fortunate. Imposing taxes after the situation normalizes
will evade this objective. The best the government can do is to assess special tax ob-
ligations based on incomes in the pandemic year but defer the dues perhaps to a later
financial year.
6 Sen (1981) estimates that in May 1943, the price of milk, fish and haircuts relative to
the price of rice in Birbhum district were 21%, 27% and 18% of their levels in Decem-
ber 1941 (p. 68, Table 6.5). Agricultural wages relative to the price of rice dropped
to 24% (p. 66, Table 6.4) of previous levels over the same period. Even this dramatic
price shift is an underestimate of the true distress since sellers of these commodities
lost in volume as well as price.
7 Chetty et al. (2020) estimate that in New York City between January and mid-April,
small business revenues declined 73% in the most severely impacted ZIP codes, but
only 13% in the least affected ZIP codes. Nationally, over the same period, employ-
ment rates fell by 37% among the bottom wage quartile and by 14% in the top wage
quartile. However, even among low wage earners, there was major geographical var-
iation. In the major cities, the decline was 70% in the most affluent areas and 30% in
the least affluent neighbourhoods.
8 The qualitative results presented below hold for any strictly convex cost function.
The quadratic form is chosen for ease of exposition.
9 The negative root arising from the first-order condition is naturally ignored for ob-
vious reasons.
10 See Ghosh and Ray (2021, forthcoming) for other, complementary reasons for the
creation of a sovereign wealth fund along the lines of resource-rich countries like
Norway and Saudi Arabia, and fiscal strategies to achieve that goal.
11 See Saez and Zucman (2020) for a similar proposal, though funded by debt,
not taxes.
References
Abedi, V., O. Olulana, V. Avula, D. Chaudhary, A. Khan, S. Shahjouei, J. Li and R.
Zand (2020): Racial, Economic and Health Inequality and Covid-19 Infection in the
United States. Journal of Racial and Ethnic Health Disparities, 8(3), 732–742.
Adams-Prassl, A., T. Boneva, M. Golin and C. Rauh (2020): Inequality in the Impact of
the Coronavirus Shock: Evidence from Real Time Surveys. Journal of Public Economics,
189, 104–245.
Barro, R., J. F. Ursúa, and J. Weng (2020): The Coronavirus and the Great Influenza
Pandemic: Lessons from the Spanish Flu for the Coronavirus’s Potential Effects on
Mortality and Economic Activity. NBER Working Paper #26866.
Borah, A., S. Das, A. Dasgupta, A. Deshpande, K. Mahajan, B. Ramaswamy, A. Saha and
A. Sharma (2020): Covid-19: Are We Ready for the Long Haul? Ideas for India, April
2. https://www.ideasforindia.in/topics/macroeconomics/covid-19-are-we-ready-
for-the-long-haul-i.html
Chetty, R., J. Friedman, N. Hendren and M. Stepner (2020): How Did Covid-19 and
Stabilization Policies Affect Spending and Employment: A New Real-Time Eco-
nomic Tracker Based on Private Sector Data. NBER Working Paper #27431.
Dewatripont, M., M. Goldman, E. Muraille and J.P. Platteau (2020): Rapidly Identifying
Workers Who Are Immune to Covid-19 and Virus Free Is A Priority for Restart-
ing the Economy. VoxEU, March 23. https://voxeu.org/article/rapidly-identifying-
workers-who-are-immune-covid-19-and-virus-free-priority-restarting-economy
Ferguson, N. et al. (2020): Impact of Non-Pharmaceutical Interventions to Reduce
Covid-19 Mortality and Healthcare Demand. Report 9 of the Imperial College Covid
Response Team. https://www.imperial.ac.uk/media/imperial-college/medicine/
sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-m odelling-16-03-
2020.pdf
Ghosh, P. (2020): We Need A Marshall Plan to Fight Covid-19. Ideas for India, April 8.
https://www.ideasforindia.in/topics/macroeconomics/we-need-a-marshall-plan-to-
fight-covid-19.html
Ghosh, P. and D. Ray (2021, forthcoming): A Sovereign Fund for India. To Appear in
Festschrift in Honor of Pranab Bardhan, Oxford University Press.
Lotka, A. (1910): Contribution to the Theory of Periodic Reaction. Journal of Physical
Chemistry, 14(3), 271–274.
162 Parikshit Ghosh
Mirrlees, J. (1971): An Exploration in the Theory of Optimal Income Taxation. Review of

Economic Studies, 38(2), 175–208.
Myerson, R. (1982): Optimal Coordination Mechanisms in Generalized Principal Agent
Problems. Journal of Mathematical Economics, 10(1), 67–81.
Ray, D. and S. Subramanian (2020): Covid-19: Is There A Reasonable Alternative To
A Comprehensive Lockdown? Ideas for India, March 28. https://www.ideasforindia.
in/topics/macroeconomics/is-there-a-reasonable-alternative-to-a-comprehensive-
lockdown.html
Roy, A. (2020): A Large Stimulus May Attract Sovereign Rating Downgrade. Business
Standard, May 9.
Saez, E. and G. Zucman (2020): The Crisis Calls for Massive Government Interven-
tion: Here’s How to Do It. The Guardian, March 17. https://www.theguardian.com/
commentisfree/2020/mar/17/governments-crisis-coronavirus-business#maincontent
Sen, A. (1981): Poverty and Famines: An Essay on Entitlement and Deprivation, Oxford Uni-
versity Press.
Volterra, V. (1926): Fluctuations in the Abundance of A Species Considered Mathemati-
cally. Nature, 118, 558–560.
Yancy, C. W. (2020): Covid-19 and African Americans. Journal of the American Medical
Association, 19, 1891–1892.
10
COVID-19 AND AN ASSESSMENT
OF SOME POLICY ACTIONS
ADOPTED BY INDIA
Hiranya Lahiri
10.1 Introduction
The Covid-19 pandemic has led to a plethora of fiscal and monetary measures
being adopted by almost all the affected countries in the world and India is no
exception. Much before the fiscal package that was announced by the Finance
Minister of the country during a five-day marathon, the Government of India
had already embarked two measures to combat the economic slowdown led by
Covid-19. These two measures were rate cuts and a hike in indirect tax rate.
While the first measure was aimed at boosting demand, the second measure was
expected to provide some cushion to the government to finance expansionary
programmes.
The objective of the chapter is to assess the efficacies of some of the policy
measures that have been adopted by the Indian economy to combat the Covid-
19-induced slowdown. And it does so by developing a short-run-open-economy
model that can emulate India’s growth.
As can be witnessed from Table 10.1 (in Appendix 10.A), India’s growth sud-
denly shot up in the year 2003–04, and it continued to maintain the trajectory
till 2010–11, barring the single year, 2008–09. However, growth rate slackened
thereafter. It is pertinent to mention that growth rates for the years 2013–14
onwards are over-estimation, since the base year has been revised from 2004–05
to 2011–12. Even overlooking this observation, it is clear that growth rates post
2011–12 have not been commensurate with the earlier phase of boom. Now, if
we delve deeper into data we can discern that the main driver of the boom and
bust is growth in Gross Capital Formation or investment (see columns 2 and 7
of Table 10.1). All the years of high (low) GDP growth were years of high (low)
growth rate of capital formation. This is all the more evident if we juxtapose the
observations that private consumption expenditure has been stable over these
DOI: 10.4324/9781003220145-12
164 Hiranya Lahiri
years and government’s expenditure has been either contractionary or neutral

(except for the year 2009–10, see column 8 of Table 10.1). Moreover, current
account deficit has either remained stable at around 2% for many years or had
widened sharply (except for the year 2009–10, see column 9 of Table 10.1), in-
dicating that net exports were clearly not the driver of growth. Thus, the main
driver of India’s growth has been investment.
Now, let us answer what caused the annual fluctuations in investment. Ta-
ble 10.1 attests the claim that years of high (depreciated) exchange rate were
years of low investment. And years of low (favourable/appreciated) exchange
rate were the years when net capital inflow was high. Let us explain. Investment
is extremely sensitive to exchange rate movements which in turn depend on
net capital inflow (primarily FPI flows, see columns 3, 4 and 5 of Table 10.1).
Exodus of capital puts pressure on the exchange rate which affects investment
detrimentally. India’s production of goods and services, and hence investment
are crucially dependent on import of intermediate goods. A rise in exchange rate
raises costs of investment, just as a rise in interest rate does. Further, Indian firms
are burdened with external debt, and a rise in exchange rate raises debt servicing
charges. Therefore, exchange rate depreciation, that is associated with net capital
outflow, decreases profitability of investment, raises external finance premium
and worsens net worth and asset quality of firms (we explain this part in greater
details in Section 10.2). As a result, exchange rate movement and investment
are inversely related. This explains why years of high investment favourable ex-
change were years of favourable (low) exchange rate.
Against this backdrop, this chapter shall try to serve two purposes. First, it
shall construct a theoretical short-run-open-economy model that can char-
acterise the present Indian economy. Second, it shall examine the effects of
some recently-embarked monetary and fiscal policies on domestic output, price
level and exchange rate by performing comparative static analysis. One salient
feature of this chapter is that it highlights the immensely important role of
exchange rate movement that plays a crucial role in yielding the results of this
model. The driver of the results of this chapter are akin to the one considered
in Leonardo (2014), where the author builds a New Keynesian model based on
the premise that exchange rate channels are critical to evaluating how exter-
nal shocks can affect monetary policy and the macroeconomic performances
linked to it.
Research works conducted on the macroeconomic effects of fiscal and mon-
etary policies embarked in India in the wake of Covid-19 are few. Chakraborty
and Thomas (2020) note that the Reserve Bank India (RBI) has adopted a
plethora of monetary and fiscal measures to combat the Covid-19-induced eco-
nomic crisis in India. Regulatory easing, widening of Monetary Policy Rate
Corridor, reduction in CRR and moratorium on loans are some of the mone-
tary measures adopted by RBI. But most important of all monetary measures are
rate cuts, manifested through reductions in policy rates. Among the fiscal meas-
ures, most notable measures include strengthening of PM Jan Dhan Yogana,
Assessment of some policy actions 165
Ujjawala Scheme, food-subsidy and liquidity infusion to ensure easy flow of

credit to small- and medium-scale enterprises (though the last measure actually
falls within the ambit of monetary policy, it was announced by the Treasury
Department). Ranjan and Chakraborty (2020) conduct an extensive survey of
growth estimates by various agencies in the awake of the Covid-19 situation for
all major developed and developing nations and conclude that there is a clear
trade-off between life and livelihood. They further conjecture that global co-
operation, and domestic macroeconomic policies with exit strategy are essential
to solve the economic disruptions in supply chains. Shah (2020) in his article
provides food for thought that can usefully be applied by the union government
and sub-national governments in promulgating new health policy to cater the
demand of Covid-19-infected patients. However, this chapter is bereft of mac-
roeconomic analysis on growth and output. Patnaik et al. (2020) conjecture a
5% decline in India’s GDP coupled with a 14.4% contraction in net tax revenue
and forecast India’s fiscal deficit to the tune of 6.2% for the year 2020–21. Thus,
we discern that the main focus of the research done on economic effects of
Covid-19 is merely a synopsis of policy actions adopted in India or forecasts of
macroeconomic variables. However, none of these studies provide any in-depth
analysis of the individual macroeconomic policy responses. This chapter aims at
discerning the possible effects of rate cut and hike in indirect tax on oil on major
Indian macro-variables.
Keeping the observations made about the Indian economy in mind, we
are now in a position to develop a macro-model that can assess the efficacy of
aforesaid policies. We proceed as follows. Section 10.2 develops a short-run-
open-economy model that can characterise the Indian macro-economy. In Sec-
tion10.3, we examine how effective rate cuts by the RBI would be to boost
aggregate demand and stimulate output. In this section, we also analyse how a
rise in indirect tax (excise duties) on petrol and diesel would affect the economy.
Finally, in Section 10.4, we conclude.
10.2 The model

The model we develop here is the standard open-economy-Keynesian model
with imperfect capital mobility as presented, in Lahiri et al. (2015) with suitable
modifications. We shall point to these modifications and give justifications for
them as we proceed. The model consists of a real and a financial sector. We con-
sider the real sector first.
10.2.1 The real sector

The output in our model is demand-determined and equilibrium in the real
sector is given as:
 M
 P
( ) ( )
Y = C (1 − t ) (1 − t ) Y ,  + I r , p, D , Y + G + NX P , Y , Y * , t (10.1)
166 Hiranya Lahiri
Apart from the standard symbols (Y , C , I , G, NX , Y * ), we define t and t as
direct and net-indirect tax rates, and M as nominal money held by households
(a proxy of household wealth). Now, p and D denote the cost of import of in-
termediate and capital goods that are used in India’s investment and short-run
eP *
external-debt-servicing charge of domestic firms, respectively. Here, p = ,
P (1 − t )
and D =
(
eD * 1 + r *) , where P *, D * and r * are foreign price (in dollar), short-run
P (1 − t )
external debt of firms (in dollar) and foreign interest rate respectively. The values
P * and r * are considered to be exogenous to the models as India is a small-open
country and, hence, a price-taker. Also, the value of eternal debt, denominated
eP *
in dollar (D *), is assumed to be fixed as we focus on short-run. P = is the
P
* *
real exchange rate. We assume P and Y to be given in the Indian context since
India is a small-open economy and cannot influence foreign price and foreign
GDP through its participation in world trade and capital flows.
Let us explain (10.1). It states that aggregate consumption is an increasing
M 
( )
function of disposable income (1 − t ) (1 − t ) Y and real money balance  . We
P 
ignore the share of foreigners in domestic income and consider them to be fixed
in the short run.
Let us explain the investment function. Since India practises interest-rate tar-
geting through the operation of LAF and MSF. The RBI attempts to confine the
call money rate within the LAF-MSF corridor, following a kind of Taylor rule
(see Taylor, 1993). Thus, to introduce this attribute, we assume interest rate is
fixed and a policy variable, denoted by r . Obviously, a rise in interest rate reduces
investment. We assume a flexible exchange rate here, as it may be more appropri-
ate at the current juncture. Now, the exchange rate movement affects Indian in-
vestment through two channels. First, India’s investment is heavily dependent on
import of intermediate and capital goods. A rise in exchange rate raises the cost
of these imports (p), lowers profitability of investment (given expectations) and
thus, lowers investment. Second, a large proportion of domestic firms have debt
denominated in foreign currency. A rise in exchange rate implies a larger portion
of output is to be used to service debt, which reduces return from investment,
and, thereby, discourages investment. Both these effects deter investment as they
worsen asset quality of firms, erode their net-worth, put pressure on firm’s mar-
gins and raise their external finance premium. When price of foreign goods and
debt servicing charge in domestic currency rise relative to the price that domestic
producers receive, there takes place profit squeeze and, consequently, erosion of
net worth. Thus, a rise in exchange rate lowers investment by raising p and D .
Lahiri (2015) argues that a rise in exchange rate can create the apprehension
that expansionary programmes will deteriorate BOP situation leading to further
depreciation of rupee, which in turn may raise the cost of production, derail
investment projects and induce capital flight and currency crisis. All these may
make the government and financial institutions nervous. Consequently, the gov-
ernment may sit on a pile of investment applications and the financial institutions
may strengthen credit standards, leaving a derogatory impact on investment. The
prospect of the above-mentioned scenario makes the investors extremely nerv-
ous too. Thus, BOP difficulties may derail investment through all the different
routes mentioned above. This explains the rationale of including p and D in the
investment function.
Note that we have deflated price of foreign goods in domestic currency (eP * )
( ( ))
and debt servicing charges eD * 1 + r * not by the domestic market price, but
( )
by the price received by the domestic producers P (1 − t ) . In calculating firms’
profit, what we have to take into consideration is not their total revenue in do-
mestic currency, but their total revenue quoted in domestic currency net of net
indirect taxes, as this is the price that is relevant for the domestic firms. The last
argument (Y ) in the investment function is standard.
The net-export function in (10.1) is standard too and it demands no explana-
tion except for the argument, t , regarding which we conjecture that consump-
tion of the relatively well-to-do section of society is highly import-intensive and
an increase in the direct tax rate will improve trade balance.
Now price level (P) is a positive function of t and e . A rise in exchange
rate raises the cost of production as India’s output and investments are highly
import-intensive (see RBI, 2012). Further indirect taxes constitute a very im-
portant source of tax revenue and imported intermediate inputs such as pe-
troleum and petroleum products, fertilizer, electronic components, etc. are
essential for production (GoI, 2012, 2013). Needlessly to say, a rise in t raises
the cost of production which is then passed on to the consumers. However,
without loss of generality, we ignore wages and output to determine price
level. Thus,
P = P (e, t ) (10.2)
We further conjecture that a rise in t or e though raise P, the adjustment in P is

imperfect due to oligopolistic interdependence among firms. The implication of
this assumption is that, following a rise in t , [P (e, t )(1 − t )] falls and an increase
in e leads to a rise in D andp .
Thus, we can write (10.1) as:
 M   eP * eD * (1 + r * ) 
Y = C (1 − t )(1 − t )Y ,  + I  r , , , Y  + G
     
P (e, t )   P (e, t )(1 − t ) P (e, t )(1 − t ) 
 eP * 
+ NX 
(10.3) , Y , Y * , t 
 P (e, t ) 
We can rewrite the above equation as
E = E (Y , e; t, t, M , G, r , Y * ) = 0 E = E − Y ⇒ E Y = EY − 1 < 0, Ee < 0 (10.4)

168 Hiranya Lahiri
Let us explain (10.4). An increase in Y raises disposable income, improves capacity

utilization and profit, and thereby raises both C and I. It, however, reduces NX.
In the net, we assume that, as standard, EY lies between zero and unity. Hence,
E Y < 0 . A rise in exchange rate lowers investment demand but improves net ex-
port. The latter improves mainly owing to the decline in investment (since, price
elasticity of both import and export is very low in India (RBI, 2012)), which is
attributable to very high import intensity of Indian export (RBI, 2012). Thus,
even though an increase in e raises NX in the Indian context, it is reasonable to
assume that the negative impact on investment significantly dominates the ex-
pansionary effect on net-exports. This explains the sign of the partial derivatives
of E and E with respect to e. This also supports the data portrayed in Table 10.1
(see column 1 in the light of movements of variables in columns 6, 7 and 9 over
the years). Finally, rises in t and t lower E and E , while a fall in r and a rise in
G, M and Y * raise E and E .
10.2.2 The financial sector

Let us discern how e is determined. The BOP consists of trade surplus (NX ) and
net inflow of foreign capital (F). Given the restrictions on capital inflow in India,
capital is imperfectly mobile. In open-economy macroeconomics literature, see,
for example Romer (2000), it is standard to assume that F depends upon interest
( )
rate differential r − r * and expected rate of currency depreciation. Now, in the
Indian context, it is prudent to assume that the expected rate of depreciation
crucially hinges on the ratings assigned by rating agencies (say, ρ ). To elucidate
with an example, in 2012–13, the Indian government attempted to introduce
General Anti-avoidance Rule and retrospective amendment to the Income Tax
pertaining to Indirect Transfer of Indian assets – both of which imply a rise in
indirect tax rate on foreign investors (τ ). Immediately credit rating agencies like
S&P, Fitch and Moody downgraded India, and that led to a huge exodus of capi-
tal from India and, consequently, exchange rate depreciation (from C49/$ in Feb-
ruary 2012, exchange rate rose to C56/$ in June and C55/$ in July, 2012). This
sent the government into jittery, and to heal the exchange rate, the government
lowered restrictions on foreign capital inflow (R) by announcing FDI in retail
(and promised some more similar reforms in the near future). This mortifies the
hapless dependence of the economy on foreign capital inflow, which in turn de-
pends upon sovereign rating.
The scenario outlined above clearly delineates the importance of the sover-
eign rating. It follows from the above that their rating depends upon govern-
ment’s attitude towards foreign investors as reflected in the tax rate that it seeks
to apply on foreign investors’ income, the level of restrictions it puts on foreign
investments. Using the logic outlined above, we can write ρ to be a decreasing
function of τ and R. Finally, there may be many exogenous factors that may
influence F. We therefore incorporate an exogenous variable ∅ as a determinant
of F. An increase in ∅ is assumed to imply an improvement in the perception
of foreign investors regarding India for extraneous reasons that are beyond eco-
nomic underpinning (for example changes in political regime, depressed business
sentiments in foreign countries, epidemics, etc.). Thus, BOP equilibrium can be
written as:
 eP * 
NX  ( )
, Y , Y * , t  + F [ r − r * , ρ (τ , R ) , ∅ ] = 0 (10.5)
 P (e, t ) 
From (10.5), we can write:
( )
B Y , e; t, t, r , τ , R, ∅, Y * = 0 (10.6)
where BY 0, Be 0 and Bt , B r , BY * , B∅ > 0; Bt , Bτ , BR < 0 .

Let us now explain (10.6) in the light of (10.5). A ceteris paribus rise in Y
decreases net exports by raising imports and thereby lowers B. Again, an in-
 eP * 
crease in e raises real exchange rate   . As a result, net exports and BOP
 P (e, t ) 
go up. An increase in t reduces NX by making domestic goods costlier and,
therefore, reduces B. A rise in τ or R discourages capital inflow by worsening
credit rating, and hence reduces B as well. Finally, a rise in foreign GDP or a
rise in ∅ raises B.
The specification of our model is now complete. It contains two key equa-
tions, viz., (10.4) and (10.6), in two unknowns, namely, Y and e. Hence, the
model is exactly determined. The solution of (10.4) and (10.6) is shown in Figure
10.1 where e is measured on the vertical axis and Y on the horizontal axis. The
lines YY and BB represent (10.4) and (10.6) respectively in Figure 10.1. The
equilibrium corresponds to the point of intersection of the two lines.
This model is completely capable of explaining year–to-year fluctuations in
India’s GDP and other macro-variables (see Lahiri et al. (2015) for a detailed
analysis on how this model explains yearly and quarterly fluctuations in India’s
GDP, investment, exchange rate, etc. from 2002 to 2003 onwards).
e
YY BB
FIGURE 10.1 Determination of Y and e.

170 Hiranya Lahiri
10.3. Comparative static results

We shall carry out some comparative static results that appear to be pertinent in
the present context for the Indian economy. First, we shall see how a rate cut
affects the Indian economy. Second, we shall see how a rise in indirect tax on
oil would affect the macro-variables of our model. Both these comparative static
results are extremely important in the present juncture, as we have seen the RBI
to pursue rate cuts very recently to combat recession, as well as the Government
of India to revise upward excise duties on petrol and diesel to provide them with
fiscal space to pursue expansionary programmes.
10.3.1 Effect of a rate cut

The RBI had revised downwards the LAF-MSF corridor in the aftermath of
Covid-19 pandemic initially on March 27, 2020, by reducing the repo rate
by 75 basis points from 5.15% to 4.40%. On May 22, 2020, the Reserve Bank
of India once again reduced the repo rate from 4.4% to 4%. Given that eco-
nomic slowdown is a greater evil compared to inflation in the present sce-
nario, the RBI is expected to maintain low interest rate regime for the time
being, not denying further rate cuts in the future. Let us see how it affects
our model.
Following a cut in r , there will be an increase in E, given Y , e and other exog-
enous variables, as can be seen from (10.4). Thus, as follows from (10.4), the goods
market will be in equilibrium at a higher e, given Y or at a higher Y , given e.
Accordingly, YY in Figure 10.1 will shift to the right. Again, as evidenced
from (10.6), following a cut in r , BOP worsens. Hence, exchange rate will de-
preciate and BOP equilibrium be at a higher e, given Y or lower Y , given e.
BB in Figure 10.1, therefore, shifts rightwards too. The situation is shown in
Figure 10.2, where the new BB and YY schedules are labelled BB1 and YY1
respectively. We see that the effect on Y is ambiguous. The really scary situation
arises, if the vertical shift in BB is larger than that in YY, as shown in F
igure 10.2,
where the end result is aggravated recession and depreciated exchange rate cul-
minating into higher inflation, or, what we call in economics, stagflation.
e
YY YY1 BB1 BB
FIGURE 10.2 Effect of a rate cut.

We further argue that ceteris paribus, if export is less sensitive to exchange

rate and net inflow of foreign capital is sensitive to interest rate differential –
which are certainly likely to be satisfied in the Indian context – then a cut in r
will lower Y . In what follows, we shall try to prove the result intuitively (see
Appendix 10.B).
Following a decline in interest rate, net capital inflow declines along with
net exports, owing to the increase in import demand, which in turn is at-
tributable to a rise in investment demand, led by interest cut. The resulting
BOP deficit depreciates the exchange rate. Let us now examine how a rise in
exchange rate, given output, restores equilibrium in the goods market, which
is now in a state of excess demand. This is because the increase in imports (led
by a spur in investment) is just a fraction of the rise in investment demand.
When exchange rate sensitivity of exports is very low and hence exports do
not increase much due to depreciation of the currency, excess demand is re-
duced by a consequent decline in investment. As we have noted earlier, a rise
in exchange rate reduces investment demand in India, which leads to a decline
in aggregate demand. Thus, at the higher e that equilibrates the goods market,
with Y at its initial equilibrium level, export is likely to be a tad above its
initial equilibrium value, investment only a little lower than its initial equilib-
rium level and import quite close to its initial equilibrium value. The reason for
import being close to its initial equilibrium value is the following. As substitu-
tion possibilities between imported goods and domestic products are extremely
limited in the short run (which is reflected in the very low price elasticity of
Indian imports (RBI, 2012)), import demand is, given Y , determined mainly
by volumes of investment and export demand. If at the e , referred to above,
investment is slightly less than its initial equilibrium value and export is just a
little higher than its initial equilibrium value, imports will be close to their ini-
tial equilibrium value as well, and this scenario is consistent with goods market
equilibrium. Thus, when investment is highly sensitive to e, but export is quite
insensitive and import (given Y ) changes mainly due to changes in investment
and exports; net export will be only slightly larger than its initial equilibrium
value in the good market, at the above-mentioned e. But F at this given e is
likely to be much less than its initial equilibrium value because of the cut in
rate of interest, as it is sensitive to interest rate differential. Thus, with Y at its
initial equilibrium value, BOP will be in deficit at the higher e that restores
equilibrium in the goods market. Therefore, with Y at its initial equilibrium
value, BOP equilibrium will occur at an e , which is much higher than the e
that equilibrates the goods market. This implies that, following a rate cut, the
vertical shift in BB will be larger than that in YY, as shown in Figure 10.2,
spelling out disastrous consequences. Thus, in the Indian context, a rate cut
by the RBI may lead to a decrease in the growth rate as well as a rise in the
exchange rate and rate of inflation.
This result in our model that we have just intuitively developed is widely dif-
ferent from the standard case, where a rate cut spurs investment, and exchange
172 Hiranya Lahiri
rate depreciation expands exports; thus, both the forces act to raise the effective
demand and output.
But India’s case is completely different from the standard one. India is depend-
ent on the Western world for its import of capital goods and technology to re-
main competitive. Thus, India’s domestic output is intensive in the use of foreign
components that are used as intermediate goods on a wide spectrum. Examples
include metals, petroleum and petroleum products, fertiliser, etc. Further, much
of the corporate sector is burdened with external debt, and thus a rise in exchange
rate dampens business sentiments and depresses investor’s morale through vari-
ous routes. First, it raises cost of imports. Second, it eats into firm’s profit as full
pass-through is not possible. Third, it raises debt-servicing charges and reduces
profitability of investment. Hence, we argue that investment is inversely related
to exchange rate. Since India’s domestic production is highly dependent on im-
ported intermediate goods, exports, too, are highly import-intensive. Therefore,
exchange rate depreciation raises supply price of exports, though incompletely.
As a result, Indian exports show little sensitivity to changes in the exchange rate.
For the same reason imports, too, respond less to ceteris paribus changes in the
exchange rate. The RBI (2012) estimates that the elasticity of India’s non-oil
exports to REER is only 0.4 with a lag of one year. This is attributable to the
high import intensity of India’s exports and slower supply responses to price
changes. Further, the same study estimates the price elasticity of non-oil imports
to be statistically insignificant while the price elasticity of POL imports is only
0.1. This quote lends evidence that the effect of exchange rate depreciation on
India’s exports and imports is very low. Therefore, the standard Marshal-Lerner
effect is unlikely to hold in the Indian context. A rise in exchange rate improves
BOP mainly through its investment-depressing effect, which reduces imports.
Therefore, in the Indian context, a rate cut would aggravate recession instead of
proving respite from it.
10.3.2 Effect of a rise in indirect tax

The Government of India hiked excise duties on petrol and diesel by C10 and
C13 per litre respectively with effect from May 2020. The additional revenue to
be generated would be spent on infrastructure development and items on devel-
opment purpose.1 In our model, this amounts to a rise in t . Let us now examine
how an increase in t is likely to affect our model. Corresponding to a rise in t
, E falls corresponding to any given (Y , e), ceteris paribus. Therefore, the goods
market will be in equilibrium at a lower Y , for any e. This implies YY in Figure
10.3 shifts left. The new YY is represented by YY1 in Figure 10.3. A rise in t
also creates a BOP deficit corresponding to any given (Y , e). The BOP, as follows
from (10.4), will be in equilibrium at a lower Y , for any e. Thus, BB in Figure
10.3 will shift upward. The new BB is represented by BB1 in Figure 10.3. In the
new equilibrium, therefore, Y will be less unambiguously, but e may change in
either direction. However, we shall focus on the case where e becomes higher in
the new equilibrium, as it fits the Indian context.
YY BB1
YY1 BB
A1
FIGURE 10.3 Effect of an increase in t .
We can discern from Figure 10.3 that e will rise if the horizontal shift of
YY is lesser than that of BB (see Appendix 10.B). It turns out that if marginal
propensities to save and invest are sufficiently high, AD does not sufficiently fall
due to a rise in t and foreign capital inflow is not overtly responsive to t , then
the horizontal shift in BB will be larger than that in YY. Intuitively, an increase
in t creates BOP deficit by making domestic goods dearer. However, it also
directly reduces demand for Y by reducing consumption and investment. The
consequent fall in Y will provide some respite to BOP pressure. However, if this
respite is not strong enough there will continue to remain BOP pressure. Thus,
to ensure BOP equilibrium, e will rise. The rise in e will crowd out investment,
putting further downward pressure on Y . Thus, a rise in t will deepen recession
and in turn lead to stagflation (through an increase in price level, brought about
by depreciated exchange rate).
10.4 Conclusion
In order to fight the recession caused and aggravated by Covid-19 pandemic, In-
dia has embarked some monetary and fiscal measures to revive her GDP. In this
chapter, we have assessed two of such measures, viz., effect of rate cut and effect
of a rise in indirect tax on oil. Contrary to standard results, we have illustrated
that a rate cut will not only deepen depression in India, but will also lead to
stagflationary outcomes. Similar result is also derived in the case where we ana-
lyse the effect of a rise in indirect tax rate. Thus, instead of expecting the Indian
economy to bounce back to a trajectory of high growth, we can expect a further
decline in GDP, coupled with some inflation.
The correct measures in this trying time would be to dole out income subsidy
to those who have lost their incomes due to Covid pandemic, financed by print-
ing money. Also, an increase in government expenditure financed by tax can be
thought of, that will stay put our YY curve in the model, but will shift to the
right the BB curve, thereby increasing output. We shall address these research
questions in our future endeavours.
174 Hiranya Lahiri
Note
1 If we introspect the fiscal measures adopted by the GoI to combat the Covid-19-
induced slowdown, we shall see that the magnitude of ‘true’ fiscal expansion is only
1.6% of India’s GDP and not 10%, as announced. The rest 8.4% is actually liquidity
measures. Now it is clear that liquidity measures will not have demand-enhancing
effect, especially in an era of muted profit and sale expectations. Thus, the scope of
expansionary effect on output is limited.
However, even if we accommodate the case of a balanced budget, it is noteworthy
that still we might get the case where the recessionary effect on consumption, invest-
ment and exports might be stronger than the expansionary effect of a rise in G. Note
that the standard result where a rise in tax under balanced budget condition leads to
an unambiguous expansion in output may not hold good here, as even though the net
effect of the decline in consumption and rise in government expenditure is positive,
there are two additional contractionary forces: a decline in investment and exports.
Given higher sensitivity of investment to price received by producer (which is differ-
ent from domestic price level) and higher sensitivity of export to domestic price, the
end result is once again likely to be deepened recession. Note that exchange rate will
unambiguously be depreciated leading to inflation.
References
Chakraborty, L., and Thomas, E. (2020). COVID-19 and Macroeconomic Uncer-
tainty: Fiscal and Monetary Policy Response. EPW, 55(15). https://www.epw.in/
journal/2020/15/commentary/covid-19-and-macroeconomic-uncertainty.html
Government of India. (2012). Union Budget of India 2012–13. Ministry of Finance (GoI).
Government of India. (2013). Press Note No. 6 (2013 Series). DIPP (Ministry of Com-
merce & Industry).
Lahiri, H., Ghosh, A.N., and Ghosh, C. (2015). India’s Recent Macroeconomic Per-
formance and Foreign Capital Inflows. Contemporary Issues and Ideas in Social Science,
March(25), 1–54.
Leonardo, V. (2014). The Simple Post-Keynesian Monetary Policy Model: An Open
Economy Approach. Review of Political Economy, 26(4), 526–548.
Ranjan, D., and Chakraborty, L. (2020). COVID-19: Global Diagnosis and Future Policy
Perspective (NIPFP Working Paper No 304). National Institute of Policy and Public
Finance. https://nipfp.org.in/media/medialibrary/2020/05/WP_304_2020.pdf (ac-
cessed on 26/02/2021).
Reserve Bank of India. (2012). The External Sector. Macroeconomic and Monetary Devel-
opment (2012–13 Second Quarter Review). RBI (Mumbai).
Romer, D. (2000). Keynesian Macroeconomics without the LM curve. Journal of Economic
Perspectives, 14(2), 1449–1469.
Shah, A. (2020). Responding to the New Corona Virus: An Indian Policy Perspective (NIPFP
Working Paper No 309). National Institute of Policy and Public Finance. www.
macrofinance.nipfp.org.inMarch2020.pdf (accessed on 26/02/2021).
Taylor, J.B. (1993). Discretion versus policy rules in practice. Carnegie-Rochester Series on
Public Policy, March 2015, 195–215.
APPENDIX 10.A
TABLE 10.1 Selected macroeconomic indicators from 2000–01 to 2018–19
Year Growth Rate of Net FDI (US $ Net Portfolio Total (US $ Annual Rupee- Gross Capital GFD (% of CAD (As % of
(1) GDP at Factor Million) Investment (US Million) Dollar Exchange Formation (% GDP) GDP)
Cost (3) $ Million) (5)=(3)+(4) Rate Growth) (8) (9)
(2) (4) (6) (7)
2000–01 4.3 3,272 2,590 5,862 45.68 0.0 9.2 2.0

2001–02 5.5 4,734 1,952 6,686 47.61 7.4 9.6 1.9
2002–03 4.0 3,217 944 4,161 47.69 6.8 9.3 1.7
2003–04 8.1 2,388 11,356 13,744 48.40 13.6 8.3 2.1
2004–05 7.0 3,713 9,287 13,000 45.95 20.7 7.2 2.7
2005–06 9.5 3,034 12,494 15,528 44.93 16.2 6.5 3.9
2006–07 9.6 7,693 7,060 14,753 44.27 13.8 5.1 3.1
2007–08 9.6 15,893 27,433 43,326 45.25 16.2 4.0 4.8
2008–09 6.7 22,372 −14,030 8,342 40.26 3.5 8.3 8.6
2009–10 8.4 17,966 32,396 50,362 45.99 7.7 9.3 0.6
2010–11 8.4 11,834 30,293 42,127 47.44 11.0 6.9 3.8
2011–12 6.5 22,061 17,170 39,231 45.56 12.3 7.8 3.6
2012–13 5.46 19,819 26,891 46,711 47.92 0.8 6.9 3.6
2013–14 6.39 21,564 4,822 26,386 54.41 −5.25 6.7 4.8
2014–15 7.41 31,251 42,205 73,456 60.50 6.13 6.7 2.6
2015–16 8 36,021 −4,130 31,891 61.14 7.04 6.9 1.1
2016–17 8.17 35,612 7,612 43,224 65.47 5.84 7.0 0.7
2017–18 7.17 30,286 22,115 52,401 67.07 12.87 6.6 1.9
2018–19 6.71 30,712 −618 30,094 64.45 – 5.9 2.1
Source: Compiled from DBIE, RBI.

176 Hiranya Lahiri
APPENDIX 10.B
10.B.1 Effect of a cut in r
Taking total differential of (10.4) and (10.6) in the text, and setting dr = −1, we
get the following system of equations:
−E Y Br + E r BY
de = > 0 (10.A.1)
−E Y Be + E e BY
−E r Be + E e Br
dY = ≥< 0 (10.A.2)
B −E r
Now, dY < 0 , when −E r Be + E e Br < 0, or r > , (10.A.3)
Be −E e
⇒Vertical Shift of BB > Vertical shift of YY.
We shall now argue that in the Indian context (10.A.3) is likely to be satisfied.
Crucial to this claim is the property that price elasticities of Indian imports and
exports are very low (RBI, 2012) and hence, following an increase in r or e, im-
ports fall, given GDP, mainly on account of the decline in investment. Now, let,
mI and m X denote import intensities of investment and export. So,
NX e = X PX PX e − mI I P Pe − m X X P PX e = (1 − m X ) X P PX e + mI (−I P )Pe > 0 (10.A.4)
NX r = mI I r < 0 (10.A.5)
−E r = −I r − NX r = (1 − mI ) (−I r ) > 0 (10.A.6)
−E e = (−I P ) Pe − NX PX PX e  = ((1 − mI ) (−I P ) Pe − (1 − m X ) X P PX e  > 0 (10.A.7)
Equation (10.A.7) is positive, as in India, as we have already mentioned, X P PX e is

very small (since PX e is very small), while the other term is likely to be relatively
quite large.
Further, we have
Br = NX r + Fr = −mI I r + Fr > 0 (10.A.8)

Be = NX P Pe + mI (−I P )Pe > 0 (10.A.9)
From (10.A.6) and (10.A.7), we obtain

−E r
=
(1 − mI ) (−I r )
−E e ((1 − m ) (−I  ) P − (1 − m ) X  P  (10.A.10)
 I P e X P Xe 
Again, from (10.A.8) and (10.A.9), we obtain:
Br −mI I r + Fr
= (10.A.11)
Be NX P Pe + mI (−I P )Pe
Let us now compare (10.A.10) and (10.A.11). Note that the numerator of (10.A.11)
is larger than that of (10.A.10), but so is the denominator. So, if, given everything
else, Fr is sufficiently large and NX P Pe is sufficiently close to zero, (10.A.11) will
be larger than (10.A.10). Since, Pe is sufficiently small in the Indian context (see
RBI, 2012), this latter condition is likely to be true. The former condition will
also hold since RBI believes that the net inflow of foreign capital is sufficiently
sensitive to interest rate differential. As a result, following a cut in interest rate,
GDP is likely to fall in India.
10.B.2 Effect of a rise in t

Taking total differential of (10.4) and (10.6) and solving for the endogenous var-
iables, we obtain:
E t Be − E e Bt
dY = < 0 (10.A.12)
E Y Bt − E t BY
de = ≥< 0 (10.A.13)
−Bt −E t
Now, de > 0 ⇒ >
−BY −E Y
Or, Horizontal Shift of BB > Horizontal Shift of YY
⇒
−NX P Pt
>
( )
C ′ {(1 − t )} + {− (I p pt )} + −NX P Pt
(10.A.14)
NX Y
{ ( ))}
 1 − C ′ (1 − t ) 1 − t + I + (−NX )
( Y Y 
From the above we conclude that if saving and investment propensities with re-
spect to Y are sufficiently large and the depressionary effect of an increase in t is
weak, then e will increase following an increase in t .
11
IMPACT OF THE CORONAVIRUS
PANDEMIC ON DE-GLOBALIZATION
Ranajoy Bhattacharyya, Anupriya Gangopadhyay
and Abhilasha Pandey
11.1 Introduction
The world witnessed two major events that had significant impact on merchan-
dise trade among countries. The first event corresponded to the period of the
Great Depression from 1929 to 1932. Just after the end of the Second World
War, global trading activities witnessed a decline of more than 65%. This was
true both for the world and for the vast majority of countries for their individual
imports and/or exports (Van Bergeijk et al. 2017). The second round of de-
globalization took place in the year 2007–08. Similar to the Great Depression,
real world trade collapsed as much as by 11% during this period (Baldwin 2009).
Though both the events led to a decline in world trade, significant differences
can be observed between the two situations. In the case of the Great Depression,
poor economic condition and tendency towards protectionism were the main
causes of the decline in trade. Whereas the housing bubble in the USA was the
cause of the Financial Crisis. As mentioned by Van Bergeijk et al. (2017), the
maximum impact of the Second World War on world trade was observed three
years post-event but the effect on world trade had reached its worst within a year
in the case of the Financial Crisis. Another striking difference was observed re-
garding the recovery. Recovery in the case of the latter crisis was faster than the
recovery from the Great Depression.
A third incident of trade collapse possibly happened during the initial phase
of the ongoing coronavirus pandemic. A lockdown was imposed in most coun-
tries as an immediate response to the pandemic. These drastic measures were
absolutely necessary to curb the progress of the disease. However, it completely
crippled economic life including port and transport activities all over the world.
Though monthly trade data on the exact level of impact of the lockdowns are
not available at the time of writing this chapter, anecdotal evidence suggests that
DOI: 10.4324/9781003220145-13
Coronavirus pandemic and de-globalization 179
during March and April 2020 world trade almost came to complete halt. The
Organisation for Economic Co-operation and Development (OECD) estimates
the fall in international trade in 2020 to 9.2%.1 Interestingly, it also predicts that
trade will recover by 7.2% in 2021 implying that the collapse will be comparable
to the collapse during the financial crisis. This estimate assumes that the pan-
demic will not trigger a recession, as many organizations predict.2 If this happens
then, of course, recovery will take time.
In the interim period between the Financial Meltdown of 2008 and the coro-
navirus pandemic of 2020, the world economy was in a prolonged period of
recession and performance of trade was much less than what was the norm prior
to 2008 (Figure 11.1). A second phase of decline took place between 2014 and
2016. Also, the average rate of growth of world exports post-crisis was slower
and performance extremely erratic. The performance of global trade during this
period has been attributed to the real business cycle ( James, 2018) implying that
the ongoing recession was the causal factor behind it.
Similar to trade in goods and services, post-financial crisis, trade in invest-
ments had also retracted, significantly (Figure 11.2). When compared to trade in
goods, investments have suffered far more due to the financial crisis. Investment
constituted 5.4%, 2.3% and 1.6% of the total world GDP in 2007, 2009 and 2018,
respectively. The share of FDI in world GDP and flow of investments across
countries have been experiencing a constant decline over the years post-crisis.
Countries have become more stringent and impose a greater degree of regula-
tions on investments to cushion themselves from the aftermath of any forthcom-
ing crisis.
As it often happens during sustained periods of underperformance of trade
and investment, countries turned to protectionist measures during this period.
Though the USA and China occupied the centre stage and caught the maxi-
mum attention of the public in this protectionist rhetoric since 2018, they were
by no means the only countries that participated in this protectionist game.
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Exports
FIGURE 11.1 World exports (2001–19).

Source: Authors’ calculations based on Trade Maps Database.
180 Ranajoy Bhattacharyya et al.
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
FDI
FIGURE 11.2 World FDI (2001–19).

Source: Authors’ calculations based on World Bank Database.
The average tariff increased from 3.48% in 2012 to 3.67% in 2016 and the total
number of non-tariff measures (NTMs), like SPS, TBT, countervailing duties,
anti-dumping duties, etc., increased over the years from 5,088 in 2012 to 6,130
in 2016, with China being one of the primary targets (270 in 2012 to 580 in
2016). Similarly, the number of regional trading agreements (RTAs) signed in
a particular year has also experienced a significant reduction over the years. For
instance, between 2012 and 2019 the maximum number of RTAs were signed in
the year 2013 (13) and the minimum number of RTAs (2) were signed in 2019.
There was a general trend of decline in such policies over the period as protec-
tionist instincts of governments took over. Finally, a new dimension was added to
the restriction on the flow of goods during this period: restriction on the flow of
immigrants. This new dimension took the discourse beyond protectionism to the
wider perspective of de-globalization. After almost four decades of continuous
globalization, the world appeared to be losing some its hard-earned “flatness”.
Given the above perspective, the purpose of this chapter is to analyse the likely
consequences of the ongoing coronavirus pandemic on the de-globalization dis-
course. Since the quantitative impact of the pandemic cannot be analysed now due
to paucity of data, we instead analyse the impact of the pandemic through the gov-
ernment policies announced during the pandemic period. Do these government
policies have an inward bias? Or do they reveal a shift in the economic thinking of
policymakers towards tighter tariff regimes in the years to come? These are some of
the issues posed and sought to be answered in a limited scale in the rest of this chapter.
As countries started emerging out of complete lockdown (though the pan-
demic progressed unabated), almost all governments announced a set of stimulus
packages to kick start their economies.3 One purpose of these packages was to
help industries open up and give them some relief in the form of tax concessions,
working capital requirements and loan availability so that they can tide over their
loss during the lockdown period. The second objective was to initiate govern-
ment transfers to allow low-income consumers to survive the pandemic and its
aftermath. In some cases implicit in these measures, however, is a fundamental
shift in the ways governments’ thoughts have evolved during the course of the
pandemic. Nowhere is this implicit shift more apparent than in India. India has
been pursuing an agenda of lowering government regulation in industries, tar-
iffs and foreign direct investments since the 1990s. The pace of this agenda has
increased manifold in the last few years of the current government that came
to power in 2014. So it was a surprise to many when, in complete contrast to
the ongoing process, the Indian government dubbed its coronavirus package
“self-reliant India” (Atmanirbhar Bharat). Given India’s long history of import
substitution and dependence on domestic demand as the key driver to economic
growth, one clear way of interpreting the tagline is to take it as an indicator of
future inward orientation – a point that should clearly interest us for this project.
We, therefore, focus on the coronavirus package announced by the Government
of India to find its possible trade policy implications. The implications are then
mapped to possible future trade policy outcomes through the well-established
trade policy results in their politico economic perspective.
The rest of the chapter is arranged as follows. The next section briefly de-
scribes the pre-pandemic trade policy scenario prevailing in the world. Section
11.3 analyses the set of packages announced by the Indian government imme-
diately after the initial phase of lockdown in India. A general discussion on the
nature of the package is followed by a more detailed discussion on the trade-
related policies and their possible outcomes. Section 11.4 attempts to find a link
between pandemic packages and globalization and evaluates the implication of
these policies in the context of the protection for sale model and its extension to
the level of government negotiations and hence its possible implications for trade
policy ahead. Section 11.5 concludes the chapter.
11.2 Pre-pandemic protectionism

Pre-pandemic de-globalization was not a large-scale phenomenon. Several
countries took a series of policies since 2018 that gradually tilted the trade policy
scenario towards protectionism. To understand the nature of these product level
and targeted policies, we give some examples below.
Canada has given retaliatory tariffs worth $16.6 billion, 25% tariff on unfin-
ished iron and steel products. European Union has challenged US tariffs at WTO
and has implemented safeguards to prevent steel and aluminium redirected from
the USA from distorting EU market and EU producers. India has filed a formal
dispute against the USA at WTO, imposed a higher tariff on goods worth $1.48
billion US exports (5.5% of US exports to India in 2017). Fruits and nuts were
the top products affected. In response to the imposition of US tariffs, Mexico
implemented retaliatory tariffs on around US$3 billion (MXN $58.6 billion)
worth of US goods. These Mexican tariffs, which went into effect on June 5,
2018, were imposed on US steel, pork, cheese, whiskey and apples, among other
goods before being lifted on May 20, 2019. Also, preferential duty treatment is
suspended for goods originating in the USA, as well as the modification of the
duties applicable to imports with rates ranging from 7% to 25% for a few tariff
codes and from 10% to 15% for a few others, which are effective as of June 5,
2018, with some exceptions. Russia has imposed additional tariffs of 25%, 30%,
35% or 40% on goods originating from the USA. Turkey proposed suspension
of substantially equivalent concessions and other obligations under GATT 1994
to the trade of the USA which takes the form of an increase in tariffs on selected
products originating in the USA. This came in the form of an increase up to
$1.78 billion by an additional duty of 5–40% from June 21, 2018 and an increase
in certain duties by 4% to 140% from August 15, 2018.
The main set of policies that attracted the attention of people were those
undertaken by the USA and China (Table A.11.1 gives a timeline of these pol-
icies). Some of these policies include global safeguard tariffs on $8.5 billion in
imports of solar panels and $1.8 billion of washing machines, 25% tariff on 1,333
products worth $50 billion and 25% tariffs worth $42 billion on the grounds of
unfair trade practices for technology and intellectual property, 25% tariff on steel
imports worth $10.2 billion and 10% on imports of aluminium worth $7.7 billion
under national security grounds, on all trading partners, a new set of tariffs on
additional $450 million worth of steel and aluminium products, an additional
10% tariff on $200 billion import good, which constitutes of intermediate goods
like computer and auto parts and consumption goods like telephone equipment,
furniture, lamps, luggage etc.
Though these policies were specific and targeted to a small number of prod-
ucts, the protectionist orientation is clear. Countries were actively using trade
policies to offer protection to their import-competing sectors and to protect
domestic jobs. While media attention was focused on the US-China trade war,
it was only a small part of the actual picture. Almost all countries were silently
pursuing what the USA was pursuing. Contrary to say, a decade earlier many
countries simultaneously were practising protectionism. It is in this sense that we
can term it as a legitimate reversal of the globalization process.
11.3 India’s pandemic package: a brief review

As we have noted in the introduction, the term self-reliance (atmanirbhar) is a
well-known term in India since its independence. One interpretation of the term,
especially to trade economists, is a strategic under-emphasis of trade openness (es-
pecially, emphasis on import substitution) as a means of economic growth. India
had essentially discarded this notion of economic progress since the economic
reforms process began in the early 1990s, shifting to a more open (lower tariff
rates) deregulated market-driven model of economic growth. And the process
was accelerated since the late 2010s by the present government. From import sub-
stitution, the discussions have shifted to greater integration with the global value
chain and an emphasis on foreign direct investment as a strategy of economic
growth. It was therefore surprising to many as the Prime Minister announced a

rescue package of C20 lakh crores for the India economy during the Covid lock-
down period with the slogan of self-reliance. Faced with a volley of questions on
whether we are going back to the old strategy, the Principle Economic Advisor
to the Government of India later came out with a clarification of its fresh use:
“decentralised localism that takes pride in local brands, emphasises resilience and
flexibility, and encourages local capacity-building and indigenisation”.4 It can be
alleged that this novel interpretation retained the import substitution flavour of
the old one (“takes pride in local brands” through “local capacity building and
indigenisation”) but was silent on some other features of the old concept that
included government regulation (also known as “Licence Raj”) and government
planning. It could also mean a professional searching for words while trying to do
a difficult job of disentangling the term from its historical context.
A possible symbolic meaning of the word self-reliance, given the extremely
distressing times during the initial months of the pandemic and the acute lack of
essential items, was to boost self-confidence during a crisis. Interpreted in that
sense the words can be read independently of its historical context and becomes a
routine exercise by a leader to emphasize that the announced package will work
in putting things right. To what extent is that true. We will shortly return to this
after we have described the kind of policies that were part of this package.
The extent of destitution faced by the people due to the pandemic was in-
versely proportional to their income status. Rural migrant workers working in
urban areas, small business and other daily wage earners were worst affected. The
package was therefore targeted to them (see Table 11.1). Among the relief meas-
ures declared for the daily wage earners were increase in MNREGA wage from
TABLE 11.1 Policies undertaken by the Government of India (apart from international
trade and investment policies)
SR. NO. Policies
01 C1.70 lakh crore relief package under Pradhan Mantri Garib Kalyan Yojana
02 Tax refunds and other direct tax measures
03 Policies towards statutory and compliance matters
04 Support to business including MSMEs
05 Extension of registration and completion date of real estate projects under
RERA
06 Support to farmers and rural economy
07 Agriculture allied activities (animal husbandry, fisheries, etc.)
08 Support to migrants and urban poor
09 MGNREGS support to returning migrants
10 Labour codes – benefits for workers
11 Civil aviation policies for cost reduction
12 Tariff policy reforms
13 Healthcare policies
14 Policies to promote online education
15 Support to state
C182 to C202, 80 crore poor people given benefit of 5 kg wheat or rice per person
and 1 kg of pulses per household per month for three months, universalization
of right of minimum wages to all workers including unorganized workers, the
introduction of the statutory concept of national floor wage to reduce regional
disparity, appointment letter for all workers in order to promote formalization,
gas cylinders provided free of cost to 8 crore families, 24% of monthly wages to
be credited to EPF account for three months for wage earners below C15,000 per
month for businesses having less than 100 workers, C3,500 crore free food grain
supply to migrant workers for two months, three meals a day for the residence of
shelters for urban homeless and others.
Business establishments severely hurt by the pandemic were given different
types of support. For example, C3 lakh crore collateral free loans for businesses
including MSMEs, C20,000 crore subordinate debt for MSMEs, C50,000 crore
equity infusion to MSME fund of funds, C25,000 crore EPF support for busi-
nesses for three months, C30,000 crore liquidity facility for NBFCs/HC/MFI,
C45,000 crore partial credit guarantee scheme for NBFCs, C90,000 crore liquid-
ity injection for DISCOMs, C50,000 crore liquidity through TDS/TCS deduc-
tion, C1,500 crore interest subvention for new business, C10,000 crore scheme for
formalization of micro-food enterprises, a new definition of MSMEs to cover
more businesses than before and others.
Given the self-reliance signal, it was expected that the governments will have
a series of policies in the trade front. However, almost nothing was announced.
Among the trade policies were prohibition on export of surgical masks, textile
materials for masks and other medical equipment; restrictions on import of diag-
nostic kits removed; exemption from customs duty for import of medical equip-
ment; the validity of import permits extended for certain products; reduction
of import of substitutable coal; reduce import of defence equipment. Given that
our main interest in this chapter is the effect of the pandemic on trade policies,
we will come back to this relatively muted response of the government. Finally,
the Reserve Bank of India announced a slew of policies to shore up the financial
sector. The most important of these are listed in Table 11.2.
After these policies were announced analysts attempted to comment on their
nature in the media. There was some debate on the relative importance given for
demand and supply. Also, there were discussions on whether the economically
poorer sections got what they deserved during this crisis. Two opinions clearly
emerged: (1) the policies had more of supply orientation and (2) benefits did not
reach the intended beneficiaries. There were also some discussions on the efficacy
of the benefits provided to the businesses as most of them were of the nature of
easier credit availability rather than direct cash transfers. All these arguments were
presented in terms of anecdotal evidence all of which were relevant at that time and
hence true to an extent. Importantly, however, there were no reports of large-scale
starvation due to the free availability of food grains through the public distribution
system and the fragmentation and hence more evenly distributed nature across rural
households stood firmly between existence and extermination of human lives.
TABLE 11.2 Policies undertaken by the Reserve Bank of India
SR. NO. Policies
01 Raising the ways and means advance limits of states by 60% and enhancing the
overdraft duration limits:
Number of days state can be in continuous overdraft from 14 days to 21 days and
Number of days state can be in overdraft in a quarter from 32 to 50 days.
02 Reduction of cash reserve ratio (CRR) which has resulted in liquidity
enhancement of C137,000 crores.
03 Targeted long-term repo operations (TLTROs) of C100,050 crores for fresh
deployment in investment grade corporate bonds, commercial paper and non-
convertible debentures.
04 TLTRO of C50,000 crores for investing them in investment grade bonds,
commercial paper and non-convertible debentures of NBFCs and MFIs.
05 Increased the banks’ limit for borrowing overnight under the marginal standing
facility (MSF), allowing the banking system to avail an additional C137,000
crores of liquidity at the reduced MSF rate.
06 Special refinance facilities to NABARD, SIDBI and NHB for a total amount of
C50,000 crores at the policy repo rate.
07 Opening of a special liquidity facility (SLF) of C50,000 crores for mutual funds
to alleviate intensified liquidity pressures.
08 Moratorium of three months on payment of instalments and payment of interest
on working capital facilities in respect of all term loans.
09 Easing of working capital financing by reducing margins
10 For loans by NBFCs to commercial real estate sector, additional time of one year
has been given for extension of the date for commencement for commercial
operations (DCCO).
It is possible to form our own opinions on the policies listed above: (1) policies
for the businesses, it can be seen, are indeed of the nature easier loan availability
and hence exchequer money was not overtly spent here; (2) there were definitely
more transfers from the exchequer towards consumers rather than producers; (3)
much of the relief to the consumers were however of the form of free availability
of food. Very little effort was given to boost incomes of people who could have
boosted demand for all goods that these people consume. The policy, in this
sense, can be alleged to be demand neutral. But neither can it be alleged that
businesses walked away with the cake (note point (1)). Fundamentally, therefore,
these are typically short-term stabilization policies intended to boost confidence
and tide over the immediate future. There does not appear to be any clear struc-
tural dimension in them – the kind of dimension that would have an influence
over such long-run economic phenomenon as self-reliance and globalization.
Looked at it from this sense, the use of the words self-reliance by the Indian
government appears to indicate something quite different from defining the na-
ture of the announced policies. It can be interpreted as an independent attempt
to rally the people through the nationalistic fervour of “we can do it ourselves”.
Appealing to the nationalistic instinct of its citizens to achieve the desired goals
is a tried and tested method used by leaders through history and is in line with
many such similar attempts in India up to recent times. It is also the in-thing in
some parts of the world including the USA (“America first”). In economic terms,
however, this can imply protectionism, as it did for the USA when the country
turned more protectionist in the late 1990s. If so, then the pandemic packages by
governments become directly relevant to defining the course of globalization or
its reverse in the future. To this, we will turn now.
11.4 The pandemic and de-globalization:

theoretical implications
The obvious thrust of India’s pandemic package was to provide immediate atten-
tion to the country’s welfare concerns arising out of the crisis. One implication of
this overwhelming concern for short-term national interest is the open disregard
for mutually agreed multilateral rules that define a wider world order. For exam-
ple, the World Trade Organization has expressed its concern5 for such policies
that disregard trading rules in the name of crisis management. The problem is, if
successful in boosting public morale during the crisis, these short-run nationalis-
tic overtures may refuse to go away with the crisis. Such long-term implications
of short-term stabilization policies allow us to conjecture about possible impli-
cations of the pandemic on trade issues such as globalization and its reverse. In
the rest of the chapter, we briefly set out to do this in the context of an extended
version of the “protection for sale” tariff levels (Grossman & Helpman 1992). In
addition to internalizing the policymaker’s (government’s/politician’s) objective
function, one great advantage of the protection for sale model is that it allows
for a differential weight on social welfare over something else (campaign funds
in this case). As we are trying to model the disproportional importance given
to social welfare during the pandemic, this is important to us. But it solves our
problem only partially as governments can set tariffs unilaterally only in the short
run. What about the long-run implications of such changes? For such tariff levels
to be optimal in the long run, exporting countries must accept them. We assume
that long-run tariffs are negotiated among countries at the bilateral level through
a simple Nash bargaining solution. The effect of the pandemic is then analysed as
a comparative static exercise in the resultant tariff profile.
To keep the analysis simple, de-globalization is represented below only in the
narrow context of protectionism through tariff escalation and the nationalistic
fervour displayed in the pandemic packages is assumed to increase the impor-
tance (weight) of social welfare over the influence of lobbying in the govern-
ment’s decision-making. The question posed therefore is: what happens to tariff
levels when importance to social welfare increases?
The list of variables used in the model is as follows:
1 P0j is the vector of all domestic prices in country j.

( )
2 G j P0j is the welfare function of the government of country j.
( )
3 Wi P0j are the total earnings of the owners of specific capital in industry i
( )
and W0 P0j is the welfare of such individuals who do not hold any unit of
specific capital of any industries in country j.
4 Bi (> 0) is the campaign contribution of an industry belonging to the organ-
ized sector, determined using the campaign contribution schedule.
5 α j (> 0) is the welfare associated with workers, whereas (1 + α j ) is the welfare
associated with organized industries.
6 Pi is the international price of good i.
7 tij, 0 is the tariff/subsidy on import good i by country j.
( )
8 P j = Pi 1 + t j is the price of good i in country j.
i , 0 i , 0
( ) is the domestic output of import good i.

j
9 yi Pi , 0
( ) = di (Pi, 0j ) L − yi (Pi, 0j ) is the import of good i, where di (Pi, 0j ) L is the

j
10 mi Pi , 0
total domestic demand for good i.
N
11 λ j = ∑
Hi
L
is the fraction of population owning specific factor in an or-
i =1
ganized industry, H i are the members of the population owning capital of
industry i.
12 δ is the bargaining power of the partner country, whereas 1 − δ is the bar-
gaining power of the home country.
13 τ 0j is the vector of trade policies (tariffs/subsidies) imposed by country j on
their import goods.
 1 i ∈ organized sector 
14 θ =  
 0 otherwise 
11.4.1 Unilateral tariffs in the short run

Suppose we consider a representative country j. In this country, there are N
goods and their respective industries. The total population is L, where each
member is endowed with only one unit of labour. Other than wage income and
consumer surplus, the individuals also earn from their holdings of specific capital
of an industry. The third component in total earnings is the re-distributed tariff
( )
revenue. Therefore, Wi P0j is the total earnings of the owners of specific capital
in industry i. And W0 (P0j ) is the welfare of such individuals who do not hold any
unit of specific capital of any industries, their welfare comprises of wage, con-
sumer welfare and re-distributed tariff revenues.
Here, the industries 1 to K belong to the organized sector and the industries
K + 1 to N belong to the unorganized sector but all goods are import goods,
as mentioned earlier. An industry belonging to the organized sector announces
their campaign contribution Bi > 0 , for all i = 1 to K , using the campaign contri-
bution schedule.
Therefore, the government forms the following welfare function, taking into
consideration the interest of both capital owners and non-owners:
( ) { ( ) ( )}
G j P0j = (1 + α j ) W1 P0j +  + WK P0j − (B1 +  + BK )
 
 j
( ) j
+ α j WK +1 P0 +  + WN P0 + W0 P0 ( ) ( )
j 
(11.1)
From the above equation, α j > 0 is the welfare associated with unorganized in-
dustries and workers, whereas (1 + α j ) is the welfare associated with organized
industries and P0j is the domestic price vector in country j. An increase in α j
therefore implies an increase in weight on all members of the society and hence
social welfare.
j
The price of an individual good i is given as Pi , 0 ( )
= Pi 1 + tij, 0 , where Pi is the
international price of the good and tij, 0 is the tariff/subsidy on import good i by
country j.
Maximizing equation (11.1) we obtain the following tariff profile for j (see
Grossman & Helpman 1992):
 


( )
 (θ − λ j ) yi Pi , 0
j 

tij, 0 = −
( )
∂m P j
 (α j + λ j ) Pi i i , 0


(11.2)
j
 ∂Pi , 0 
 
Since λj is the proportion of workers in the organized sector and αi is the weight
for social welfare in the government’s welfare function and since the partial de-
rivative with respect to import demand is negative, we immediately have the fol-
lowing two results: (1) as weight on social welfare increases tariffs fall, however;
(2) the higher the proportion of workers in the organized sector, or the more
labour intensive the organized sector, the higher the tariff.
11.4.2 Long-run tariffs

In the Nash bargaining model, countries bargain with each other in pairs to
determine the optimal long-run level of tariff. There are three countries, i.e.,
a home country and two foreign countries; each pair of countries is bilaterally
negotiating trade policies, over their existing tariff/subsidy vector, on the import
goods which are exported by the partner country. In what follows, we report
only the bilateral bargaining outcomes individually rather than simultaneously.
Effects of changes in the terms of trade due to bargaining by a pair of countries
on the third country are thus not considered.
For negotiation between the home country “h” and the foreign country “ f 1”:
Max
∆G h1∆G f 1 (11.3)
τ h1τ f 1
( ( ) ( )) ( ( ) ( ))
1−δ δ
where ∆G h1 = G h1 P*h1 − G h1 P0h1 and ∆G f 1 = G f 1 P*f 1 − G f 1 P0f 1
are the changes in gain functions of the respective countries as they move from
the protection for sale tariff to the negotiated tariff.
Maximization of equation (11.3) yields the following tariff profiles for the
home country (h1) and the first foreign country ( f1):
 
 (θ − λ ) y h1 P h1 +  δ   ∆G   α + 2λ e f 1 P h1 − α Ld f 1 P h1  
h1
 h1 i ( )i ,*
 1 − δ   ∆G f 1  
( f 1 f 1) i,* i,* ( ) ( )
f 1 i ,* i ,* 

h1
ti , * = −  



Pi (αh1 + λh1 )
( )
∂ mih1 Pih,*1  δ   ∆G h1 
+   (α f 1 + λ f 1 )
∂ ei ,*( )
f 1 h1 
Pi ,*



h1 f
 1 − δ   ∆G 1 h 1
  ∂Pi ,* ∂Pi ,*  
(11.4)
(See the appendix for the formal derivation of equation (11.4).)
On comparing equations (11.2) and (11.4), we see that the effect on tariff addi-
tionally depends on the bargaining power of the two countries. In particular, an
increase in the bargaining power of the home country increases tariff.
11.5 Implications and concluding comments

Given the conclusion that the pandemic has forced governments to put more em-
phasis on social issues, the actual outcome on globalization is conditional. Two
conditions are particularly important: in the short run, whether large sections
of populations are involved in the organized sector that contributes funds to
governments and in the long run, the extent of bargaining power of the country
in trade negotiations. The effect on globalization is thus conditional. In general,
countries attaching more importance to social welfare will reduce tariffs and
ease out of the de-globalization process if they have a relatively capital-intensive
import-competing sector and low bargaining power in trade negotiations. How-
ever, countries with labour-intensive import-competing sectors and some bar-
gaining power in trade negotiations may increase tariffs and further contribute
to the de-globalization process.
If we apply these observations to the Indian scenario, some confusing post
lockdown signals become a bit less so. For instance, the self-reliance signal that
created such furore can be rationalized in terms of the large sections of the pop-
ulations who have lost their jobs due to the lockdown. Our analysis shows that a
large section of populations being involved in the organized sector is pro-tariff.
Thus, the pro-protectionist undertone of the self-reliance signal is an expected
outcome. However, as we have seen, in spite of the signal nothing concrete has
yet been observed at the policy level in terms of large-scale increase in tariff. In
our analysis a social welfare bias is anti-tariff. Hence, the concerted effort of the
government to show pro-social welfare bias during the pandemic has made them
fall short of taking concrete action of actual tariff rates. Our analysis supports the
conjecture that given the rather weak bargaining power of India in trade negoti-
ations (see Bhattacharyya & Mandal (2016) for the example of the India ASEAN
free trade agreement) may have played an implicit role in refraining from actually
imposing tariffs. Knowing that the low bargaining power will force later nego-
tiations to roll back tariffs, the government may have saved itself from that em-
barrassment in future. Looked at it from this angle, the self-reliance is completely
expected during these times but is unlikely to be protectionist in the long run.
Notes
1 https://www.wto.org/english/news_e/pres20_e/pr862_e.htm.
2 See, for example, Bloomberg’s recovery tracker graphics.
3 See https://www.wto.org/english/news_e/news20_e/covid_07oct20_e.htm.
4 Financial Express, June 6 (https://www.financialexpress.com/economy/atmanirbhar-
bharat-not-about-import-substitution-heres-what-the-modis-self-reliant-india-
means/1983347/).
5 See, for instance, https://www.brookings.edu/blog/order-from-chaos/2020/07/10/
the-post-covid-19-world-economic-nationalism-triumphant/.
References
Van B rgeijk, P. A., Brakman, S., & van Marrewijk, C. (2017). Heterogeneous economic
resilience and the great recession’s world trade collapse. Papers in Regional Science,
96(1), 3–12.
Baldwin, R. E. (Ed.). (2009). The great trade collapse: Causes, consequences and prospects. CEPR.
Bhattacharyya, R., & Mandal, A. (2016). India–ASEAN free trade agreement: An ex post
evaluation. Journal of Policy Modeling, 38(2), 340–352.
James, H. (2018). Deglobalization: The rise of disembedded unilateralism. Annual Review
of Financial Economics, 10, 219–237.
WITS database.
Trade Maps database.
World Bank database.
UNCTAD TRAINS database.
WTO database.
UN database.
/(https://...)www.india.gov.in/spotlight/building-atmanirbhar-bharat-overcoming-
covid-19.
/(https:// ...)home.kpmg/xx/en/home/insights/2020/04/india-government-and-institution-
measures-in-response-to-covid.html.
APPENDIX
Derivation of equation (11.4) using equation (11.3).
Assumption: As the bargaining model is a game of complete information,
exporters of country j receive subsidies equivalent to the tariffs imposed by the
partner country.
Max
∆G h1∆G f 1
τ hτ f 1
The FOCs are as follows:
∂∆G h1∆G f 1 ∂∆G h1

( ( ) ( ))
−δ
= ∆G f 1 (1 − δ ) G h1 P*h1 − G h1 P0h1
∂Pih,*1 ∂Pih,*1
(11.5)
( ( ) − G ( ))
δ −1 ∂∆G f 1
h1
+ ∆G δ G f1
P*f 1 f1
P0f 1 =0
∂Pih,*1
where
∂∆G h1
∂Pi , *
( )
f 1 = (1 − λh1 ) yi Pi , * + Pi (α h1 + λh1 )
h1 h1  (11.6)
and
∂∆G f 1
∂Pih, *1
( ) f1
( )
= (α f 1 + 2λ f 1 ) e 2B Pih, *1 − α f 1Ldi , * Pih, *1 + (α f 1 + λ f 1 ) . (11.7)
Substituting equations (11.6) and (11.7) into equation (11.5), we obtain:

 
 (1 − λ ) y h1 P h1 +  δ   ∆G   α + 2λ e f 1 P h1 − α Ld f 1
h1
 h 1 i ( )
i ,*
 1 − δ   ∆G f 1  
( f 1 f 1) i,* i,* ( ) f 1 i ,* ( )
Pih,*1  

tih,*1 =−  



Pi (αh1 + λh1 )
( )
∂ mih1 Pih,*1  δ   ∆G h1 
+     (α f 1 + λ f 1 )
∂ ei ,* ( )
f 1 h1
Pi ,* 



h1 f 1 h
  ∂Pi ,*  1 − δ   ∆G  ∂Pi ,*1  
Likewise, we can derive the tariff profiles for partner countries f1 and f2.
TABLE A.11.1 Timeline of US-China trade war
Date Country Policy
22.1.2018 USA Safeguard tariffs on imports of washing machine and solar

panels, based on the report published by USITC.
1.03.2018 USA On security grounds, based on the report from the
Department of Commerce, tariffs on steel and aluminium
worth $48 billion.
2.04.2018 China Retaliatory tariffs on aluminium waste and scraps, pork and
other US products worth $2.8 billion.
3.04.2018 USA Based on USTR investigation, 25% tariffs on 1,333 Chinese
products.
4.04.2018 China Threats to retaliate by imposing 25% tariffs on US export
goods, mostly focused on autos, aircraft and agriculture.
5.04.2018 USA Considers additional tariffs on import worth $100 billion
from China.
17.04.2018 China Preliminary tariffs on imports of sorghum from the USA.
18.05.2018 China End of tariffs on US sorghum export during negotiation.
29.05.2018 USA The official release of a statement that tariffs will be imposed
on imports worth $50 billion from China.
15.06.2018 China Revised retaliation list considers a two-phase approach for
the list. Aircraft taken off the list and many other products
like mineral fuels, some consumption goods were added.
15.06.2018 USA Revision of $50 billion tariff list. The share of intermediate
goods in the list was increased from 41% to 52%, whereas
the share of consumer goods in the list decreased from
12% to 1%.
18.06.2018 USA Threat to impose 25% tariffs on an additional $200 billion
worth of Chinese imports, a retaliatory measure to tariffs
on US exports to China.
10.07.2018 USA Tariffs go into effect from both sides.
16.07.2018 USA and A new set of tariffs (10%) on goods worth $200 billion
China Chinese imports. A significant rise in tariffs imposed on
consumer goods makes for 23% of the list compared to 1%
previously.
20.07.2018 USA Threat to levy higher tariffs on all the remaining products
imported from China.
01.08.2018 USA Trump wants 25% tariff instead of 10% on the list of goods
worth $200 billion imports.
03.08.2018 China Threats to add duties of 5–25% on imports with $60 billion
from the USA.
07.08.2018 USA Release a revised list of the second phase of $50 billion list.
08.08.2018 China Revised the second tranche of $50 billion list.
14.08.2018 China Files WTO dispute against US solar panel tariffs.
23.08.2018 USA and Both impose second phase of $50 billion by tariffs.
China
17.09.2018 USA $200 billion tariff list finalized. The rates will increase from
10% to 25% from January 1 2019.
24.09.2018 USA and Tariffs announced on September 17 goes into effect.

China Similarly, retaliatory tariffs imposed by China on
September 18 goes into effect.
24.02.2019 USA Tariff increase on $200 billion worth of goods delayed, as
earlier planned to go into effect from March 1, 2019.
10.05.2019 USA Raises tariff rate on the previous list of goods, from 10% to
25%.
13.05.2019 China Increase in tariff on imports of US goods, a retaliatory
measure to the hike in tariffs imposed by Trump on May
10.
01.06.2019 China Higher tariffs imposed on imports worth $60 billion.
01.08.2019 USA Announcement of additional 10% tariffs on all remaining
Chinese imports.
13.08.2019 USA Plans to impose additional 10% tariff on $112 billion goods
from September 1, 2019 and then $160 billion from
December 15, 2019.
23.08.2019 USA and China retaliates and Trump plans to impose higher tariffs.
China
11.09.2019 USA and China removes 16 products from its retaliatory tariff list and
China Trump delays the date to impose additional tariffs from
October 1 to October 15.
11.10.2019 USA Cancels additional tariff, as announced earlier. Talks
initiated regarding “phase one” of deal with China.
13.12.2019 USA December tariffs called off, in anticipation of a trade deal.
15.01.2020 USA and Both leaders sign phase one deal, where China agrees to buy
China additional exports from the USA worth $200 billion.
14.02.2020 USA Phase one deal goes into effect, a tariff on Chinese imports
six times higher than the beginning of the trade war in
2018.
Source: Peterson Institute for International Economics.

12
GREEN OR BROWN
Global environment in the post-COVIDian
era, and challenges for India
Runa Sarkar
12.1 Introduction
In a world that was trudging along the well-beaten path of economic growth and
prosperity, as measured by limited parameters such as GDP, with limited concern
for social and environmental issues such as inequality and climate change, the
Covid-19 crisis has come as a wakeup call. It has provided every nation in the
world the opportunity to take a step back and review its progress, with future
outcomes in mind, some of which may not appeal to those with a short-term
outlook. This is more true for countries in the developing world such as India,
which, hitherto, have given economic growth primacy over all other concerns to
ensure that its entire population can enjoy a reasonable quality of life, at par with
the rest of the world. Each nation now realises that the environmental, socioeco-
nomic and policy paths that they choose to take now may result in two diametri-
cally opposite outcomes in the future. There are strategies that could lead a nation
on a path towards more sustainable development, if other nations also choose
similar paths, or along a road that would lead to more instability, inequality and
further environmental pressures, of which the Covid-19 pandemic is an instance.
A rapid increase in population, coupled with strong economic growth over
the last several decades has brought about significant pressure on the carrying
capacity of India’s ecosystem to continue to support such an activity. Agricul-
tural land use has gone up, the material footprint of production has increased
manifold, as has our energy use intensity, leading to (irreversible) changes to
the ecosystem that throws up several challenges, such as air and water pollution,
problems of waste and climate change. The consistent pressure on biodiversity
has ratcheted upwards due to increased land use and domestication and rearing of
a few species of animals, often in close quarters. This situation is as true of India
as of the rest of the world.
DOI: 10.4324/9781003220145-14
Green or brown 195
Over the past few decades, the species conflict resulting from humankind
destroying and disregarding biodiversity and forestland coupled with climate
parameters has manifested itself in the form of viral crossovers, leading to out-
breaks such as SARS, MERS, dengue, ebola and chikungunya affecting humans.
Such diseases are referred to as zoonotic diseases (United Nations Environment
Programme & International Livestock Research Institute, 2020; World Wildlife
Fund, 2020). According to the WWF (2020), “Of all the emerging diseases, zo-
onosis of wildlife origin represent one of the most significant threats to the health
of the world population”. Covid-19 is the outcome of a new strain of virus found
in bats and pangolins (who serve as natural reservoirs of the virus) abundant in
some areas of Asia that has never affected humans. The SARS-CoV-2 (Covid-19)
outbreak appears to have originated in the large animal market in Wuhan, which
trades in both bats and pangolin scales, although the mechanism of transmission
and whether there was an intermediate species involved are not clear.
While there is no clear evidence linking the spread of Covid-19 with climate
change directly, many of the root causes of climate change are also responsible
for the spread of pandemics in general. For instance, we could focus on deforest-
ation and loss of wild habitat as one of the root causes of climate change as they
lead to a loss in natural carbon sinks, thus leading to global warming. Increasing
temperatures and changing climates could lead to increased transmission of dis-
ease through a combination of (a) direct action on infectious agents (malarial par-
asites thriving in the mosquito at higher temperatures is an instance), (b) impact
on vectors (for instance, a longer active season for the mosquito) and (c) changes
in host behaviour, the most common instance being shifts in migratory bird
patterns (Curseu et al., 2010). Climate change influences bird habitat, migration
routes and stopover routes, and could be a factor in global distribution of avian
virus agents and/or the emergence of new strains.
Increased consumption of meat and poultry is another root cause of climate
change, given that animal-based foods have much higher carbon footprints, pri-
marily because rearing livestock in farms for meeting society’s meat and dairy
needs leads to the release of substantial amounts of methane, a potent greenhouse
gas. Further, widespread animal farms have caused changes in land use patterns
and a loss in biodiversity, which affects climate change, but also can increase the
risk of pandemics manifold. The intensification of livestock production facili-
tates disease transmission by increasing population size and density within the
same farm. Use of antimicrobials for growth promotion and disease prevention
promotes the evolution of antimicrobial resistance in zoonotic pathogens and
low genetic diversity favours increased transmission and adaptation. Greater fre-
quency of movement of people on and off the farms, poor ventilation systems,
improper waste management systems and use of animal waste for aqua culture all
increase the risk of pathogenic transmissions and serve as a source for a spillover
of infections from one species of animals to others and to people due to their
coming in close proximity of one another ( Jones et al., 2013).
196 Runa Sarkar
Further, research has shown that people who live in places with poor air qual-
ity are more likely to die from Covid-19 even when accounting for other factors
that may influence the risk of death, which once again links a cause of climate
change to the impact of the pandemic (Wu et al., 2020). In its 2018 report, the
Intergovernmental Panel on Climate Change clearly expresses the concern that
one of the likely impacts of global warming is the acceleration of the emergence
of new viruses (Intergovernmental Panel on Climate Change, 2018); perhaps
Covid-19 is just the tip of the iceberg! Moreover, greater interconnectedness via
global trade and international migration has made it very difficult to contain
rapidly communicable diseases like Covid-19 locally.
In this backdrop, this chapter seeks to first compare the ongoing acute public
health crisis generated by the Covid-19 pandemic to the chronic climate change
crisis. We then assess the impact of Covid-19 on the environment, and more spe-
cifically on the progress towards meeting commitments made by nations towards
climate change with a focus on India. As the series of lockdowns and then unlock
measures unfolded in the country, different strata of society reacted differently to
the perceptible changes in the environment. We try and draw some lessons from
these reactions. We then assess how firms may react to what many would like
to call the “new normal”. This inevitably brings us to the policy measures put
in place to counter the impact of the pandemic and we examine the impact of
these measures and other related government decisions. Since there is a need for
coordinated actions across all nations to arrest climate change to a reasonable ex-
tent for the survival of humankind, we then look at some of the measures taken
by other countries to combat the pandemic. We note that while countries such
as Germany and South Korea pledged to use the pandemic to revisit and grad-
ually discontinue some of their economic activities with brown environmental
consequences, others such as the USA used it to emphasise economic activity ex-
clusively regardless of environmental implications. If the paths to recovery differ
across nations, the world would be far from a sustainable path. We follow this up
with a brief discussion on the climate change implications of India’s economic
stimulus package as a response to the pandemic. Finally, we conclude with some
ideas as to how to promote green economic growth for a sustainable future with
lower risks of pandemics such as the Covid-19.
12.2 The dual challenges of the Covid-19 pandemic

and climate change
The World Health Organisation declared Covid-19 as a pandemic in early March
2020, expressing deep concern over its severity, alarming levels of spread and
disturbing levels of inactivity among governments in reacting to the disease. The
pandemic hit the world, up close and personal, so to say, with the world popula-
tion in a state of panic fearing for their lives. Moreover, the general feeling was
that once a vaccine is discovered and the virus is researched better, solutions for
cure would appear and the threat of the disease would disappear. The optimist
Green or brown 197
believed that this would take just a couple of months, while the pessimist felt that
it could last a year or two. Committing to rigid shutdown measures to contain
the spreading of the coronavirus was undertaken by different nations on the tacit
assumption that these measures would be temporary. They could be loosened
when the public health system is strengthened and therefore less overwhelmed by
Covid-19 patients, and once Covid-19 infection rates decreased. Finally, all pre-
cautionary measures could be discontinued once vaccines were made available.
Some countries have been successful in eliminating the virus from their borders
by gravely restricting international travel, both into and out of the country. The
short time horizon and the private costs of Covid-19 made the populace accept-
ing of drastic policy measures to contain the pandemic, although it came at a
significant economic and social cost. Moreover, while cooperation across nations
is desirable to bring an early end to the pandemic, it is not a necessary condition.
The Covid-19 pandemic arrived in what had been billed as a “banner year”
for climate action with commitments for the Paris Agreement kicking in. The
climate change crisis has been looming over the world for over two decades,
its dangers seem impersonal and its causes diffuse. Several nations have tried to
address the challenge across many contentious global meets over the years. Mit-
igating climate change and achieving ambitious temperature targets as set out in
the Paris Agreement require commitments by nations towards long-term struc-
tural change. The end objective would be a shift away from the current carbon-
intensive economy to a zero-carbon economy soon and, ideally, in the very long
term, a net-negative carbon economy. Such a shift would come at a large cost as
several legacy production systems would cease to be in operation and consumers
would have to change their consumption patterns, akin to the economic and
social costs borne during the lockdowns imposed by the Covid-19 pandemic.
Moreover, there does not appear to be any possibility of a clear sure shot solution
or a “perfect vaccine”, whether in the form of harnessing solar energy or wind
energy or converting all waste to energy being made available in the future to
combat climate change. Instead, sustained incremental measures and efforts are
required from all nations, developed and developing, which together must trans-
late into a permanent, ongoing form of commitment to tackle the longer term
and more complex problem of climate change.
Pandemics and climate risk are similar in that they both represent physical
shocks, which can be remedied only by understanding and addressing the un-
derlying physical causes. Both are systemic, that is, they propagate fast across an
interconnected world. They are non-linear, as their socioeconomic impact grows
disproportionally, and even catastrophically once certain thresholds are breached
(such as hospital capacity to treat pandemic patients). They are both risk multi-
pliers because they exacerbate hitherto untested vulnerabilities inherent in the
economy. Addressing pandemics and climate risk requires a fundamental shift
from optimising for the shorter-term performance of systems to ensuring their
longer-term resilience. Neither pandemics nor climate hazards can be confronted
without true global coordination and cooperation (Pinner et al., 2020, p. 108).
198 Runa Sarkar
Given that the root causes of climate change coincide with the severity of the
current pandemic and the possibility of more such pandemics in the future, it
is imperative that climate action is taken to prevent the next outbreak, whether
by preventing further deforestation, stemming biodiversity loss, rethinking ag-
ricultural practices and livestock rearing measures or reducing air pollution by
minimising the use of fossil fuel. Health and environmental policy makers must
therefore work together, not only because of the interrelatedness of the two chal-
lenges, but also because several of the costs of climate change are manifested as
health-related costs (Bernstein, 2020). As Sterner et al. (2020) note, the climate
crisis requires policy changes that are less disruptive, economically, socially and
culturally, than the measures being taken right now to tackle Covid-19. For
tackling climate change, we do not need to shut down the economy, but we need
to transition towards a low-carbon economy that supports public and private
investments in renewable, energy-efficient and circular, technologies and infra-
structure. The feasibility of such a transition is relatively high as these technolo-
gies exist, are cost-efficient and even barriers such as insufficient energy storage
installations are being overcome.
12.3 Impact of Covid-19 and consequent lockdowns

on India’s environment
Due to the initial uncertainty surrounding both the impact of Covid-19 and its
spread through society overburdening hospitals, policy makers across the world
quickly decided to shut down interpersonal interactions by restricting local, na-
tional and international mobilities. The “lockdowns” had pervasive impacts on
economic activity across the globe, with significant reduction in production,
increase in unemployment, inequality and poverty, fall in international migra-
tion and trade, significant increases in bankruptcy filings and large ripple effects
down the supply chains.
Beginning with the first lockdown from March 25th, 2020, with a notice
of just four hours, India imposed arguably one of the most stringent pandemic
control measures to curtail the spread of the Covid-19 virus. Despite the con-
comitant economic hardship, such measures were the need of the hour, given
India’s population density and its poor public health infrastructure. The series
of lockdown-imposed measures initially focused on preserving life. However,
soon the attention shifted to livelihoods of people, leading to a series of unlock
“measures”, fuelled primarily by the migrant crisis and rampant unemployment.
Environmentalists were euphoric at the early stages of the lockdowns. The
resulting clear skies, improvements in local water quality, reductions in noise
and air pollution, substantial numbers of employees working in home offices,
a new-found interest in health and wellness all contributed to their optimism.
There appeared to be a greater understanding of what is non-essential for living
reasonably comfortably as humankind forcibly learnt to live on less and practise
responsible consumerism, the proverbial Holy Grail for environmentalists. Even
Green or brown 199
underwater noise came down due to reduced marine transportation activities,

implying a better living environment for aquatic species. They believed that a
new social order was now being created, one that valued the environment and
would do everything possible to limit their activities which resulted in a negative
environmental impact.
There was a visible improvement in air quality with blue skies for all to see
and people in Jalandhar in Punjab, and Saharanpur in Uttarakhand, could see
the distant Himalayas for the first time in almost 50 years due to the absence of
smog and dust haze. New Delhi’s air quality index (AQI), which ranges around
200 in general, fell to below 20, and average concentration of PM2.5 plunged
by 71% during the first lockdown (Goswami, 2020). Even in the southern part
of the country, reduction in particulate matter concentrations was of the order
of 50% to 60% (Press Trust of India b, 2020). Water pollution also came down,
primarily because of industrial effluents not being released into the rivers. The
Ganga looked blue at Rishikesh, and its waters were fit for bathing and support-
ing aquatic life at Haridwar, with at least a 50% improvement in dissolved oxygen
(DO) and 500% less total dissolved solids (TDS) levels. Several towns witnessed
wild animals roaming free on the streets as people were jailed indoors owing to
the lockdown. It was a vindication that if anthropogenic activity slowed down,
nature had the capability of healing herself, thereby restoring a more pristine
environment. Such was the optimism that Manish Kumar Singh, Assistant Pro-
fessor, Earth Sciences, Indian Institute of Technology (IIT) Gandhinagar, stated,
“In my opinion, the kind of recovery shown by mother nature during the lock-
down, there is no harm in using lockdown as a possible emergency measure to
combat severe air pollution episodes like we witness in Delhi winters” (Press
Trust of India b, 2020).
These hopes were, however, short-lived, both in India and globally (McClo-
skey & Heymann, 2020). Not only have, so far at least, all countries that have
(mostly) overcome the first wave returned to business as usual, but companies
have also used the opportunity to lobby for relaxed emission standards, to in-
crease subsidies or receive financial aid and, therefore, been able to quickly re-
turn to the pre-Covid-19 status quo (Ing & Nicolaï, 2020). Side effects are that,
already in the short run, many initial improvements in environmental quality
succumbed to a relaxation of environmental regulations and a catching-up pro-
cess that aims to bring countries back to their original economic growth levels.
In Delhi, crop burning in adjoining regions and the increase of private ve-
hicles on the road due to limited public transportation services have once again
resulted in an AQI of 255 (https://aqicn.org/city/delhi/) on October 16th, 2020,
with fears that the quality will deteriorate further to severe as the wind pattern
changes to the north-westerlies. Moreover with insufficient availability of public
transport with enough physical distancing and people fearful of availing public
transport in general, over time, the sales of private two wheelers and cars have
gone up in all the cities leading to increases in air pollution. Even when the
population was experiencing improved air quality outside, the quality of indoor
200 Runa Sarkar
air could worsen for families living in poorly ventilated cramped living spaces or
using wood fires for cooking.
There has been a huge setback in progress made with respect to reducing or
eliminating the use of single-use plastics over the last two years. Safety concerns
related to shopping during Covid-19 led to a preference of consumers and pro-
viders for fresh-food packaged in plastic containers (to avoid food contamination
and to extend shelf-life), and for the use of single-use food packaging and plastic
bags to carry groceries. Many stores implemented additional health and safety
measures such as social distance, cleanliness, hygiene and, in some cases, by pro-
viding home delivery and/or a pick-up service. Even the local panipuri seller
adapted to the “new normal” by packaging the puri, pani and stuffing separately
in transparent single-use plastic bags to service customers wary of contagion.
Once littered in open environments, plastic waste will likely induce sewage sys-
tem blockage in towns and negatively affect water percolation and normal agri-
cultural soil aeration, with repercussions on land productivity. The persistence of
plastic debris is known to impose serious threats to biodiversity as well.
The use of single-use and disposable PPEs and masks is contributing to
mountains of medical waste, often beyond the region’s capability to process,
aggravating their impact in the natural ecosystems and compromising potential
mitigation/remediation measures. According to a study of the World Wildlife
Fund in Italy, improper disposal of just 1% of face masks translates to more than
10 million items for that country alone, weighing 30,000 to 40,000 kg (Adyel,
2020)! This waste would then persist for a long time in nature. While some
municipalities will manage alternatives to treat medical waste properly, others
(with less economic and waste management resources) might be forced to apply
inappropriate management strategies. Some Indian municipalities are reported
to be “following a flawed system of medical waste disposal and management,
which mostly rely on landfilling and local burning strategies” (Patrício Silva
et al., 2021). With medical and municipal solid waste (MSW) generated being
considered as potentially infectious during the Covid-19 pandemic, incineration
and landfilling are being prioritised over recycling, which leads to poorer air
quality. Open burning of plastic waste is also responsible for the emission of
heavy metals, PCBs, dioxins and furans, all of which are linked to an increased
risk of respiratory disorders among the exposed population.
Disinfection campaigns for public spaces like hospitals, offices, clinics, uni-
versities, airports, and even public gardens and zoos are a must. However, it is
possible that the choice of the chemical disinfectants and the places for disin-
fection may not be entirely scientific and inappropriate disinfectants could lead
to chronic obstructive pulmonary disease among healthcare workers or others
repeatedly exposed to disinfectants, and have negative impacts on the local fauna
and flora.
The panic over the Covid-19 pandemic has clearly outshone the perceived
threat of plastic pollution, leading to a sudden shift in the hierarchy of values,
i.e. where health is considered as something of greater value independent of the
Green or brown 201
environment, and there is a clear decrease in perceived importance of a clean

environment (Grodzińska-Jurczak et al., 2020). This regression in behaviour is a
huge step backwards for environmental awareness and can have severe long-term
implications.
The economic impact of the lockdown has meant that several migrant work-
ers have come back to their villages from which they fled because of the lack
of employment opportunities. This has serious implications for deforestation
and denudation of woods around rural areas as hungry villagers forage nature’s
bounty for sustenance and firewood. Although it is too early to be able to see
a noticeable impact of the pandemic on deforestation, the likelihood of a sig-
nificant negative impact is high, which, in turn, leads to greater possibilities of
zoonosis in the future.
12.4 Reactions of the corporate world

The pandemic, coupled by the series of lockdowns and then unlock initiatives,
hit industry hard. On the one hand, it had to deal with severely depressed de-
mand; on the other, its supply chains were disrupted, it lost several of its skilled
labour as migrant workers returned to their villages and existing employees were
wary to come to work. Then, there were unclear regulatory norms to comply
with, with respect to restarting manufacturing operations. Many (small) firms
struggled with survival with senior employees taking severe salary cuts to tide
over the bad times and retrenching excess employees. Others shut down. Some
larger firms with deep pockets could try to view the pandemic as an oppor-
tunity to bring about strategic change looking at a new future, knowing that
stakeholder pressures to focus on social and environmental issues would ratchet
upwards once the pandemic was over.
Firms that were focused on climate risk prior to the pandemic evaluated reg-
ulatory risk and transition risk. The Covid-19 pandemic brought to their at-
tention the immediacy and the scope and the scale of physical climate risk, in
terms of availability of resources and human capital. Physical climate risk has
been increasing over the years and is localised and non-stationary in nature. It is
non-linear, has indirect and cumulative effects on the interconnected socioec-
onomic and financial systems and impacts different strata of society differently,
with the poorest communities being more vulnerable (Woetzel et al., 2020).
The experience of firms in dealing with the Covid-19 pandemic will help them
be better prepared to deal with a continuous onslaught of challenges posed by
climate change. As an example, firms in India will take into cognisance that
given current projections, 160–200 million people in India (75% of the labour
force) could be at risk of experiencing lethal heat wave conditions by 2030. Fur-
ther, such intense heat would ensure that it would be unsafe to work outside for
around 30% of annual daylight hours (Woetzel et al., 2020, p. 11). Hence, they
would consider building cooling shelters for outdoor work, shifting work hours
for outdoor labourers and automating outdoor work. It is of concern that many
202 Runa Sarkar
firms may choose to automate their production processes to reduce reliance on

labour, given the risk posed by the pandemic.
Going forward in a post-pandemic world, firms may choose to embrace cli-
mate risks and rework their strategies around it, or lobby with regulatory author-
ities to delay the onslaught of regulations related to climate change and stay in
business for a little while longer. Many of the firms adopting the second approach
would be (or soon turn into) zombie firms (What to Do about Zombie Firms,
2020), that is, firms that are marginally profitable for a long period of time and
should be shut down. While it is welcome, the loan moratorium granted by the
Indian government may have the perverse effect of prolonging the life of some
of the zombie firms as well.
Firms that adopt the first approach would review how businesses worked dur-
ing the pandemic to draw lessons in efficiency from their functioning with lim-
ited resources and mobility constraints. For example, the concept of work from
home and flexible work hours may be formalised into regular firm operations
as it appears to have provided greater flexibility to employees while not having
much of an adverse effect on productivity. This may contribute to more gender
parity at the workplace while also reducing scope 1 (direct) GHG emissions due
to reduced transportation demands. This is also true about the extent of business
travel, which plummeted during the lockdowns, but may not recover even when
the pandemic is over as the pandemic demonstrated that several business trips
could be easily substituted with the greener option of videoconferencing. The
supply chain shock faced by several firms may lead them to consider repatriating
their supply chains wherever possible. This would also reduce environmental
footprint in terms of scope 3 emissions.1 Moreover, a better understanding of
physical risks because of lessons learnt from the pandemic may help firms’ better
price in those risks. This would create the potential for additional near-term
business-model disruptions and broader transition risks but also offer greater in-
centives for accelerated change (Pinner et al., 2020). Embracing the idea of work
from home could also lead to savings in terms of reduced office space. Firms
could also explore shifting their offices from overpriced centrally located spaces
(to ease of daily commuters) to more open and cheaper locations in the sub-
urbs, which could free up resources for more sustainable investments. Firms will
have to aim at boosting demand through changing the mind-set of the reluctant
consumer and one can now see advertising that focuses on the health benefits
of a product (including goods such as plywood and cement). Many firms have
intensified their presence in the e-marketing channels downplaying their brick
and mortar retail stores and would continue to stay in that space even after the
pandemic is over, thus becoming less resource-intensive.
The impact of the pandemic and post-pandemic responses of firms will have
different impacts on different sectors. While the automobile sector in India ap-
pears to have recovered from the pandemic with a huge increase in sales of two
wheelers and four wheelers for private consumption, the aviation and hospitality
sector is still in doldrums as people are hesitant to travel. These industries would
Green or brown 203
see consolidation in the near future and their business models would have to
adapt to a scenario of limited business travel, and to one where the willingness
to pay for access to pristine environments for leisure travellers is on the uptick.
Although the pandemic should have been a boom time for the healthcare sector,
infrastructure and human resources were in short supply impeding their ability
to service the high level of demand. The financial sector also has concerns going
forward as several loans may turn out to be bad debts, although firms may emerge
unscathed if sufficient provisions were made for the same.
The pandemic may be used as an opportunity by regulators to raise environ-
mental standards. However, acceptability of environmental policies by compa-
nies is a necessary condition for their successful implementation, as in general,
environmental efforts are costly and may lead to higher product prices. Social
acceptability for higher prices may be higher as the pandemic may have helped
society realise the imminent danger posited by climate risks. Some regulators
have chosen to make aid to firms conditional on efforts to protect the environ-
ment. For instance, in France, Renault’s bailout requires it to increase the share
of electric vehicles, while the 7 billion euro Air France bailout imposes the use of
at least 2% of alternative jet fuel by 2025, a target of 50% emission cuts by 2030
and a 50% decrease in domestic flights by 2024, especially those that compete
with high-speed trains (Ing & Nicolaï, 2020).
With respect to investing in new projects, in the short term, firms may choose
to invest in brown projects rather than green projects without policy interven-
tion. At present, the cues from the global oil market have raised the opportunity
cost of transitioning away from fossil fuels and for investing in green technology.
The increase in opportunity costs of energy transition will facilitate the reallo-
cation of funds away from renewable projects towards post-pandemic economic
stimuli. Recovering lost investments in energy transition projects would be un-
likely because the resource rents that fund them are likely to continue to be low
in the future, given low resource prices and demand pressures of climate change
mitigation, rendering firms involved in such projects uninteresting to investors
(Shehabi, 2020). Thus, there is a need for regulators to step in to ensure that the
global movement to substitute fossil fuel with green energy does not suffer a
setback.
12.5 E
ffectiveness of stimulus measures towards greener
economic growth
The lockdown measures imposed by many governments gave the opportunity
to several people to take a step back and smell the proverbial roses, as they were
confined to their home with little to do. Many of them suggested that in the
aftermath of the Covid-19 crisis, economic recovery stimulus packages be built
around low-carbon investments, such that they could be climate-friendly while
trying to restore the level of pre-Covid-19 economic activity and well-being.
The climate change sceptics, however, advocated the removal or postponement
204 Runa Sarkar
of climate mitigation-related taxes, levies and regulations to unleash the ani-

mal spirits of economic growth. Environmental concerns could come later; the
focus must be on ensuring livelihoods for all. Indeed, after the financial crisis of
2009, economic growth was the single-minded focus of policy makers, because
of which many of the gains from the Kyoto protocol with respect to reducing
CO2 equivalent emissions were squandered. This is not to say that one must
disregard economic growth, but that broad guidelines must be in place such
that new investments meet a comprehensive list of social and environmental
criteria so as to minimise interest-group-driven project proposals being ap-
proved. For instance, such criteria could include expected economic multi-
plier as a short-term criterion, and reduction in CO2 equivalent emission over
the lifetime of the investment as a long-term criterion. Moreover, the crisis
has revealed that populations will accept drastic measures (lockdown, wearing
face masks) and a rapid change in social norms when confronted with a crisis.
Increased environmental awareness and communication of the link between
climate change and the possibility of an increased frequency of emergence of
pandemics such as the Covid-19 would lead the population to demand more
stringent environmental policies to combat climate change. This is an oppor-
tunity that one must not let go of easily, especially at a time when people are
amenable to change.
As Hepburn et al. (2020) state (emphasis added):
Green fiscal recovery packages can act to decouple economic growth from
GHG emissions and reduce existing welfare inequalities that will be exac-
erbated by the pandemic in the short-term and climate change in the long-
term. Short-term reductions in GHG emissions resulting from lockdowns
will themselves have minor long-term effects, unless they facilitate deeper
and longer-term human, business, and institutional changes. Urgent res-
cue packages have been necessarily ‘colourless’ and focused on preserving
liquidity, solvency, and livelihoods, but their climate impact is also unlikely to
be positive.
Notwithstanding this observation, some of the themes around which sustainable

recovery efforts of countries around the world are clustered include
a decarbonising electricity through a focus on efficient buildings and sustain-

able transportation, including aviation,
b nature-based solutions such as forest management, restoring wetlands,
c a thrust on digital and circular economy systems through digital currency,
e-commerce digitisation and industrial ecology,
d better governance through multi-level systems that foster inter-ministerial
collaboration and community-driven approaches and
e greening of finance through public disclosures, limiting investments and
lowering subsidies in fossil fuel-based industries.
Green or brown 205
In the subsequent subsection, we provide a brief overview of how green the stim-
ulus packages of select countries are, followed by an analysis of India’s stimulus
package from the point of view of climate change.
12.5.1 International
As the pandemic’s first wave slowly receded, governments worldwide introduced
plans for economic stimulus worth trillions of dollars (Evans & Gabbatiss, 2020),
which, in turn, would have an impact on consequent CO2 emissions and climate
change-related targets. Monetary resources have been pledged through meas-
ures such as lowering base rates and/or quantitative easing by central banks or
through direct government spending to pay peoples’ wages, invest in specific
programmes or give loans to distressed companies. Governments also intervened
to stimulate the economy, by making it easy to add more renewable energy ca-
pacity, increasing taxes on fossil fuel-powered vehicles, providing tax relief to
oil firms or moving to ease other environmental rules to facilitate ease of doing
business. Despite calls for a “green recovery” voiced by world leaders, including
the Presidents of Germany, Austria and Switzerland, several major firms and
even the Pope, stimulus spending overall has been colourless rather than green.
In what follows, we provide a quick summary of the bailout packages of select
countries within the context of how green they are.
Germany’s stimulus package, hailed as “the world’s greenest stimulus plan”
(Lombrana & Rathi, 2020), focuses, among other things, on energy transition
(development of renewable hydrogen, subsidies for hybrid and electric vehicles)
and sustainable mobility. The Canadian government’s three-year infrastructure
plan assigns 60% of investments to green projects, including home retrofits, clean
energy projects and zero-emission buses. Sweden also proposed a large three-
year plan, including extensive government credit guarantees, direct investments
in energy-efficient homes, wetland restoration and low-emission transport, and
the removal of fossil fuel subsidies. Further, environmental taxes on fossil fuel-
powered vehicles are to be raised and made up for by lower taxes on labour and
enterprise. The French government announced a “France re-launch” plan, 30%
of which focuses on four “ecological” key sectors: Building renovations, trans-
port, agriculture and energy (including a focus on hydrogen), overlapping with
another 40% to bail out French industry, including aviation and the automotive
sector through spending, loans and loan guarantees (some of which have climate
conditionality).
Spain announced a recovery package for its automotive industry, of which
70% comprised “soft loans” to bail out companies with green conditionality,
with the rest made up of direct spending to support Spain’s sustainable transport
sector. As with many other European countries, it also promised to help its na-
tional airlines recover. South Korea put forward a “Korean new deal” to reinvig-
orate the economy with a focus on “green” and “digital”, with a commitment
to create 133,000 jobs. However, its bailout of coal plant manufacturer Doosan
206 Runa Sarkar
Heavy Industries & Construction Co. with an emergency loan has not been
looked upon kindly, as have its provisions to bail out the aviation and ship build-
ing sector. The Nigerian government’s plan, titled “bouncing back”, includes
expanding solar infrastructure, a gas expansion programme (relatively cleaner
than oil) and scrapping of gasoline subsidies. Chile updated its nationally deter-
mined contribution under the Paris Agreement in April 2020, and in its recovery
plan titled “step by step, Chile recovers”, it proposed an increased budget for its
public investment programme comprising a wide range of projects (including
expansion of airport runways and roads) with a commitment that 30% of these
projects would be green. However, its lack of specificity has been criticised. At
the other end of the spectrum, the USA announced the largest amount of fiscal
support without any considerations for sustainability, providing support to its
fossil fuel-based industry and rolling back environmental regulations (Dagnet &
Jaeger, 2020).
12.5.2 India
India had two choices when designing its economic stimulus package to revive
the economy because of the Covid-19 pandemic. It could catalyse a “green”
recovery by foregrounding measures to support renewable energies, public trans-
port, energy efficiency, etc., or postpone and/or abandon climate measures and
environmental regulations, depending on vested interests, with a focus on just
economic recovery, that is, a return to a business as usual scenario. It appears to
have chosen the latter option. Vivid Economics, a public policy think tank oper-
ating out of London, ranked India as the fifth worst performer of 320 countries
on the “Green Stimulus Index”, trailing South Africa, Mexico, China and the
USA, with respect to the net impact the stimulus is expected to have on the envi-
ronment. According to their analysis, stimulus packages of 16 of the 20 countries
would have a negative environmental impact (Vivid Economics, 2020).
While India’s Rs. 20-lakh crore ($260 billion) stimulus package to revive the
economy itself is a lot of hot air (Press Trust of India a, 2020, for example), its
focus on carbon-intensive sectors such as manufacturing and the extraction of
fossil fuels and their use for power generation, and policy stance to rapidly ap-
prove forest clearances for industries resulted in this poor score. With measures to
privatise the mining of coal and other minerals, the economic package is not in
consonance with India’s intended nationally determined commitments to reduce
fossil fuel use to mitigate climate change. It has earmarked $6.6 billion for coal
infrastructure and actively promotes coal gasification with tax incentives That
the emphasis of the bailout package, however small, was on improving the ease
of doing business above all else is evident. Neither the farmer nor the migrant
worker, or the hapless tribal subsisting on forest resources, stands to gain from
the stimulus package. Conditional bailouts for industry, contingent on them be-
coming greener, have not been considered at all in the proposed package with
the argument that there is not enough fiscal room for such proposals. There are
Green or brown 207
positive elements of the stimulus package, for instance, afforestation measures

and creation of green jobs for vulnerable tribal communities. However, many
of these were pre-existing schemes that were just repackaged. In addition, the
decision to hike excise duty on petrol and diesel as global oil prices fall could
be construed as green although the intent of the rise in taxes is to raise revenue
rather than curtail investments.
12.6 Way forward

The dual crises of the Covid-19 pandemic and climate change have provided
us a unique opportunity to capture a moment in world history to revisit our
policies so as to deliver economic growth to generate job opportunities along
with environmental well-being. While job creation, boosting consumption, and
managing the risks associated with the pandemic may be the need of the hour
for economic stimulus packages, they must not stop with just that. It would be
a pity if stimulus packages imposed large stranded asset costs on an economy for
decades to come because of an incorrect choice of technology, for example. The
packages must support projects with long-term growth potential that are resilient
to future shocks and which support a nation’s intended nationally determined
commitments to combating climate change as outlined in the Paris Agreement.
While subsidies to sectors particularly hit by the crisis, such as tourism, avi-
ation, and agriculture, should be provided, they should be made conditional on
meeting minimum environmental standards. Innovative green business models,
which may be at risk if banks limit loans in the presence of the current uncertain-
ties, must be encouraged through government intervention. Wherever possible,
public expenditures for a green recovery programme should at least partly be
funded by polluters by implementing a carbon price and abandoning ecologically
harmful subsidies.
More specifically, for the Indian economy, the Rocky Mountain Institute
and the NITI Aayog have charted a path forward in a report titled TOWARDS
A CLEAN ENERGY ECONOMY Post-Covid-19 Opportunities for India’s
Energy and Mobility Sectors (NITI Aayog & Rocky Mountain Institute, 2020).
The report outlines four broad principles to guide investments in clean energy
with specific recommendations for the transport sector (making public transport
safe, enhancing and expanding non-motorised transport infrastructure, reduc-
ing vehicle kilometres travelled, adopt electric vehicles, become an automotive
export hub) and the power sector (improving electricity distribution, enabling
renewable and distributed energy resources, promoting energy resilience and
local manufacturing of renewable energy and energy storage technologies). In
addition, it is important to underline that any bailout package for India has to be
inclusive in nature, so that the millions who have been pushed into poverty as a
result of the pandemic can find a way out of their misery.
To conclude, Hepburn et al. (2020) identify five policies that can both boost
an economy and positively impact climate change, while suggesting that their
208 Runa Sarkar
relative importance could vary depending on a country’s needs. These include

the development of clean infrastructure, building efficiency retrofits, investment
in education, investment in natural capital and investment in clean R&D. In the
Indian context, spending on rural support, whether through clean infrastructure,
investments in education or natural capital, would be the most potent policy
measure.
Without decarbonisation, a faster path towards global warming will only has-
ten climate change and along with it more instances of zoonosis leading to a
greater risk of public health episodes. By integrating scientific, economic and
socio-political knowledge, we can design policies which not only address the
immediate impacts of the Covid-19 pandemic, but also address the root cause of
the existential threats of future pandemics by meeting the challenges of climate
change, biodiversity loss and the crossing of planetary boundary limits, head on.
Note
1 Those emissions were caused because of a company’s value chain but were not asso-
ciated with its direct emissions or with the generation of energy consumed by it.
References
Adyel, T. M. (2020). Accumulation of plastic waste during COVID-19. Science, 369(6509),
1314–1315. https://doi.org/10.1126/science.abd9925
Bernstein, A. (2020). Coronavirus, Climate Change, and the Environment: A Conversation on
COVID-19 with Dr. Aaron Bernstein, Director of Harvard Chan C-CHANGE. Retrieved
from https://www.hsph.harvard.edu/c-change/subtopics/coronavirus-and-climate-
change/, last accessed 3rd March 2021.
Curseu, D., Popa, M., Sirbu, D. & Stoian, I. (2010). Potential impact of climate change
on pandemic influenza risk. In I. Dincer, A. Hepbasli, A. Midilli & T. Karakoc (Eds)
Global Warming. Green Energy and Technology (pp. 643–657). Boston, MA: Springer.
https://doi.org/10.1007/978-1-4419-1017-2_45
Dagnet, Y. & Jaeger, J. (2020). Not Enough Climate Action in Stimulus Plans. World Re-
sources Institute. Retrieved from https://www.wri.org/blog/2020/09/coronavirus-
green-economic-recovery, last accessed 3rd March 2021.
Evans, S. & Gabbatiss, J. (2020, June 16). Coronavirus: Tracking how the world’s ‘green
recovery’ plans aim to cut emissions. International Policy. Carbon Brief. Retrieved
from https://www.carbonbrief.org/coronavirus-tracking-how-the-worlds-green-
recovery-plans-aim-to-cut-emissions, last accessed 3rd March 2021.
Goswami, K. (2020, May 2). Covid-19: 4 unbelievable environmental changes seen
in India since lockdown. India Today. Retrieved from https://www.indiatoday.in/
education-today/gk-current-affairs/story/covid-19-4-vital-environmental-changes-
evidenced-in-india-since-lockdown-1673726-2020-05-02, last accessed 3rd March
2021.
Grodzińska-Jurczak, M., Krawczyk, A., Jurczak, A., Strzelecka, M., Rechciński, M. &
Boćkowski, M. (2020). Environmental choices vs. Covid-19 pandemic fear – plastic
governance re-assessment. Society Register, 4(2), 4–66. doi: 10.14746/sr.2020.4.2.04.
Green or brown 209
Hepburn, C, O’Callaghan, B., Stern, N., Stiglitz, J. & Zenghelis, D. (2020). Will
COVID-19 fiscal recovery packages accelerate or retard progress on climate
change? Oxford Review of Economic Policy, 36(Supplement_1), S359–S381, https://doi.
org/10.1093/oxrep/graa015
Ing, J. & Nicolaï, J. (2020). COVID‑19 and Climate Change: Should Governments Tie
Corporate Bailouts to Environmental Efforts or Strengthen Current Environmental
Policies? In S. Schumacher (Ed.) Perspectives on the Economics of the Environment
in the Shadow of Coronavirus, Environmental and Resource Economics, 76. 447–517,
https://doi.org/10.1007/s10640-020-00493-2.
Intergovernmental Panel on Climate Change. (2018). Global warming of 1.5°C. An IPCC
Special Report on the impacts of global warming of 1.5°C above pre-industrial levels and related
global greenhouse gas emission pathways, in the context of strengthening the global response to
the threat of climate change, sustainable development and efforts to eradicate poverty. Retrieved
from https://www.ipcc.ch/site/assets/uploads/sites/2/2019/06/SR15_Full_Report_
High_Res.pdf, last accessed 5th March 2021.
Jones, B. A., Grace, D., Kock, R., Alonso, S., Rushton, J., Said, M. Y., McKeever, D.,
Mutua, F., Young, J., McDermott, J. & Pfeiffer, D. U. (2013). Zoonosis emergence
linked to agricultural intensification and environmental change. Proceedings of the Na-
tional Academy of Sciences of the United States of America, 110(21), 8399–8404. https://doi.
org/10.1073/pnas.1208059110
Lombrana, L. M. & Rathi, A. (2020). Germany Just Unveiled the World’s Greenest Stim-
ulus Plan. Bloomberg Green. Retrieved from https://www.bloomberg.com/news/
articles/2020-06-05/germany-s-recovery-fund-gets-green-hue-with-its-focus-on-
climate, last accessed 5th March 2021
McCloskey, B. & Heymann, D. (2020). SARS to novel coronavirus - old lessons and new
lessons. Epidemiology and Infection, 148, e22. doi: 10.1017/S0950268820000254
NITI Aayog & Rocky Mountain Institute (RMI). (2020). Towards a Clean Energy Econ-
omy: Post-COVID-19 Opportunities for India’s Energy and Mobility Sectors. Retrieved
from https://rmi.org/wp-content/uploads/2020/07/India-Green-Stimulus-Report_
NITI_VF_ June_2020.pdf, last accessed 5th March 2021
Patrício Silva, A. L., Prata, J. C., Walker, T. R., Duarte, A. C., Ouyang, W., Barcelò, D.
& Rocha-Santos, T. (2021). Increased plastic pollution due to COVID-19 pandemic:
Challenges and recommendations. Chemical Engineering Journal (Lausanne, Switzerland:
1996), 405, 126683. https://doi.org/10.1016/j.cej.2020.126683
Pinner, D., Matt, R. & Samandari, H. (2020). Addressing climate change in a post-
pandemic World. In D. Pinner (Ed.) McKinsey on Climate Change (pp. 106–111). Retrieved
from https://www.mckinsey.com/~/media/McKinsey/Business%20Functions/
Sust a inabi l it y/Our %20Insig ht s/McK insey %20on%20Cl i m ate%20Change/
McKinsey-on-Climate%20Change-Report.pdf, last accessed March 5th, 2021.
Press Trust of India a (2020 May 20). India’s ‘mammoth’ COVID-19 package much
smaller than it seems, says Fitch Solutions. The Economic Times, Policy. Retrieved from
https://economictimes.indiatimes.com/news/economy/policy/indias-mammoth-
covid-19-package-much-smaller-than-it-seems-says-fitch-solutions/articleshow/
75823604.cms, last accessed 5th March, 2021
Press Trust of India b (2020 June 05). COVID-19 lockdown-like interventions may
help combat air pollution in India, say scientists, Hindu, Scitech. Retrieved from
https://www.thehindu.com/sci-tech/energy-and-environment/covid-19-lockdown-
like-inter ventions-may-help-combat-air-pol lution-in-india-say-scientists/
article31746151.ece, last accessed 5th March, 2021
210 Runa Sarkar
Shehabi, M. (2020). Novel policy research on the resource exporters‑energy policy.

Perspectives on the economics of the environment in the shadow of coronavirus.
Environmental and Resource Economics, 76, 447–517. https://doi.org/10.1007/s10640-
020-00493-2
Sterner, T., Cantillon, E., Kåberger, T. & Henry, C. (2020). COVID response: Use pub-
lic finances to support change. Springer Nature Sustainable Community. Retrieved from
https://sustainabilitycommunity.springernature.com/posts/66541-covid-response-
use-public-finances-to-support-change, last accessed 5th March, 2021
United Nations Environment Programme & International Livestock Research Institute.
(2020). Preventing the Next Pandemic: Zoonotic Diseases and How to Break the Chain of
Transmission. Nairobi, Kenya. Retrieved from https://www.unep.org/resources/
report/preventing-future-zoonotic-disease-outbreaks-protecting-environment-
animals-and, last accessed 5th March, 2021
Vivid Economics. (2020). Greenness of Stimulus Index. Retrieved from https://www.
vivideconomics.com/wp-content/uploads/2020/09/GSI_924.pdf, last accessed 5th
March, 2021
What to Do about Zombie Firms. (2020 September 26). The Economist, Leaders. Re-
trieved from https://www.economist.com/leaders/2020/09/26/what-to-do-about-
zombie-firms, last accessed 5th March, 2021
Woetzel, J., Dickon, P., Samandari, H., Engel, H., Krishnan, M., Boland, B. & Powis, C.
(2020). Climate Risk and Response: Physical Hazards and Socioeconomic Impacts. Retrieved
from https://www.mckinsey.com/business-functions/sustainability/our-insights/
climate-risk-and-response-physical-hazards-and-socioeconomic-impacts, last ac-
cessed 5th March, 2021
World Wildlife Fund. (2020). The Loss of Nature and the Rise of Pandemics- Protecting Hu-
man and Planetary Health. Italy. Retrieved from https://wwfeu.awsassets.panda.org/
downloads/the_loss_of_nature_and_rise_of_pandemics___protecting_human_
and_planetary_health.pdf, last accessed 5th March, 2021
Wu, X., Nethery, R. C., Sabath, M. B., Braun, D. & Dominici, F. (2020). Air pollution
and COVID-19 mortality in the United States: Strengths and limitations of an eco-
logical regression analysis. Science Advances, 6(45), p.eabd4049, doi: 10.1126/sciadv.
abd4049
PART III
Sectoral perspectives
13
COPING WITH COVID-19 AND ITS
CONSEQUENCES
Household adjustments in India
Arnab Mukherji and Arjun Shatrunjay
13.1 Introduction
Epidemics and infectious diseases have had profound impacts on how people
live their lives since recorded history. One part of this is driven by people’s own
attempts to avoid exposure to the epidemic, while a second part is how public
policy impacts people’s lives and livelihoods (Neustadt 2011; Roy 2012). Figures
13.1 and 13.2 present a temporal picture of the spread of Covid-19 (C19) in India
as well as the stringency of Government of India’s public policy response to this.
India’s current caseload is near 81 lakh confirmed cases; with 0.5 lakh new cases
every day, India is set to soon overtake the count of cases in the USA – currently,
the country with the highest incidence of C19. Little is known about the con-
sequence of this on how households have dealt with the twin crises of epidemic
outbreak and economic lockdown, and we present some preliminary context and
numbers to these unfolding phenomena in this chapter.
The “twin E crisis” of the simultaneous and co-dependent crisis of the epi-
demic outbreak and economic lockdown present a challenge that has stretched
India in ways that are unheralded – at the one end, significantly fragmented
health systems are forced to deal with an infectious disease whose management
is still something that we are figuring out globally. At the other end, India’s
largely informally employed workforce faces significant challenges as economic
activities close and there are major policy and access gaps in the existing social
security infrastructure (Ghosh 2020). Just as the outbreak of C19 was unantici-
pated, particularly its local outbreak in cities, neighbourhoods, communities and
households, the lockdown announcements and how its imposition was managed
locally saw a reasonable amount of uncertainty. For example, during the initial
lockdown, the shaded part in Figure 13.2, movement of essential goods and ser-
vices and those needed to support these value-chains were permitted. However,
DOI: 10.4324/9781003220145-16
214 Arnab Mukherji and Arjun Shatrunjay
Daily
FIGURE 13.1 counts of Covid-19 new confirmed cases, additional active cases
and recoveries as reported on 28th October 2020.
FIGURE 13.2 Oxford C19 government response tracker.

Coping with Covid-19 and its consequences 215
there appears to have been widespread confusion about what constitutes essential
goods and services (Onmanorama Staff 2020; Scroll Staff 2020).
India’s stringent policy response has been widely discussed in a debate if
“lives lost” is more important than “livelihoods lost” or vice versa (see Ray and
Subramanian 2020). By 2019, India’s economy was already slowing down and
resembled a typical demand-constrained economy. The Center for the Mon-
itoring of the Indian Economy’s (CMIE) unemployment data shows that the
lockdown led to a rise in unemployment rate from the usual 7–8% between
January and March 2020, to 23% in April! By June 2020, unemployment fell
to 10% and this further reduced to 8.3% by August 2020. Ministry of Statistics
and Program Implementation quantifies the April–June quarter GDP growth
to have declined by 23.9%. This constitutes the largest contraction that the
Indian economy has ever seen in a quarter and the first economy-wide contrac-
tion since 1980 – 40 years ago. Two other consequences of the lockdown are
worth mentioning – air pollution across cities recorded drastic improvements
during this time (Mahato et al. 2020; Sharma et al. 2020) and river waters
became cleaner and potable (Lokhandwala and Gautam 2020; Picheta 2020).
These measures simply reinforce the observation that the economy shut down
suddenly, and substantively.
Dasgupta and Rajeev (2020) characterize this phenomenon using a macro-
economic model to argue that the demand-deficient economy combined with
the lockdown meant that a demand-deficient economy was transformed into
a supply constrained economy. A lockdown meant a separation from jobs in
most instances, particularly in the informal sector where incomes were low
and savings very slim. As the lockdown began relaxing, the Indian Railways
began operations to ferry home those migrant workers who were stuck at their
workplace. Thus, the lockdown-induced supply shocks in many sectors, and the
resultant labour migration meant that many supply chains and their associated
support ecosystems were no longer operative leaving supply chains in disarray.
Dasgupta and Rajeev (2020) argue that the current circumstances will need
both demand side and supply side measures; on the demand side, it will be
important to re-establish consumer confidence, ensure social security and im-
proved liquidity, while on the supply side, measures to reduce search costs and
re-match skills in a largely informal sector can play an important role. Mukherji
and Basu (2020) document some of the concerns arising in the specific con-
text of the lockdown-induced separation of the worker from the employer in a
largely informal economy.
The lockdown not only shut down most of the economy, but it also meant
that households dependent on daily wages were suddenly not able to finance
their consumption. The largely informal nature of employment in India im-
plies that this unanticipated economic shock will have meant significant adjust-
ments to household budgets, dissavings, and increased borrowing and financial
dependency for most households (Deaton 1992). These adjustments are well
documented in the economic literature on how households manage risk and
uncertainty in their livelihoods (Morduch 1995). When formal institutions to
manage risks for households such as social security and a wide variety of bor-
rowing and insurance markets are missing, informal strategies arise to smooth
consumption. Some examples of this include migration, reduced consumption,
depleting assets, changing time-use and work, and increased informal borrow-
ing. Many of these are associated with long-term negative consequences for
income and welfare. In the context of the C19-driven lockdown and subsequent
policy, the natural question that arises is how was consumption expenditure of
households impacted?
13.2 The CPHS data

Understanding the consequences of economic shocks requires carefully collected
data on consumption expenditure. While collection of consumption data has a
rich history in India, it needs to be collected carefully and systematically; ex-
penditure data is not collected at the time of expenditure and hence households
have to recollect data where the length of the recall period, and thus biases,
varies with the frequency with which the expenditure is made and cues from the
data collector are very important (Beckett et al. 2001). The Consumer Pyramids
Household Survey (CPHS), organized by the CMIE, is a year-long longitudinal
survey that tracks consumption expenditure of households. Vyas (2018) carefully
documents CMIE household survey’s sampling design and discusses important
trends in consumption expenditure that can be attributed to the policy of de-
monetization. The CMIE is meant to be a nationally representative survey with a
stratified, multi-stage sampling design based on identifying geographically inde-
pendent units that it terms as homogenous regions (HRs). These HRs represent
one or more neighbouring districts within a state, with similar agro-climatic
conditions, female literacy, household size and urbanization levels. HRs are split
into urban and rural strata with town and villages forming the primary sampling
units (PSU) in each of the respective stratums. Systematic simple random sam-
pling within each PSU is used to select the sample of households to be studied
(Vyas 2020).
Two further aspects of the data make this unique: (a) Being a longitudinal
dataset, this allows us to track the same set of households over time to enable us
to see how the same household responds to different macroeconomic conditions;
and (b) data collection is done throughout the year and thus, it is possible to
produce monthly estimates; further, the entire sample consists of four separate
cohorts that are sampled once every year. Thus, the January–March cohort in
years t and t + 1 would be the same after accounting for loss due to attrition and
non-response. We work with the October–December quarterly sample and the
January–March quarterly sample for all our estimates; in addition, we also use
data from the month of April. Ideally, we would like to include the April–June
quarter as well, but this is yet to be released as this fully overlaps with the lock-
down period. We find some initial responses already setting into consumption
patterns already in the earlier quarter.
13.3 Nation-wide quarterly trends in household

economic aggregates
Household consumption is key to assessing their welfare and poverty estimates
in India which continue to be based on household consumption expenditure.
Table 13.1 presents quarterly estimates for income and expenditure for the last
two quarters of the financial years 2017–18, 2018–19 and 2019–20 and the first
month of the 2020–21 year. The CPHS data reports both household income (Y )
and household expenditure (E) and using this we calculate the residual category
(R = Y − E) to measure the excess of income over expenditure. Each of these
aggregates shows somewhat different trends:
Income: The average monthly income reported by each of the quarter-specific
samples is Rs. 22,581 and this reports significant variation within the sample
with an average coefficient of variation (CoV) of 105%. Thus, income is dispersed
quite widely with the mean and standard deviation being roughly equal. Across
quarters, the mean and CoV also vary significantly; income varies from a max-
imum of Rs. 24,257 in the October–December quarter of 2018 to a minimum
of Rs. 19,473 in the January–March 2020 quarter. The January–March 2020
TABLE 13.1 Quarterly trends in income and expenditure
Quarter/Month Income (Y) Expenditure (E) Residual (R)
Mean COV Mean COV Mean COV

(₹) (%) (₹) (%) (₹) (%)
October– 21,374 114 11,918 67 9,456 233

December 2017
January–March 22,175 96 12,327 65 9,849 190
2018
October– 24,257 105 13,355 68 10,902 206
December 2018
January–March 23,970 103 13,169 62 10,802 204
2019
October– 24,237 111 13,418 62 10,819 221
December 2019
January–March 19,473 101 11,023 68 8,450 200
2020
Month of April 13,956 176 7,526 59 6,429 363
2020
Source:
1 Author calculations from CMIE’s CPHS data.
2 The month of April includes April–June quarter households who were interviewed in April
2020. All other rows include households in all three months of the quarter.
3 COV = [(standard deviation/mean) *100] is the coefficient of variation.
4 CPHS data reports household income and consumption expenditure. We construct R = Y − E
as the excess of income over current consumption expenditure.
5 All ₹s are in nominal terms and are not inflation-adjusted.
quarter is the last quarter for which we have data and the first quarter where
households were partially exposed to the lockdown during the last two weeks.
The average income is the lowest for the month of April 2020 at Rs. 13,956 and
within the month, the CoV rises to 176% indicating that not only did the average
incomes fall, but also that the distribution of income expanded significantly.
Expenditure: The average monthly expenditure is lower than the average in-
come for each of our quarterly samples indicating that the average person in the
sample is above subsistence levels. The average expenditure is Rs. 12,535 across
the six quarterly samples with a maximum in the October–December 2018 quar-
ter at Rs. 13,355 and a minimum in the January–March quarter of Rs. 11, 023.
It falls further to Rs. 7,526 in the month of April, i.e. the first full month of
lockdown. The CoV for expenditure is smaller than that of income indicating
that income varies within and across the cohort much more than expenditure
does. The average quarterly CoV for expenditure is 65% and with lockdown this
declines to 59% as expenditure levels decline.
Residual: The average monthly residual measures the average monthly income
after accounting for monthly current expenditures for the households and collec-
tively reflects household savings, or transfers that reflect gifts, payment towards
past borrowings, etc. The mean value of the residual is Rs. 10,046 across the six
quarters and is the lowest in the January–March 2020 quarter at Rs. 8,450; it
further declined to Rs. 6,429 during April 2020. The average CoV for this resid-
ual category is much higher than for income or expenditure at 209% effectively
indicating that the standard deviation is almost twice the mean. Further, this
sample-wide dispersion is accentuated significantly in the month of April 2020
and rises to 363% indicating even greater dispersion suggesting that household
budgets and expenditure patterns would have been widely affected.
The preliminary evidence from Table 13.1 shows that household budgets have
been deeply affected by C19 – incomes declined by 14% in the January–March
2020 cohort from the average and it further reduced by 38% in April 2020.
This is in line with the 23.9% decline reported by MoSPI for the April–June
quarter. Further, the increase in CoV suggests that this loss of income has ac-
centuated the dispersion of income; some households must have been impacted
more than other households. However, while monthly consumption expenditure
shows a 40% decline from its average levels in April 2020, this decline does not
see too much change in dispersion. Thus, while mean consumption and hence
welfare have gone down, this has been a variance-preserving decline, and the
entire sample has been affected uniformly. Finally, the residual income absorbs
all the variation in income and is crucial in ensuring that households maintain
their monthly consumption and hence both the mean levels decline and the CoV
increases – it almost doubles! Thus, with the variance of consumption expendi-
ture being smaller than the variance of household incomes, households appear to
be relying on residual incomes to try and maintain their consumption expendi-
ture. However, this dependence is not sufficient to insure against income losses
faced by households.
TABLE 13.2 Quarterly trends in the Gini Coefficient
Quarter/Month Income (Y) (%) Expenditure (E) (%)

October–December 2017 42.8 28.3
January–March 2018 42.1 27.7
Month of April 2020 66.5 25.5
Source:
Clearly, the impact of C19 on the distribution of income and expenditure is

quite significant. One way to measure this is to look at the Gini Coefficient for
the same quarterly samples as we saw in Table 13.1. We do so in Table 13.2 where
we see that both income and consumption expenditure show an increase in the
Gini Coefficient. Table 13.2 clearly establishes that income distribution has been
the most unequal in January–March 2020 quarter among all the quarters. This
trend is further accentuated if we look at April 2020 – income inequality has
risen sharply for households in this month as the Gini Coefficient is 66.5%. This
sudden and shocking rise in the Gini Coefficient captures an astonishingly large
distortion in the income distribution process within the sample and parallels the
100% lockdown recorded by the Oxford Government response tracker.
From an expenditure perspective, the Gini Coefficient is much smaller re-
flecting a much more egalitarian distribution of consumption in the sample and
while there is a rise in the January–March 2020 quarter, this level of inequality
is observed in earlier quarters as well – the October–December quarter of 2018
and 2017 and hence less startling. Unlike with income, expenditure inequality
records a decline in April 2020 and reduces to 25.5% from the 28.2% recorded for
the January–March 2020 cohort. Thus, while expenditure declined significantly,
the distribution of expenditure became somewhat more equitable.
Thus, at the national level, the economy had strong effects on the budget of
households and their ability to manage their routine activities – not only can
we expect income to have declined significantly, but that a wide range of other
expenditures have been reduced, including routine consumption. This decline
in consumption would have been much larger had they not utilized household
savings and forgone on other expenditures. The lack of a broad social security
system for much of the informally contracted workforce essentially is reflected in
this data showing that income risk is fully absorbed by households as they adapt
to the breakdown of income-generating process in the economy due to C19.
Households vary significantly from each other, and the means show important
gradients on different key demographic indicators (Gould 2013). Very helpfully,
CPHS classifies households on two important demographic gradients – education
of all members of the household and occupations of the key income earner in
the household. Table 13.3 presents how the mean income varies across different
educational groups, while Tables 13.4 and 13.5 present the gradients for different
occupational groups. With households able to account for some of the income
shock in their consumption profiles, we focus on just the income shock for the
subsequent discussion.
TABLE 13.3 Income trends in households, by educational categories
Education Categories Quarter/Month Mean CoV

(₹) (%)
All Graduates household January–March 2019 46,920.7 75.7

January–March 2020 47,098.0 81.9
April 2020 20,774.1 99.8
All Matriculates household January–March 2019 23,334.6 93.6
January–March 2020 18,216.7 90.8
April 2020 15,271.2 229.6
Graduates dominated household January–March 2019 63,085.7 65.4
January–March 2020 63,720.7 59.6
April 2020 35,961.9 95.4
Graduates majority household January–March 2019 34,741.9 78.1
January–March 2020 31,672.8 71.2
April 2020 19,790.3 114.0
Graduates minority household January–March 2019 34,062.5 98.3
January–March 2020 27,875.4 69.1
April 2020 18,202.7 181.8
Households of all illiterates January–March 2019 11,728.4 144.7
January–March 2020 10,469.7 93.3
April 2020 7,166.6 222.8
Households of all literates January–March 2019 16,138.9 95.7
January–March 2020 13,616.2 92.6
April 2020 10,746.7 231.6
Households of some literates January–March 2019 15,734.8 115.3
January–March 2020 12,847.2 94.2
April 2020 12,414.5 164.6
Matriculates dominated household January–March 2019 32,975.5 94.4
January–March 2020 26,079.0 66.8
April 2020 13,124.8 228.1
Matriculates majority household January–March 2019 21,049.5 99.5
January–March 2020 16,984.0 91.4
April 2020 16,850.6 224.4
Matriculates minority household January–March 2019 23,662.2 92.7
January–March 2020 18,252.5 85.3
April 2020 14,841.2 224.5
Source:
TABLE 13.4 Income trends for households by employment categories who faced income
losses
Employment Categories Quarter/Month Mean CoV

(₹) (%)
Agricultural Labourers January–March 2019 13,407.6 61.8

January–March 2020 11,272.4 63.6
April 2020 6,058.0 128.1
Business & Salaried January–March 2019 40,971.1 67.2
Employees
January–March 2020 32,236.2 76.3
April 2020 21,316.5 91.6
Entrepreneurs January–March 2019 37,875.5 80.4
January–March 2020 29,120.4 83.2
April 2020 2,557.91 303.7
Home-based Workers January–March 2019 13,608.9 62.2
January–March 2020 14,250.6 54.5
April 2020 2,052.9 175.3
Organized Farmers January–March 2019 30,574.3 168.8
January–March 2020 24,440.5 177.9
April 2020 7,478.5 199.4
Qualified Self-Employed January–March 2019 43,758.8 82.6
Professionals
January–March 2020 31,583.3 75.5
April 2020 14,241.5 141.9
Retired/Aged January–March 2019 21,896.6 93.5
January–March 2020 19,150.7 100.7
April 2020 3,782.4 291.5
Self-employed Entrepreneurs January–March 2019 18,805.2 78.5
January–March 2020 15,726.0 161.3
April 2020 3,655.5 138.7
Support Staff January–March 2019 19,880.0 67.6
January–March 2020 17,689.6 73.7
April 2020 1,716.7 308.3
White-collar Professional January–March 2019 49,751.3 62.4
Employees
January–March 2020 47,962.5 56.3
April 2020 6,058.0 128.1
Managers/Supervisors January–March 2019 61,882.4 64.5
January–March 2020 47,753.9 71.0
April 2020 8,932.5 75.6
Source:
TABLE 13.5 Households by employment categories who experienced income gains
Employment Categories Quarter/Month Mean CoV

(₹) (%)
Industrial Workers January–March 2019 25,403.3 62.9

January–March 2020 23,524.9 67.4
April 2020 29,380.1 97.1
Legislators/Social Workers/ January–March 2019 21,465.1 94.1
Activists
January–March 2020 20,181.5 151.0
April 2020 41,037.3 41.9
Non-industrial Technical January–March 2019 19,725.9 67.1
Employees
January–March 2020 17,336.9 64.5
April 2020 28,707.4 170.9
Small Trades/Hawkers January–March 2019 14,979.1 57.7
January–March 2020 13,131.5 58.2
April 2020 28,081.5 159.8
Small/Marginal Farmers January–March 2019 14,994.8 96.7
January–March 2020 12,184.7 109.8
April 2020 14,886.4 88.0
Wage Labourers January–March 2019 13,905.7 54.9
January–March 2020 11,540.8 62.5
April 2020 28,617.4 47.4
Source:
13.4 Heterogeneity along the education gradient

Table 13.3 shows that across all possible education classifications of households
from all adult members being illiterate, to all adult members being a graduate,
income in April 2020 is less than the incomes for households in the January–
March 2020 quarter. The data shows that incomes are the highest for house-
holds where 75% or more are graduates, i.e. graduate-dominant households.
Graduate-dominant households make around Rs. 63,500 per month in both the
January–March 2019 and 2020 quarters; however, in April 2020, this group’s
incomes reduce to Rs. 35,961, i.e. a 56% reduction in incomes. Similarly,
households with all adult members with a graduate degree make Rs. 47,000 per
month in both the January–March 2019 and 2020 quarters; however, in April
2020, this group’s incomes reduce to Rs. 20,774, i.e. less than 50% of what
it was. Households, where all members were illiterate, on average made Rs.
11,724 in the January–March 2019 quarter, Rs. 10,469 in the January–March
2020 quarter and Rs. 7,166 in the month of April 2020. The educational gra-
dient slopes in just one direction – for all educational levels, households receive
lower incomes during the lockdown.
While all households see declines, declines were relatively minor for some of
the educational groups whose incomes were not very high in non-C19 condi-
tions. For example, a household where some members are literate would receive
an income of Rs. 15,734 in the January–March 2019 quarter, Rs. 12,847 in
the January–March 2020 quarter and Rs. 12,414 in the month of April 2020.
Thus, while these educational groups make low incomes, their incomes are not
significantly affected during the lockdown. Some other examples of this are –
matriculate-majority households (households with between 50% and 75% of its
adults with a matriculate degree), and matriculate-minority households (house-
holds with between 25% and 50% of its adults with a matriculate degree) and
even households where all adults are literates.
The CoV seems to also vary with the level of education. Households with
graduate-dominant households (60–65%) or all graduate households (75–81%)
tend to have lower values than other households, particularly when compared
to households of all illiterates (93–144%), or households with some literates
(94–114%). Looking at the month of April 2020, we see that incomes across
all educational categories report a high CoV; however, its rise has also been
the lowest in the graduate-dominant households. Households with lower levels
of education, beginning with households of all illiterates, report CoV in the
range of 220–225% indicating significantly more variation during the period
of lockdown.
Thus, there appears to be a non-linear relationship between education level
and income – households with high education tend to make more income; how-
ever, they also suffer large average declines and this seems to have the entire
group in approximately similar ways. In comparison, lower education house-
holds make much less income, but they also do not suffer as large a proportionate
decline in income, even though the absolute values suggest significant deepening
of poverty. It is also the lower education group that reports higher income dis-
persion in April 2020 as some people make far less income than others.
13.5 Heterogeneity along the occupation gradient

Unlike education, the occupational gradient slopes downwards for many income
groups, and slopes upwards for others indicating that employment cuts the sam-
ple in a very different way. Occupations such as agricultural workers, home-
based workers, organized farmers, self-employed entrepreneurs, entrepreneurs,
qualified self-employed professionals, support staff, white-collar professional
employees, retired, mangers or supervisors, and business and salaried employ-
ees record large declines in April 2020 and most of them also record declines
in the January–March 2020 quarter when compared with the January–March
2019 quarter. However, household where the primary earner was working as an
industrial worker, as a legislator or a social worker or activist, as a non-industrial
technical employee, in small trades or as a hawker, small or marginal farmer, or
as a wage labourer, all see important rises in income.
Surprisingly, the size of the losses for households in occupations that lost income
and the size of the gains for households in occupations that saw gains are compa-
rable. While some occupations that saw incomes fall saw a decline by as much as
80% or more (see entrepreneurs, support staff, white-collar professionals, managers
or supervisors), most other report halving of their incomes or less (e.g. qualified
self-employed, business & salaried employees, agricultural workers). However,
households that were in occupations that saw increases, for many categories, saw
their incomes double, for example, wage labourers, small trades or hawkers.
By and large, the dispersion of income within each employment category
appears to have increased during the lockdown. However, for some employment
categories where incomes saw a rise during the lockdown, the CoV appears to
have actually declined during the lockdown, for example, wage labourers, small
or marginal farmers, non-industrial technical employees and legislators/social
workers/activists. These occupation categories see both an increase in salaries
during April 2020 and a systematic and large decline in the CoV of income for
households in these income categories.
There is clear evidence that across formal and informal employment, across
high-skilled and less-skilled occupation categories, and across rural and urban
geographies, households were impacted in very different ways. A more careful
analysis of the underlying data is needed to disentangle how income distribution
varies differently across these income groups. However, it is reasonable to expect
that if access to social security is incomplete, the impact of fluctuations in income
would also be experienced in very different ways in the economy.
13.6 Coping with income shocks

While there is some similarity in experiences in the spread of C19 in the pop-
ulation, there are also important differences. However, even though the disease
outbreak may have varied across the country, the policies adopted, at least for the
month of April appear to have been reasonably universal in their roll-out under
the Epidemic Diseases Act of 1897. One initial sign of crisis almost always is
hoarding of cash due to the collapsing economy; RBI data reported on the NPCI
platform suggests that cash in circulation rose by 23% by September 2020 in
comparison to a year ago, and over 13% from March 2020. With digital payments
registering an initial decline, they quickly returned to pre-C19 levels by August
2020 in terms of volumes (25,000 million transactions) and value (nearly 14,000
billion rupees). Digital payments have absorbed some of the liquidity-driven pre-
cautionary motives to hold cash in hand. However, as the economy collapsed
with the spread of disease and policy-induced lockdowns, people also began suf-
fering income losses and the size of some of these income shocks was clearly
larger than what household could insulate against.
All households have some inherent ability of insurance against limited shocks
and when they are idiosyncratic, households even in informal settings have a
stronger ability to weather income shocks through a complex web of ex-post

risk-mitigating mechanisms such as dissaving, sale of assets, gift-giving and bor-
rowing. However, when income shocks affect everyone, the fiscal space for com-
munity members to help each other out shrinks enormously and households are
faced with strategies that are forced, and some with larger negative consequences
than others.
Hunger: The Global Hunger Index (GHI) placed India 94th in a set of 107
countries for which data was available for their 2020 rankings. The HGI value of
27.2 is squarely in the middle of their classification of a country facing “serious
hunger” (20–34.9). It is not apparent how well this reflects conditions on the
ground today since at least some of the data used, for example, stunting or wast-
ing for children, is drawn from data before C19.The literature clearly documents
that reductions in food intake and increasing calorie deficits are widely resorted
to when facing income shocks induced either by the market, diseases or through
public policy (Currie 2000; Patnaik 2007; Barrett and Lentz 2017; Mukherji
2013). This chapter already documents that even though declines in household
consumption were less than that of household incomes, but these declines are im-
portant enough to expect that hunger deepened significantly in India during the
lockdown. A breakdown of the consumption basket of households would reflect
both the calorie deficit and the micronutrient deficits that households faced due
to the C19 and policy-induced economic collapse.
Migration: As soon as public policy permitted travel, many households took the
opportunity to reverse the internal migration driven by economic reasons. One
measure of this is that Indian Railways recorded more than 4,620 Shramik Trains
between 1st May and 31st August 2020 to help migrants return home (Patel and
Behera 2020). At an estimated cost of Rs. 36,960 crores, this would probably
amount to close to 63 lakh migrants and their family members who returned
home using railways alone. Many undocumented migrations took place through
buses, on cycles and many who simply walked home over hundreds of kilometres –
another clear metric of the collapse of the economy. The massive labour out-
flow from labour-deficient states such as Gujarat and Maharashtra implies that
their policy challenges are very different from labour-surplus states such as Uttar
Pradesh and Bihar.
The pattern of migration and the running of trains allow us to classify the
states of India into three categories – (a) labour-surplus states such as Assam,
Bihar, Jharkhand and Madhya Pradesh, (b) labour-deficient states such as Delhi,
Gujarat, Karnataka, Maharashtra and Punjab and (c) both such as Rajasthan and
Uttar Pradesh. States that are labour-surplus are the destination of much of the
migration that took place once travel was permitted, while labour-deficient
states are those that are the places from where migration originated. States such
as Rajasthan and Uttar Pradesh represent cases wherein there is either significant
migration within the state or there is significant inflow and outflow of migrants
across the state boundaries.
Household incomes and expenditures pre-C19 are the highest in labour-

deficient states, followed by states that exhibit both labour-surplus and -deficient
patterns, and the lowest are in labour-surplus states. While there are specific
differences, this regularity in the data is quite clear. It is unclear how well the
CPHS is able to track migration during the C19 phase, but these patterns would
be further accentuated if one were to look at per capita income and per capita
expenditure after accounting for labour migration as many migrants choose to
return to their rural homes that they had supported with income from economic
activities in labour-deficit states. Dispersion of income increases across all states
independent of where they are in the migration flow and while everyone ex-
periences declines in incomes, it is the dispersion in expenditures and residual
incomes that masks the underlying adjustments that households make in the face
of loss of incomes.
Demanding Greater Employment: As the income crisis deepened due to declining
economy activity, households seek alternative ways to support themselves. The
Mahatma Gandhi National Employment Guarantee Act of 2005 (MGNREGA)
plays an important role in these kinds of crisis since it allows households to reg-
ister their demand for work since the employment guarantee is demand-driven.
Figure 13.3 documents the monthly demand for work across five labour-surplus
states for the period April 2019 to June 2020. It is only when these demands are
registered do subsequent steps get taken in terms of its sanction and grant of work
to those who demand it. Using administrative data from MGNREGA portal, it
is immediately obvious that there is a large spike in the demand for work across
all states in May 2020 – the first month in which households return to their
homes. The MGNREGA allows the data at greater geographical and temporal
3000
2500
2000
1500
1000
500
0
19
0
19
20
20
M 9
0
Ja 9
M 0
9
19
Au 9
9
-2
r-1
-1
r-2
-1
2
-1
l-1
-1
-2
p-
n-
n-
n-
b-
g-
ay
ct
ec
ov
ay
ar
Ap
Ap
Ju
Se
Ju
Ju
Fe
O
M
N
Madhya Pradesh Bihar Odisha Assam Jharkhand
FIGURE 13.3 Monthly demand for work under MGNREGA.

granularity and looking at district or block level data shows significantly larger
spikes, particularly when looking at districts affected by migration flows.
While the law grants a maximum of 100 days of work in a year at the mini-
mum wage rates, the actual utilization of the scheme has rarely approached this
number. While the demand for work in the months of May, June and later has
registered historic highs, these demands have not translated into equivalent addi-
tional person-days of work being provided under the scheme (Mukherjee 2020).
In recognition of the crisis, the Union government released an additional Rs.
40,000 crores making the MGNREGA’s annual allotment Rs. 100,000 crores
for the year – a first ever. The level of income and consumption distress are so
high that workers in many parts of the country staged protests to ensure release
of work, raise minimum wages and expand coverage from 100 to 200 days of
the year. Thus, this spike is no more than a signal of distress that was registered
through official channels – an indication that households are urgently looking to
augment their incomes as their routine sources of employment were lost.
13.7 Discussion
The C19 outbreak and its associated policy response have led to large changes
in the income distribution of households and resultant expenditures that house-
holds can support. As economic activity shrank both at the national level so did
average household incomes and household expenditures. As mean income and
expenditure declined, its dispersion too increased. To a limited extent, house-
holds were able to insulate from the income risk in their household expenditures;
however, almost everywhere average household consumption expenditure de-
clined by 40%. There is clear and worrying evidence to show that these declines
in income increased poverty and led to sharp increases in the Gini Coefficient –
the post-C19 income distributions are very different from the pre-C19 income
distribution reflecting an almost 50% worsening of the Gini Coefficient.
Households at different levels of education and occupation experience very
different effects of the lockdown. Households at all levels of education, from a
household where every adult is illiterate, to a household where every adult has
at least a graduate degree, show large declines in income during the lockdown.
Households with less education experience larger declines in their mean income
and greater increases in CoV. Unlike education, households in many employ-
ment categories saw significant declines in their incomes, while in other catego-
ries, households saw an increase in average incomes as well. Further exploration
of what separates employment groups that saw increases from those that saw
declines when the economic activity is slowing down is needed. On first pass,
it would appear that households that saw gains are employed in professions such
as wage labour and small trades, are all informal in nature, do not need signifi-
cant investment in capital and serve very local needs with shallow supply chains.
However, households that saw declines include high and low skills, high and low
salaries, and some are formal, while others are not. These effects have panned
out differently in states that have been labour-surplus and those that have been
receiving labour from other states and while the labour-surplus states are worse
off in an absolute sense today, it is not clear that the labour shortage states or
employers will find their pre-C19 incomes any time soon.
Much of the inference is drawn from the April 2020 data in this chapter.
While this data is valid, however, since the other units of analysis are quarters,
ideally, we should be using the data for the entire April–June 2020 cohort. This
would enable us to be more confident of the preliminary and indicative nature
of findings that we currently have. This would accommodate a comparison of
the pre-C19 quarter to the first full post-C19 quarter; it would also enable us to
consider the changing progression of the disease. As is clear from Figure 13.1,
the daily peak for C19 happens much later approximately during the week of the
16th of September. However, by this time, the economy is also unlocked and
thus, the narrative built here, while indicative of the reaction to the shock, is
both preliminary and necessarily an incomplete one. As part of the world head
towards their second lockdowns, there remains much uncertainty about what the
future will bring.
Low probability events such as the outbreak of C19 are expected to occur
infrequently, but that does not mean that they do not happen. Such situations
inevitably require major policy responses and investments by the government.
Tumbe (2020) provides a good description of government responses in the con-
text of cholera, plague and influenza in India spanning 1817–1920, i.e. during
the British Raj in India. Looking at other countries today, many countries have
initiated large fiscal stimuluses in response to C19 to trying and supporting their
economy. Countries such as Japan, Canada and Brazil have initiated a fiscal stim-
ulus to the tune of 21%, 16.4% and 12%, respectively, of their GDP as per IMF
data; India’s stimulus at 6.9% of GDP has been weaker in comparison. Indian
policy approach has, however, attempted to initiate institutional reforms in ex-
isting welfare schemes in response to C19. These may have long-term bene-
fits for household facing risk if these reforms take root; for example, having
one ration card across the country would create nation-wide access to India’s
public distribution system that has always been locally available at the migrant’s
home address. If done successfully, this may truly be transformative and protect
households in important ways. Similarly, expanding MGNREGA to urban areas
or to newer activities can create new opportunities for employment. However,
separation from work is a problem that will not be resolved quickly. Those on
daily wages will not have the luxury to wait for their employers to re-hire them,
and the employers may also not be in business to hire them; there is little in
place for all those separated from work to return to the pre-C19 status quo. Our
workforce-contracting norms and our social security systems are not designed
to be robust to changes in status quo. This may be a chance to reframe labour
policy, expand public distribution systems and build institutions to support social
security and clearing of the labour market. This will build both long-term resil-
ience and a return to much needed work for many.
References
Barrett, Christopher B., and E. Lentz. “Hunger and food insecurity.” in David Brady and
Linda M. Burton (Ed.) The Oxford Handbook of Poverty and Society (2013), Chapter 26;
23 pages; DOI: 10.1093/oxfordhb/9780199914050.013.27.
Beckett, M., J. Da Vanzo, N. Sastry, C. Panis, and C. Peterson. “The quality of retro-
spective data: An examination of long-term recall in a developing country.” Journal of
Human Resources 36, no. 3 (2001): 593–625.
Currie, B. The Politics of hunger in India: A study of democracy, governance, and Kalahandi’s
poverty. Springer, 2000.
Dasgupta, D., and M. Rajeev. “Paradox of a supply constrained Keynesian equilibrium.”
Economic & Political Weekly 55, no. 22 (2020): 23.
Deaton, A. Understanding consumption. Oxford University Press, 1992.
Ghosh, P. “We Need a Marshall Plan to Fight Covid-19.” Ideas for India (Apr. 8, 2020).
url: https://www.ideasforindia.in/topics/macroeconomics/weneed-a-marshall-plan-
to-fight-covid-19.html; Last Accessed: 27th February 2021.
Gould, S. J. “The median isn’t the message.” AMA Journal of Ethics 15, no. 1 (2013):
77–81.
Lokhandwala, S., and P. Gautam. “Indirect impact of COVID-19 on environment: A
brief study in Indian context.” Environmental Research 188 (2020): 109807.
Mahato, S., S. Pal, and K. G. Ghosh. “Effect of lockdown amid COVID-19 pandemic on air
quality of the megacity Delhi, India.” Science of the Total Environment 730 (2020): 139086.
Morduch, J. “Income smoothing and consumption smoothing.” Journal of Economic Per-
spectives 9, no. 3 (1995): 103–114.
Mukherjee, S. “Covid-19 impact: Demand for work under MGNREGA saw 21% jump
last month”, Business Standard, 1st July, 2020; url: https://www.business-standard.
com/article/economy-policy/21-per-cent-rise-in-households-demanding-mgnrega-
work-in-june-over-may-120070101721_1.html; Last Accessed: 27th February 2021.
Mukherji, A. “Evidence on community-driven development from an Indian village.”
Journal of Development Studies 49, no. 11 (2013): 1548–1563.
Mukherji, A. and S. Basu. “Rebuilding broken employment networks in India” Turn,
Turn, Turn ( July 6, 2020). url: https://www.iimb.ac.in/turn_turn/rebuilding-broken-
employment-networks-india.php; Last Accessed: 27th February 2021.
Neustadt, R. E. “Thinking in time: The uses of history for decision makers.” Simon and
Schuster (2011).
Onmanorama Staff. “COVID-19 lockdown 3.0: Central guidelines unclear, confusion pre-
vails” (May 04 2020). url: https://www.onmanorama.com/news/nation/2020/05/03/
covid-19-lockdown-3-confusion-prevails-as-more-shops-set-to-re-o.html; Last
Accessed: 27th February 2021.
Patnaik, U. The republic of hunger and other essays. Three Essays Collective, 2007.
Patel, D. V., and K. C. Behera. “Shramik special train operation: Largest domestic evac-
uation”, Turn, Turn, Turn ( June 2, 2020). url: https://www.iimb.ac.in/turn_turn/
shramik-special-train-operation-largest-domestic-evacuation.php; Last Accessed:
27th February 2021.
Picheta, R. “People in India can see the Himalayas for the first time in ‘decades’ as the
lockdown eases air pollution.” CNN Travel 9th April (2020). url: https://edition.cnn.
com/travel/article/himalayas-visible-lockdown-india-scli-intl; Last Accessed: 27th
February 2021.
Ray, D., and S. Subramanian. India’s lockdown: An interim report. No. w27282. National
Bureau of Economic Research, 2020.
Roy, T. Natural disasters and Indian history. Oxford India Short Introductions. OUP Cat-
alogue, 2012.
Scroll Staff. “Coronavirus lockdown: Supply of essential items hit as confusion about ex-
emptions prevail” (March 25, 2020). url: https://scroll.in/latest/957174/coronavirus-
lockdown-supply-of-essential-items-hit-as-confusion-about-exemptions-prevail;
Last Accessed: 27th February 2021.
Sharma, S., M. Zhang, J. Gao, H. Zhang, and S. H. Kota. “Effect of restricted emissions
during COVID-19 on air quality in India.” Science of the Total Environment 728 (2020):
138878.
Tumbe, C. Age of Pandemics (1817–1920): How they shaped India and the World HarperCollins
India 2020. https://harpercollins.co.in/product/age-of-pandemics-1817-1920/
Vyas, M. “Using fast frequency household survey data to estimate the impact of demone-
tisation on employment.” Review of Market Integration 10, no. 3 (2018): 159–183.
Vyas, M. “Consumer pyramids household survey: Survey design and sample”, Technical
Report (12 March 2020), Center for Monitoring Indian Economy Pvt. Ltd: Mumbai.
14
FIRM FRAGILITY
A re-interpretation in the times of the
Covid-19 outbreak
Debdatta Saha
14.1 Introduction
The Covid-19 pandemic, which started in Wuhan, China in November/De-
cember 2019, has now covered almost all countries in the world. While this
pandemic, like the recent pandemic episodes of SARS (Severe Acute Respiratory
Syndrome in 2003) and MERS (Middle East Respiratory Syndrome in 2012), is
an exogenous shock to the system, recent estimates show that the quantum of
government dole-outs is approximately three times that of the 2008 global fi-
nancial crisis that developed endogenously from within the system of sub-prime
mortgages in the US.1 The large effect on most economies is seen as a result of an
interconnected, hyper-globalized world, where shock transmissions occur with-
out friction through supply chains and financial ties (Fombrun and Wally, 1992;
Cheema-Fox et al., 2020; Manheim, 2020; Bartik et al., 2020; Bostrom, 2013;
Demirguc-Kunt et al., 2020). The virus has brought with it the ‘new normal’
terminology, which refers to the collective of non-pharmaceutical interventions
such as lockdowns, social distancing and accentuated hygiene protocols (wearing
masks in public spaces, thermal screening to name a few). These signify radical
changes in the conduct of business in ways that endogenously generated financial
shocks have not been able to effect in recent times.2
At the time of writing this chapter, the pandemic is far from over in most
countries of the world.3 As with any ongoing event, it is difficult to conclude
any takeaways regarding specific impacts on the economy, other than very gen-
eral findings. Most trajectories, be it in the supply side or the demand side, are
changing in different countries, that too at different paces. This depends on the
particular pandemic phase that the region is witnessing.4 However, a common set
of challenges have emerged in every nation regarding firm survival, such as supply
chain disruptions5 (Ivanov and Dolgui, 2020; Cheema-Fox et al., 2020), structural
DOI: 10.4324/9781003220145-17
232 Debdatta Saha
changes in consumer demand,6 potential anti-trust concerns due to abuse of

dominance by online retailers7 and the rise of e-commerce firms (see Dannen-
berg et al. (2020) for evidence on German online retail in food and Hasanat et al.
(2020) for the impact of e-commerce in Malaysia during the pandemic).
Of these, we focus on the issue of firm fragility. Cross-country evidence
evolving during the pandemic points toward difficulties faced by small- and
medium-sized firms due to non-pharmaceutical interventions, such as lock-
downs across different cities of the world (see Bartik et al. (2020) for a study based
on a sample drawn from the Alignable network of small firms in North Amer-
ica,8 Cowling et al. (2020) for SMEs in the UK,9 Shafi et al., (2020) for SMEs
in Pakistan, Thorgren and Williams (2020) for 456 SMEs from the Norbotten
region of Sweden and Brown et al. (2020) for steep declines in seedcorn finance
for entrepreneurship in the UK as well as in China). However, existing literature
points not only at small-firm, but also large-firm vulnerability. For instance, Al-
faro et al. (2019) find that large firms are financially fragile for a large number of
emerging market economies and Bottazzi et al. (2006) find not only small-, but
also evidence of large-firm fragility for Italy across all economic sectors between
1996 and 2003. Similarly, Hsieh and Olken (2014) found that emerging econo-
mies like China and Brazil, as well as India to some degree, are characterized by
large-firm rather than small-firm fragility. Of course, neither Alfaro et al. (2019)
nor Hsieh and Olken (2014) were writing keeping in mind an exogenous coordi-
nated shock like that of Covid-19. The commentary about what exactly changes
in times of shocks like these in terms of the vulnerability of a particular firm of a
given size requires further research and elaboration.
Not only firm size, there are other dimensions about firm fragility that require
further clarity. The mainstream macroeconomics literature focuses mostly on
financial fragility of firms. Using the representative agent framework, this line of
the literature works out economy-wide fragility resulting from firm-level vul-
nerabilities (Alfaro et al., 2019; Minsky, 1992; Nishi, 2019; Schroeder, 2009 to
mention a few). However, the absence of any explicit modeling of the interaction
effect between agents raises questions about the validity of these modeling struc-
tures in a hyper-globalized world. A few exceptions like Gallegati et al. (2003)
move away from the representative agent framework into the domain of agent-
based models, but the fragility of heterogeneous firms is again limited to balance
sheet variables (different equity ratios) and firm size (small, medium and large).
In this chapter, we begin with a definition of firm fragility that is relevant
for current times. Our intention is to investigate the validity of balance-sheet-
based financials to assess firm-level fragility for a sub-class of firms that are dis-
advantaged in determining contract terms linking them with others in a supply
network. We demonstrate that a typical firm of this class (we refer to them as
‘agent firms’) might face bankruptcy and very high fragility due to the nature
of the contracts it signs. We simplify these interlinkages as being either of the
short-term/non-relational contract (type n) or the longer-term relational type
of contract (type b). The latter allows principal firms issuing these contracts (we
Firm fragility 233
refer to them as ‘principal firms’) to retain liquidity for a period of length T ,

in return for guaranteed ‘large’ purchase that defrays the entire fixed cost of
any agent firm. However, a single non-relational contract does not cover the
agent firms’ fixed cost. For a given network configuration, which can result from
profit-maximizing or behavioral reasons, we show situations where a firm with
an ex ante ‘non-fragile’ balance sheet configuration faces fragility at the end of the
period of length T following a shock like Covid-19.
We do not assume fully connected supply networks for these results. With
some simplifying assumptions, we infer that shorter the length of this period and
larger the probability and magnitude of the coordinated negative shock in the
supply network for all principal firms, agent firms that hope to diversify their
production through multiple non-relational contracts are more fragile than those
agents with relational contracts. We also show the possibility of fragility arising
from relational contracts in the same framework. Invocation of contract clauses
like force majeure strategically in supply networks by firms with bargaining power
is of central importance in our theoretical framework, and availability of relevant
data on the specifics of contracting is of paramount importance for commenting
on policy issues regarding firm survival.
The structure of this chapter is as follows: Section 14.2 presents a few
pandemic-related stylized facts that have emerged across countries, with par-
ticular relevance for firm fragility. The harder part about precisely pinning down
what firm fragility entails at present times is set aside for Section 14.3, where we
discuss how firm fragility has been studied in the existing literature before pre-
senting our perspective on this issue. Section 14.4 works out a theoretical setup
that provides a micro-economic foundation for the genesis of fragility at the firm
level in a supply network using the stylized facts from Section 14.2. While this
is a generalized description of firm fragility (sanitized from issues of empirical
measurability), Section 14.5 elaborates on difficulties of empirical estimation of
our theory, along with some anecdotal evidence to evaluate the limitations of the
model in the earlier section. Section 14.6 concludes with a discussion on policy
implications regarding firm fragility and survival.
14.2 Stylized facts about Covid-19: implications for

industrial ecosystems
We have mentioned earlier, albeit in brief, that Covid-19 presents some unique
challenges for a globalized world, bringing to the fore nomenclature like the
‘new normal’ or ‘black swan’ event.10 What is the radicality of this event as far as
firm fragility is concerned? We present a set of stylized facts that have emerged
till now below:
S1 Negative demand shock due to the initial reaction to the pandemic: Ini-
tial survey results, particularly for emerging economies, show a significant
reduction in economic activity, as demand fell sharply. Beck et al. (2020)
234 Debdatta Saha
find that a significant majority of nearly 500 listed firms in ten emerging
economies experienced negative outcomes since April 2020. The first half
of the financial year 2021 has seen a decadal low in terms of credit profiles
for most Indian companies, as the rating agency Crisil notes.11 This has been
attributed to the fall in economic activity and the concomitant business dis-
ruption, which results from demand-side curtailment for multiple services
and products (induced by lockdowns and necessities of physical distancing).
Whether this translates into a long-term decline in demand remains to be
seen.
S1a The supply shock was transmitted from the retail end up the entire supply
network: Due to the manner in which non-pharmaceutical interventions
played out, the demand shock transmitted itself from the retail end higher
up to intermediate goods and other related products in the intertwined sup-
ply chain between firms. Beck et al. (2020) find that medium- to large-
sized firms in emerging markets, which contribute between 0.5% and 8%
of the GDP in India, reacted to the pandemic beginning early April 2020
by rolling back investments rather than reducing expenditure on salaries of
employees. This is more likely to impact upstream as well as laterally placed
firm in the network of a particular medium to large firm.
S2 Changing outlook regarding the economic prognosis of the pandemic:
Compared to the initial outlook regarding the effect of the pandemic, the
current expectation about the pandemic has changed.12 The final trajectory,
in terms of economic outcomes, is likely to depend on the timing of im-
plementing the non-pharmaceutical interventions, such as lockdowns. For
instance, Demirguc-Kunt et al. (2020) mention empirical evidence showing
that countries that implemented lockdowns early on during the outbreak,
in Central Asia and Europe, suffered less in terms of short-term economic
impacts and lower mortality rates than countries which implemented stricter
lockdowns further into the pandemic. The overall average estimated eco-
nomic impact by Demirguc-Kunt et al. (2020) is estimated to be around
10%. Note, however, that India which implemented strict economic lock-
downs early on in the pandemic (with only 500 odd cases in the country)
in late March of 2020 suffered a severe contraction in GDP to the tune of
a shrinkage of 24%, much larger than other comparable economies which
implemented lockdowns. The economic outlook has worsened through the
denouement of the pandemic in India, but not evenly across all sectors, as
the next point highlights.
S3 Business opportunity destruction as well as creation: Consider the case of
personal protective equipment (PPE), which has seen a drastic transition in
India.13 At the beginning of the outbreak, India was short of PPEs and was
importing the bulk of it. Within three months, India was producing a sur-
plus of these to the point of being able to export them abroad. Globally, new
product demand, such as the case for electric vehicles,14 has surged during
the pandemic. Transitions in the purchasing medium from physical outlets
Firm fragility 235
to online mechanisms have been a boon for many e-commerce entities.15

Sectors such as education have largely found a new avatar through online
teaching and learning avenues (see Rapanta et al., 2020). Unsurprisingly,
the travel tourism segment has seen a near rout.16 Some industrial sectors in
India have seen partial revivals, after country-level lockdowns were lifted,
such as some green shoots in the automobiles industry,17 whereas transitions
to alternative digital platforms are a reality for a few (such as Netflix and
Amazon Prime releasing original content as well as new movie releases on-
line).18 Interestingly, though liquidity shortage dogged most firms, the rate
of credit downgrades was limited by government policy buffers such as li-
quidity windows available through the corporate bond market, the morato-
rium on debt servicing permitted by the Reserve Bank of India and Security
and Exchange Board of India (SEBI)’s temporary relaxation in the norms for
default recognition for credit rating agencies.19 Till date, the rating agency
Crisil finds that it has been a case of a higher decrease in credit upgrades than
an increase in credit downgrades.20 Sectors such as pharmaceuticals have
seen steady demand and improved liquidity, whereas others such as textiles,
real estate, hospitality, auto ancillaries and construction 21 have faced chal-
lenges in meeting revenue targets.
S4 Coordinated supply shock: Due to the manner in which nation-wide lock-
downs were implemented, most supply networks experienced coordinated
simultaneous shocks, most of which negative, with very few exceptions as S3
mentions. These were mostly due to shortfall in final demand, as S1 details.
This coordinated shock across supply networks (that we define precisely in
Section 14.4) has been the most radical aspect of the pandemic. It is very
difficult to create risk-sharing mechanisms through contractual or other
means when shocks of the same magnitude and sign affect the entire supply
networks.22
In this paper, we use these stylized facts, which characterize the core features of
the ongoing pandemic, to motivate our definition and theoretical exploration of
firm-level fragility in a connected supply network in Sections 14.3.1 and 14.4,
respectively. Prior to that, we investigate the existing mainstream literature in
economics regarding firm fragility in Section 14.3.
14.3 Firm fragility: revisiting existing literature in the

light of Covid-19
As we pointed out in the introduction to this chapter, most of the macro-finance
literature identifies fragility synonymously with financial fragility. Table 14.1
provides a summary of some existing definitions in the literature, with particular
reference to the context in which it is relevant. This set is by no means exhaus-
tive, but it shows the diversity of measures and their different applicability de-
pending on the specifics of the industrial and market configuration under study.
236 Debdatta Saha
TABLE 14.1 Definition and expression of fragility in economic systems in the existing
literature
Source [Short Description] Definition of Fragility Variables Reflecting Fragility

in Economic Systems
(Unit Affected)
Altman (1993) [First Z-score is a linear Ratios of working capital,

multivariate bankruptcy combination of five retained earnings, and
prediction model with corporate income and operating income to
continued popularity balance sheet values to total assets, the book
in finance, accounting, measure the financial value of assets to total
and macroeconomics health of a company; liabilities and the sales to
research: Z-Score Z-score statistics total assets.
model is used as correspond to AAA to
either the central or BBB for the safe zone,
a supporting tool for BBB- to B- for the grey
bankruptcy or financial zone and CCC+ and
distress prediction below for the distress
with approximately zone (measure applies
80%–90% accuracy in to all modern corporate
predicting bankruptcy firms)
one year prior to
event, with a Type II
error of approximately
15%–20%]
Alfaro et al. (2019) Financial fragility Z-score computed from
[Empirical estimation measured using the four ratios from firms’
of financial fragility of Altman (2005) Emerging financial statements,
non-financial firms from Market Z-score using; viz. working capital to
1992 to 2014 for most lower Z score indicates total assets (X 1), retained
countries classified as greater vulnerability earnings to total assets
emerging markets by (measure applies to (X 2 ), operating income
the MSCI’s emerging corporates in emerging to total assets (X 3 ), and
market index: Argentina, markets) book value of equity to
Jordan, Morocco, total liabilities (X 4 ), as:
Slovakia, Slovenia and EM Z-score = 6.56X1
Vietnam included but + 3.26X 2 + 6.72X 3 +
Greece, Egypt, Qatar 1.05X4 + 3.25
and UAE excluded]
Bartik et al. (2020) Financial fragility of small Firms’ cash-in-hand (in
[Survey results from firms using a measure savings and checking
5819 US firms, which of liquidity (measure accounts with banks)
are members of the applies to small firms, without resort to friends
Alignable network with particular reference and family for additional
of small firms in to the US) cash infusion into their
March 2020; measure business normalized by
of financial fragility expenses of firms as on
relevant for small firms] 31 January 2020
Firm fragility 237
Deli Gatti et al. (2002) Financial fragility of firms Equity ratio (the ratio of
[Theoretical exploration which are heterogeneous net worth or equity base
of financial fragility in terms of degree of to capital stock of the
through patterns of entry financial robustness firm); firms with low
and exit of firms that (theoretically applies equity ratio values are
change the underlying to any representative financially fragile and
distribution of size of firm, results are based vice versa.
firms in the industry] on simulations of the
theoretical framework)
As is evident, fragility measures are mostly concentrated around ex ante (and

some ex post) balance sheet and profit and loss statements of firms. Nishi (2019)
summarizes these stock and flow variables, following the Minskian tradition as
follows:
1 financial fragility index based on a cash-flow accounting framework, similar

to Schroeder (2009):
i profit rate (r) measured by operating profits divided by capital stock, where
capital stock is the average value of tangible fixed assets minus land;
ii capital accumulation rate (g) measured by investment divided by capital
stock, where investment is the sum of the increase in capital stock and
depreciation;
i ii debt service per capital (iD ) measured by interest expense divided by capital
stock;
iv dividend payments per capital (d) measured by cash dividends divided by
capital stock;
2 financial fragility index based on margins of safety, as defined in Minsky
(1992) as (i) cash-flow margin (τ ), (ii) capital value margin ( µ ) and (iii) mar-
gin provided by a liquid asset kicker (η).
Note that Nishi (2019) uses the margin of safety as a fragility measure, which
comprises the capital value margin 23 (defined as µ ) and the liquid asset kicker24
(defined as η). Using these two different measures of financial fragility, Nishi
(2019) gives a measurable expression of the different kinds of firm-level fragility
(hedge, speculative and Ponzi) that was identified originally by Minsky (1992).
In order to identify macroeconomic fragility, Nishi (2019) adopts the rep-
resentative firm framework 25 by first classifying them as either being a hedge
unit (least fragile) or a speculative (relatively more fragile) or a Ponzi one (the
most fragile). Other papers, such as Schroeder (2009) or Mulligan (2013), employ
a similar methodology. Apart from the fact that balance-sheet-based financial
measures are sensitive to the time at which their values are observed, they are
more reflective of the resultant effects of the condition of financial fragility rather
238 Debdatta Saha
TABLE 14.2 Classification of firm-level fragility in existing macroeconomics literature
Fragility of an Hedge Speculative Ponzi

Economic Unit
Cash-flow (r − g − i D − d ≥ 0 ) (r − g − i D − d < 0 ) (r − g − i D − d < 0 )

accounting & &
framework (FF-1) (r − i D − d ≥ 0 ) (r − iD − d < 0)
in Nishi (2019)
Margin of safety µ ,η > 1 µ ≥1 & 0 <η ≤1 µ <1∀η
framework (FF-2)
in Nishi (2019)
Source: Author’s own creation based on definitions in Nishi (2019).
than being able to reveal the actual conditions that lead to these values on the
balance sheet. Therefore, to identify fragility of a firm purely on the basis of its
financials measured at different times can only be equated with taking different
temperature readings for a feverish patient, without commenting anything on
the nature of the disease that results in those measurements.
In the current context, interlinkages between firms across supply chains and
across countries, the determinants of fragility for a firm are not often limited to
the firm itself. One dimension of firm fragility is the extent of risk a firm inher-
its from the industrial ecosystem, which consists of vertically and horizontally
linked entities connected through contracts of various kinds. That these contrac-
tual linkages matter is evident in the Minskian line of research on financial fra-
gility, where various margins of safety or capital use the concept of ‘contractual
cash payment commitments on debt’ (measured by interest rates and dividends).
However, it is not only the cash flows generated by these contracts that have
the potential to create a vulnerable situation for a firm in a network with other
firms. The vast literature on contractual interlinkages between firms indicates
various reasons why these links are formed: Production links in a supply chain
(which can be very complicated, leading to intertwined supply networks and
non-linear supply chains; see Ang et al., 2017; Braziotis et al., 2013; Ivanov and
Dolgui, 2020 for a detailed discussion), trade credit as well as other contrac-
tual payments shared through connected firms in industrial districts (see Cainelli
et al., 2012) or to share technology as well as for market access (refer to Chen
and Chen, 2002). Therefore, these linkages between firms are sensitive to the
particular context for which they are formed in the first place.26 Nonetheless,
the moment an alliance is formed between two separate firms, risk spills over
from one to another (refer to Harland et al., 2003); the manner in which it does
so might be context-specific27 and the resultant effects reflected in an individual
firm’s financials. The evolving literature on firm responses to Covid-19 is mak-
ing direct references to this theme: The role of inter-firm links in transmitting or
mitigating shocks (Majumdar et al., 2020; Manheim, 2020; Yogaananthan and
Theva, 2020; Ivanov and Dolgui, 2020 to name a few).
Firm fragility 239
Therefore, we believe that discussions on firm fragility need the dimension

of contractual links to be explicitly built into the existing scoreboard of firm-
specific financials, particularly in a post-Covid-19 world. Note that this comes
at the cost of lack of generalizability: The specific nature of links matter for the
manner in which a firm within a network becomes fragile. For instance, credit
linkages in trade credit networks give rise to a form of co-sharing of financial
risks that is absent in other forms of risk-sharing networks.28 This integration
of network-specific attributes in the concept of firm-specific fragility requires a
precise definition, which we attempt in Section 14.3.1, along with its implica-
tions for measuring firm-specific fragility in Section 14.3.2.
14.3.1 Defining firm fragility in a network context

We define firm fragility as follows:
Definition of firm fragility

Firm fragility indicates a condition of vulnerable existence of a firm, where it is
near extinction but continues to function, due to its contractual relationships based
on pre-determined supply arrangements with other firms in a supply network.
We use the term supply network in our definition of firm fragility along the lines
of Ang et al. (2017), Choi and Krause (2006) and Käki et al. (2015), which
mention specific hierarchical relationships made through supply contracts be-
tween buyer-firms (we refer to them as principal-retailers or ‘principal firms’)
and supplier-firms (we refer to them as agent-producers or ‘agent firms’). These
linkages need not only be between downstream firms and upstream suppliers, as
firms at a particular supply layer can have mutual links with themselves, as the di-
amond network connection between the lowest sub-tier suppliers and higher tier
suppliers in Ang et al. (2017). We use a simple hierarchy: Only one downstream
layer (the principal-retailers) and one upstream layer (the agent-producers). Our
focus is more on the quality of the links between these two layers and the result-
ant impact on fragility for a firm i in the upstream layer. Incorporating this single
tier is sufficient to show our claim that the dimension of inter-firm links has to
be explored to understand the fragility of a single firm in a supply network. Of
course, we can make the structure richer and incorporate some of the other hier-
archies as is the case with Ang et al. (2017) or Choi and Krause (2006).
Additionally, note a few qualifiers resulting from our definition. First, we
do not restrict the network to be an industrial district as is the case for Cainelli
et al. (2012); it can be an interlinked system cutting across supply chains and
geographical locations through production-based outsourcing contracts, such as
the ones discussed in Ang et al. (2017) or Wang et al. (2015). Note also that
240 Debdatta Saha
we do not investigate the optimal hierarchical structure that is resistant to sup-

ply disruptions, as investigated in Ang et al. (2017) or Choi and Krause (2006).
For a given network structure that has emerged either due to optimizing or
due to behavioral reasons between different layers of the supply network, our
research question investigates the fallout on an individual agent firm (without
the bargaining power to write the terms of the contract29) in terms of fragility.
Rather than differentiating between agent firms as either innately less or more
reliable (as is the case with Ang et al., 2017), we underscore the fact that sub-tier
risks in the supply network can emerge from the nature of interlinkages between
firms through different types of contracts. This does not require some agent
firms to be innately risky entities.
This leads to the second implication of our definition: The condition of fragil-
ity might not show up immediately in the financials of the firm, i.e. in liquidity or
insolvency measures. The expression of vulnerability due to contractual linkages
might become apparent after a coordinated synchronous shock to the network,
but otherwise remain as a hidden condition. Covid-19 presents a shock like this,
where the force majeure clause is being invoked by multiple firms (which are in
the nature of ‘principal firms’) to deal with unforeseen disruptions in the supply
network.30 Depending on the nature on the contract (short-term or long-term),
the survival strategy of the firm which is embedded in a supply network changes
(see Harland et al., 2003 and Harland et al., 1999 for a detailed discussion). Rela-
tional contracts, as mentioned by Jeffries and Reed (2000) or Milgrom (1992), are
unstructured arrangements that allow for firms to flexibly determine the exact
contours of exchange and the rules governing them within a standardized frame-
work. As opposed to these flexible arrangements, there are tighter, narrower and
more short-term contracts of the kind that Ang et al. (2017) or Wang et al. (2015)
study. Our central contribution is, therefore, to provide a common platform to
study the implications of firm-level fragility arising from different kinds of con-
tractual links in a network of firms in the presence of a coordinated shock like
Covid-19. Of course, we make several simplifying assumptions in our theoretical
structure in order to build this common platform in Section 14.4. Additionally,
our definition raises empirical caveats regarding appropriate measures of firm
fragility in a supply network.
14.3.2 Empirical import of firm fragility definition

We identify fragility with a state of vulnerability for a firm.31 This means that the
condition should be diagnosed from among the population of currently opera-
tional firms in the industry, which are on the brink of exit due to their weakness;
this avoids the problem of having to get measurements for firms which have
exited and for whom data is difficult to gather.
That is the only easy implication regarding empirical measurement. The
network-based fragility definition requires data on both relational and narrow
supply contracts that firms sign with each other. In the specific context of India,
secondary data on different kinds of contracts that even formal sector firms enter
Firm fragility 241
into is not available. The closest match that we have is in terms of data regard-
ing ‘related party transactions’ (RPT henceforth). The Companies Act of 2013
incorporated a change in accounting standards for formal sector registered firms,
making it mandatory for them to report important aspects of RPTs, as these
transactions are with closely allied entities of the promoters of the firm. Such close
links can potentially lead to tunneling of the firm’s resources. Given the closeness
of these links (some of the linked firms are subsidiaries of one particular firm),
one can potentially use them as proxies for relational contracting. However, the
data on narrow supply contracts that firms regularly enter into with other firms
in their network is not available. Our theoretical model rests on the relative im-
portance of relational contracting-type inter-firm links in contrast to narrower
and shorter duration contracts. The lack of data on the latter make any empirical
exercise difficult for the population of firms in the formal sector in India. We
discuss this data issue again in Section 14.6, as some aspects of measurability need
to be precisely framed to test our formal model that we present in Section 14.4.
14.4 M
odeling firm fragility: nesting individual firm
fragility within the industrial ecosystem
The central question that we intend to address is:
To what extent can identify an agent firm’s fragility in the context of

Covid-19 on the basis of the firm’s individual pre-pandemic financial indi-
cators (based on balance sheet and profit and loss statements)?
To address this question, we draw up a theoretical framework that nests an in-

dividual agent firm’s fragility within the industrial ecosystem. We model a net-
work of firms N , where a firm is connected with another through contractual
links of two different kinds. In terms of any balance-sheet-based measure θi for
any firm i in the network, all firms are ex ante ‘non-fragile’ at the beginning of
period of length T . However, due to the underlying contracts that the firms have
with each other, some of them become fragile (financially) in the presence of
shocks like Covid-19 affecting the supply network, whereas others do not. The
ex ante appearance of non-fragility, based on an individual agent firms’ financials,
might result in an incorrect impression of fragility that the firm will experience
in uncertain times. Note again that we keep our focus on fragility of agent firms
which are in a disadvantaged position for dictating contractual terms to principal
firms. This is an assumption we make to link the issue of fragility with the kind
of ties chosen by agent firms. An extension of our research is to examine cases
where such disadvantage is missing or reversed.
14.4.1 Model specification

Assume that there is a supply network N with firms as nodes (some are prin-
cipal nodes, which face the final retail demand and some are agent nodes, that
242 Debdatta Saha
produce a part of the final demand for the principal). The contractual links be-
tween the principal and agent nodes are edges. The network is not necessarily
complete: Not all principal firms are connected to all the agent firms; rather,
there are dense neighborhoods of links between these two types of firms/nodes
as sub-graphs in the entire supply network. These densely connected sub-graphs
g in the network N are such that an agent firm i has a neighborhood with at
least k0 principal firms which it can contract with. We consider two types of
links that an agent i can form with principal nodes j, where { j : ij ∈ g ∀i}, both
of which require firm i (the agent) to invest in a fixed investment F prior to
signing the contract.
1 Narrow or short-term (non-relational contract type n): The contract be-

tween i and j is a price-quantity tuple pn , qn such that firm i supplies to
firm j the specific quantity qn within a time period of duration less than T .
Payment of pn = p0n is per unit of qn only for the specific quantity contracted
and it is zero otherwise. This payment is made instantaneously upon receipt
of delivered goods, hence the subscript 0.
2 Broad or generic/longer-term (relational contract type b): The contract be-
tween i and j is considered a generic or broad contract pb , qb such that
firm i gets paid pb per unit of qb > qn at the end of a period of length T and
not instantly upon receipt of delivered goods. Note that pb = p1n , i.e. the same
payment of pn is made under contract b, just that the payment is made at the end of a
period32 of length T , so that the relevant subscript here is 1. With a discount factor of
0 < δT < 1 for time length T , we have pb = δT pn < pn.
Note that the fixed cost F incurred by each agent firm caps the maximum num-
ber of type n contracts it can enter to ku and to one for the type b contract.
FIGURE 14.1 Alternative supplier network configurations.

Firm fragility 243
Figure 14.1 shows an abstract depiction of two supply networks N 1 and N 2 with
different kinds of contractual relations between firms (principal-retailers and
agent-producers). In N 1, we show sub-graphs of firms (principals and agents)
connected by contracts of type n exclusively; there are two distinct sub-graphs
here.33 However, there are only relational contracts of type b in N 2.
Note that F has to be borne by the agent firm i, who is in the role of the agent
picking up the contract given to it by principal firm j to produce a fixed level of
output at a pre-negotiated per-unit price, with a force majeure clause to cover un-
foreseen shocks. As we assume perfect information between agent i and principal
j in the sub-graph g ∈ N and that F is not specific investment for a particular
contract, the question of strategic under-investment in the fixed cost F by agent
firm i is removed. We make the following assumption regarding the returns to
the two types of contracts:
Assumption 1
The fixed cost of production (F) for any agent firm i cannot be recovered through
a single contract type n, but it is possible through contract type b. The marginal
cost c is the same for agent firm i under both the contracts.
In order to make our problem interesting and to show the irrelevance of

b alance-sheet-based financials regarding fragility, assume that any agent firm
i has a maximum provisioning θi = F , captured as an index of financial fra-
gility of the second type (FF-2) discussed in Nishi (2019), such that µ > 1 and
all agent firms are ‘economic hedges’. 34 So, in terms of financial provisioning
per period T , every agent firm is non-fragile in the traditional sense at the be-
ginning of a period of length T . These contracts are signed at the beginning
of the period. The return to firm i from a single contract type n with the jth
principal firm is:
( )
π in (k = 1) = p0n − c qijn − F < 0 (14.1)
where the principal firm purchases qijn from the agent through the contract
type n.
Assume that achieving break-even through contract type n requires firm i to
jk0
sign at least k0 such contracts (with ∑ qijn = qn ), so that:

i:ij ∈ g ; j = j1
jk0
π in (k = k0 ) = ( p0n −c ) ∑ ( )
qijn − F = p0n − c k0qn − F = 0
i:ij ∈ g ; j = j1
244 Debdatta Saha
As mentioned earlier, with the per period fixed cost of F, any firm can pick up at
most ku > k number of contracts of type n. Each of these contracts is identical to
each other. Further, we assume that:
Assumption 2
The quantity contracted through the b-type contract is large enough to ensure
break-even on fixed costs F for agent firm i, even with k = 1, where qb (k = 1) is
given by
jk0
b n
q (k = 1) = Qk = ∑ qnij = k0qn (14.2)
i:ij ∈ g ; j = j1
The two core differences for the two different contract types are (1) in terms
of payment deadlines and (2) in terms of quantity contracted. The first type of
contract ensures payment without delays, but the second type is qualified by a
built-in deferral in payment against delivery. The second difference, in terms of
quantity transacted, is that a single generic-type contract gives as much output
off-take to the agent as k0 > 1 non-relational contracts. Note that the generic
contract pays out the same per-unit price as the narrow-type contract, but it
saves the search cost of the agent firm i in order to form at least k0 different ties
with different principal firms. However, it extracts a commitment from the
agent firm i: The deferral in payment for work done, such that the time cost
of δ shaves the per-unit price of the generic contract lower than the specific
contract.
A built-in assumption for signing the generic type of contract is the financial
cushioning that an agent firm i has so that it can survive a length of time T with-
out any financial stress. Hence, the trade-off in the choice between the two types
of links is primarily in terms of deferred payment in return for insurance in terms
of off-take in terms of quantity. In network terminology commonly used in so-
ciology, type n contract results in a ‘weak tie’, whereas type b contract is a ‘strong
tie’, as the two firms have to stay committed for the duration of the quarter for
the execution of the contract.35 The reason for an agent firm i to prefer contract
n over b can be one of many reasons, such as:
1 Standard incentive compatibility with low search costs: With densely con-
nected sub-graphs g with more than k0 firms in the supply network, the
search cost becomes trivial in relation to c and F. Agent firm i can choose
among multiple possible trading partners to supply goods to.36 In this case,
an agent firm will prefer to sign type n contracts with k0 firms to picking up
a single type b contract if the incentive compatibility constraint for signing
type n contract(s) is satisfied for agent firm i:
Firm fragility 245
( p0n − c ) qb − F ≥ ( pb − c ) qb − F ⇒ 1 > δT (14.3)
which is trivially satisfied given the assumption that δ < 1.

2 Behavioral reasons: Aversion to delayed payments by agent firm i, assuming
that there are no uncertainties in the payment of pn upon delivery of qn . The
delay of T in getting paid with type b contract might be a dis-incentive for
agent firms. If agent firms believe that the number of principal firms in the
network which will face excess demand following an exogenous shock is
large enough, then the flexibility of type n contracts will appear tempting to
the agent, as we discuss later.
We do not try to isolate which of these reasons drive link formation, unlike
in Ang et al. (2017) or Käki et al. (2015). Link formations are often not done
from a purely profit-maximization or efficiency-seeking motive. We believe that
it is difficult to elicit the former reason from the latter from a given network
structure.
Note a simplifying assumption that we are making in our analysis in this
benchmark model: A firm picks up either one type b contract per period of
length T or at least k0 contracts of type n, each issued by a different principal
j ∈ g . The constraint here is that of capacity; in a period, with a fixed F, a maxi-
mum number ku narrow contracts can be signed, but only a single type b contract
can be supported. Of course, with the same fixed F, we provide no background
as to why only one relational contract can be signed. One can argue that rela-
tional contracting might require no-compete agreements and exclusive supply
conditions, which is an additional cost of guaranteed off-take. Building trust
through relational contracting of type b often comes with these clauses that limit
the total number of contracting that a seller can do. We can relax this assumption
in a more refined analysis.
A related question is the benefit to the principal from offering either type
of contract. As long as we assume that the principal will need supplies from at
least k0 firms with type n contracts to substitute for a single relational contract b,
the latter will also prefer to sign k0 non-relational contracts to a single type b
contract. Note here that F does not matter for the principal; it is solely borne by
the agent.
14.4.1.1 Modeling shocks in the supply network

In the previous setup, we now introduce an exogenous coordinated shock at
the retail level, keeping in mind the stylized facts S1–S4 presented in Section
14.2. The ‘black-swan’-like event that Covid-19 presents for agent firms arises
from the fact that the shock is coordinated for all principal firms in the supply
network N , albeit with some differences in sign and magnitude. To show this
in a simple manner, we assume that all principal firms j in the network face an
exogenous shock to their final demand. With probability β > 0, this results in
246 Debdatta Saha
a negative demand shock and with probability (1 − β ), the principal firm faces
an excess demand from the retail level.37 Note that the uncertainty is about the
sign of the shock (positive or negative) for the principal firms, which then affects
agents through contractual links. The magnitude of the shock matters when
both positive and negative shocks affect different principal firms in the same
supply network.
A principal firm, facing a negative demand shock of size ε > 0 with probability
β , will invoke the force majeure clause in either type of contract b or n and instead
of the promise to off-take an amount q = {qn , qb } will now only ask for (q − ε )
through the contract. For an agent firm i that has diversified its earnings through
( )( )
k̂, different links now face a loss of p0n − c kˆε , where instead of a single type b
contract, the firm can sign k̂ number of type n contracts.38 However, the demand
( )
drop for the type b contract is qb − ε , leading to a loss of δT p0n − c ε . Note also that
not all principal nodes face a negative demand shock; some face an excess demand
for their product, and need to sign contracts of type n or b quickly to scale up.
Proposition 1
As β → 1, T → 0 and ε → qn for each principal firm j with which firm i has a
type n or type b contract in the neighborhood g ∈ N , diversification strategy for
agent firm i through k̂ number of type n contracts for an agent firm i results in
losses much larger than through a single type b contract.
Proof. In the presence of a coordinated quantity shock of the kind we describe
(with near certainty, all principal firms j face a negative shock ε → qn ), note that
the net profit for agent firm i under k̂ type n contracts is
( ) ( ) ( ) ( ) ( )
π in:ij ∈ g k = kˆ,ε = p0n − c kˆ qn − ε − F = π in k = kˆ − p0n − c kˆε
As ε → qn , we shall get, in the limit:
( )
limε →qn π in k = kˆ,ε = −F
However, when agent firm i has a single type b contract with principal firm j
while facing a quantity shock ε , the agent firm i sees a net profit:
( )( ) ( )
π ib:ij ∈ g (k = 1,ε ) = pb − c qb − ε − F = π ib (k = 1) − δT p0n − c ε (14.4)
As ε → qn , we shall get:
( )( )
limε →qn π ib (k = 1,ε ) = δT p0n − c qb − qn − F (14.5)
Firm fragility 247
Given our assumption that qb = k0qn, with k0 > 1, subtracting equation from,
we get:
 (
limε →qn π ib (k = 1,ε ) − π in k = kˆ,ε ) = (δ T n
)
p − c (k0 − 1) qn > 0 (14.6)
This requires that the lock-in period in type b contracts be insignificant, as T → 0 ,

( ) ( )
so that δT p0n − c → p0n − c > 0.
From this discussion, the following lemma follows immediately:
Lemma 1
As β → 1 for all principal firms j in the supply network, the losses for contract
type n relative to contract type b for an agent firm i will be higher the larger is
the value of k0 , as long as T → 0 .
The important point about the previous lemma is that the actual number of type
n contracts that a firm i signs with firms in its immediate neighborhood does not
matter in the limit. What matters is k0 : The minimum number of links that a
firm has to form in order to break even. Higher is this number, higher are po-
tential losses in the face of coordinated shocks. Note also that smaller is δ , the
amplification of the loss from type n contracts will be larger, in comparison to a
type b contract.
A coordinated negative shock for all type j firms (for us, this happens when
β → 1 for all principal firms) in the supply network with either contract type
might seem implausible in normal situations; the dramatic reduction in eco-
nomic activities due to coordinated lockdowns at the national level for most
countries due to the current pandemic is reminiscent of this scenario, as we
discussed in our stylized facts S1, S1a and S4. The larger the value of k0 , the
larger would be the connectivity among agent and principal firms in a network.
In a traditional sense, higher connectivity in a network can achieve higher di-
versification of risk, as discussed in the context of risk-sharing networks (see the
discussion in Cainelli et al., 2012). However, even in a fully connected network,
Cainelli et al. (2012) find that risk cannot be fully diversified for a shock above
a certain threshold. In this sense, a firm which requires a large number
of type n contracts to break even is truly fragile, even if its balance sheet
figures do not indicate any vulnerability prior to the shock. The manner
in which a firm becomes co-dependent on the network is a micro-measure of
fragility that one needs to study shocks like Covid-19. We show a possibility
(Proposition 1) when these types of shocks can be catastrophic for non-relational
contracts, as all principal firms issuing such contracts roll back upon them using
force majeure to preserve their own liquidity. The fallout on relational contracts
248 Debdatta Saha
due to negative shocks is lower, as these do not require immediate cash payouts
by the principal firms.
Proposition 1 and Lemma 1 are text-book applications of our definition of
firm-level fragility in a supply network. The net worth of an agent firm i under
Proposition 1 is completely wiped out following a shock like the current pan-
demic, as its provisioning of θi = F is the quantum of its negative profits. How-
ever, the balance sheet fundamental of µ > 1 for an agent firm i is maintained in
the first place by forming the large number of type n contracts when there is no
shock to the supply network. The lack of bargaining power of agent firms in type
n contracts is the biggest source of fragility in this instance, which is ex post going
to be expressed in its balance-sheet-based financials.39
However, this does not mean that an agent firm i signing contracts of type n is
generally more fragile than those which sign the other type of contract. The condi-
tions of Proposition 1 and Lemma 1 are stringent. Note that the Covid-19 shock
has also provided new opportunities for some new businesses, as discussed in S3.
The simple framework that we propose can be used to capture the possibility of op-
portunity creation despite the exogenous shock, under certain specific conditions.
Let 0 < β < β < 1, so that (1 − β ) is not insignificantly small for at least some of
the principals in the network and let T > 0 , i.e. a finite lock-in period for type
b contracts. These principals face a non-zero probability of a positive demand
shock, rather than a negative one. In the event of that positive shock, the principal
firm will try to fulfill additional demand by forming new ties using contracts of
type n, which can be executed within the time period T . One such situation will
prevail when this is true for all the principal firms in the network. In that case, the
result of Proposition 1 is reversed. Sketching out a precise proposition with this
result requires specific assumptions about the number of principal firms facing the
probability of a positive shock. We settle for a weaker claim, with a weak assump-
tion about the density of the sub-graph in which agent i is embedded:
Claim 1
For every agent firm i in the network with k0 type n contracts, as long as the
number of principal firms in { g ∈ N : ij ∈ g } with which any agent firm i does
not have a contract is greater than k0 and all of these new principal firms expe-
rience an excess retail demand condition ((1 − β ) > 0 ), firm i can switch to a new
production contract within the period T itself. There will always exist an unex-
ploited link in the densely connected sub-graph g for a new link ik to be formed
even if an old link ij ∈ g fails as ε → qn . Such shifts in suppliers are not possible
with type b contracts within time period T .
This proposition requires that the density of principals in the neighborhood of

an agent firm i be at least 2* k0 , so that even if existing k0 links fail to provide
Firm fragility 249
liquidity to agent firm i, another set of k0 can be formed instantaneously at no

additional search costs. In Figure 14.1, we can see that it might hold if the agent
firms in g1 can form links with the single principal firm in g 2 experiencing a pos-
itive demand shock, only if the magnitude of the positive shock for is 2* qn . This result
holds trivially for links that are in the same sub-graph g ∈ N , where the same
number of principal firms face opposite demand shocks of the same magnitude.
The precise manner in which these semi-exits40 occur depends on the structure of
the sub-graphs in the network.
One way to motivate Claim 1 for the current pandemic is as follows. While
some product segments, such as textiles, saw a coordinated demand shortage,
there was excess demand for some other product segments, such as PPEs and
other hygiene products. Firms producing these protective equipment and sani-
tization material would require to quickly scale up through contracts of type n
within the time period T to cater to the excess demand. As long as any principal
firm j does not share a common agent firm i, the fragility for the principal is dis-
tributed to k different type n contract agents or is passed onto a single agent with
contract type b. This is the diamond network structure that Ang et al. (2017)
mention, just that the shock is not from the upstream tiers to the downstream,
but the other way around.
Using this simple framework, we can also highlight the potential diffi-
culties that can arise with type b relational contracting for an agent firm i.
This is a function of the lock-in period in the contract, where the agent firm
has to wait to get paid by the principal firm, as explained in the following
proposition:
Proposition 2
For a given 0 < δ < 1, as T → ∞, higher fragility can be generated for the agent
firm i by a single type b contract rather than k0 contracts of type n.
Proof. With a single type b contract, as T → ∞, δT → 0 and the net profits from
( )
a single type b is − cqb + F , whereas the profits from k0 contracts of type n lead
to a break-even condition for fixed costs F.
In this case, it is relational contracts that create fragility for the agent firm, due
to the extended delay in payments through the contract. The transfer of risk by
the principal to the agent in case Proposition 2 holds does not come from the size
and the magnitude of the shock; it holds for any shock where payments can be
withheld by the principal firm for a significantly long period such that the agent
faces erosion of wealth. Presumably, such contracts will not be in the interest of
the principal either. However, this theoretical possibility shows why an agent
firm i might prefer type n contracts to relational contracts, particularly when
their individual bargaining power with the principal is low.
250 Debdatta Saha
14.4.2 Firm-specific fragility measurement: network-based measures

At the beginning of any period of length T , our model shows the problems of
labeling firms fragile using balance sheet or profit and loss statement variables
standalone. Despite lack of fragility at the beginning of the period, the (1) length
of the period T , (2) density of the connected sub-graph g of N and, of course,
(3) strength of the shock ε as well as (4) its coordinated effect on firms with
which a particular entity contract determines the true condition of fragility of
the agent firm. Ex ante, in an interconnected industrial ecosystem, with con-
nected sub-components, firm fragility has multiple dimensions; only a part of it
(pre-determined reserves for cushioning θi ) depends solely on the agent firm. A
firm’s individual fragility is a function of not only its own reserves, but also the
extent of shock experienced by those with which it forms contractual links with.
Note that our abstraction which requires the agent to fully bear the fixed costs
of contracting and the presence of densely connected sub-graphs in the supply
network, where there are enough principals, as well as agents, picking up these
contracts, might seem a bit extreme. However, any qualifications that we build
onto this structure will make the results hold in very narrowly defined contexts.
Using our simple framework, we can sketch some patterns of the agent firm’s
fragility as well as survival, depending on the individual financial reserves as well
as the network effect:
1 ‘Blessing in Disguise’: For an agent firm i able to enact a semi-exit using type
n contracts (Claim 1 holds but not Proposition 1);
2 ‘Bare-Bones Survival’: For an agent firm i facing unfavorable liquidity terms
in relational contract b signing k0 type n contracts (Proposition 2 holds);
3 ‘Chokehold’: For an agent firm i with relational contract b (Proposition 1
holds but not Claim 1);
4 ‘Exit with Contagion’: For an agent firm i in a supply network characterized
by paths between connected sub-graphs such that in each sub-component,
all agent firms are signed up with contracts of type n and there is a negative
shock of size ε ≥ qn with probability β → 1 for all the principal firms (Prop-
osition 1 and Lemma 1 in connected graphs).
Note that there are other patterns possible, if we make suitable assumptions about
the nature and quantum of the supply shock and the distribution of the contract
types among different agent firms.
14.5 Application of firm fragility measurement for

specific networks
As our simple theoretical model indicates, a measure of firm-level fragility can-
not be based on financials alone, without a microfoundation regarding the chan-
nel that a firm takes to achieve those financials. In order to identify the different
trajectories that a firm i can take, despite the same θi in a sub-graph g ∈ N , we
Firm fragility 251
need information regarding the kinds of contracts that the firm signs with other
firms. Lemma 1 in Section 14.4 indicates that what matters is also not only actual
number of connections that a firm has of type n, but rather the minimum num-
ber k0 of such ties. This has implications for the manner in which we can conduct
empirical investigations regarding fragility of firms, which are in a disadvanta-
geous position when signing supply contracts with principal firms.
In the absence of secondary data regarding firm-level contracts (other than
information on RPTs), one has to rely on primary surveys of firms of both
types in a supply network to be able to apply the theory empirically. Absent
reliable survey results, we are limited to anecdotal evidence from India. Based
on the early evidence from news reports and some primary evidence (such as
Beyer et al., 2021), most small firms at all levels of the supply network have
suffered financial losses during the initial days of the pandemic. The possibility
of enacting switching strategies which we see for the ‘Blessing in Disguise’
has been limited to some anecdotal evidence involving large principals, as we
discuss below. Note, however, that this is very early to make predictions about
changes in patterns of inter-firm links with lock-ins more than six months,
though the importance of relational contracting and stakeholder-friendly strat-
egies on stock returns for some principal firms has been reported (see Cheema-
Fox et al., 2020, though Beck et al., 2020 only find partial evidence in devel-
oping economies).
14.5.1 Limitations of theory in predicting supply network outcomes

Consider the product line switch made by the brand Coloressence (promoted
by Nature Magic World), a ‘Made in India’ firm engaged in the manufacture
of cosmetic products (in our terminology, a principal firm) with more than a
500-strong distributor network, during Covid-19: From cosmetic products to
premium quality hand sanitizers. This is an example of a principal firm, while
facing a negative demand shock in its main product line of cosmetics, which itself
switches to another product line with excess demand. We have not modeled this
possibility, and we limit ourselves to modeling principal firms as passive shock
recipients who pass it on through contractual clauses to upstream agent firms. It
is quite likely that large principal firms would themselves migrate out of product
lines that face negative demand shocks to ones with excess demand.41 We intend
to extend our model to account to incorporate this scenario to see the extent to
which the theoretical predictions regarding financial fragility hold robustly.
Analysts have pointed out business opportunities which have opened up in
various sectors, such as courier services, bookstores, e-commerce entities like
online grocery sales and sanitary paper product manufacturing.42 The last cat-
egory of sanitary paper products has seen such excess demand that Australian
majors like Kimberley-Clark have ramped up production. This creates the op-
portunity for trajectories like ‘Blessing in Disguise’. However, linking these an-
ecdotal examples with our theory requires data on contracts between firms and
their suppliers.
252 Debdatta Saha
14.6 Discussion and policy implications

Covid-19 has clearly demonstrated that the possibility of systemic and coordi-
nated shocks to the supply network of firms can be a reality.43 In fact, even prior
to this pandemic, the manner in which technology and forces of globalization
have been advancing in recent years has raised questions about such possibilities
(Bostrom, 2013; Fombrun and Wally, 1992).44 We have shown a theoretical pos-
sibility that fragility experienced by an agent firm in the presence of such shocks,
which to some extent would be considered inevitable in a Beckerian ‘risk society’
sense,45 is a function of the nature of links that these firms form. While some of
those links can create the space for eventual exit from existing relationships to
pursue new opportunities, they can also foment fragility through financial bank-
ruptcy for a single firm leading to a network-wide contagion.
In the specific context of Covid-19, the patterns of ‘Exit with Contagion’,
‘Chokehold’ and ’Bare-Bones Survival’ are of special relevance, as ‘Blessing in
Disguise’ does not raise issues of unintended exit. The problem of unintended
exit of agent firms in connected graphs raises survival issues for the principal
firms as well, who issue such contracts. The very existence of a network of such
firms is threatened, and we have seen some examples of this in the wake of the
2008 financial crisis: The bailout of the automobile industry in the US, with the
largest principal firms like Chrysler and GM facing bankruptcy.46
Fragility of principal firms with large-scale bailouts is a fractious public
policy debate. However, the Covid-19 shock is likely to unfairly be passed
on from principal firms to upper tier suppliers through contractual clauses,
as discussed in our stylized fact S1a. As the economy re-opens, this is, then,
likely to restart the production engine for the principal-type firms. However,
under conditions of Proposition 1 and Lemma 1, as we noted earlier, this
will result in a fragile condition for agent-type firms in the upper tiers of
manufacturing.
The significance of the CARES Act in the US (Beck et al., 2020) and the
relief package for micro, small and medium enterprises (MSME) in India47 need
to be evaluated against the patterns of linkages between the principal and agent-
type firms we discuss. Even large-sized agent firms shall not be spared potential
liquidity and bankruptcy threats. Higher the number of non-relational contracts
relative to relational contracts (with a reasonably finite T ), the more important
will such liquidity infusion be for sustaining robust participation of illiquid agent
firms in the network.
This kind of fragility, ex ante, is not necessarily evident through financial
measures based on leverage or liquidity/solvency ratios from balance sheets
of individual firms. Inferring this kind of fragility requires granular data on
the kind of contracts that firms sign with each other in a network of firms.
While the Companies Act of 2013 in India made necessary changes in report-
ing transactions of firms attributable to related parties (RPT), not all kinds
of contractual information are available through publicly available secondary
Firm fragility 253
data on the population of functional formal sector firms in India. Limiting

research to consider only financials to comment on fragility would be akin
to measuring what can be easily measured, but not what we would like to
measure.48
The Covid-19 shock has rendered balance sheet fundamentals of almost all
firms in the Indian economy fragile in a traditional sense, at least in the first
half of the fiscal year 2021. How does one identify the truly fragile ones from
those that are likely to pull through and therefore, unlikely to require gov-
ernment bailouts? A reasonable answer requires not only policy measures to
support immediately ailing firms in the space of the small- and medium-sector
enterprises that are not fragile, in temporary hibernation due to the pandemic
(see Didier et al., 2020 on this issue of financing hibernating firms during
Covid-19), but also changes in reporting standards for firms. While ex post
measurements of financial ratios do not reveal the channels through which
fragility expresses itself for a firm, ex ante financial ratios might provide a
misleading picture, as our analysis shows. Reliable secondary data on contract
durations, value exchanged through contracts and termination details for all
contracts (and not only RPTs) for at least large firms across all sectors is the
need of the hour.
Notes
1 See McKinsey’s September 2020 Outlook report (available at https://www.
mckinsey.com/~/media/McKinsey /Business%20Functions/Risk/Our%20Insights/
COVID%2019%20Implications%20for%20business/COVID%2019%20July%2023/
COVID-19-Facts-and-Insights-July-23-vF.pdf. Last accessed January 2021.
2 See Beyer et al. (2021) and Kanitkar (2020) for evidence on decline in economic ac-
tivity in India following one of the strictest national lockdowns from 24 March 2020
for 21 days and Beck et al. (2020) for evidence regarding the financial hurdles faced
by small- and medium-scale firms in emerging markets due to Covid-19.
3 The World Health Organization reports that there are 33,249,563 confirmed cases
(in various stages of infection) of Covid-19, including 1,000,040 deaths as on 29
September 2020; https://www.worldometers.info/coronavirus/ shows that as on 30
September 2020, there are 7,691,734 active cases worldwide, though only 1% are in a
serious condition. Last accessed March 13, 2021.
4 As per the World Health Organization, at present, India is leading the daily reported
case count numbers, with 70,589 cases reported in the last 24 hours for 29 Septem-
ber, 2020. However, the status of infectious spread is limited to clusters, whereas the
cumulative case count is the highest in the US, with community transmission as per
WHO data available at https://covid19.who.int/table. Last accessed March 13, 2021.
5 ht t ps://w w w.theat la nt ic.com /idea s/a rch ive/2020/03/supply- cha i n s-a nd-
coronavirus/608329/ notes the fragility of the supply side based on outsourcing and
just-in-time inventory management in the specific context of supply disruptions due
to over-reliance on Chinese manufacturing. For some product groups such as global
electrical equipment and motor vehicle parts, this dominance of Chinese manufac-
turing is as large as 33–53%; see Bouey (2020). Last accessed March 13, 2021.
6 McKinsey’s September 2020 Outlook report (available at https://www.mckinsey.
com /~/med ia/McK i n sey/ Busi ness%2 0Fu nct ion s/ R isk /Ou r %2 0In sig ht s/
COVID%2019%20Implications%20for%20business\/COVID%2019%20July%2023/
254 Debdatta Saha
COVID-19-Facts-and-Insights-July-23-vF.pdf. Last accessed January 2021) notes

recovery of non-discretionary consumer expenditure as opposed to discretionary ex-
penditure, which remains depressed globally as well as increased consumer and B2B
transactions through e-commerce instead of the physical marketplace.
7 See the OECD discussion paper on exploitative pricing during the pandemic: https://
www.oecd.org/daf/competition /Exploitative-pricing-in-the-time-of-COVID-19.
pdf, which mentions the investigation by Thailand’s anti-trust authority regarding
alleged inflated prices of surgical facemasks being sold on an online platform amid the
outbreak of Covid-19. It also mentions that the anti-trust authority in Australia, the
ACCC, coordinated with online re-sellers to prevent price gouging for products such
as facemasks, hand sanitizers and some grocery items under its consumer protection
powers. Last accessed January 2021.
8 The survey results in Bartik et al. (2020) find close to 43% of their sample of small
firms being closed temporarily due to the pandemic, an effect that is more severe than
the economic effect of the 1918 influenza epidemic.
9 The paper comes with a dire prediction that 61% of small businesses in the UK may
run out of cash, including an 8.6% that had no retained earnings whatsoever; micro
firms being particularly at risk.
10 Note that the ‘black swan’, a term popularized by Taleb is referred to unforeseen events
with large effects. The identification of Covid-19 outbreak as a ‘black swan’ event
is not without controversy. While some medical practitioners claim it to be ‘novel’
and unpredictable enough to qualify as a ‘black swan’ (see the discussion in Lüscher,
2020]), others, including Nassim Taleb himself (who popularized the term ‘black
swan’; see Taleb (2007)), refute this claim (see https://jfsdigital.org/2020/03/18/
neither-a-black-swan-nor-a-zombie-apocalypse -the-futures-of-a-world-with-the-
covid-19-coronavirus/. Last accessed March 13, 2021). However, one cannot prevent
the Covid-19 outbreak as being recorded for posterity as a very radical event; its rad-
icality is not beholden to its lack of predictability, rather to the effect it has produced
on hyper-connected supply network of firms.
11 https://www.livemint.com/companies/news/credit-ratio-of-indian-c ompanies-
at-decadal-low-in-april-september-11601558137801.html. Last accessed March 13,
2021.
1 2 As discussed in the McKinsey report of September 2020, the expectations about eco-
nomic outcomes have changed since the first few days of the pandemic, where many
enterprises in developing countries foresee a gloomier future economic scenario
in September 2020 than in March of this year; see https://www.mckinsey.com/~/
media/McKinsey/Business%20Functions/Risk/Our%20Insights/COVID%20
19%20Implications %20for%20business/COVID%2019%20July%2023/COVID-19-
Facts-and-Insights-July-23-vF.pdf. Last accessed January 2021.
1 3 https://www.deccanherald.com/national/india-exported-23-lakh-ppe-kits-to-5-
countries-in-july-health-ministry-873251.html. Last accessed January 2021.
14 https://www.mckinsey.com/business-functions/risk/our-insights/covid-19-
implications-for-business#. Last accessed March 13, 2021.
1 5 https://tech.economictimes.indiatimes.com/news/internet/online-sales-doubles-
for-top-consumer-brands-in-india-due-to-covid-19/76541776. Last accessed March
13, 2021.
16 See the UNWTO situation report on the travel and tourism industry: https://www.
unwto.org/impact-assessment-of-the-covid-19-outbreak-on-international-tourism;
this has been one of the worst affected sectors due to the current pandemic. Last
accessed January 2021.
17 https://www.thehindubusinessline.com/news/partial-revival-of-auto-industry-
gives-life-to-machine-tool-sector/article32620714.ece shows the interlinkages be-
tween firms and industries: Partial revival of the automobiles industry has boosted
demand in the machine tools sector. Last accessed January 2021.
Firm fragility 255
18 See the BBC News report of the UK music magazine publication NME owner’s
assessment of the impact of Covid-19 on the recording industry as an event which
promises some positive outcomes; available at https://www.bbc.com/news/business-
54138180. Last accessed March 13, 2021.
19 For further details, see https://www.livemint.com/companies/news/credit-ratio-
of-indian-companies-at-decadal-low-in-april-september-11601558137801.html.
Last accessed March 13, 2021.
20 Ibid.
21 Ibid.
22 This theme is partly explored in the context of fully connected production networks
with credit interlinkages: For shocks above a certain threshold, even completely con-
nected graphs cannot sustain risk-sharing objectives for its component nodes (which
are firms). More on this in Section 14.3.
23 The measurement of µ , defined as µ =

( )
K Q − λσ Q2
, where K is the discount oper-
K (CC )
ator of Minsky (1992) to obtain the present value of variables, Q is average/expected
quasi-rents measured by operating profits by Nishi (2019, σ Q2 is its variance, the
Minskian scalar λ to measure the impact of fluctuations in quasi-rent is normalized
to 1 and CC are contractual cash payment commitments on debt measured by interest
expense plus cash dividends. K (CC ) , however, is an indicator of debt and is measured
by liabilities. Values of capital stock, capital assets, debt and equity are used in the
paper by taking averages of the values of these variables at the beginning and the end
of the period.
K (CC ) + Eq − Pk K
24 This variable is defined as η = , where Pk K is an indicator of cap-
K (CC )
ital assets and is measured by fixed assets, and Eq is equity measured by net assets.
25 Some research, such as Vives (2014), incorporates variables from the industrial eco-
system to discuss fragility at the macro-level. In the context of the post-2008 finan-
cial crisis scenario, Vives (2014) mentions that fragility of the industrial ecosystem
depends on the balance sheet composition of financial intermediaries, precision of
information signals as well as market stress parameters.
26 For instance, Chen and Chen (2002) report that for a sample of 1,597 Taiwanese firms
between 1990 and 1997, the major reason to form alliances was to get market access
and not to diversify risk or share costs. Thus, most of the ties formed in the network
of firms were asymmetric alliances: Small Taiwanese firms forming links with large
and well-established foreign firms to access new markets and technology. One might
argue that this is specific to the set of Taiwanese firms under observation and not a
general truism for all kinds of strategic inter-firm alliances.
27 Consider the network-based risk-sharing among firms, as discussed in Bramoullé and
Kranton (2007) or Cainelli et al. (2012). The central result on full sharing of risks due
to production shocks in Cainelli et al. (2012) is specific to fully connected networks,
commonly observed in industrial districts (their particular reference point is such a
cluster in Italy). In general, as Kirman (2010) points out, most firms are not members
of completely connected networks and Cainelli et al. (2012) mention the lack of gen-
eralizable results when there are peculiarities in link formation in the network, such
that not all firms are connected to others.
28 Bramoullé and Kranton (2007) discuss theoretical aspects of risk-sharing networks
in the context of a developing economy, particularly village-based ties, which we
feel are not relevant in the context of firm-specific fragility in a network. Nor is the
discussion in Cainelli et al. (2012) fully relevant, as the general theoretical results
are valid for fully connected networks.
29 We assume that the downstream principal-retailer firms dictate contract terms to
the upstream agent-producers. We also assume that the principal firms do not have
256 Debdatta Saha
any commitment problems in honouring the terms of the contract, other than un-
foreseen exogenous shocks like the current pandemic. In such circumstances, the
downstream principal invokes the force majeure clause and passes on to the upstream
the full extent of the supply shock. For details on force majeure clause invocation in
the context of public-private partnerships (PPPs) due to the Covid-19 outbreak, re-
fer to Casady and Baxter’s 2020 article available at: https://www.researchgate.net/
publication/344232465_Pandemics_Public-Private_Partnerships _PPPs_and_Force_
Majeure_COVID-19_Expectations_and_Implications. Last accessed January 2021.
3 0 For instance, see the discussion in Ward (2020) on the South African public electric-
ity utility Eskom’s Covid-19 force majeure invocation.
31 We limit the discussion to formal sector registered units. As our purpose is to exam-
ine the extent to which existing stock and flow measures are based on balance sheets
and profit and loss statements, we consider this universe of firms and not the entities
which are a part of unregistered manufacturing. Most units of the latter type do not
maintain these accounts, though they are in the majority in economies like India.
Additionally, the nature of their fragility is likely to be qualitatively very different
from formal sector firms.
3 2 This period can be the length of a financial quarter. The difference between the con-
tracts is that an agent firm i picking up a b-type contract has to wait the entire period
before it can be renegotiated. A feature of relational contracting is that it reduces the
necessity to renegotiate the terms of the contract frequently.
33 This is an abstraction to show our model graphically. We can generalize our results to
graphs where both types of contracts are present in the same network.
34 We find that liquidity issues dominate the discussion around firm survival in post-
Covid-19, as mentioned in the survey of Bartik et al. (2020). We feel, therefore, that
using an index like FF-2 makes sense across all firm sizes, as the shock to the system
is not idiosyncratic; rather, there is a synchronous effect on all firms.
35 This terminology comes from Mark Granovetter’s work in sociology on information
sharing between different kinds of relationships between individuals (strong versus
weak); see Granovetter, M.S. (1973). “The Strength of Weak Ties”, The American
Journal of Sociology, 78(6):1360–1380. doi:10.1086/225469.
3 6 Note that we do not assume a fully connected supply network N to derive this result.
All we need is densely connected sub-graphs with more than k nodes for this assump-
tion to go through.
37 Refer to Figure 14.1, where we show these shocks with the sign (−) or (+) to denote
negative or positive demand shocks.
38 Note that ku > kˆ > k0, where k0 is the minimum number of type n contracts needed by
a firm in order to break even: π n (k0 ) = 0, and ku is the maximum number of type n
contract that can be supported by the fixed capacity cost F .
3 9 Interestingly, the literature on gig economy contracts in the service sector discusses
this problem; see Aliosi (2016)’s discussion on commoditized workers and labor law
issues (available at Heinonline:https://heinonline.org/HOL/Page?handle=hein.
journals/cllpj37&id=689 &collection=journals&index=#). We are investigating this
in a general context, including manufacturing. Last accessed January 2021.
4 0 We use the term semi-exit here to distinguish between exit as a failure by an agent
firm from exit from a contractual relationship due to a new opportunity for the agent.
With the same investment in F , the agent firm shifts from one set of type n contracts
to another set of the same type of contracts, due to their short validity and payment
cycle. This type of exit we label semi-exit.
41 Another anecdotal evidence from India is Tata Group of Companies, which built
a PPE manufacturing plant during the lockdown; see https://www.tata.com/
newsroom/covid19/covid-19-healthcare-equipment-tata-steel-motors-taco-tcs for
further details. Last accessed January 2021.
4 2 See https://www.ibisworld.com/industry-insider/ coronavirus-insights/five-industries-
set-to-outperform-due-to-covid-19-part-2/. Last accessed March 13, 2021.
Firm fragility 257
43 The nation-wide lockdown in India, which was announced on 24 March 2020, has
gone through a process of phased ‘unlocks’. Many services and production facilities
have faced systematic shutdowns; some like entertainment theatres and schools con-
tinue to be in a state of lockdown at many places of India at the time of writing this
chapter.
4 4 In a globalized world, we should become conditioned to the fact that more such shocks
will occur in the near future. The time gap between successive pandemics, such as
the SARS, MERS and then Covid-19 has been decreasing over time; refer to https://
jfsdigital.org/2020/03/18/neither-a-black-swan-nor-a-zombie-apocalypse-the-
futures-of-a-world-with-the-covid-19-coronavirus. Last accessed March 13, 2021.
45 The treatise of the German sociologist, Ulrich Beck, is a classic text on the external-
ities caused in the process of industrialization in post-industrial societies, marked by
calamitous risk. The entire focus of such societies then becomes mitigation of such
risks, particularly to the environment; for further details, refer to Beck, U. (1992).
Risk Society: Towards a New Modernity. Translated by Ritter, Mark. London: Sage
Publications. ISBN 978-0-8039-8346-5.
46 This is an example of a spillover of financial shock from one sector of the economy
to another (in manufacturing); for further details, see https://spia.princeton.edu/
system/files/research/documents/ Krueger_A%20Retrospective%20Look%20at%20
Rescuing\ %20an%20Restructuring%20General%20Motors%20and%20Chrysler.pdf.
Last accessed January 2021.
47 Further details are at https://msme.gov.in/sites/default/files/ActiononCOVID-19
ReliefforMSMESector.pdf. Last accessed March 13, 2021.
48 Prof. Subramanian discusses the issues of measurability in the context of multidimen-
sional aspects of poverty measurement; see https://www.theindiaforum.in/article/
poverty-india. Last accessed March 13, 2021.
Bibliography
Alfaro, L., Gonzalo, A., Chari, A., and Panizza, U. (2019). Corporate debt, firm size
and financial fragility in emerging markets. Journal of International Economics, 118: 1–9.
Altman, E. I. (2005). An emerging market credit scoring system for corporate
bonds. Emerging markets review, 6(4): 311–323.
Ang, E., Iancu, D. A., and Swinney, R. (2017). Disruption risk and optimal sourcing in
multitier supply networks. Management Science, 63(8): 2397–2419.
Bartik, A. W., Bertrand, M., Cullen, Z. B., Glaeser, E. L., Luca, M., and Stanton, C. T.
(2020). How are small businesses adjusting to covid-19? Early evidence from a survey.
Working Paper 26989, National Bureau of Economic Research.
Beck, T., Flynn, B., and Homanen, M. (2020). Covid-19 in emerging markets: Firm-
survey evidence. COVID Economics Vetted and Real-time Papers, Centre for Eco-
nomic Policy Reforms (CEPR) Press Issue 38, CEPR.
Beyer, R. C., Franco-Bedoya, S., and Galdo, V. (2021). Examining the economic impact
of covid-19 in India through daily electricity consumption and nighttime light inten-
sity. World Development, 140:105287.
Bostrom, N. (2013). Existential risk prevention as global priority. Global Policy, 4:15–31.
Bottazzi, G., Secchi, A., and Tamagni, F. (2006). Financial fragility and growth dynam-
ics of Italian business firms. Technical report, LEM Working Paper Series.
Bouey, J. (2020). Assessment of covid-19’s impact on small and medium-sized enterprises:
Implications from China: Testimony of Jennifer Bouey, before the US house commit-
tee on small business, March 10, 2020. In RAND Corporation. Rand Corporation.
Bramoullé, Y., and Kranton, R. (2007). Risk-sharing networks. Journal of Economic Behav-
ior & Organization, 64(3–4):275–294.
258 Debdatta Saha
Braziotis, C., Bourlakis, M., Rogers, H., and Tannock, J. (2013). Supply chains and
supply networks: Distinctions and overlaps. Supply Chain Management: An International
Journal, 18(6): 644–652.
Brown, R., Rocha, A., and Cowling, M. (2020). Financing entrepreneurship in times of
crisis: Exploring the impact of covid-19 on the market for entrepreneurial finance in
the United Kingdom. International Small Business Journal, 38(5):380–390.
Cainelli, G., Montresor, S., and Marzetti, G. V. (2012). Production and financial linkages
in inter-firm networks: Structural variety, risk-sharing and resilience. Journal of Evo-
lutionary Economics, 22(4):711–734.
Cheema-Fox, A., LaPerla, B. R., Serafeim, G., and Wang, H. S. (2020). Corporate resil-
ience and response during covid-19. Available at SSRN 3578167.
Chen, H., and Chen, T.-J. (2002). Asymmetric strategic alliances: A network view. Jour-
nal of Business Research, 55(12):1007–1013.
Choi, T. Y., and Krause, D. R. (2006). The supply base and its complexity: Implications
for transaction costs, risks, responsiveness, and innovation. Journal of Operations Man-
agement, 24(5):637–652.
Cowling, M., Brown, R., and Rocha, A. (2020). Did you save some cash for a rainy
covid-19 day? The crisis and SMES. International Small Business Journal, 38(7):593–604.
Dannenberg, P., Fuchs, M., Riedler, T., and Wiedemann, C. (2020). Digital transition
by covid-19 pandemic? The German food online retail. Tijdschrift voor economische en
sociale geografie, 111(3):543–560.
Demirguc-Kunt, A., Lokshin, M., and Torre, I. (2020). The sooner, the better: The
early economic impact of non-pharmaceutical interventions during the covid-19 pan-
demic. World Bank Policy Research Working Paper, (9257).
Didier, T., Huneeus, F., Larrain, M., and Schmukler, S. L. (2020). Financing firms in hiber-
nation during the COVID-19 pandemic. The World Bank.
Fombrun, C. J., and Wally, S. (1992). “Global entanglements: The structure of corporate
transnationalism.” V. Pucik, N.M. Tichy and C.K. Barnett (eds.), Globalizing manage-
ment: Creating and leading the competitive organization. New York. John Willey & Sons.
Gallegati, M., Giulioni, G., and Kichiji, N. (2003). Complex dynamics and financial fra-
gility in an agent-based model. Advances in Complex Systems, 6(03):267–282.
Harland, C., Brenchley, R., and Walker, H. (2003). Risk in supply networks. Journal of
Purchasing and Supply Management, 9(2):51–62.
Harland, C. M., Lamming, R. C., and Cousins, P. D. (1999). Developing the concept
of supply strategy. International Journal of Operations & Production Management, 19(7):
650–674. doi:10.1108/01443579910278910
Hasanat, M. W., Hoque, A., Shikha, F. A., Anwar, M., Hamid, A. B. A., and Tat, H. H.
(2020). The impact of coronavirus (covid-19) on e-business in Malaysia. Asian Journal
of Multidisciplinary Studies, 3(1):85–90.
Hsieh, C.-T., and Olken, B. A. (2014). The missing “missing middle”. Journal of Economic
Perspectives, 28(3):89–108.
Ivanov, D., and Dolgui, A. (2020). Viability of intertwined supply networks: Extending
the supply chain resilience angles towards survivability. A position paper motivated
by covid-19 outbreak. International Journal of Production Research, 58(10):2904–2915.
Jeffries, F. L., and Reed, R. (2000). Trust and adaptation in relational contracting. Acad-
emy of Management Review, 25(4):873–882.
Käki, A., Salo, A., and Talluri, S. (2015). Disruptions in supply networks: A probabilistic
risk assessment approach. Journal of Business Logistics, 36(3):273–287.
Firm fragility 259
Kanitkar, T. (2020). The covid-19 lockdown in India: Impacts on the economy and the
power sector. Global Transitions, 2:150–156
Kirman, A. (2010). Complex economics: Individual and collective rationality. Routledge.
Lüscher, T. F. (2020). Covid-19: (Mis)managing an announced black swan. European Heart
Journal, 41(19): 1779–1782. doi:10.1093/eurheartj/ehaa435.
Majumdar, A., Shaw, M., and Sinha, S. K. (2020). Covid-19 debunks the myth of socially
sustainable supply chain: A case of the clothing industry in South Asian countries.
Sustainable Production and Consumption, 24:150–155.
Manheim, D. (2020). The fragile world hypothesis: Complexity, fragility, and systemic
existential risk. Futures, 122:102570.
Milgrom, P. (1992). Economics, organization and management. Prentice-Hall.
Minsky, H. P. (1992). The financial instability hypothesis. The Jerome Levy Economics
Institute Working Paper, (74).
Mulligan, R. F. (2013). A sectoral analysis of the financial instability hypothesis. The
Quarterly Review of Economics and Finance, 53(4):450–459.
Nishi, H. (2019). An empirical contribution to Minsky’s financial fragility: Evidence
from non-financial sectors in Japan. Cambridge Journal of Economics, 43(3):585–622.
Rapanta, C., Botturi, L., Goodyear, P., Guardia, L., and Koole, M. (2020). Online uni-
versity teaching during and after the covid-19 crisis: Refocusing teacher presence and
learning activity. Postdigital Science and Education, 2(3):923–945.
Schroeder, S. (2009). Defining and detecting financial fragility: New Zealand’s ex-
perience. International Journal of Social Economics, 36(3): 287–307. doi:10.1108/
03068290910932765.
Shafi, M., Liu, J., and Ren, W. (2020). Impact of covid-19 pandemic on micro, small, and
medium-sized enterprises operating in Pakistan. Research in Globalization, 2:100018.
Taleb, N. N. (2007). The black swan: The impact of the highly improbable, volume 2. Random
house.
Thorgren, S., and Williams, T. A. (2020). Staying alive during an unfolding crisis: How
SMES ward off impending disaster. Journal of Business Venturing Insights, 14:e00187.
Vives, X. (2014). Strategic complementarity, fragility, and regulation. The Review of Fi-
nancial Studies, 27(12):3547–3592.
Wang, Y. I., Li, J., and Anupindi, R. (2015). Risky suppliers or risky supply chains? An em-
pirical analysis of sub-tier supply network structure on firm performance in the high-tech sector.
deepblue.lib.umich.edu.
Ward, M. (2020). Eskom’s covid-19 force majeure. Emerald Emerging Markets Case Studies.
Yogaananthan, S., and Theva, O. (2020). Building critical supply chain resilience in the wake
of covid-19. think-asia.org.
15
IMPACT OF COVID-19 ON
MANUFACTURING INDUSTRIES
IN INDIA
Networks and trade dependence
Partha Chatterjee and Shweta Jain
15.1 Introduction
The Covid pandemic has resulted in an unprecedented health and economic cri-
sis worldwide. There is hardly any sector or segment in any part of the world that
remains unaffected by this raging virus. The world economy is facing a shock
like no other in recent past. The impact is evident in the contraction of the world
economy in the second quarter of 2020. India has been among the countries
hardest hit by the pandemic. The GDP in India contracted by a massive 23.9%
in April–June quarter in the Financial Year 2020–21. While all sectors except
agriculture contracted, manufacturing contracted by a whooping 39.3% in that
quarter.
Manufacturing sector in India contributes 17.42% of GDP in India. Employ-
ment in the manufacturing sector in India has steadily increased from 14.81% in
1991 to 23.70% in 2017. It is likely that this number has increased a bit by 2020.
As India goes through the demographic transition, it is expected that manufac-
turing will play a critical role in generating employment for the large youth base
of the country. This is also recognized by the government which is bringing out
a slew of policies through Make in India and more recent Atmanirbhar Bharat to
expand the manufacturing base in India. It is therefore important to understand
the impact on manufacturing.
Even within manufacturing, the impact will not be similar for all industries.
It is important to understand the heterogeneity across industries and how the im-
pact is likely to be different for different industries. That may help in better miti-
gation and policy making. Our objective in this chapter is to study the impact of
Covid on the manufacturing sector in India and see which industries are likely
to be impacted more. In particular, we focus on the supply side of the economy.
We analyze the heterogeneity across industries based on how inputs are sourced,
DOI: 10.4324/9781003220145-18
Impact of Covid-19 on manufacturing 261
domestic or international, and also how they are placed in the supply network
within industries. We then look at the impact caused by the disruptions on the
supply side, both domestic and via trade.
The pandemic has resulted in both demand and supply shocks in the economy.
Our focus in this chapter will be on the supply side. We study how various in-
dustries in the manufacturing sector have been impacted in the aftermath of the
pandemic. The upheaval in the manufacturing sector is visible in each industry.
However, there is a differential impact across industries. We point out that the
impact on an industry will depend on two factors, how much it relies on im-
ported goods for its inputs and how dependent the industry is in the network of
manufacturing industries.
International trade requires several linkages to be formed and transactions
to happen. There are a variety of steps that need to be taken to ensure that the
product reaches the recipient from the supplier. The possibility of disruption is
high because of this complexity. Moreover, in today’s world where the global
value chain (GVC) is severely fragmented, the possibility of disruption goes up –
particularly in a situation like this. Not only is the supply chain vertically frag-
mented, a particular nature of the GVC has been lack of diversification and even
dependence on single suppliers. While this has resulted in significant cost saving
and allowed prices to remain low, it also has significantly increased the risk of
disruption. Further, in case of disruption in the international supply chain, it
may not be easy to move to an alternative supplier. Hence, industries where
there is larger dependence on imported inputs are likely to experience higher
contraction.
The second aspect that will cause a variation in the impact across industries
is the dependency of the industry. For each industry, we find out which other
industries supply inputs to it. The higher the number of industries supplying to
a particular industry, more dependent the latter is. Higher is the dependency of an
industry, greater is the chance of disruption. Note that we also calculate central-
ity. This is based on supply rather than use. We create a network using the same
procedure we used to create the dependence network, except that it is based on
how many industries a particular industry supplies to.1 Interestingly, there is a
high correlation between dependency and centrality. This implies that industries
that supply to a lot of industries also source inputs from a large number of other
industries. This means that a supply disruption will have ripple effects and the
shock will be amplified.
Thus, both these channels will amplify the shock – industries that are de-
pendent on imports or on several other industries for their inputs will experience
a bigger shock. Whether one channel will moderate the effect of the other, or
strengthen each other will depend on whether share of imports in the total inputs
and dependency are positively correlated, negatively correlated or uncorrelated.
We find that there is a weak positive correlation between the two.
That there is very little correlation between dependency and import shares
implies that these two channels are somewhat separated. High dependency
262 Partha Chatterjee and Shweta Jain
industries rely mostly on domestic sources and trade disruption will not impact
them. However, as noted above, trade disruption will have a direct impact on
industries that has high reliance on imported inputs. In the literature, there is no
direct counterpart for an analysis similar to us, but several papers have noted the
importance of networks and imports in the GVC. For example, Vandenbussche
et al. (2019) look at the production networks in European countries and study
the impact of Brexit in terms of using the input-output data. Similarly, Los et al.
(2015) look at the network of inputs in the manufacturing sector across coun-
tries. Acemoglu et al. (2012) and Boehm et al. (2016) also study interlinkages
of firms using a network. However, Baldwin et al. (2014) look at the Canadian
manufacturing sector and the role of imported inputs.
To understand how trade shock will impact manufacturing output, we look at
different industries and see how much of their inputs are imported. We find that
there is a great degree of variation in how much of their inputs are imported. We
compute the value of imported inputs and value of inputs for the years 2011–12
to 2017–18 using Annual Survey of Industries (ASI) data at two-digit National
Industrial Classification (NIC) 2008 code. The share of imported inputs in total
inputs varies from 72.63% for Coke and Petroleum derivatives to only 1.32% for
Tobacco industry. We find that there is a positive correlation between output
share (in the total manufacturing output) of the industries and import share.
In panel regressions between the two variables, the coefficient is positive and
significant.
Next, we turn our attention to how dependency plays a role. We construct
networks of industries based on their use of inputs from other industries. Based
on that network, we calculate a dependency score for each industry. We then rank
industries by their dependency score. At the top of ranking are industries like
Chemical and Machinery which uses inputs from a large number of other indus-
tries and at the bottom are Leather and Tobacco which needs few inputs for their
products. We find that dependency is positively correlated with the output share
of the industries, though the correlation is not that high. In panel regressions, the
coefficient is positive but significant only at 10% significance level when a year
fixed effect is included.
Thus, both these dimensions of heterogeneity in industries are important in
understanding how the shock due to Covid impacts various industries differently.
An industry with high import share, or with high dependency, or both (we find
positive, but not high, correlation between the two), will be disrupted more than
others. As a result, there will be bigger fall in total output. In the first quarter
of the Financial Year 2020–21, the first full quarter after the spread of Covid in
India, output fell by as much as −23.9%. While all industries fared badly, there
was a variation in the performance of the industries. Using industry-wise Index
of Industrial Production (IIP), we find that the fall is strongly correlated with
the dependency score of the industry and the industry’s share of imports in total
inputs. The fall is significantly higher for industries which have high dependency
score and high import share.
Then, using a difference-in-difference strategy, we look at whether capital-

intensive or labor-intensive industries are more impacted by these channels.
We find that higher import share impacts labor-intensive industries more than
capital-intensive industries, while it is the reverse for dependency.
Finally, we create a vulnerability index (VI) for industries, which is the product
of import share and dependency score. An industry which is heavily reliant on
imports might find it harder to restore the supply chains at a competitive cost.
Similarly, an industry which uses inputs from a large number of industries, it is
exposed to the vagaries of all of them and recovery can be relatively fragile and
difficult. If an industry is reliant on imports and uses multiple inputs, it becomes
significantly more difficult for it to recover. We use the data on IIPs from April
2020 to July 2020 and see that while almost all industries continue to perform
poorly, industries with lower VI score have relatively done better with recovery.
15.2 Production function, data and methodology

To understand how inputs, whether they are domestic or international, and the
variety will have an impact on output, let us consider a general production func-
tion with multiple intermediate inputs. Industries are indexed by s, the num-
ber of intermediate inputs that industry uses. We follow Halpern et al. (2015)
in defining the production technology. Plants in industry s ∈ S are indexed by
i = 1,..., I s . The output of plant i is given by the production function:
s ∈S
Qi = ωi K iα Li β ∏X γj
ij (15.1)
j =1
In a departure from Halpern et al. (2015) paper, we assume that X is intermediate

input is imported, while the rest are sourced domestically. Also, notice that the
number of inputs used differs as s varies across industries. Thus, if there is a shock
propagating through the economy, the impact on output will depend on the
share of X is in the total input as well as the value of s. In the sections that follow,
we estimate the share of aggregate X is in total inputs in an industry and s and
characterize industries by that.
To construct the networks, we use supply and use tables provided by the
Ministry of Statistics and Programme Implementation (MOSPI) for the years
staring from 2011–12 to 2015–16. We created the network maps which indicate
the interdependence of each industry on others (Acemoglu et al., 2012; Boehm
et al., 2016). For centrality, “Supply” table is used and for dependency, “Use”
table is used. Supply table gives the information on the value of output by each
industry (supplier) across columns and use table provides the value of the inputs
consumed by each industry.
The data on value of imported inputs, value of output and value of exports
by industry is estimated for the years 2009–10 to 2017–18 using ASI unit level
data of two-digit NIC 2008 code. ASI is also published by MOSPI. We also
computed the year-on-year change in the IIP for the months April–July 2020.
IIP depicts the changes in the composition of the industry. We have matched the
industry classification of three data sets, namely ASI data, Supply and use tables
and IIP. We have in total 19 two-digit industries. However, in our analysis, we
do not include other manufacturing, so we have 18 industries.
We have done various regression analyses using different sets of variables.
First, the relationship between output share for each industry was computed
using the value of output available for all industries. Output share of industry is
calculated by dividing value of output of each industry by total value of output of
that year. Relationship between output share and imported inputs share is exam-
ined using econometric methodologies, namely OLS regression and regression
with industry and year fixed effects.
Further, we also evaluate the relationship between change in IIP, dependency
and imported inputs share using difference-in-difference strategy. For this anal-
ysis, we took April to July 2020 as post and November 2019 to February 2020 as
pre-shock. We also calculate capital intensity for each industry using ASI data.
15.3 Imported inputs

Over the last few decades, globalization has steadily increased. One key aspect of
this wave of globalization has been fragmentation of the production process and
the value chain being distributed globally. While India is on the lower side of
openness, nevertheless it is a part of the GVC. However, there is a great degree
of variation across industries in how globally connected they are. In particular,
there exists huge variation across industries in the share of imported inputs over
total inputs. Import share is calculated as the ratio of value of imported inputs to
total inputs.
In Table 15.1, import share for each industry from the year 2009–10 to 2017–
18 is given. We find that the import share varies from 75.48% (for Coke) to 1.36%
(beverages) in 2009–10, and similarly from 72.63% (Coke) to 1.32% (Tobacco
Products) in 2017–18. It is easily noticed that the share has remained somewhat
stable over the years from each industry, except for Wood industry which saw the
share of input fall from 23.26% in 2009–10 to just 6.8% in 2017–18.
We also look at the output share of the industries. That is, we find out what
share of the total output of all industries is each industry producing (we exclude
other manufacturing). Here too, we find that their relative contribution remains
more or less stable across years. The share of output in total output for each in-
dustry from the year 2009–10 to 2017–18 is given in Table 15.2.
The two, import share and output share, by industry are plotted in
Figure 15.1. As we can see, import share for all industries, except Coke, is below
25%. The output share is also well diversified. Output share for any industry does
not exceed 15%. Thus, there is no dominant industry as such.
What we are interested to know is that whether there is any correlation be-
tween output share and import share.
TABLE 15.1 I mport share
Industry 09–10 11–12 12–13 13–14 14–15 15–16 16–17 17–18

Coke 75.48 75.48 80.50 77.08 61.67 62.44 52.12 72.63
Chemicals 22.45 21.29 19.16 21.73 19.08 20.89 18.01 22.05
Machinery 16.82 16.34 16.78 18.40 14.71 17.16 15.52 21.19
Basic Metals 17.64 14.55 17.57 12.93 13.26 22.09 12.83 19.64
Pharmaceuticals 16.32 16.58 12.59 16.81 13.40 12.44 13.82 14.70
Rubber 16.15 14.17 13.62 16.92 12.09 12.66 10.11 12.67
Paper 12.62 13.92 12.18 14.96 10.85 11.27 9.82 11.69
Printing 10.41 14.17 11.02 11.31 7.94 11.32 9.33 11.37
Transport 13.27 11.73 14.10 12.81 10.77 12.09 9.84 11.30
Leather 8.12 8.92 7.01 8.59 7.71 10.51 7.98 10.63
Non-Metallic 7.47 8.27 8.73 7.75 6.83 7.05 6.50 6.85
Wood 23.26 14.35 22.69 17.43 8.89 7.77 7.55 6.80
Fabricated 10.67 9.11 8.72 7.74 5.33 10.13 4.81 5.81
Metals
Food 7.53 6.88 7.06 7.06 3.98 5.94 4.84 5.75
Furniture 6.45 8.51 5.02 6.98 6.67 5.27 4.64 4.54
Textiles 4.75 4.80 4.67 4.44 3.52 5.06 3.18 3.99
Beverages 1.36 1.59 1.47 3.13 1.25 2.33 1.27 1.71
Tobacco 4.25 1.58 2.53 1.09 1.09 2.10 1.19 1.32
Data Source: Annual Survey of Industries (ASI).
TABLE 15.2 Output share
Industry 09–10 11–12 12–13 13–14 14–15 15–16 16–17 17–18

Food 12.21 12.87 12.70 12.96 13.75 13.80 15.31 14.90
Basic Metals 14.84 15.76 13.98 14.41 13.64 13.29 12.49 14.09
Coke 16.10 17.10 19.11 18.64 16.55 13.42 13.15 12.87
Machinery 12.53 10.62 10.45 9.87 10.04 10.75 11.48 11.72
Transport 9.40 9.58 9.65 8.95 9.91 10.90 10.85 10.99
Chemicals 8.26 8.77 8.57 8.61 8.95 9.72 8.75 8.87
Textiles 7.69 7.05 7.29 8.31 7.73 7.94 7.82 7.40
Pharmaceuticals 3.32 3.44 3.46 3.54 3.73 4.23 4.08 3.83
Rubber 3.71 3.44 3.53 3.61 3.84 3.86 3.93 3.74
Non-Metallic 3.71 3.38 3.55 3.24 3.71 3.54 3.78 3.74
Fabricated 3.14 3.42 2.95 2.90 2.78 3.05 3.06 2.83
Metals
Paper 1.31 1.28 1.25 1.41 1.48 1.51 1.48 1.52
Beverages 0.98 1.01 0.95 1.03 1.16 1.13 1.07 1.03
Leather 0.91 0.68 0.75 0.80 0.81 0.96 0.86 0.74
Tobacco 0.66 0.52 0.60 0.58 0.65 0.76 0.66 0.59
Printing 0.62 0.59 0.62 0.52 0.70 0.61 0.59 0.56
Wood 0.32 0.31 0.34 0.36 0.33 0.33 0.35 0.30
Furniture 0.27 0.21 0.27 0.26 0.25 0.21 0.28 0.28
Data Source: Annual Survey of Industries (ASI).

FIGURE 15.1 Output share and import share in 2015–16.
TABLE 15.3 Correlation between import share and output share
Year Correlation Year Correlation
2011–12 0.56 2015–16 0.55

2012–13 0.65 2016–17 0.48
2013–14 0.61 2017–18 0.50
2014–15 0.59
We look at year-wise correlation in the cross-section data and also use the
panel of industries over time. Table 15.3 lists the correlation between import share
and output share for each year. In the cross-section data for each year, we find the
correlation to be between 0.48 and 0.65. This suggests that industries that have
a higher share of imported inputs contribute a higher share of the total output.
This is also corroborated in the panel regression between the two variables. As
we can see in Table 15.4, there is a positive and significant relationship between
import share and output share. Thus, industries, which have a higher share of
their imports in their inputs, contribute a higher share of the total output. This is
an important finding. This signifies that any disruption to trade due to pandemic
will have a significant impact on the total output of manufacturing goods in India.
15.4 Dependency
The heterogeneity of the impact of the pandemic shock on industries also comes
from another source. Industries will be differentially impacted based on the fact
how they depend on other domestic industries for their inputs. In particular, if
an industry sources inputs from a large number of other industries, the chance of
disruption goes up significantly compared to an industry that sources inputs from
TABLE 15.4 Panel regression between output share and import share
Variables Import Share Import Share Import Share Import Share
Output Share 1.683*** 1.738*** 1.683*** 1.742***

[0.24] [0.50] [0.25] [0.45]
Constant 4.466** −8.634 5.220 −7.934
[1.87] [7.24] [3.26] [6.48]
Year FE N N Y Y
Industry FE N Y N Y
Observations 90 90 90 90
R-squared 0.352 0.973 0.358 0.980
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
TABLE 15.5 R anking (dependency)
Industry 2009–10 2011–12 2012–13 2013–14 2014–15 2015–16
Chemicals 1 1 1 1 1 1
Machinery 2 2 2 2 2 2
Rubber 3 4 3 3 3 2
Furniture 4 13 15 13 10 14
Textiles 5 3 4 4 5 12
Non-Metallic 6 9 9 9 4 6
Fabricated Metals 7 6 6 6 7 4
Food 8 15 16 11 11 15
Paper 9 7 7 5 9 7
Wood 10 5 5 8 6 5
Pharmaceuticals 11 10 10 10 12 13
Coke 12 8 8 7 14 8
Basic Metals 13 14 13 15 13 10
Beverages 14 12 12 14 16 17
Transport 15 17 17 17 17 9
Printing 16 11 11 12 8 11
Leather 17 16 14 16 15 16
Tobacco 18 18 18 18 17 18
a fewer domestic industries. We use use value for the input-output tables and cre-
ate a network. The use gives the value of the inputs consumed by each industry.
First, we estimate the share of input of industry j used by industry i to produce
the output by dividing the consumption of intermediate input of industry j by
total value of inputs usage of industry i. This will give us the matrix which con-
tains the share of inputs for each industry. We compute the dependency scores by
evaluating the networks which depicts the dependency of each industry on other
industries for purchasing the intermediate inputs. Finally, we rank these depend-
ency scores and sort these ranks by year 2009–10. This is tabulated in Table 15.5.
The network diagrams are included in the Appendix.
The ranks have remained more or less stable except for two industries –
urnitures and Fixtures and Textiles. They started out as a relatively high de-
F
pendency industry, but their ranks dropped significantly by the year 2015–16.
Figure 15.2 plots dependency and output share, and illustrates this point.
We then look at correlation between output share and dependency. First, we
look at the correlation between these two variables for each year. Table 15.6
shows those correlations. There is a positive correlation, though the correlation
is not that high.
We then look at the panel regression between output share and dependency
across industries. In pure OLS without any fixed effect, we find that there is a
significant and positive correlation between the two. When we use industry
fixed effect, we still get a positive significant relationship. However, with time
fixed effects, there is no significant relationship. Thus, there seems to be some
indication that more dependent industries have a higher output share, but the
relationship is not that strong.
Note that we can also calculate the centrality of an industry which depends on
the supply of that particular industry. More number of industries it supplies to,
FIGURE 15.2 Output share and dependency.
TABLE 15.6 Correlation between

output share and
dependency
Year Correlation
2012 0.185
2013 0.177
2014 0.194
2015 0.161
2016 0.219
more central that industry is. The details of this as well as the network diagrams
are available in Chatterjee, Dey and Jain (2020). We investigate if there is a cor-
relation between the two – dependency and centrality. It turns out, as we can
see in Table 15.8, that they are very highly correlated. This is interesting since it
means that there will be a strong ripple effect.
TABLE 15.7 Panel regression between output share and dependency
VARIABLES Dependency Dependency Dependency Dependency
Output Share 0.389* 0.065 0.389* 0.068

[0.22] [0.36] [0.22] [0.36]
Constant 4.733*** 0.302 4.940* 0.473
[1.69] [5.13] [2.94] [5.15]
Year FE N N Y Y
Industry FE N Y N Y
R-squared 0.034 0.975 0.036 0.977
* p < 0.1.
TABLE 15.8 Correlation between centrality

and dependency
Year Correlation
2011–12 0.923
2012–13 0.812
2013–14 0.942
2014–15 0.899
2015–16 0.867
TABLE 15.9 Panel regression between centrality and dependency
Centrality 0.504*** 0.033 0.505*** 0.023

[0.03] [0.03] [0.03] [0.03]
Constant 0.944 1.046 1.028 1.305
[0.67] [0.90] [1.35] [0.99]
Year FE N N Y Y
Industry FE N Y N Y
R-squared 0.768 0.976 0.769 0.977
* p < 0.1.
TABLE 15.10 Correlation between

import share and
dependency
Year Correlation
2012 0.139
2013 0.107
2014 0.173
2015 0.102
2016 0.180
TABLE 15.11 Panel regression between import share and dependency
Import Share 0.103 0.156** 0.102 0.121

[0.08] [0.08] [0.08] [0.09]
Constant 5.467*** −1.275 5.615* −0.602
[1.59] [1.49] [2.93] [1.75]
Year FE N N Y Y
Industry FE N Y N Y
R-squared 0.020 0.977 0.020 0.977
* p < 0.1.
In a panel regression between centrality and dependency, we get a similar

result. We get a significant (at 5%) positive relationship between centrality and
dependency. The significance, however, is lost if we add an industry fixed effect,
though it survives a time fixed effect.
Finally, we look at the correlation between import share and dependency.
There seems to be a weak correlation between the two. Thus, it is likely that
the two channels amplify the shocks independently and impact a different set of
industries.
We get a similar result in the panel regression between import share and de-
pendency. We run a regression without any fixed effect and then a combination
of year and industry fixed effects. The relationship is positive and significant
when there is an industry fixed effect but no time fixed effect. It is not significant
otherwise.
15.5 Impact on manufacturing output (IIP)

Covid has impacted all countries and practically all sectors. However, the impact
on the Indian economy has been particularly hard. The Indian economy was
slowing down even before Covid spread to the country. The growth rate in the
quarters preceding the pandemic was below the long-run trend. That, coupled
with the full impact of the nationwide lockdown announced by the central gov-
ernment on March 24, 2020, meant the Indian economy severely contracted
in the April–June quarter of 2020 (Q1 of Financial Year 2020–21). The GDP
at constant (2011–12) prices grew at −23.9%. All sectors, except Agriculture,
Forestry and Fishing, contracted. Manufacturing contracted severely, manufac-
turing GVA by 39.3% and manufacturing IIP by −40.7%.
The national lockdown lasted from March 25, 2020 to May 31, 2020, in four
different phases. Beginning July 1, 2020, unlock of the economy started. That
led to some recovery. As we can see in the monthly IIP data, the severity of the
contraction decreased over the months, but it remains in the negative zone. We
look at the change in IIP for each industry for the months of April–July 2020. In
Figure 15.3, we plot the year-on-year change for each industry for each of the
months. We see that in the wake of the strict national lockdown implemented
from March 25, 2020, April was the worst hit. Several industries were completely
shut, hence contracted close to 100%. May 2020 was slightly better, but not by
much. The nationwide lockdown was lifted by May 31, 2020. That allowed some
of the industries to scale up production compared to the previous months. The
recovery continued in July. Yet, even in these two months, almost all industries
contracted. Manufacturing IIP contracted −38.4% in May, −16% in June and
−11.1% in July. All types of industries, except Pharmaceuticals and Tobbaco, have
continued to contract.
In the graphs in Figure 15.3, industries are sorted by the share of imports. It
appears that industries with higher import share have fared worse than others.
Similarly, in Figure 15.4, industries are sorted by dependency and we see that the
industries which have higher dependency have witnessed a bigger contraction.
60.00
40.00
20.00
0.00
–20.00
–40.00
–60.00
–80.00
–100.00
–120.00
Tabacco Beverages Textiles Furniture Food
Non- Metallic Wood Fabricated Metals Leather Paper
Printing Transport Pharmaceuticals Rubber Machinery
Chemicals Basic Metals Coke
FIGURE 15.3 oY change in IIP for the months April–July 2020 sorted by import
Y
share.
60
40
20
–20
–40
–60
–80
–100
–120
Tabacco Beverages Leather Food Furniture Pharmaceuticals
Textiles Printing Basic Metals Transport Coke Paper
Non-Metallic wood Fabricated Metals Rubber Machinery Chemicals
FIGURE 15.4 YoY change in IIP for the months April–July 2020 sorted by dependency.
TABLE 15.12 Correlation between change in IIP,

import share and dependency
Month Import Share Dependency
Apr-20 0.57 0.80

May-20 0.33 0.40
Jun-20 0.16 0.20
Jul-20 0.03 0.06
If we look at the correlation between change in IIP and the import share, we
find that they are positively correlated. The correlation is the highest in April
2020, but weakens as the months progress, suggesting that industries are adjust-
ing and finding alternative supply channels. We get a similar picture when we
look at the correlation between change in IIP and dependency. Again, we find
a very strong correlation in April as expected, but it weakens by July. It is likely
that the supply channels were restored to a certain extent. Note that what we are
pointing out is that the correlation is weakening, while almost all the industries
are still experiencing contraction.
Thus, we get what we expected. That the industries have experienced a mas-
sive negative shock, the degree of impact to an extent depends on the nature
of industry. In particular, the supply channels matter. Whether an industry is
reliant on imports matters. What also matters is the number of inputs necessary
to produce the output.
Finally, to understand how the impact differentially impacted the industries
post- Covid, we adopt a difference-in-difference strategy. We take advantage
of the fact that in the production function, capital and labor are inputs apart
from the intermediate inputs. So, industries differ by their capital intensities.
We divide industries into capital-intensive and labor-intensive industries based
on the median capital intensity. The impact of Covid started to be felt in March
2020. The national lockdown also started in March 2020. We, therefore, look
at four months before March and four months after March in our difference-in-
difference regressions. We leave March 2020 out of this analysis and consider
November 2019 to February 2020 as the months prior to the pandemic and April
2020 to July 2020 as the months post-pandemic.
We run three sets of difference in difference. One in which we include only
import share other than the interaction term and fixed effects, another in which
we include only dependency and the third in which we include both.
When we include only import shares, we find that the coefficient on import
share is positive and significant when industry fixed effects are included. Thus,
import share aggravates the impact of the Covid shock for labor-intensive indus-
tries. For those industries, higher import share leads to a higher fall in IIP. The
results are shown in Table 15.13.
In Table 15.14, we tabulate the results from the difference-in-difference re-
gression where we include only the dependency variable. Here, we find that de-
pendency moderates the impact for the labor-intensive industries and aggravates
that for the capital-intensive industries. The coefficient for the dependency is
negative and significant with industry fixed effects.
In the difference-in-difference regression where we include both import share
and dependency, we get similar signs for both the variables, but only import
share is significant at 5% significance level. The results are shown in Table 15.15.
Thus, having established that these two factors, import share and dependency,
result in a heterogeneous impact of Covid, we create a VI. We scale the import
share and the dependency score and take the product of it. We scale the prod-
uct so that the value is between 0 and 100. If an industry has high import share
and also has several inputs, it is most vulnerable. It will not be easy to substitute
several inputs which depend on GVC. However, in those industries which have
low import share and also low number of inputs, it is likely to recover the fastest
TABLE 15.13 Difference in difference: IIP and import share
VARIABLES IIP IIP IIP IIP
Import Share 0.218 2.765** 0.218 2.765**

[0.16] [1.13] [0.16] [1.13]
Interaction term −12.797 −13.072* −12.817 −13.099*
[7.80] [7.58] [7.83] [7.61]
Constant −3.663 −53.163*** −4.472 −54.112***
[4.66] [20.21] [5.48] [20.47]
Year FE N N Y Y
Industry FE N Y N Y
R-squared 0.455 0.542 0.455 0.542
* p < 0.1.
TABLE 15.14 Difference in difference: IIP and dependency

[7.84] [7.58] [7.86] [7.61]
Dependency 0.152 −72.755** 0.152 −72.755**
[0.18] [29.64] [0.19] [29.74]
Constant −1.040 175.244** −1.810 174.296**
[4.03] [74.05] [4.95] [74.37]
Year FE N N Y Y
Industry FE N Y N Y
R-squared 0.450 0.542 0.450 0.542
* p < 0.1.
TABLE 15.15 Difference in difference: IIP, import share and dependency
Import Share 0.205 2.649** 0.205 2.649**

[0.16] [1.09] [0.16] [1.09]
[7.82] [7.58] [7.84] [7.61]
Dependency 0.122 −3.034 0.123 −3.034
[0.19] [2.18] [0.19] [2.19]
Constant −4.436 −43.639** −5.257 −44.588**
[4.81] [18.04] [5.62] [18.31]
Year FE N N Y Y
Industry FE N Y N Y
R-squared 0.457 0.542 0.457 0.542
* p < 0.1.
as it may be possible to find domestic sources and restore the supply chain. For
an industry with one high import share, there is a risk of disrupted supply chain,
but if it has low number of inputs, its vulnerability will be moderated. Similarly,
vulnerability is low for cases where import share is low but uses high number of
inputs. We find Chemical industry to be the most vulnerable using this measure
and Tobacco the least. Table 15.16 gives the full list.
Even though only a few months have passed since the onset of the pandemic
and almost all industries are still contracting, we can see that those industries
with low vulnerability score have recovered more. If we look at the difference in
the growth rate of an industry in April 2020 and July 2020, that gives us a sense
about the recovery process of that industry. Tobacco industries have the lowest
TABLE 15.16 Vulnerability index
Industry VI Score VI Rank
Chemicals 9.86 1
Machinery 2.23 2
Coke 2.17 3
Rubber 1.65 4
Fabricated 1.05 5
Basic Metals 0.51 6
Paper 0.50 7
Wood 0.45 8
Transport 0.32 9
Non-Metallic 0.32 10
Printing 0.22 11
Pharmaceuticals 0.16 12
Textiles 0.10 13
Leather 0.06 14
Furniture 0.04 15
Food 0.04 16
Beverages 0.01 17
Tobacco 0.00 18
vulnerability score (VI rank 18) and they have made a stunning recovery in a
few months. In April 2020, they almost completely shut production, their con-
traction being −99.59%; yet by July 2020, they grew at a rate of 6.05%. On the
other end of the spectrum, an industry like Machinery with high vulnerability
score (VI rank 2) contracted in April 2020 by −92.54%; by July, they recovered
a bit but still contracted by −22.4%. The correlation between the difference in
growth rate in April 2020 and July 2020 and the VI score is −0.22. As expected,
the sign is negative. The correlation is on the lower side because of two industries
in particular, Chemical (VI rank 1) and Food (VI rank 16). Both these industries
were not hit by the same full force of national lockdown at the beginning because
of them (or, part of those industries) being considered as essential commodities.
Hence, their initial fall was relatively moderate. If we leave those two industries
out, the correlation goes up to −0.48. Thus, overall, even short data that we
have, VI seems to be a good indicator of the ease or difficulty in recovery.
15.6 Conclusion
In this chapter, we have looked at the supply side of the manufacturing industry
in India and studied the channels through which disruption can happen. This
gives us a better sense of understanding of the heterogeneity across industries and
therefore helps in planning and policy making. However, it is worth mention-
ing that we have not looked at the demand side of the economy. This economic
shock, due to Covid, is all-encompassing and there is a big demand shock too.
So, in recovery, some of the industries may find it harder because of the demand
not picking up. Demand for some products may simply vanish, while some other
products may be in more demand than earlier, so there will be reallocation too.
This process will also be dynamic, and may only stabilize well after the pandemic
is over. The supply side disruptions are, however, already visible and firms might
benefit in working toward mitigation of this. Similarly, government policy needs
to be nuanced keeping in mind the heterogeneity of the industries.
Note
1 For a detailed analysis of centrality, please refer to Chatterjee et al. (2020).
Bibliography
Acemoglu, D., Carvalho, V. M., Ozdaglar, A., & Tahbaz-Salehi, A. (2012). The network
origins of aggregate fluctuations. Econometrica, 80(5), 1977–2016.
Baldwin, J. R., Yan, B., et al. (2014). Global value chains and the productivity of Canadian
manufacturing firms. Citeseer.
Boehm, J., Dhingra, S., Morrow, J., et al. (2016). Swimming upstream: Inputoutput link-
ages and the direction of product adoption. CEP Discussion Paper, (1407).
Chatterjee, P., Dey, S., & Jain, S. (2020). Lives and livelihood: An exit strategy from
lockdown. Economic and Political Weekly, 55(22),38–44.
Halpern, L., Koren, M., & Szeidl, A. (2015). Imported inputs and productivity. American
Economic Review, 105(12), 3660–3703.
Los, B., Timmer, M. P., & de Vries, G. J. (2015). How global are global value chains?
A new approach to measure international fragmentation. Journal of Regional Science,
55(1), 66–92.
Vandenbussche, H., Connell Garcia, W., & Simons, W. (2019). Global value chains, trade
shocks and jobs: An application to brexit. CESifo working paper.
APPENDIX
FIGURE 15.5 Dependency networks.

16
MEASUREMENT OF EFFICIENCY LOSS
IN INDIAN AGRICULTURAL SECTOR
DUE TO COVID-19
Evidence from rice production
Dipyaman Pal, Arpita Ghose and Chandrima Chakraborty
16.1 Introduction
Although the share of agriculture in Gross Domestic Product (GDP) is one-
seventh, around 48% workforce is engaged in the agricultural sector. The annual
growth rate in real terms in agriculture and its allied sectors was 2.88% from
2014–15 to 2018–19. The estimated growth rate of this sector is 2.9% during
2019–20 (Economic Survey 2019–20). In case of India, the first case of Covid-19
was noted in Kerala on 31st January, 2020. During the nationwide lockdown pe-
riod from 25th March, 2020 to 31st May, 2020, a huge number of migrant labour
migrated from the informal industrial sector to their own state and faced major
problems (near about 40 million migrant workers of India, according to the re-
port published in The Economic Times (dated April 23, 2020) and pointed out by
World Bank). In case of India, migrant workers are mostly engaged in informal
or unrecognized sector like construction, hospitality, textile, domestic work, etc.
According to the report published by the Central Statistical Organisation
(CSO), the third advance estimate of India’s GDP for 2019–20 is Rs. 47 lakh
crore. The estimate of GDP per week turns out to be Rs. 3.6 lakh crore, of
which 17% is received by casual labour, 21% by regular wage employees and 62%
by self-employed workers. As per the data available from the Periodic Labour
Force Survey (PLFS) for 2017, near about 25% of total workers in India are casual
labourers (i.e. almost 93 million people) who earned about Rs. 1,754 per week.
Regular wage earners who acquire on average Rs. 4,063 per week are 23% of the
total labour force. The rest, 52% workers, are self-employed ones that include own
account workers, the self-employed and helpers in household enterprises and a
self-employed person earns about Rs. 3,460 per week. The nationwide lockdown
affected all these workers on a different extent; the casual labourers are the most
affected groups, as their income depends on working days available in a week.
DOI: 10.4324/9781003220145-19
Efficiency loss in agricultural sector 279
In this Covid-19 situation, the picture of Indian economic growth is almost

disastrous like most of the other countries. According to the press note by NSSO
on 31st August, 2020, the GDP at a constant price declined by 23.9% in the first
quarter of 2020–21 as compared to 5.2% growth in the first quarter of 2019–20.
However, agricultural sector is slightly better placed. The percentage change in
GVA at Basic Prices in the first quarter of 2020–21 as compared to the first quar-
ter of 2019–20 is negative in all the sectors except agricultural sector. In case of
agricultural sector, there is an increase in 3.4% of GVA at basic Prices in the first
quarter of 2020–21 as compared to the first quarter of 2019–20. However, the
huge influx of migrant labour in rural India may affect the performance of agri-
cultural sector in the reverse way. The main reason behind this is that these huge
numbers of migrant labourers who are bound to choose the agriculture-related
activities are not equally efficient as compared to the usual labourers who nor-
mally work in the agricultural sector. In fact, this huge migration from informal
or unrecognized sector to the agricultural sector may increase the surplus labour
in the agricultural sector which, in turn, may decrease the extent of efficiency
of the agricultural sector. So, the actual loss of output may be much more as it is
predicted. It is very much needed to predict the actual loss of output in the agri-
cultural sector in Indian economy keeping in mind the loss of efficiency.
There are some studies related to the impact of Covid-19 on the Indian ag-
ricultural sector. Dev and Sengupta (2020) pointed out the health risk in rural
areas. They also indicated that as some migrant workers are back in rural areas,
there may be a risk of Covid-19 for the cultivators, agricultural labourers, work-
ers and others working throughout the food supply chains. Sahoo and Rath
(2020) concluded that to get rid of the aggressive effect of Covid-19 and to
keep the agricultural business unaffected, the administration should give precise
rules for horticulture activity, the value structures have to be thought about
and changed, such that the farmers are not under any sort of monetary troubles,
the cultivating system alongside the workers should be protected from Corona
infection by keeping up social separating together with managing the agitation
and mental misfortunes during this time and obviously giving the farmers a stage
for direct advertising to fulfil the need of the shoppers by alternative measures
and avoid any sort of wastages in the reaped items. Arumugam, Kanagavalli and
Manida (2020) concluded that during this Covid-19 situation, powerful spread
of the rules and usage on the ground will be urgent. They also said that this
Covid-19 situation will have a negative implication on the farming group of
people and will be undergoing serious inimical and mind losses. Pohit (2020)
recommended that the direct money transfer at the hand of the poor people
may be more advantageous because poor people will probably spend the entire
amount as their saving propensity is negligible. This kind of spending may in-
crease the demand of the economy. Dev (2020) pointed out that there may be a
mix of private market and government procurement as access of the markets by
the farmers is much needed. More help is needed for the small poultry and dairy
farmers. The farmers and the agricultural workers should be included in the
280 Dipyaman Pal et al.
assistance package given by the Government of India and in any social protection
programmes addressing the crisis. Carberry and Padhee (2020) pointed out that
the state governments may play a vital role by strengthening their mechanism
for smooth procurement operations of farmer’s marketable surplus at minimum
support price in this Covid-19 situation. Not only that, budget allocation in the
agricultural sector is not crowded out because agricultural sector is the backbone
of food security. They also concluded that productive investment, including re-
search and innovation, would be very helpful. Kumar, Padhee and Kumar (2020)
concluded that the Covid-19 situation has affected the India’s agricultural food
system and emphasized the agricultural market reforms and digital solutions to
connect farmers to markets. They also pointed out that decentralization of ag-
ricultural food system makes farmers more strong. Reardon et al. (2020) argued
that due to Covid-19, there is an increase in the transaction costs and uncertainty
in India’s transformed food supply chains which leads to a high risk in food
security because 92% of food consumption in India is purchased mainly from
the private sector. They also recommended that the government may focus on
sustaining the food supply chains towards ultimate ricochet because government
food distribution cannot replace even a tenth of the market. All the above studies
measured the impact of Covid-19 on the agricultural sector in India either by
the reforms in the agricultural market or by the strengthening mechanism for
smooth procurement operations and also some are concerned with the health risk
of the rural sector.
However, there is dearth in the study measuring the impact of Covid-19 on
Indian agricultural sector by taking into consideration the efficiency in the agri-
cultural sector. This type of study is much needed because huge influx of migrant
labour in the agricultural sector from the informal or unrecognized sector may
increase the problem of surplus labour and may, in turn, affect the efficiency of
the agricultural sector. This chapter tries to fill up this gap in the literature and
attempts to contribute to the literature in the following directions:
I To get an estimate of the loss in efficiency in the Indian agricultural sector

due to mammoth inflow of the agricultural labour, as huge amount of labour
migrated from the informal or unrecognized sector to the agricultural sector
in 2020,
II To measure the loss of output due to loss in efficiency,
III To find out the extent of the casual labour force that can be dispensed in
order to keep the output level unchanged, considering rice production as a
case study.
Rice is chosen because rice happens to be the major item of food produc-
tion. Authors own calculation based on the data provided by the Ministry of
Agriculture & Farmers Welfare, Government of India suggests that the aver-
age share of rice production in total food grains production during 2010–11
to 2018–19 is 40.52% and the average share of rice in total agricultural ex-
port during 2014–15 to 2018–19 is 22.85%. Further, among all the countries
around the globe, India is the second largest producer of rice in 2018–19
(116.42 metric tons) (Data Source: https://www.statista.com/).
The present study is based on 16 major rice-producing states in India.
The rest of the chapter is as follows: Section 16.2 deals with methodology and
data sources. Results are discussed in Section 16.3. Section 16.4 represents the
concluding remarks.
16.2 Methodology and data sources

In efficiency analysis, the objective is to generate the maximum output using a
given amount of inputs. Technical efficiency (TE) refers to a situation in which it
is impossible to produce more output from the given level of inputs unless reduc-
ing the level of other output (termed as output-oriented TE) and the same level of
output by using less of one or more inputs unless increasing the amount of other
inputs (termed as input-oriented TE). In the case of output-oriented measure,
the TE of a firm can be computed by comparing its actual output with the max-
imum producible quantity from its observed inputs, i.e. by how much can output
quantities be proportionally expanded without altering the input quantities used.
In input-oriented measure, the TE of a firm can be measured by comparing its
actual input in use with the minimum input that would produce the targeted
output level, i.e. by how much can input quantities be proportionally reduced
without changing the actual output bundle. Both the input- and output-oriented
measures of technical efficiencies in the case of single input and output can be
visualized from Figure 16.1.
In Figure 16.1, input x is measured along the horizontal axis and output y
along the vertical axis. Point A ( x0 , y0 ) represents the actual input-output bundle
of a firm, A. Now, y * = f ( x0 ), where y* is the maximum output producible
Output(y)
y* f(x)
y0 A(x0,y0)
x* x0
FIGURE 16.1 The input- and output-oriented measure of technical efficiency.

y0
from input x0 . The output-oriented measure of TE of firm A = which is the
y*
comparison of actual output with the maximum producible quantity from the
observed input. Now for the same output bundle y0, the input quantity can be
reduced proportionately till the frontier is reached. So, y 0 can be produced from
x*
input x*. Thus, the input-oriented TE measure for firm A = . The TE score
x0
of a firm takes a value between 0 and 1. A value of 1 indicates that the firm is
fully, technically efficient.
16.2.1 Measurement of technical efficiency (TE)

There are two major approaches of estimating TE: (i) Parametric approach, i.e.
by estimating production function, and (ii) Non-parametric data envelopment
approach (DEA). This study adopts non-parametric DEA (Banker, Charnes and
Cooper (1984) model of variable returns to scale (VRS)). The Data Envelop-
ment Analysis was originally formulated by Charnes, Cooper and Rhodes [CCR
(1978)]. The original CCR model was applicable only to technologies charac-
terized by constant returns to scale (CRS) globally. Later, Banker, Charnes and
Cooper (BCC) (1984) extended the CCR model to accommodate technologies
that exhibit VRS. The major advantage of this approach is that it does not require
any prior specification of production function. Given the data on actual input and
output, the frontiers are generated by using a programming method. Relevantly,
it may be mentioned that the misspecification of the functional relationship in
case of the use of parametric specification may result in a wrong estimate.
For evaluation of performance of production unit, a benchmark production
function has to be constructed which is known as frontier, and is supposed to be
perfectly efficient.
Comparison of the observed performance of production unit with the postu-
lated standard of perfect efficiency is the basic approach of measuring efficiency.
In DEA, a benchmark technology is constructed from the observed input-
output bundles of the firm in the sample without any assumption regarding the
production frontier. The general assumptions made about the production tech-
nology are as follows: (i) All actually observed input-output combinations are
feasible. (ii) The production possibility set is convex. (iii) Inputs are freely dis-
posable. (iv) Outputs are freely disposable. These are weak assumptions. These
assumptions hold for all technologies represented by quasi-concave and weak
monotonic production function.
16.2.1.1 Output-oriented TE
It is supposed that there are N firms. Each of them is producing ‘g’ outputs using
‘h’ inputs. The firm t uses input bundle x t = ( x1t ,x 2t ,xht ) and produces the out-
put bundle y t = ( y1t ,y 2t ,y gt ). Technology can follow either CRS or VRS.
The production possibility set corresponding to CRS can be defined as
 N N 
 
T CRS
= ( x , y ) : x ≥

∑ t
λ jx ; y ≤ ∑
λ j y ; λ j ≥ 0; ( j = 1,2, N ) (16.1)
t

 j =1 j =1 
The specific production possibility set under VRS is given by
 N N N 
 
T VRS = ( x, y ) : x ≥

∑ λ j xt ; y ≤ ∑ λ j yt ; ∑ λ j = 1; λ j ≥ 0; ( j = 1,2, N ) (16.2)

 j =1 j =1 j =1 
The output-oriented measure of TE of any firm t under CRS technology requires
the solution of the following LP problem.
max φ
N
Subject to ∑λ y j rj ≥ φ yrt ; (r = 1,2, g ) ;
j =1
∑λ x j ij ≤ xit ; (i = 1,2,h ) ;
j =1
φfree λ j ≥ 0; ( j = 1,2, N ) (16.3)
Output-oriented TE of firm t can be determined by using equation (16.4).

1
TEoct = TEoct ( x t , yt ) = (16.4)
φ*
where φ * is the solution of equation (16.3) showing that the maximum value
of φ .y * is the maximum output bundle producible from input bundle xt and is
defined as y * = φ * y t . Under VRS, max φ , φ • can be determined by solving equa-
N
tion (16.3) along with the constraint ∑λ j = 1, taking into account the VRS
j =1
frontier (equation (16.2)). Knowing φ • , TE of the firm can be solved.
16.2.2 The data and its sources

The study envisaged one output and multi-input production process. The pro-
duction of rice (in thousand tons) as output and (i) fertilizer (Kg. Nutrients), (ii)
manure (Qtl.), (iii) animal labour (Pair Hrs.) (iii) human labour (Man Hrs.) as
inputs are considered. Since the main concern of the present chapter is to find
out the effect of casual labour, total human labour is decomposed into three
components: Family labour, attached labour and casual labour. Thus, altogether,
there are seven inputs.
The data on production of rice (in thousand tons) and area of rice production
(in thousand hectares) are collected from the Ministry of Agriculture and Farm-
ers Welfare, Government of India. The data on fertilizer (Kg. Nutrients), manure
(Qtl.), animal labour (Pair Hrs.) and human labour (Man Hrs.) are obtained
from the cost of cultivation data, the Directorate of Economics and Statistics,
Government of India. The data source of migration of labour is the census 2011,
Government of India.
For measurement of data in 2020, only those migrant labours that migrated
from the rural area of a particular state to the other state and within the same
state are considered. As per the census of India 2011, around 64.1% of total
workforce is engaged in agriculture in rural areas. Accordingly, it is assumed
that the 64.1% of the total migrant workers will be absorbed in the agricultural
sector due to the Covid-19 effect. Those migrant labour are considered who are
illiterate or have education below graduation level because this kind of migrant
labours are mainly migrated from the informal sector or unorganized sector to
the rural areas. As because of their education level, this kind of migrant labours
have no option apart from the agricultural activity. It is also assumed that as
these labours are migrant in nature, there is an increase in the casual labour only
among all the inputs. Male migrant labour and total migrant labour are consid-
ered separately. That is, efficiency in 2020 is estimated separately by assuming
that (i) only male migrant labours are engaged in the agricultural sector and (ii)
total migrant labours (both male and female) are engaged in the agricultural
sector.
For estimating the output and input figures of rice sector in 2020, it is con-
sidered that production of rice increased by 3.4% (according to the press note by
NSSO on 31st August, 2020, in the case of agricultural sector there is an increase
of 3.4% of GVA at Basic Prices in the first quarter of 2020–21 as compared to the
first quarter of 2019–20) in 2020 from 2017–2018 (latest available data).
The data on wholesale price indices of rice crops has been collected from the
Agricultural prices in India published by the Government of India.
The sample states are Andhra Pradesh, Assam, Bihar, Gujarat, Haryana, Hi-
machal Pradesh, Jharkhand, Karnataka, Kerala, Madhya Pradesh, Maharashtra,
Odisha, Punjab, Tamil Nadu, Uttar Pradesh and West Bengal. The sample pe-
riod is from 2004–05 to 2017–18.
For comparison of (i) changes in efficiency score and (ii) consequent changes
in output as well as (iii) to find out whether there is any possibility of getting some
figures regarding the extent to which casual labour force can be dispensed, in
order to keep output level unchanged, between the period 2004–05 to 2017–18
and the year 2020, after obtaining the results for the period 2004–05 to 2017–18,
the results for 2020 will be re-estimated using the obtained figures of inputs and
output for 2020, using the method just mentioned above.
16.3 Results
All the estimated results are presented in Tables 16.1–16.3.
From the results of Table 16.1, one can conclude that the average efficiency
of rice production in India from the period 2004–05 to 2017–18 is 0.91, imply-
ing that there is a scope of increase of 9% of output by using the same amount
of inputs. The efficiency of rice production over the sample period is lowest in
Himachal Pradesh (0.49) followed by Tamil Nadu (0.62). Among the 16 states,
11 states have efficiency of rice greater than 0.95.
It is evident that the overall decrease in the efficiency of rice production in
2020 is 1.54% and 2.65% under the above assumptions i.e. (i) Only male migrant
labours are engaged in the agricultural sector and (ii) total migrant labours (both
male and female) are engaged in the agricultural sector, respectively.
The state level estimates suggest high value of loss in efficiency occurs for
Jharkhand and Tamilnadu. In case of Jharkhand and Madhya Pradesh, there is a
TABLE 16.1 Change in efficiency due to Covid-19
State Average Efficiency Efficiency % Change by % Change by

Efficiency in 2020 by in 2020 by considering only considering Total
2004–05 considering considering Male Migrant Migrant Labour
to only male only total Labour
2017–18 migrant migrant labour
labour
Andhra 0.9955 1 1 0.452034154 0.452034154

Pradesh
Assam 1 1 1 0 0
Bihar 0.922 0.922 0.922 0 0
Gujarat 0.780846 0.811 0.801 3.861688504 2.5810265
Haryana 0.999071 1 1 0.092943447 0.092943447
Himachal 0.471 0.462 0.492 −1.910828025 4.458598726
Pradesh
Jharkhand 0.804071 0.686 0.698 −14.6841965 −13.19179177
Karnataka 0.952071 0.952071 0.952071 0 0
Kerala 0.731714 0.731714 0.731714 0 0
Madhya 0.632143 0.623 0.628 −1.446327684 −0.655367232
Pradesh
Maharashtra 0.971 0.95 0.95 −2.162718847 −2.162718847
Odisha 0.943 0.897 0.89 −4.87804878 −5.620360551
Punjab 1 1 1 0 0
Tamil Nadu 0.926857 0.627 0.627 −32.35203453 −32.35203453
Uttar Pradesh 1 1 1 0 0
West Bengal 1 1 1 0 0
Average 0.88308 0.853862 0.855737 −3.314218016 -2.899854381
fall in the efficiency of rice production by 13.19% and 32.35%, respectively, if it is

assumed that total migrant labours are engaged in the agricultural sector.
From the analysis, it is possible to estimate the extent of input slack for the
casual labour, showing the amount of casual labour that can be dispensed with,
in order to keep the amount of output level unchanged. Such an estimate is ob-
tained under both assumptions (i) and (ii). The existence of surplus labour is evi-
dent, in the sense that the decrease in casual labour force will not reduce the level
of output. In fact, the percentage amount of casual worker that can be dispensed
with, in order to keep the amount of output remaining unchanged, actually in-
creases in 2020 as depicted in Table 16.2. In 2020, such increase is 465.9429%,
if total migrant labourers are absorbed and are 258.4814%, and if only male mi-
grant labours are involved in the agricultural sector. This result clearly indicates
that there is an increase in surplus labour in the agricultural sector if migrant la-
bours are engaged in the agricultural activities. In fact, the system produces more
surplus labour if the female migrant labourers are taken into account.
NSSO predicts 3.4% growth in agricultural output in the first quarter of 2020.
But the results of the present model support that there is inefficiency in the system,
which basically shows penalty to the production process and, in turn, implies that
the output should have been much more in the absence of such inefficiency. The
estimation of the model suggests that if only male migrant workers are absorbed,
then the resulting inefficiency of the model implies that the output should have
been 17.11% in 2020. Thus, here basically we are admitting the loss of output to
the extent of 13.71% in the form of penalty. However, if only total migrant work-
ers are absorbed, then the resulting inefficiency of the model supports that the
output should have been 17.36% in 2020. Hence, here the penalty in the form of
output loss is 13.96%. These results are presented in Table 16.3.
TABLE 16.2 Change in input slack due to Covid-19
Casual
Average Input Slack up to 2018 0.034123

Input Slack By Considering Male Migrant Labour in 2020 0.122324
Input Slack By Considering Total Migrant Labour in 2020 0.193116
% Change by considering only Male Migrant Labour in 2020 258.4814
% Change by considering Total Migrant Labour in 2020 465.9429
TABLE 16.3 Percentage change in output due to Covid-19
Output Loss Output Loss Due To

(in %) Efficiency Loss (in %)
Percentage Change in Output by considering 17.11 13.71

only Male Migrant Labour in 2020
Percentage Change in Output by considering 17.36 13.96
only Total Migrant Labour in 2020
16.4 Conclusion
During the Covid-19 pandemic situation, there is a huge downward pressure in
the Indian economy, which is accentuated by the problem of migrant labour. A
huge number of migrant labourers migrated from the informal industrial or un-
organized sector to the agricultural sector during the Covid-19 situation, which
may, in turn, adversely affect the performance of the Indian agricultural sector.
Appreciating this problem, the Government of India has also taken some
initiatives to boost up the performances of the Indian agriculture sector. The
problem is that these huge numbers of migrant labourers who are bound to
choose the agriculture-related activities are not equally efficient as compared to
the usual labourers who normally used to work in the agricultural sector. For
this reason, there may be a decrease in the extent of efficiency of the agricultural
sector. Such decline in efficiency will pay a growth penalty in the form of lower
benefit and hence leads to the economic loss of lower growth of output. This
chapter attempts to measure the extent of decline in efficiency, the consequent
loss of output and also the extent of casual labour that can be dispensed with in
the Indian agricultural sector arising out of the employment of inefficient labour
due to the Covid-19 problem. The analysis is carried out considering 16 major
rice-producing states over the period 2004–05 to 2017–18. To find out the loss
in efficiency, an attempt has been taken to estimate the efficiency in 2020 on
the basis of some usual assumptions. The results suggest that if one assumes that
only male migrant labours are engaged in the agricultural sector, the overall de-
crease in efficiency of rice production in 2020 is 1.54% and the consequent loss
of output due to the loss in efficiency is 13.71%. While if one assumes that total
migrant labours (both male and female) are engaged in the agricultural sector,
then the overall decrease in the efficiency in 2020 is 2.65% and the consequent
loss of output due to the loss in efficiency is 13.96%.
Again, it is also evident that the percentage change in the input slack for casual
labour, i.e. the amount of casual labour that can be dispensed in order to keep
output level unchanged, has increased by 456.9429% in 2020 if total migrant
labours are involved in the agricultural sector. However, input slack for casual
labour has been amplified by 258.4814% in 2020 if only male migrant labours are
involved in the agricultural sector. This result clearly indicates that there is an
increase in surplus labour in the agricultural sector, if migrant labourers are ab-
sorbed in the agricultural activities after the Covid-19 situation. The system pro-
duces more surplus labour if the female migrant labourer is taken into account.
According to the press note by NSSO on 31st August, 2020, in case of ag-
ricultural sector there is an increase of 3.4% of GVA at Basic Prices in the first
quarter of 2020–21 as compared to the first quarter of 2019–20. But this growth
rate may be more if there is no loss in efficiency due to reverse migration in the
agricultural sector during the Covid-19 period. So, it can be concluded that al-
though the performance of Indian agricultural sector is more or less satisfactory
in the first quarter of 2020, we are committing a possibility of loss of output due
to the loss in efficiency. So, in order to check the loss of output, government may
take some policies regarding the re-employment of the migrant labour in the
rural sector in informal or unorganized sector. Otherwise, this problem of mi-
grant labour may increase the problem of the agricultural sector and may affect
the growth of output in the agricultural sector in the reverse way. This loss of
output in the agricultural sector may in the long term become a threat towards
the food security in India.
References
Arumugam, Dr. U. and Kanagavalli, Dr. G. and M. Manida, (2020). COVID-19. Impact
of Agriculture in India (May 14, 2020). Aegaeum Journal. 8(5). Available at SSRN.
https.//ssrn.com/abstract=3600813
Banker, R. D., Charnes, A. and W. W. Cooper (1984). Some Models for Estimating
Technical and Scale Efficiencies in Data Envelopment Analysis. Management Science.
30(9), 1078–1092.
Carberry, Dr. P. and Padhee, Dr. A. K. (2020). Containing Covid19 Impacts On In-
dian Agriculture. https.//www.icrisat.org/containing-covid19-impacts-on-indian-
agriculture/. 03.03.2021.
Charnes, A., Cooper, W. W. and E. L. Rhodes (1978). Measuring the Efficiency of
Decision- Making Units. European Journal of Operational Research. 2(6), 429–444.
Dev, M. S. (2020). Addressing COVID-19 Impacts on Agriculture, Food Security, and
Livelihoods in India. https.//www.ifpri.org/blog/addressing-covid-19-impacts-
agriculture-food-security-and-livelihoods-india. 03.03.2021.
Dev, M. S. and Sengupta, R. (2020). Covid-19. Impact on the Indian Economy. Indira
Gandhi Institute of Development Research, Mumbai April 2020.
Kumar, A., Padhee, A. K. and S. Kumar (2020). How Indian Agriculture Should Change
after COVID-19. Food Security. 12. https.//doi.org/10.1007/s12571-020-01063-6
Pohit, S. (2020). Recession, COVID-19, and Aftermath. A Macroview. In book.
COVID-19. Challenges for the Indian Economy - Trade and Foreign Policy Effects (pp.
71–73), ASEAN-India Centre (AIC) - Engineering Export Promotion Council of
India (EEPC).
Reardon, T., Mishra, A., Nuthalapati, C. S. R., Bellemare, M. F. and D. Zilberman (2020).
COVID-19’s Disruption of India’s Transformed Food Supply Chains. Economic &
Political Weekly. 55(18), 18–22.
Sahoo, P. P. and Rath, S. (2020). Potential Impact of Corona Virus on Agriculture Sec-
tor. Biotica Research Today. 2(4). www.bioticainternational.com. 03.03.2021.
17
THE PANDEMIC AND THE INFORMAL
SECTOR OF INDIA
Probable impacts and strategic intervention1
Anirban Kundu, Manojit Bhattacharjee, Pratip Kumar Datta,

Kasturi Sadhu and Saumya Chakrabarti
17.1 Introduction
The world has been overwhelmed by Covid-19 since the beginning of 2020.
A new species of coronavirus (SARS-COV-2, later named Covid-19) has been
identified in the city of Wuhan in the Hubei province of China and has spread
rapidly across the world. The pandemic has brought tremendous economic crisis
worldwide (IMF, 2020). According to Jonas (2013), this type of crisis has multi-
dimensional impacts, which ultimately brings in production losses due to death
and illness of the labour force. Baldwin (2020) describes the economic impact of
Covid-19 in a more comprehensive manner. He has tried to capture the nature of
the crisis by identifying the demand- and supply side income chains and showed
that labour is the most affected factor of production. According to him, labour
becomes vulnerable due to disrupted household income that ultimately reduces
consumption, savings and hence investment and aggregate output at the macro
level. This, in turn, has an adverse multiplier effect on labour income itself (see
Figure 17.1 in the Appendix).
Among the labourers, those who are involved in the informal sector are cer-
tainly trapped in the most precarious condition. India has the largest informal
sector of the globe. Around 70.5% of the total non-agricultural workforce of
India takes refuge in this informal sector (Mehrotra, 2019). Dasgupta and Rajeev
(2020) in this context have opined that the large informal sector of India has
become most vulnerable due to the lockdown. According to their theoretical
position, the global economy, along with India, was suffering because of the
effective demand problem even before the pandemic. The lockdown measure
has further added the problem of supply inelasticity and thus made the economy
more vulnerable.
DOI: 10.4324/9781003220145-20
290 Anirban Kundu et al.
Along with theoretical analyses, empirical evidence also shows that the im-
pact of the pandemic on the informal sector of India is going to be severe and
can be compared with the severity of the Great Depression of the 1930s (IMF,
2020). Khan and Mansoor (2020) predicted that this pandemic would expose
millions of informal workers and their families to starvation, hunger and even
death. Estupinan, Sharma, Gupta and Birla (2020) estimate a monthly wage loss
of casual workers and regular and salaried employees of Rs. 33.8 thousand crores
(as per 2017–18 prices) due to this pandemic. Abraham and Khan (2020) show
that 430.4 million (93.4% of the Indian workforce) are already affected by the
lockdown in terms of wage/earning loss. Mukherjee, Ray & Bag (2020) opined
that around 40 million people are going to lose their jobs in the country and most
of them are informal workers. With a telephonic survey of over 4,942 households
all over India, Kesar, Abraham, Lahoti, Nath and Basole (2020) show that 66%
of the workforce (in the sample) lost jobs during the lockdown and among them,
the urban self-employed are the worst affected.
In this context, our study gives a special emphasis on the most vulnerable
section of the informal sector, viz., own account enterprise (OAE), i.e. the self-
employed. The informal sector, in general, have been suffering due to economic
distress even before the outbreak of the current pandemic; however, the pan-
demic and the following nation-wide lockdown exacerbated the vulnerability of
these large masses. Our present study captures the extent of vulnerability both at
the enterprise- and at the household-level, based on the various rounds of NSSO
data. Subsequently, the study proposes a strategic intervention in the informal
sector with a series of short-, medium- and long-run measures (Chakrabarti,
Dutta, Kundu, & Sadhu, 2020).
We argue for universal food transfer as a short-run policy intervention in
the wake of probable job loss and the already existing low level of average in-
come for a large section of the self-employed. As a medium-run policy measure,
we advocate in favour of un/conditional cash transfer to the informal sector
workers (along with petty agriculture), which would strengthen the micro-level
production linkages particularly within the informal sector, and also with agri-
culture. Hence, we argue that such a measure would ease both the supply- and
demand-side rigidities in the economy and therefore, this un/conditional cash
transfer is non-inflationary. Finally, we advocate for a long-run strategy of form-
ing/inducing clusters of informal firms: either physically for manufacturing or
virtual clustering for services.
17.2 A tale of acute vulnerability
17.2.1 Vulnerable rural-urban self-employed with abysmally

low per-capita income
Let us start our discussion by presenting the enormity and deep deprivation of
the Indian non-agricultural informal sector. Table 17.1 shows that the sectoral
TABLE 17.1 Total employment (million) and share (%) in different sectors in India over time
Year Agriculture Informal Sector Formal Construction Others

Manufacturing
Total Rural Self- Rural Urban Self- Urban Sector
employment Establishment employment Establishment
using hired using hired labour
labour
1999–2000 242.72 77.11 32.96 5.33 21.43 17.40 8.17 17.39 52.51
(61) (19.38) (8.28) (1.34) (5.38) (4.37) (2.05) (4.37) (13.20)
2010–11 224.30 107.98 41.70 11.48 28.06 26.73 12.69 42.42 48.01
(51.52) (24.80) (9.58) (2.64) (6.45) (6.14) (2.92) (9.74) (11.03)
2015–16 205.46 111.27 38.13 11.74 30.96 30.45 14.30 47.94 70.86
(45.67) (24.74) (8.48) (2.61) (6.88) (6.77) (3.18) (10.66) (15.75)
Source: Authors’ computation based on various databases.

Note: Agricultural sector and construction sector data are taken from ILO; total employment data is taken from ILO; formal manufacturing data from ASI; informal
sector data from NSSO enterprise survey, 55th (1999–2000), 67th (2010–11) and 73rd (2015–16) rounds. “Others” primarily includes formal services and is com-
puted by subtracting employments in agriculture, informal sector, formal manufacturing and construction sector from total employment.
The pandemic and the informal sector 291
share of the informal sector workforce, including construction (that is consider-

ably informal), is large and expanding. Although agriculture is the largest seg-
ment, its share is declining. The informal sector (including construction) is a
major driving force for this sectoral shift of labour.
However, if we consider the informal sector as a whole and the large self-
employment segment, in particular, we observe that the gross value added per
worker (partial labour productivity) is mostly lower than the average income
of the low-productive/low-income agricultural sector (see Table 17.2). On the
contrary, productivity levels of formal manufacturing and services are very high
compared to that of agriculture and non-agricultural informality, and these ab-
solute differences have been widening over time, deepening deprivation.
More disaggregated analysis on the self-employed enterprises (i.e. OAEs)
across labour-size class helps to capture the extent of vulnerability among the
informal workers in a better way (Tables 17.3 and 17.4). From Table 17.3, the
lowest two size classes (of tiny/petty OAEs) together (i.e. size classes 1 and 2
together) cover around 84% of rural and 80% of urban workforce that depend
on manufacturing activities for their livelihood. However, the concentration of
workers within rural-urban petty services, for labour-size class of one and two
together, is 80% and 88%, respectively. Thus, considering both manufacturing
and services, most of the firms are found to be tiny with 1–2 workers only.
Now, during 2010–11, almost 53% and 21% of the rural self-employed work-
ers of informal manufacturing and informal service enterprises (respectively),
who are operating with only one person, earn an average income even lower
than the poor and marginal farmers. Again, these shares of rural self-employed,
engaged in informal enterprises with absorption-capacity of two workers, are
around 65% and 50%, respectively. In urban settlements, the scenario is only
marginally better. Consequently, the condition of a large section of the tiny
self-employed firms, having 1–2 labourers, is even worse than petty agriculture.
We can see from Table 17.4 that the situation further deteriorates in 2015–16.
This vulnerability of the tiny firms indicates their inability to survive during any
catastrophe – be it natural or economic.
17.2.2 Low savings and exposure to economic shocks

due to pandemic
To understand the nature and depth of the shock, we examine whether sav-
ings are sufficient to sustain consumption spending. However, due to the lack
of a comparable database, we resort to two different time points in our anal-
ysis, viz. 2011–12 (68th round NSSO Consumption Expenditure survey) and
2012–13 (70th round NSSO Debt and Investment Survey). Moreover, due to the
non-availability of the Consumption Expenditure Survey and Debt-Investment
Survey during 2010–11 and 2015–16, we cannot directly compare the average
income derived from enterprise survey (as depicted in Tables 17.3 and 17.4)
with consumption, savings and assets across informal households (as depicted in
TABLE 17.2 Gross value added per worker (in Rs.) in 2017–18 prices
Year Agriculture Informal Rural Informal Rural Informal Urban Informal Urban Informal Formal Construction Others
Self-employed Establishment Self-employed Establishment Sector Manufac.
using Hired using Hired Labour Sector
Labour
1999–2000 56,410 38,073 70,898 76,294 135,110 57,457 532,939 187,385 331,618
2010–11 83,309 46,941 100,565 88,509 160,201 72,150 812,823 235,342 949,526
2015–16 115,398 59,029 121,364 103,622 179,592 87,538 847,053 221,834 1,036,524
Source: Authors’ computation based on various databases.

Note: (1) Agriculture and construction sector gross value-added data are taken from NAS, CSO, GOI; Informal sector GVA data taken from NSSO’s Enterprise Survey
as mentioned in Table 17.1; and data on Organized Manufacturing’s GVA is taken from ASI. “Others” data obtained as residual.
(2) CPI for agricultural workers (2017–18 base price) is used to deflate the GVA of the agriculture and rural informal sector. CPI for industrial workers (2017–18 base
price) is used to deflate the GVA of the urban informal sector. WPI for all commodities (2017–18 base price) is used to deflate the GVA of the formal manufacturing
sector.
TABLE 17.3 Total employment and vulnerable employment across labour-size class
among self-employed during 2010–11
Labour- Rural Rural OAE Labour-Force Urban Urban OAE Labour-Force

Size Class OAE with Labour Productivity OAE with Labour Productivity
of OAE Labour- Less Than or Equal to Labour- Less Than or Equal to
Force Marginal Farmers Force Marginal Farmers
Manufacturing
1 5999383 3194717 (53.25) 3624716 1371455 (37.84)

(45.46) (47.58)
2 5027732 3263767 (64.92) 2436544 1012126 (41.54)
(38.09) (31.98)
3–4 1774324 1395198 (78.63) 1255002 679538 (54.15)
(13.44) (16.47)
More 396622 345044 (87) 302457 227755 (75.3)
than 4 (3.01) (3.97)
Total 13198061 7618719 –
(100) (100)
Services
1 13384292 2813866 (21.02) 11237168 1220191 (10.86)
(47.09) (55.07)
2 9360596 4654936 (49.73) 6775893 1410304 (20.81)
(32.94) (33.2)
3–4 2069329 1491712 (72.09) 1811870 592924 (32.72)
(7.28) (8.88)
More 3606880 3534702 (98) 581502 410124 (70.53)
than 4 (12.69) (2.85)
Total 28421097 20406433 –
(100) (100)
Source: Computed Using NSSO 67th Round Unit records data (2010–11).
Note: (a) Values in the parenthesis in the 3rd and 5th columns show the percentage of OAE workers
in that particular labour-size class.
(b) Computed labour productivity of marginal farmers during 2010–11 is Rs. 31,118.22 (2017–18
base price). Marginal farmer’s labour productivity during 2010–11 is obtained by imputing the
output share of the same category of farmers in the year 2012–13 (NSS 70th round) to overall
agricultural labour productivity obtained from Table 17.2.
Tables 17.5, 17.6 and 17.7), but we can understand the general trends based on
the nearby data points. It is to note that Tables 17.6 and 17.7 depict consumption,
savings and assets for the year 2011–12 and hence, these components are perfectly
comparable. We have found from the above-mentioned unit level NSSO data
that the self-employed and casual labour in non-agricultural households across
rural and urban settlements constitute the nonfarm informal households. On the
contrary, regular salaried households involve both formal and informal employ-
ment, but the size of regular salaried jobs in the informal sector is very low (less
than 10%).
TABLE 17.4 Total employment and vulnerable employment across labour-size class
among self-employed during 2015–16
Labour- Rural Rural OAE Labour Urban Urban OAE Labour

Size Class OAE with Labour Productivity OAE with Labour Productivity
of OAE Labour- Less Than or Equal to Labour- Less Than or Equal to
Force Marginal Farmers Force Marginal Farmers
Manufacturing
1 7657263 4857758 (63.44) 4710997 2172169 (46.11)

(54.1) (55.32)
2 4546698 3345176 (73.57) 2438348 1087557 (44.6)
(32.12) (28.63)
3–4 1714414 1458905 (85.1) 1107424 586155 (52.93)
(12.11) (13)
More 235802 209205 (88.72) 258717 160885 (62.19)
than 4 (1.67) (3.04)
Total 14154177 8515486 –
(100) (100)
Services
1 13312324 3501805 (26.3) 12692739 1779962 (14.02)
(55.54) (55.32)
2 8615381 4754157 (55.18) 7894488 2027242 (25.68)
(35.95) (28.63)
3–4 1739330 1238441 (71.2) 1680444 497830 (29.62)
(7.26) (13)
More 300280 176324 (58.72) 175468 99205 (56.54)
than 4 (1.25) (3.04)
Total 23967315 8515486
(100) (100)
Source: Computed using NSSO 73rd Round Unit records data (2015–16).
Note: (a) Values in the parenthesis in the 3rd and 5th columns show the percentage of OAE workers
in that particular labour-size class.
(b) Computed labour productivity of marginal farmers during 2015–16 is Rs. 43,104.35 (2017–
18 base price). Marginal farmer’s labour productivity during 2010–11 is obtained by imputing
the output share of the same category of farmers in the year 2012–13 (NSS 70th round) to overall
agricultural labour productivity obtained from Table 17.2.
Our analysis (Tables 17.5 and 17.6) reveals that, in both rural and urban areas,
a large number of informal households do not possess savings at all; more than
one-quarter (around 27%) of the self-employed in non-agricultural households
in the rural segment and 18% of the similar category in the urban segment do not
possess any savings. The extent of severity in terms of zero savings is more prom-
inent in the case of casual labour in non-agricultural households in rural settle-
ments (more than 40% of the households). The figure stands around 25% in the
case of casual labour households in urban settlements. Further, we can observe
that irrespective of household categories, around 31.8% of the rural households
TABLE 17.5 Share (%) of households owning financial assets (in Rs.) other than shares
and debentures as of 30.06.2012 in rural areas
Household type Zero Savings 0 to 3,000 3,000 to Above

25,000 25,000
Self-employed in agriculture 27.5 35.0 26.4 11.1

Self-employed in 26.6 28.1 29.0 16.4
non-agriculture
Regular wage/salary earning 20.4 22.6 27.3 29.7
Casual labour in agriculture 48.7 32.8 15.5 2.9
Casual labour in 40.5 37.0 18.0 4.5
non-agriculture
Others 52.9 20.2 16.6 10.3
Total 33.5 31.8 23.3 11.3
Source: Computed using 70th round unit level NSSO data on All India Debt and Investment Survey
(AIDIS).
Note: Others include households primarily having unemployed individuals.
TABLE 17.6 Share (%) of households owning financial assets (in Rs.) other than shares
and debentures as of 30.06.2012 in urban areas
Household type Zero Savings 0 to 3,000 3,000 to Above 25,000

25,000
Self-employed 17.8 27.1 33.2 22.0

Regular wage/salary 11.9 23.3 50.0 14.9
earning
Casual labour 24.7 22.0 10.7 42.6
Others 19.5 21.8 22.3 36.5
Total 16.6 24.1 35.2 24.1
Source: Computed using 70th round unit level NSSO data (AIDIS).
and around 24% urban households have savings below Rs. 3,000. Certainly,
most of these households cannot sustain their consumption expenditure beyond
a month or two since the figures stand at Rs. 1,429 and Rs. 2,630, respectively,
in rural and urban areas (see Table 17.7).2
17.3 What needs to be done: not an act of benevolence

but basic responsibility of the state
No doubt, the impact of lockdown has turned out to be enormous and can be
expected to expose millions of informal workers and their families to starva-
tion, hunger and even death (Khan & Mansoor, 2020; Mehta & Kumar, 2020).
Though the government has announced institutional credit and market-linked
economic relief package for the MSMEs, the same has been criticized by several
TABLE 17.7 Monthly per-capita consumption expenditure of households

(in Rs.) by broad item groups, All India, 2011–12
Items Rural Urban
Cereals & cereal substitutes 154 175

Pulses & their products* 42 54
Milk & milk products 115 184
Edible oil 53 70
Egg, fish & meat 68 96
Vegetables 95 122
Fruits 41 90
Sugar, salts and species 76 94
Beverages, refreshments & proc. food 113 236
Total Food Expenditure 756 1121
Clothing and Foot Wear 100 167
Medical 95 146
Education 50 182
Other Non-Food Expenditure 428 1014
Total Non-food Expenditure 673 1509
Total Expenditure 1429 2630
Source: Computed using 68th round consumption expenditure data NSSO.
researchers due to its inadequacy as well as ineffectiveness (Abraham & Khan,

2020). Rather, we need to undertake some truly radical steps, not only in the
short and medium runs but also in the long run.
17.3.1 Short run: the need of the hour

We bolster the argument of a universal food transfer because a significant number
of households have either zero or very low level of savings that could hardly act as
a cushion to maintain the level of consumption needed to combat the economic
shocks associated with the nation-wide lockdown.
Apart from providing food and essentials through a universal public distri-
bution system at the doorsteps, a major complementary mechanism needs to be
designed for nation-wide provisioning of “mid-day-meal”. Thus, (strictly san-
itized) “mass kitchen” providing basic meals could be universalized (e.g. open-
ing up of community kitchen in Kerala; see Raghunath, 2020). The existing
institutions and mechanisms should be beefed up and the huge buffer stock with
the government should be utilized in this time of acute distress. Food should be
distributed freely, for at least a year, to prevent deaths and hunger and arrest the
anticipated increase in the incidence of acute malnutrition, especially among
children and elderly people (Dreze, 2020).
Given the difficult nature of the situation and a constant requirement of ag-
ricultural supplies, a cash transfer to all the marginal and small farmers together
with extensive arrangements for procurement is extremely necessary.
Most crucially, as has been shown theoretically, even if all these processes
(farmer’s-advance-cash-support, procurement, free ration to all, mass-kitchen,
etc.) are financed by printing new money (and not through redistribution), it will
not be inflationary; rather strikingly, it could induce overall economic growth
(Chakrabarti, 2010; Chakrabarti & Datta, 2019). Given the large volume of sur-
plus food grains, this injection of money and the resultant purchasing power in
the hands of the poor should not trigger price rise in agriculture, provided the
states in consultation with the Union Government ensure seamless supply of food
grains. Further, in the wake of acute shortfalls of demand in the post-Covid-19
situation, this excess money supply in the system would in no way create non-
agricultural inflation either. Contrarily, this should induce the almost standstill
industrial economy to kick start, post-lockdown. However, there should be a few
notes of caution as well:
First, labour mobilization should be taken care of, which might pose a serious
problem owing to the sudden lockdown; erstwhile migrants would hesitate to
migrate creating excess supply-demand situations across the labour markets of
the country. Second, instead of full procurement, partial procurement should
be undertaken to avoid food inflation (Chakrabarti, 2010). Third, institutional
frictions, especially in the supply chain that might arise mainly out of post-Covid
lockdown, should be mitigated through monitoring as well as financial support.
Finally, post-Covid-19 sanitary norms should be strictly adhered to.
17.3.2 Medium run: rejuvenating life

As the economy moves out of lockdown, demand needs to be boosted and supply
needs to be freed. A universal and unconditional cash transfer, financed through
new money printing and/or by using idle balance, seems to be appropriate, with
additional focus on petty nonfarm producers (-cum-consumers). Contrary to
a cash transfer to the industrial workers who are solely consumers and to the
large entrepreneurs who are mainly producers, cash injection towards the petty
farm and nonfarm (self-employed) agents will boost up both consumption and
production. This is because of the uniqueness of petty commodity production,
where an individual (informal) producer (with mostly household-based produc-
tion unit) is simultaneously a decision-maker in respect of both production and
consumption. More concretely, due to the meagre volume of production, a size-
able part of the surplus, if any, will be consumed; in fact, a very large part of the
petty production sector participates in production (and thereby generates so-
cial surplus) mainly for survival and/or betterment of standard of living (Sanyal,
2007). They mostly use indigenously produced inputs and tools for production
and consumption goods (Chakrabarti, 2016). So, the production and consump-
tion demand of one informal unit helps running the other informal businesses.
Thus, the very urge for survival itself can put production and supplies back on the
rails. Consequently, any injection of purchasing power into the informal sector
boosts both consumption and production and further strengthens the backward
and forward production linkages within the nonfarm informal sector itself at
the village/town level (or micro-level) and thereby induces the macro-level
intra-linkages too.
This simultaneous push for demand as well as supply through cash transfer to the
poor producers (even through new money creation) is, thus, neither inflationary
nor does it affect adversely the growth potentials of the economy – driven mostly
by the big corporations (Chakrabarti & Datta, 2019). A universal cash transfer to
the farm and nonfarm petty self-employed producers could induce a “balanced
growth” process by not only bridging the demand-supply gaps but also boosting
expectations inducing a self-fulfilling process. Thus, if demand is induced and the
supply line smoothened, the economy moves on a dynamic equilibrium path.
In this context, we can use the Leontief input-output coefficient between
and within the formal-informal sectors, including agricultural and allied activ-
ities for the year 2003–043 (Kundu, 2019). Table 17.8 shows the intra-sectoral
production linkages, both forward and backward, among the informal sectors,
especially agro-based manufacturing and services, as well as the linkages with
agriculture which is much stronger compared to that with the formal sector. As
agricultural growth in the present scenario is robust, a strategic intervention is
required to boost demand and supply from within the informal sector infusing
financial resources. Lack of savings as observed earlier makes this sector more
vulnerable and hence, unconditional cash transfer is one of the best measures to
revive the informal economy.
17.3.3 Long run: building new institution

However, in the long run, we need to think of implementing something more
constructive, rather than (only) a supportive policy/scheme, for these informal
sector workers, that can help them fight against their odds and smoothen the
functioning of their enterprises. Since the petty nonfarm population faces a
threat of disintegration, its comprehensive rehabilitation should be a prime pol-
icy target. This is a ripe moment for such an upheaval, especially when, there
would be resistance to migration and the erstwhile migrants would try to acquire
an economic space within their native regions. The situation calls for product-
specific cluster formations (Mukherjee, Saha, Datta, & Chakrabarti, 2019) involv-
ing millions of rural-urban petty informal units. This could not only strengthen
their backward and forward production linkages but also create dynamic produc-
tion processes through a mechanism of realizing synergic potentials by sharing
tangible-intangible resources and tapping large markets, which, in turn, would
foster competitiveness (Cucculelli & Storai, 2018). Finally, this major rehabilita-
tion programme should not only boost the micro-units but also generate a robust
macroeconomic multiplier process through a major infrastructure development
providing support to the clusters.
Keeping these aspects in mind, we draw the following milestones for the suc-
cessful development of the micro informal firm-based clusters in India.
TABLE 17.8 L eontief Input-Output Coefficient 2003–04
Agriculture Agro- Non-agro Infrastructure OthrSrvice-F Agro- Non-agro Infrastructure OthrSrvice-I

based- based- Service-F based- based- Service-I

Manf-F Manf-F Manf-I Manf-I
Agriculture 0.52 0.27 0.01 0.01 0.01 0.34 0.03 0.03 0.23
Agro basedManf-F 0.02 0.14 0.01 0.01 0.02 0.05 0.01 0.01 0.09
Non-agrobased-Manf-F 0.13 0.05 0.45 0.23 0.09 0.17 0.52 0.50 0.11
Infrastructure-F 0.06 0.08 0.15 0.22 0.13 0.03 0.01 0.01 0.01
OtherSrvc-F 0.04 0.11 0.12 0.14 0.31 0.04 0.04 0.01 0.04
Agro based-Manf-I 0.01 0.05 0.00 0.00 0.01 0.15 0.02 0.02 0.06
Non-agro based-Manf-I 0.03 0.00 0.04 0.03 0.03 0.08 0.32 0.29 0.10
Infrastructure Srvc-I 0.10 0.15 0.13 0.22 0.25 0.04 0.03 0.10 0.27
OtherSrvc-I 0.09 0.15 0.10 0.15 0.16 0.09 0.02 0.03 0.09
Source: Computation based on Kundu (2019).

Note: F: Formal Sector, I: Informal Sector, Manf: Manufacturing Sector and Srvc: Service Sector.
17.3.3.1 Mapping the region

At the outset, one needs to demarcate the region/area based on the “socio-territorial
entity which is characterized by the active presence of both a community of people and
a population of firms in one naturally and historically bounded area” (Becattini,
1990 emphasis added, p.38). International experience suggests that agglomeration
economies along with non-economic factors helped small-scale enterprises in
Italy to enhance their competitiveness. However in India, to form these clus-
ters, one needs careful mapping of regions/districts not only based on certain
economic indicators but also other intangible cultural aspects (UNESCO, 2011)
embedded in local knowledge and traditional skills. Further, local-level resource
availability and tastes and preferences could also be crucial determining factors;
otherwise, reaping the benefit from regional/micro-level production interlink-
ages may not be successful. This is pertinent since a significantly large number of
informal manufacturing units reported that shrinkage/fall of demand is the most
severe problem faced during 2015–16 followed by the problems related to credit
and shortage of raw materials (see Table 17.9).4 This calls for an urgent need of
mapping the regions based on the regional tastes and preferences along with the
availability of inputs.
However, similarly, by mapping local-level resource availability and tastes,
preferences and demand for the informal services too, we can think of forming
TABLE 17.9 Percentage of Manufacturing Enterprises Facing Severe Problems during

2015–16
Nature of Problem Informal Rural Informal Urban Informal Informal

Self-empl. Self-empl. Rural est. Urban est.
Erratic Power Supply/Power 9.1 9.4 20.2 19.7

Cut
Shortage of Raw Materials 9.5 9.3 6.9 3
Shrinkage/Fall of Demand 37.5 42.7 24.3 29
Non-Availability/High Cost 11.5 11.5 12.5 8.8
of Credit
Non-recovery of Financial 9.8 8 12 11.6
Dues
Non-availability of Labour 0.3 1.5 7.3 7.4
as and When Needed
Non-availability of Skilled 0.4 1.1 7.2 11.2
Labour as and When
Needed
Labour Disputes and Related 0 0 0.6 0.5
Problems
Others 21.9 16.5 8.8 8.9
Source: Computation is based on NSSO 73rd Round Unit records data (2015–16).
Note: It is reported that 43% (33.9%) of rural (urban) OAEs and 55.4% (45%) of the rural (urban)
Establishments faced any problems during 2015–16.
a virtual cluster using an online/telephonic app-based platform for the local in-
formal service providers/entrepreneurs. It helps to strengthen social capital and
increase local market networks among the service providers.
17.3.3.2 Tapping indigenous skill through government support

Primary survey-based research in India reveals that informal micro-units within
a cluster withstand fierce competition by pursuing various innovative methods
(Kundu & Chakrabarti, 2008) that help to develop indigenous (informal) skill
among the workers. However, at present, there is almost no comprehensive in-
stitutional mechanism that helps and could foster this indigenous skill formation.
Hence, there is a need to tap and institutionalize these indigenous skills through
active government support. Indigenous skill formation along with fostering so-
cial capital may be helpful to reap the benefits out of agglomeration economies.
It could turn the clusters into endogenous growth poles.
17.3.3.3 Government support to provide basic infrastructure

There is a need for direct government support that ensures availability of cheap
raw materials, power, water supply and other necessary inputs, including the
seamless flow of credit that would help to reap the benefits (economies of scale)
out of agglomeration of micro informal firms within a cluster. It is notewor-
thy from Table 17.9 that non-availability/high cost of credit is a major problem
that the informal non-agricultural manufacturing units face. Not only physical
infrastructure but also social infrastructure, which is essential for the stability
and growth of the clusters. Adequate support of social infrastructure like artisan
training institutes, health care centres could be critical.
At the present juncture, this could also accommodate a large number of
construction workers, who have been affected severely due to Covid-19 and
lockdown.
17.3.3.4 Government support to market access

Apart from various supply side supports, the government should help the infor-
mal firms within the clusters to have access to output markets so that small firms
could reap the entire surplus without sharing it with various intermediaries who
have better information and access to the global/national market.
17.3.3.5 Strengthening data bank through government’s

direct intervention
Instead of providing only financial support to industry associations/NGOs to set
up the subcontracting and open market exchanges, there is a scope of govern-
ment’s active intervention to strengthen the data bank on the capacities of the
micro-units in terms of products (manufacturing/services) rendered; availability

of idle capacities on particular processes/machines; and buyers’ requirements in
terms of product/components/sub-assemblies/services. This is important since
most of the informal enterprises lack sufficient knowledge. Hence, without the
government’s active involvement, there will always be a possibility of a lack of
detailed documentation of these small but extremely important features.
Data bank can be improved by strengthening the collection of information
through the Census and Sample Survey. Although NSSO conducts follow-up
enterprise surveys on the informal sector, based on the Economic Census (EC)
conducted every five years, nevertheless, sample surveys alone are not sufficient
to provide information about tiny enterprises.
A comprehensive survey needs to be designed that not only captures the en-
terprise aspects of these micro- and small-units but also captures their socio-
economic households-centric aspects. This is important as more than 80% of the
informal enterprises are household-based.
17.3.3.6 Land user right rather than land transfer right to informal units
So far as informal enterprises are concerned, a significantly large portion con-
sists of trading (almost 38% during 2010–11) and other services (35% during
2010–11). Among all these units, on average, 86% are own account firms. Gov-
ernment intervention through the provisioning of land use right for the work-
space and provisioning of other complementary resource use rights (like the
right to use common infrastructure, water, energy, etc.) could facilitate these
service units to deliver their services from one platform. Land use right could
help these own account service enterprises to form a cluster-type group under
one roof of “Service Park” which will function similar to the conventional
Industrial Parks.
Similar user right could also be extended to petty manufacturing units (en-
gaged in metal-based manufacturing, especially repairing, petty food processing,
handloom and handicrafts production, tailoring, pottery, carpentry, etc.). How-
ever, land use right should by no means be extended to land transfer right. Oth-
erwise, the transfer of land from micro-units to the large agents may lead to
the use of this land for the unproductive purpose and the mere proliferation of
rent-seeking activities.
This issue is extremely important especially in the present situation when
there is a deepening of the vulnerability of the micro-enterprises and hence, a
probable widespread crisis of pessimism, anxiety and maybe mistrust. This pro-
visioning of user rights for land (and other complementary resources like water,
forest, infrastructure, energy, to name a few) could be a significant “big push”
particularly for the rural-urban informal manufacturing and repairing and in-
numerable urban service sector units. This could instil much-needed confidence
among the millions of petty entrepreneurs and would significantly reduce the
transaction costs of these units (vis-à-vis the broader economy).
17.3.3.7 Government support to develop a brand value

The sustainability of a cluster is crucially dependent on its branding and the cor-
responding brand value that is acquired over time. The existence and progress of a
cluster are conditional upon its acceptance in the broader market and for that, the
government has an important role especially during the initial years of the cluster.
17.3.3.8 Linking of clusters with the broader value chain

Clusters could get additional impetus through their linkages with the so-called
modern economy. In this respect, the government could play a very important
role (as a facilitator of contracts). Thus, providing the clusters with a variety
of support (physical and social infrastructure, technology, user rights, brand-
ing, information, etc.) and acting as a facilitator for linking them with the long
value chain is the crux of the policy bundle that is essential for a cluster-based
development model ensuring a job-oriented, inclusive, endogenous and hence
sustainable growth process.
However, at the present moment, as the vulnerability of the informal sector
has increased manifold, there is a higher possibility that these units would be
exploited severely by the larger (contracting) firms. Hence, government moni-
toring should be stricter.
17.4 Epilogue: this is not our “war”

In this study, we show the extent of vulnerability of the informal sector, in terms
of their abysmally low level of income and lack of savings. We argue that the
government must undertake a massive institutional effort and inject an enormous
amount of money to reinvigorate the economy from the bottom, as with monu-
mental “entitlement failure”, market miserably fails to generate a truly inclusive
development process. Contrarily, waiting for market corrections could lead to
economic calamity as well as to political Armageddon.
More fundamentally, although the fight against Covid-19 has been repeatedly
projected as a unifying and unidirectional war against a natural calamity and
a crippling burden is thrust upon the poor through an impromptu imposition
of nation-wide lockdown5 (like the collateral damage of a so-called nationalist
war), the truth is that it is the greed of global neo-liberal capital, jeopardiz-
ing the balance inherent within nature and threatening significantly the very
notion of public health, coupled with the mismanagement of the governments
with a distinct bias towards the (high flying) rich, which have been instrumen-
tal in inviting the disease, creating the pandemic and its associated insecurity,
fear, all-round suspicion and deep pessimism (Sengupta & Choudhury, 2020;
Wallace, Liebman, Chaves, & Wallace 2020). This is sure to leave in its trail
long-run debilitating effects on the socio-economic and political lives of the
millions of working people, in general, and legions of the impoverished and the
marginalized, in particular. The onus cannot be put on the suffering multitudes

for something for which they are never responsible. The state has to bear the re-
sponsibility and, if anybody has to pay a price for this appalling human tragedy,
it is the global capital! Life, livelihood and dignity of labour have to be saved, not
merely because it is a basic responsibility of the state, but most importantly, it is a
moral and natural right of labour. The state can’t hide behind the rhetoric of war,
social distancing, self-reliance, austerity, fiscal crisis, etc.
These days, the Ramayana is being propagated as the ideal for India. At the
beginning of the Epic, when Bandit Ratnaakar is urged to go back to his family
members and ask whether they would share the burden of his sin, the answer
was a resounding, No! Time has come, when labour (from doctor to day la-
bourer), who run this globe (and this lockdown has proved it better than any
theory), should pronounce: Be it war or not, we are not responsible for the “im-
ported” pandemonium thrust upon us and therefore, we must be compensated
for unconditionally!
Maybe the lockdown is necessary because every single life is infinitely pre-
cious and involuntary experiment with human life violates the inalienable
right to live. But just as lockdown sans tracing, testing, secluding and treating
is almost a futile exercise, without a universal, comprehensive and short- and
long-run humanitarian and economic support, lockdown becomes dangerously
counterproductive.
Notes
1 We are grateful to the editors for giving us this opportunity to contribute. We would
like to express our deepest gratitude to Professor Soumyen Sikdar for his invaluable
encouragement. We thank Krishnakanta Das of Gokhale Institute of Politics and
Economics, for his sincere research support.
Kasturi Sadhu would like to thank Professor B.V. Phani for his valuable cooperation
and would also like to gratefully acknowledge the crucial support of Syndicate Bank
Entrepreneurship Research and Training Centre at IIT Kanpur.
2 A similar argument has been proposed by Kesar et al. (2020).
3 Although the coefficient matrix is constructed for the year 2003–04, this is to show
the intra- and inter-industry linkages in support of our argument. No other study is
found for the years 2010–11 or 2014–15.
4 We get a similar result for 2010–11 data.
5 https://in.video.search.yahoo.com/search/video?fr=mcafee&ei=UTF-8&p=
ndtv+sen+covid+war&t ype=E211IN714G0#id=1&vid=f4ac4f 045789c63c1d
0225b9558e0419&action=click.
References
Abraham, A., & Khan, M. I. (2020, July 20). Covid-19: How Long Can Consumption be
Sustained in Lockdown? Ideas for India. Retrieved August 3, 2020, from https://www.
ideasforindia.in/topics/poverty-inequality/covid-19-how-long-can-consumption-
be-sustained-in-lockdown.html?f bclid=IwAR3RjriHaqeiN6yf XvtpFw7Vhdr8OS
9D9TORgjHAZ0atXmDca7HSdfxF8yI
Baldwin, R. (2020, March 13). Keeping the Lights On: Economic Medicine for a Med-
ical Shock, VoxEU CEPR. Retrieved April 28, 2020 from https://voxeu.org/article/
how-should-we-think-about-containing-covid-19-economic-crisis
Becattini, G. (1990). The Marshallian Industrial District as a Socio-Economic Notion.
In F. Pyke, G. Becattini, & W. Sengenberger (Eds.), Industrial Districts and Inter-firm
Co-operation in Italy (pp. 37–51). International Institute for Labour Studies, Geneva.
Chakrabarti, S. (2010). A Macroeconomic Structure of Employment: Rural-Urban Con-
flict in a Kaleckian Framework, Review of Radical Political Economics, 43(2), 172–197.
Chakrabarti, S. (2016). Inclusive Growth and Social Change: Formal-Informal-Agrarian Rela-
tions in India, Oxford University Press, New Delhi, India.
Chakrabarti, S., & Datta, P. K. (2019). Problems of Financing Development Manage-
ment: A Kaleckian Analysis, Review of Radical Political Economics, 51(3), 457–476.
Chakrabarti, S., Datta, P. K., Kundu, A., & Sadhu, K. (2020, May 25). India and the Covid
19 Crisis: Radical Measures Needed, Globaldev. Retrieved June 4, 2020 from https://
www.globaldev.blog/blog/india-and-covid-19-crisis-radical-measures-needed
Cucculelli, M., & Storai, D. (2018). Industrial Districts, District Effect and Firm Size:
The Italian Evidence, Cambridge Journal of Economics, 42(6), 1543–1566.
Dasgupta, R., & Rajeev, M. (2020). Paradox of a Supply Constrained Keynesian Equilib-
rium: The COVID-19 Case, Economic and Political Weekly, 55(22), 22–27.
Dreze, J. (2020, April 14). Migrant Workers Badly Treated, More Needs to Be Done to
Help Them Now: Jean Dreze on lockdown, India Today. Retrieved April 22, 2020
from https://www.indiatoday.in/india/video/migrant-workers-badly-treated-more-
needs-to-be-done-to-help-them-now-jean-dreze-on-migrant-crisis-1667019-2020-
04-14?f bclid=IwAR1CcltB2ZvB47uVUq1HxfQRgLvyyVaXp1hS2I1seVE6j04gkg
vMkqQjunQ
Estupinan, X., Sharma, M., Gupta, S., & Birla, B. (2020). Impact of COVID-19 Pan-
demic on Labour Supply and Gross Value Added in India, IGIDR Working Paper No.
022. Retrieved July 20, 2020 from http://www.igidr.ac.in/pdf/publication/WP-
2020-022.pdf
IMF. (2020, April 14). World Economic Outlook: The Great Lockdown. Retrieved April
30, 2020 from https://www.imf.org/en/Publications/WEO/Issues/2020/04/14/
World-Economic-Outlook-April-2020-The-Great-Lockdown-49306
Jonas, O. B. (2013). Pandemic Risk, World Development Report, Background Paper. Re-
trieved October 3, 2020 from https://www.worldbank.org/content/dam/Worldbank/
document/HDN/Health/WDR14_bp_Pandemic_Risk_ Jonas.pdf
Kesar, S., Abraham, R., Lahoti, R., Nath, P., & Basole, A., (2020). Pandemic, Informal-
ity and Vulnerability: Impact on COVID-19 on Livelihoods in India, CSE Working
Paper 2020–01, Azim Premji University.
Khan, F., & Mansoor, K. (2020, May 11). COVID-19 Impact: Informal Economy Work-
ers Excluded from Most Govt. Measures, be it Cash transfers or Tax benefits, Firstpost.
Retrieved May 15, 2020 from https://www.firstpost.com/business/covid-19-impact-
informal-economy-workers-excluded-from-most-govt-measures-be-it-cash-
transfers-or-tax-benefits-8354051.html
Kundu, A. (2019). Revisiting Formal-Informal-Agriculture Interlinkages in India: A Computable
General Equilibrium Analysis, Unpublished PhD Dissertation, Centre for Development
Studies, Trivandrum.
Kundu, A., & Chakrabarti, S. (2008). Some Microeconomic Aspects of Informal Sector
Enterprises: Field-based Evidence from West Bengal, Artha Vijnana, 50(3), 209–238.
Mehrotra, S. (2019). Informal Employment Trends in the Indian Economy: Persistent In-
formality, but Growing Positive Development, ILO Working Paper No. 254. Retrieved
May 15, 2020 from https://www.ilo.org/wcmsp5/groups/public/---ed_emp/---ifp_
skills/documents/publication/wcms_734503.pdf
Mehta, B. S. & Kumar, A. (2020). Covid-19 Lockdown Impact: Estimating The Job
Losses in India’s Informal Economy, Swarajyamag, Retrieved April 22, 2020 from
https://swarajyamag.com/economy/covid-19-lockdown-impact-estimating-
the-job-losses-in-indias-informal-economy
Mukherjee, A., Saha, N., Datta, P.K., & Chakrabarti, S. (2019). Informal Sector: Pro-
gression or Persistence? A Study of Four Traditional Clusters of West Bengal, India.
In S. Bandyopadhyay & M. Dutta (Eds.), Opportunities and Challenges in Develop-
ment (pp. 117–148), Springer, Singapore. Retrieved from https://link.springer.com/
chapter/10.1007/978-981-13-9981-7_7
Mukherjee, T., Ray, N., & Bag, S. (2020, October 20). Opinion: Impact of Covid-19 on
the Indian Economy, ETGovernment.com. Retrieved October 25, 2020 from https://
government.economictimes.indiatimes.com/news/economy/opinion-impact-
of-covid-19-on-the-indian-economy/75021731
Raghunath, A. (2020, March 27). Coronavirus: Kerala Govt Starts Door-delivery of Free
Food, Deccan Herald. Retrieved April 28, 2020 from https://www.deccanherald.com/
national/south/coronavirus-kerala-govt-starts-door-delivery-of-free-food-818325.
html
Sanyal, K. (2007). Rethinking Capitalist Development: Primitive Accumulation, Governmental-
ity and Post Colonial Capitalism, Routledge, New Delhi.
Sengupta, R., & Choudhuri, C. (2020). COVID-19 – Neo-Malthusian Event? Ecological
Links, Challenges and Opportunities, International Journal of Ecology and Environmental
Sciences, 46(2), 141–154.
UNESCO. (2011, November). 2003 Convention for the Safeguarding of the Intangible Cul-
tural Heritage, Sixth session of the Intergovernmental Committee for the Safeguarding
of the Intangible Cultural Heritage, Bali, Indonesia. Retrieved September 26, 2020
from https://ich.unesco.org/doc/src/15164-EN.pdf
Wallace, R., Liebman, A., Chaves, L. F., & Wallace, R. (2020). COVID-19 and
the Circuits of Capital, Monthly Review. Retrieved April 27, 2020 from https://
monthlyreview.org/2020/04/01/covid-19-and-circuits-of-capital/
APPENDIX
Decrease in
Reduced household Low
demand for income savings
imports
Reduced
domesc
demand Reducon Supply-
in side
investment chain
Labour becomes more
Demand
vulnerable than capital (lay
-side Reducon
off, forced reducon in
chain in the
working hours)
world's Reducon
income in capital
stock
Reducon in the use of factors

of producon
Reducon
in
country’s
export
Fall in output
FIGURE 17.1 Impact on income.

18
COVID-19 PANDEMIC
Implications for widowed and elderly
women in India
Sutirtha Bandyopadhyay and Bipasha Maity
18.1 Introduction
For large parts of the developing world, the female to male ratio has remained
much lower than the biologically determined natural level over the years. This is
a matter of concern and has been known as the “missing women” phenomenon
in the literature (Sen, 1989; Sen, 1990; Coale, 1991; Klasen, 1994; Klasen &
Wink, 2003; Anderson & Ray, 2010; Anderson & Ray, 2019). Focusing on es-
pecially India, Sen finds worsening of sex ratio (against females) right from the
beginning of the twentieth century and sharp regional differences within the
country (Sen, 1982; Kynch & Sen, 1983; Dreze & Sen, 1995). The recent data re-
veals the number of males per 100 females to be abnormally high for some Asian
countries. These figures are 108 for India and Bhutan, 113 for Maldives and 153
for Bahrain (The World Factbook, 2020).
Although the computation methodology and measures of missing women
vary across studies, a predominant reason of low female to male ratio is plausibly
on account of women’s relative deprivation. The differences in the female to
male ratio in the total population are mainly due to dissimilarities in death rates
and the forces that govern higher female mortality. Ultimately, they potentially
reflect differential provision of nutrition, medical attention and care. There are
numerous studies for India and other developing countries which firmly high-
light discrimination against girls and women and the consequent impact on ex-
cess female mortality and morbidity (see, for example, Bardhan, 1974; D’Souza
and Chen, 1980; Chen, Huq & D’Souza, 1981; Das, 1987; Gupta, 1987; Pande,
2003; Jayachandran & Kuziemko, 2011; Jayachandran, 2015).
The existing literature finds that women face discrimination in several di-
mensions irrespective of their age group. However, the disadvantages are likely
to be higher for the elderly and widowed women. A large number of studies
DOI: 10.4324/9781003220145-21
310 Sutirtha Bandyopadhyay and Bipasha Maity
focusing mostly on developing countries show that the elderly and especially
widowed women are more vulnerable than their younger and married counter-
parts (Rahman, Foster & Menken, 1992; Chen & Drèze, 1995; Dreze & Srini-
vasan, 1997; Chen, 2000; Jensen, 2005; Van De Walle, 2013; Lambert & Rossi,
2016; Bandyopadhyay & Maity, 2020; Calvi, 2020). The vulnerability due to
widowhood is also closely linked to the “missing women” phenomenon. Ander-
son and Ray (2019) find that around 35% of missing women of adult age can be
attributed to not being married. In developing countries like India, marriage at
young ages is largely universal and divorce is rare. Therefore, unmarried adult
women are typically widows. Anderson and Ray (2019) find that India accounts
for the largest proportion of missing adult women on account of not having a
husband. This naturally puts the focus on India for examining vulnerability due
to widowhood. According to the Census of India 2011, there are at least 55 mil-
lion widowed women in India and among individuals aged 60 years and above,
there are almost 3.4 times as many widowed women as there are widowed men.
Now, from the onset of the Covid-19 pandemic, researchers have emphasized
the negative impact of the pandemic and its induced lockdowns on different
vulnerable sections of the population. For example, short-term migrants and
people working in the informal sector are vulnerable due to the negative income
shock on account of cessation of economic activities (Kapoor, 2020; Abraham,
Basole & Kesar, 2021). However, the elderly are at risk due to their low immu-
nity level and higher prevalence of comorbidities on account of their advanced
age. Researchers have also pointed out that the pandemic could have different
implications by gender. For example, Alon et al. (2020) mention that unlike
usual recessions, social distancing measures due to the current pandemic could
adversely affect sectors with large female share in employment in the context
of the USA. Deshpande (2020) finds that although overall men’s employment
declined in India due to Covid-19-induced lockdowns as the female share of
employment was low to start with, women were less likely to be employed if
they were also working pre-lockdown. There is a concern of increased burden
of childcare work on working women due to school and daycare closures in
the short run, with the possibility of emerging equitable allocation of childcare
responsibilities between parents in some richer countries (Alon et al., 2020; Del
Boca et al., 2020; Sevilla & Smith, 2020). However, using the US Current Pop-
ulation Survey, a very recent study by Collins et al. (2020) shows that mothers
with young children have reduced their work hours four to five times more than
fathers. Consequently, the gender gap in work hours has grown by 20–50%.
Further, Ravindran and Shah (2020) find significant increase in domestic vi-
olence-related complaints in districts with the strictest lockdown measures in
India. The shift in resources towards addressing the public health emergency can
entail disruptions to key health services for women and girls, such as reproduc-
tive and sexual health services (de Paz et al., 2020).
Although the current emerging literature on Covid-19 has emphasized
how younger, currently married women and working women are likely to be
Covid-19 pandemic 311
disadvantaged on account of the pandemic, the implications for the elderly and
especially widowed women have remained largely understudied. As the afore-
mentioned literature on widowhood has demonstrated that widowhood is an
additional source of vulnerability for women in especially developing countries,
examining the implications for widowed women in the context of the Covid-19
pandemic is of paramount importance. There are also additional reasons why
studying the implications for widowed women is important. First, a new study
by Joe et al. (2020) using crowdsourced data finds that although more men are
found to be infected than women by Covid-19, conditional on being infected,
women are more likely to die from it than men in India. Although it is possible
that women are less likely to have access to testing than men (and therefore lesser
women are counted as being infected with Covid-19 than men), the unequal
access to nutrition and healthcare for women in general in India are already
likely to place women at a significant disadvantage in terms of contracting and
recovering from the disease. These disadvantages in terms of access to nutrition
and potentially poorer pre-existing health condition are likely to be higher for
widowed women in India, given the overall deprivation that widows already
face in Indian society in terms of their diet and health ( Jensen, 2005). Therefore,
conditional on contracting Covid-19, there is a significant possibility that wid-
owed women could be at a higher risk of mortality than other women in general.
Second, in contrast to the study by Joe et al. (2020), most of the existing studies
have documented a higher risk of male mortality (de Paz et al., 2020; Sharma
et al., 2020). Therefore, this implies an increased risk of widowhood for women.
Anderson and Ray (2019) have found that India accounts for the largest share of
excess female mortality/“missing women” on account of women not having a
husband. Therefore, women who are likely to survive now but become widows
due to spousal death from Covid-19 could face an increased risk of mortality
due to widowhood-induced vulnerability in the long run. Negative shocks to
overall household income due to the pandemic-induced lockdown can further
exacerbate the already unequal and low access to resources within the household
for widowed women (both newly widowed due to Covid-19 and existing wid-
owed women). Therefore, apart from its direct impact, Covid-19 can indirectly
increase the pool of “missing women” through the channel of deprivation related
to widowhood.
In this chapter, we cannot directly examine the impact of the pandemic on
widow mortality rate or discrimination faced by widowed women as current
data are not available to conduct such analyses. Instead, we examine the dep-
rivation of widows in terms of consumption, asset ownership and analyse their
poor health status from Indian data collected before the pandemic. Based on our
findings, we qualitatively assess and discuss the vulnerability due to widowhood
triggered by the Covid-19 pandemic.
A preview of our analysis and main findings are as follows. First, we analyse
the deprivation of widowed women in terms of consumption and thereafter in
terms of asset ownership using nationally representative survey datasets. We then
show that widowed women are more likely to report having chronic ailments, on
average, relative to their married counterparts; however, no such differences can
be found between widowed and currently married men. Therefore, these depri-
vations in consumption resources and control over assets are likely to contribute
to poor health status of widowed women (Bandyopadhyay & Maity, 2020; Calvi,
2020). The deprivation and consequent poor health status can potentially lead to
higher mortality among widowed women in general and the Covid-19 pandemic
might increase this disadvantage. Hence, a poverty alleviation strategy that aims
to boost overall household income/consumption may not be enough and that
widowed women are likely to need additional support from the government
during and after the pandemic.
18.2 Widowhood and deprivation in terms of consumption

At first, we document the deprivation that widowed women face in terms of ac-
cess to consumption within the household. Calvi (2020) has found that women’s
resource share in terms of consumption within households declines with age in
India and has relatedly found that women’s resource share is lower in households
where widows are present. Further, lower resource share and associated higher
poverty rate among older women are shown to be associated with poor health
outcomes (Calvi, 2020). However, the central focus of this research was not on
the implications of widowhood on women’s well-being, but to provide for a
potential explanation of excess female mortality/“missing women” among older
women in India.
In a recent paper, Bandyopadhyay and Maity (2020) study how the presence of
at least one widowed woman affects the consumption of female and male assigna-
ble goods in a household using multiple rounds of large, nationally representative
consumption expenditure surveys from the National Sample Surveys (NSS) of
India. In particular, the monthly spending on women’s (men’s) clothing per adult
female (male) is considered as indicative of consumption of female (male) assign-
able goods in a household. The paper finds that the average monthly expenditure
on female assignable goods is lower in households that have at least one widowed
woman relative to households that have no widows in every round of the NSS.
In particular, the monthly expenditure on women’s clothing per adult female is
found to be Rs. 48 in households that have a widow, while the corresponding
figure is Rs. 64 for households with no widowed women in the 2011–12 round
of the NSS. Importantly, the study finds that average monthly expenditure on
men’s clothing per adult male is slightly higher in households that have at least
one widowed woman relative to households that have no widows (although the
difference is very small). In the spirit of Deaton (1989), this indicates poten-
tially higher discrimination against women in households that have at least one
widow. Moreover, when one restricts the sample to include households with
elderly women (that is, women aged 60 years or older), it is found that house-
holds without any widowed women have around 22% higher average monthly
spending on women’s clothing per adult female than households that have at least
one widow. This finding indicates that widowhood continues to be an important
channel of vulnerability for women that is distinct from the effect of older age
(Bandyopadhyay & Maity, 2020). These findings continue to hold in a regression
estimation with a rich set of controls (see Bandyopadhyay & Maity, 2020 for the
detailed analysis).
Bandyopadhyay and Maity (2020) find that the average difference in spending
on female assignable goods between households that have a widow and those
without any widowed women is higher for urban households than for their rural
counterparts. Our computation from the NSS data shows that the average spend-
ing on female assignable clothing per adult woman is 28% higher in households
without a widowed woman relative to a household with a widowed woman in
rural India. The difference turns out to be much higher in urban areas where per
adult woman spending on female assignable clothing is 46% higher on an aver-
age in a household without a widowed woman compared to a household with a
widowed woman. This finding related to urban-rural disparity in consumption
is immensely important from the perspective of the Covid-19 pandemic as the
disease’s worst effects are closely linked with urban areas, where death rates tend
to be higher because of a complex combination of factors, including population
density as well as national and international connectivities (see, for example,
Biswas, 2020; Lee et al., 2020; Parth, 2020; Auerbach & Thachil, 2021 in the
context of urban slums). The pandemic has had a huge impact on urban poverty,
unemployment and standard of living. The condition of widowed women in
urban areas may, therefore, worsen even further due to the pandemic if they are
already found to be more discriminated against relative to their rural counter-
parts in the pre-pandemic scenario.
There are many important institutions, behaviours and practices surround-
ing widowhood that adversely affect the well-being of widows. These are well
documented in the literature (Chen & Drèze, 1992; Chen & Drèze, 1995; Chen,
2000; Jensen, 2005; Czerenda, 2010). Societal constraints and discrimination
against widows exist in various forms, including restrictions on diet and cloth-
ing, i.e. what widowed women are mandated to eat and wear (Chen and Drèze,
1995; Jensen, 2005). But, in this chapter, we just focus on the quantitative meas-
ure of consumption discrimination and data on female/male assignable clothing
helps us to capture this partially. Although we have used private assignable goods
(like female and male assignable clothing) in order to measure the discrimination
faced by the elderly and widowed women relative to their married and younger
counterparts in terms of consumption, it has its wider application in measuring
resource share disparity within a household. In recent research done on col-
lective household models, resource shares of individuals within a household
have been computed by estimating the Engel curve of private assignable goods
(Dunbar, Lewbel & Pendakur, 2013; Lechene, Pendakur & Wolf, 2019; Calvi,
2020). Finally, the differential treatment against widowed and elderly women
in consumption is not only limited to consumption of private assignable goods.
Bandyopadhyay and Maity (2020) examine whether currently married and wid-
owed women differ in terms of frequency of food consumption across a number
of items using the Demographic and Health Surveys (DHS) 2015–16 and find
that currently married women are indeed more likely to frequently (that is, daily
or weekly) consume various food items.
Since a number of studies mentioned above have clearly documented the dis-
advantage of widowed women in terms of consumption within the household
even during the pre-pandemic time period, decrease in household income on
account of the pandemic-induced lockdowns could make households allocate
even less of the now-limited consumption resources to the elderly and widowed
women.
18.3 Widowhood and deprivation in terms of asset ownership

We examine whether widowhood for women is also associated with decline in
land and livestock ownership, which are important assets in developing econ-
omies like India with a large agricultural sector. For this purpose, we consider
data from the India Human Development Survey (IHDS) as it is a panel dataset
and thereby enables us to compare how household asset ownership changes for
women who transition from being married to widowhood vis-à-vis those who
remain married across the two rounds of the survey. For our comparison, we
only consider women who were married during the first round of the survey
(2005) and who have either remained married or have become widows in the
second round of the survey (2011–12). Further, we only consider those house-
holds that have not split across the survey rounds for ease of comparison. We only
consider the sample from rural India here. Our sample consists of close to 17,000
women and about 10% of the sample has become widowed by 2011–12.
We find that ownership/cultivation of agricultural land as well as livestock has
declined for households where women transitioned from being married to wid-
owhood between 2005 and 2011–12, in contrast to households where women re-
mained married between the two rounds. Panels A and C in Table 18.1 show that
land and livestock ownership declined by 1 and 6 percentage points, respectively,
for households where married women became widows by 2011–12. In contrast,
households where women continued to be married saw an increase in land owner-
ship/cultivation and livestock ownership by 4 and 2 percentage points, respectively.
Panel B shows that in 2005, amount of land holding (in acres) is higher for
households where women have transitioned to widowhood relative to their
counterparts who have remained married in both rounds. However, we are in-
terested in the change in asset ownership status between the rounds as opposed
to the amount of the asset owned in any round. Therefore, when we consider
the change in the average land ownership between the rounds, we find that,
households where women have been widowed witnessed nearly 16% decline in
the amount of land holding, while households where women stayed married
saw only a 2% decline in average amount of land holding. Similarly, we find an
TABLE 18.1 L and and livestock ownership status of households of women who were
married in 2005 and who are widowed or remained married in 2012
Panel A: Proportion of Households Who Own or Cultivate Agricultural Land
2005 2011–12
Households where women transitioned from being 64% 63%
married in 2005 to being widowed in 2011–12
Observations 1,694 1,694
Households where women continued to remain married 63% 67%
in 2005 and 2011–12
Panel B: Average Land Holding by Households (Acres)
2005 2011–12
Households where women transitioned from being 4.25 3.56
Households where women continued to remain married 3.46 3.40
in 2005 and 2011–12
Panel C: Proportion of Households who Own Livestock
2005 2011–12
Households where women transitioned from being 61% 55%
Households where women continued to remain married 62% 64%
in 2005 and 2011–12
Panel D: Average Count of the Number of Animals Owned
2005 2011–12
Households where women transitioned from being 4.87 4.35
Households where women continued to remain married 4.79 4.89
in 2005 and 2011–12
Note: Authors’ calculations based on the IHDS data (2005, 2012). Households where women tran-
sitioned from being married in 2005 to being widowed in 2011–12 imply households where at least
one woman transitioned from being married to being widowed between the two periods of the
survey. We limit the sample to households with less than or equal to 100 acres of land holding in
both rounds in Panel B and to households owning 50 or less number of animals in both rounds in
Panel D. Land holding is the sum of the amounts of land owned and land leased/rented/mortgaged
in net of land leased/rented/mortgaged out. Livestock animals include milch cows, milch buffaloes,
draft animals, goat, sheep, poultry and other animals.
average decline of around 11% in the number of animals owned by households

where women became widows by 2011–12, whereas households where women
stayed married saw an increase in the average number of animals owned by
nearly 2% between the rounds (Panel D of Table 18.1). Therefore, these findings
appear to be similar to the findings in the existing literature where household

asset ownership is found to decline for women following their husband’s death
(Hurd & Wise, 1989; Chapoto et al., 2011).
Hence, if men are more likely to die from Covid-19 ( Joe et al., 2020), then the
potential increased risk of widowhood could exacerbate loss of assets for women.
The possible decline in asset ownership could, in turn, reduce widowed women’s
ability to seek and afford healthcare services on becoming ill. This could then
disproportionately raise morbidity and mortality among widowed women from
a number of diseases, including Covid-19.
18.4 Widowhood and deprivation in terms of health

We consider the NSS 71st round of Social Consumption: Health Survey (2014)
to understand the prevalence of chronic ailments among individuals aged 60
years and older. There are 6,627 males and 3,011 females who were currently
married at the time of the survey. 1,011 men and 3,226 women were widowed
at the time of the survey.
Table 18.2 shows that the percentage of men aged 60 or above who report hav-
ing chronic ailments is almost similar to those who were married and widowed at
the time of the survey. However, we see that widowed women reporting having
chronic ailments are higher than that of married women by nearly 3 percent-
age points. Further, widowed women are found to report higher prevalence of
chronic ailments relative to widowed men by nearly 6 percentage points. There-
fore, it appears that prevalence of chronic ailments is higher among widowed
women not only relative to their currently married counterparts, but also relative
to widowed men. Therefore, even among the elderly, widowed women are likely
to face the greatest mortality risk from newly emerging infectious diseases such as
Covid-19 on account of higher prevalence of comorbidities among them.
TABLE 18.2 If suffering from any chronic ailment (self-reported)

among individuals aged 60 years or older
Currently married Widowed
Males:
Yes 28.34% 28.39%
Females:
Yes 30.85% 33.42%
Note: Authors’ calculation. Data source is NSS 71st round Social Consump-
tion: Health Survey. Only those reports are considered which are reported by
the respondents themselves and not their proxy. Sample restricted to include
individuals whose ages do not exceed 100 years.
Further studies by Kapoor et al. (2019), Mondal and Dubey (2020) and Mo-
radhvaj and Saikia (2019) document gender disparities in healthcare access such
as outpatient care, hospitalization and in-patient health expenditures in India.
These studies clearly point out that older women and widowed women are at
a distinct disadvantage in terms of healthcare access and spending. Mondal and
Dubey (2020) also find that households spend more on diagnostic tests for older
adult males relative to their female counterparts and widowhood among women
appears to drive a lot of this disadvantage. Moradhvaj and Saikia (2019) find
that households are more likely to resort to distress financing to meet health
expenditures when it pertains to older adult males than females. Given that em-
ployment and job losses are likely to cause significant declines in overall health
spending (Gupta & Kishore, 2020) and the Covid-19 pandemic has resulted in
significant job losses in India (Abraham, Basole & Kesar, 2021), it is possible that
these can exacerbate gender disparity in healthcare expenditure, including on
Covid-related care, as households would need to decide how to allocate limited
resources available for health spending. In the light of the existing literature,
older and widowed women, therefore, are likely to be greatly disadvantaged in
this situation, given large pre-existing gender disparities in healthcare spending
among older adults.
18.5 Policy implications

An important insight from the above discussion is the multiple sources of dep-
rivation that widowed women face. As has been mentioned, widowhood con-
tributes to an additional source of disadvantage to women in India. Therefore,
studying the potential gendered effects of the current pandemic would be incom-
plete without understanding its implications on a specific subsection of women,
namely the elderly and especially widowed women. Importantly, given the lim-
ited access to consumption resources within the household, poor control of as-
sets, greater reported prevalence of chronic ailments for widowed women and
gender disparity in healthcare expenditures for older adults, poverty alleviation
programmes such as income transfers that seek to increase overall household
income would not be enough. Income transfers that are specifically directed at
widowed women could be a potential avenue through which mitigation of the
negative income and health shocks due to the pandemic may be achieved. Fur-
ther, the above discussion finds that lower access to consumption resources for
widowed women is more pronounced in urban areas. Therefore, it is important
for policymakers to focus on poverty mitigation strategies for widowed women
in urban India, especially in the light of the Covid-19 pandemic.
Acknowledgement
Maity acknowledges financial support from Ashoka Research Grant 2019–21,
Ashoka University. All remaining errors are our own.
References
Abraham, R., Basole, A., & Kesar, S. (2021). Tracking Employment Trajectories during
the Covid-19 Pandemic: Evidence from Indian Panel Data. Centre for Sustainable Em-
ployment Working Paper #35, Azim Premji University, Bangalore.
Alon, T. M., Doepke, M., Olmstead-Rumsey, J., & Tertilt, M. (2020). The impact of
COVID-19 on gender equality (No. w26947). National Bureau of Economic Research.
Anderson, S., & Ray, D. (2010). Missing women: Age and disease. The Review of Economic
Studies, 77(4), 1262–1300.
Anderson, S., & Ray, D. (2019). Missing unmarried women. Journal of the European Eco-
nomic Association, 17(5), 1585–1616.
Auerbach, A. M., & Thachil, T. (2021). How does Covid-19 affect urban slums? Evidence
from settlement leaders in India. World Development, 140, 105304.
Bandyopadhyay, S., & Maity, B. (2020). Widowhood and consumption of female assign-
able goods in India. Unpublished Manuscript, Ashoka University.
Bardhan, P. K. (1974). On life and death questions. Economic and Political Weekly,
1293–1304.
Biswas, S. (2020). Coronavirus: The Race to Stop the Virus Spread in Asia’s ‘Biggest
Slum.’ BBC News, April 6. Retrieved from https://www.bbc.com/news/world-asia-
india-52159986, last accessed on February 25, 2021.
Calvi, R. (2020). Why are older women missing in India? The age profile of bargaining
power and poverty. Journal of Political Economy, 128(7), 2453–2501.
Central Intelligence Agency (Ed.). (2020). The World Factbook 2020. Central Intelligence
Agency.
Chapoto, A., Jayne, T. S., & Mason, N. M. (2011). Widows’ land security in the era of
HIV/AIDS: Panel survey evidence from Zambia. Economic Development and Cultural
Change, 59(3), 511–547.
Chen, L. C., Huq, E., & D’Souza, S. (1981). Sex bias in the family allocation of food and
health care in rural Bangladesh. Population and Development Review, 7(1), 55–70.
Chen, M. A. (2000). Perpetual mourning: Widowhood in rural India. Philadelphia: University
of Pennsylvania Press.
Chen, M., & Drèze, J. (1992). Widows and health in rural north India. Economic and Po-
litical Weekly, WS81–WS92.
Chen, M., & Drèze, J. (1995). Recent research on widows in India: Workshop and con-
ference report. Economic and Political Weekly, 2435–2450.
Coale, A. J. (1991). Excess female mortality and the balance of the sexes in the popula-
tion: An estimate of the number of “missing females”. The Population and Development
Review, 17(3), 517–523.
Collins, C., Landivar, L. C., Ruppanner, L., & Scarborough, W. J. (2020). COVID-19
and the gender gap in work hours. Gender, Work & Organization, 28, 101–112.
Czerenda, A. J. (2010). The meaning of widowhood and health to older middle-class
Hindu widows living in a South Indian community. Journal of Transcultural Nursing,
21(4), 351–360.
Das, N. (1987). Sex preference and fertility behavior: A study of recent Indian data. De-
mography, 24, 517–530.
Deaton, A. (1989). Looking for boy-girl discrimination in household expenditure data.
The World Bank Economic Review, 3(1), 1–15.
Del Boca, D., Oggero, N., Profeta, P., & Rossi, M. (2020). Women’s Work, Housework and
Childcare, before and during COVID-19. IZA Discussion paper Series, Discussion Paper
No. 13409.
de Paz, C., Muller, M., Munoz Boudet, A. M., & Gaddis, I. (2020). Gender Dimensions of
the COVID-19 Pandemic. World Bank Group Policy Note. Washington, DC: The World
Bank.
Desai, S., Vanneman, R., & National Council of Applied Economic Research, New Delhi.
India Human Development Survey Panel (IHDS, IHDS-II), 2005, 2011–2012. Inter-
university Consortium for Political and Social Research [distributor], 2019-11-19.
https://doi.org/10.3886/ICPSR37382.v1
Deshpande, A. (2020). The Covid-19 Pandemic and Lockdown: First Order Effects on Gender
Gaps in Employment and Domestic Time Use in India. GLO Discussion Paper No. 607.
Dreze, J., & Sen, A. (1995). India economic development and social opportunity. Oxford: Clar-
endon Press.
Dreze, J., & Srinivasan, P. V. (1997). Widowhood and poverty in rural India: Some in-
ferences from household survey data. Journal of Development Economics, 54(2), 217–234.
D’Souza, S., & Chen, L. C. (1980). Sex differentials in mortality in rural Bangladesh.
Population and development review, 257–270.
Dunbar, G. R., Lewbel, A., & Pendakur, K. (2013). Children’s resources in collective
households: identification, estimation, and an application to child poverty in Malawi.
American Economic Review, 103(1), 438–471.
Gupta, M. D. (1987). Selective discrimination against female children in rural Punjab,
India. Population and Development Review, 13(2), 77–100.
Gupta, M., & Kishore, A. (2020). Unemployment and household spending in rural and
urban India: Evidence from panel data (2019). IFPRI Discussion Paper 01978. Wash-
ington, DC: International Food Policy Research Institute.
Hurd, M. D., & Wise, D. A. (1989). The wealth and poverty of widows: Assets before
and after the Husband’s Death. In The economics of aging (pp. 177–200), edited by D.A.
Wise. Chicago: University of Chicago Press and NBER.
Jayac handran, S. (2015): The roots of gender inequality in developing countries. Annual
Review of Economics, 7, 63–88.
Jayachandran, S., & Kuziemko, I. (2011). Why do mothers breastfeed girls less than boys?
Evidence and implications for child health in India. The Quarterly journal of economics,
126(3), 1485–1538.
Jensen, R. T. (2005). Caste, culture, and the status and well-being of widows in India.
In Analyses in the economics of aging (pp. 357–376), edited by D.A. Wise. Chicago: Uni-
versity of Chicago Press.
Joe, W., Kumar, A., Rajpal, S., Mishra, U. S., & Subramanian, S. V. (2020). Equal risk,
unequal burden? Gender differentials in COVID-19 mortality in India. Journal of
Global Health Science, 2(1). https://doi.org/10.35500/jghs.2020.2.e17
Kapoor, M., Agrawal, D., Ravi, S., Roy, A., Subramanian, S. V., & Guleria, R. (2019).
Missing Female Patients: an observational analysis of sex ratio among outpatients in
a referral tertiary care public hospital in India. BMJ Open; 9:e026850. https://doi.
org/10.1136/bmjopen-2018-026850
Kapoor, R. (2020). Covid-19 and the State of India’s Labour Market. Indian Council for
Research on International Economic Relations (ICRIER) Policy Series Paper No. 18, New
Delhi.
Klasen, S. (1994). “Missing women” reconsidered. World Development, 22(7), 1061–1071.
Klasen, S., & Wink, C. (2003). “Missing women”: Revisiting the debate. Feminist Eco-
nomics, 9(2–3), 263–299.
Kynch, J., & Sen, A. (1983). Indian women: Well-being and survival. Cambridge Journal of
Economics, 7(3/4), 363–380.
Lambert, S., & Rossi, P. (2016). Sons as widowhood insurance: Evidence from Senegal.
Journal of Development Economics, 120, 113–127.
Lechene, V., Pendakur, K., & Wolf, A. (2019). OLS estimation of the intra-household distri-
bution of consumption (No. W19/19). IFS Working Papers.
Lee, R., Raphael, E., & Snyder, R. (2020). A billion people live in slums. Can they
survive the virus? New York Times, April 8. Retrieved from https://www.nytimes.
com/2020/04/08/opinion/coronavirus-slums.html, last accessed on February 25, 2021.
Mondal, B., & Dubey, J. D. (2020). Gender discrimination in health-care expenditure:
An analysis across the age-groups with special focus on the elderly. Social Science &
Medicine, 258, 113089.
Moradhvaj, & Saikia, N. (2019). Gender disparities in health care expenditures and fi-
nancing strategies (HCFS) for inpatient care in India. SSM-Population Health, 9,
100372.
Pande, R. P. (2003). Selective gender differences in childhood nutrition and immuniza-
tion in rural India: The role of siblings. Demography, 40(3), 395–418.
Parth, M. N. (2020). Social distancing during the coronavirus? Not when a family of six
shares one room. Los Angeles Times, April 7. Retrieved from https://www.latimes.
com/world-nation/story/2020-04-07/social-distancing-family-one-room, last
accessed on February 25, 2021.
Rahman, O., Foster, A., & Menken, J. (1992). Older widow mortality in rural Bangla-
desh. Social Science & Medicine, 34(1), 89–96.
Ravindran, S., & Shah, M. (2020). Unintended consequences of lockdowns: Covid-19 and the
shadow pandemic (No. w27562). National Bureau of Economic Research.
Sen, A. (1982). Coromandel Lecture. New Delhi: As published in Food and Nutrition (1984),
10, 81–89.
Sen, A. (1989). Women’s survival as a development problem. Bulletin of the American Acad-
emy of Arts and Sciences, 43(2), 14–29.
Sen, A. (1990). More than 100 million women are missing. The New York Review of Books,
37(20), 61–66.
Sevilla, Almudena, and Sarah Smith. (2020). Baby Steps: The Gender Division of Childcare
during the COVID-19 Pandemic, IZA Discussion paper Series, Discussion Paper No.
13302.
Sharma, G., Volgman, A. S., & Michos, E. D. (2020). Sex differences in mortality from
COVID-19 pandemic: Are men vulnerable and women protected?. JACC: Case Re-
ports, 2(9), 1407–1410.
Van de Walle, D. (2013). Lasting welfare effects of widowhood in Mali. World Develop-
ment, 51, 1–19.
PART IV
Health, education and labour
19
POLICY FAILURE BEFORE COVID-19
PANDEMIC AND OPTIONS FOR THE
GOVERNMENT IN NEAR FUTURE
Arijita Dutta and Montu Bose
19.1 Introduction
Covid-19 has been a kind of structural break in our life. It’s a highly contagious
and communicable disease that spread fast across East Asian countries and Eu-
rope, then hitting North and South America and finally the rest of Asia. Owing
to the pandemic, the response of any government needs to be two-pronged. On
the one hand, a strict disease surveillance system with health care facilities is
the conditio sine quo non of control of disease spread and treatment. On the other
hand, there is a need of support to compensate the loss of lives and livelihoods
to be offered to the general mass. This assumes particular importance on the
background of lockdown and suspension of almost all economic activities. In
the present context, though most of the state governments in India have been
at least partially successful, majority of them failed in disease surveillance. One
of the prime reasons behind this is a near-complete policy void in health system
and coordination failure. This chapter attempts to analyse the historical context
of health policy matrix of pandemic preparedness and post-epidemic policy in-
terventions in different countries and then look into the sheer neglect of pub-
lic health in general and epidemic preparedness in particular in India, which
resulted in a near-complete collapse of health system within a month of the
Covid-19 pandemic. With Covid cases rising fast in the country, the surveillance
system was poor across the states with low coverage of tests. A state-wise analy-
sis of available data on test coverage, test positivity and case fatality might offer
insights into the policy failure and future direction of health policy itself. Like
any disease, not just the number of cases, but also its fatality matters a lot for the
current pandemic. Depending upon stage of development within the state, both
reporting of diseases and case fatality among them ideologically depend on the
health infrastructure available and per capita income. Improved infrastructure
DOI: 10.4324/9781003220145-23
324 Arijita Dutta and Montu Bose
would contain the spread of disease, while reducing the number of deaths too.
In the absence of further disaggregated data, simple statistical analysis of state-
level data might offer interesting observations on morbidity-mortality debate.
However, one must understand that in a pandemic situation, not just the public
policies make a difference, but the private demand for preventive care changes
with the nature and degree of the disease prevalence. Given this, the chapter also
attempts to identify the policy options that the country has in the aftermath of
Covid to strengthen the health system and public health framework altogether.
The flow of the chapter is as follows: Section 19.2 outlines the historical traits
of public health policy across the globe, while in Section 19.3, we deal with
the similar issue for India. Section 19.4 discusses the issue of private demand
for preventive services on the face of a spreading pandemic like the Covid-19.
The morbidity and mortality trends and patterns from the disease are analysed
using the state-level data to find whether the usual health economics theory of
morbidity-mortality movements with development holds for Covid-19 or not.
19.2 The history of public health policy and pandemic

preparedness
In developed countries, the need for effective public health policies, including
disease surveillance and prevention, was triggered partly by military concerns
and the rest by the demand from elites till the cure was uncertain. As the loss of
army life was higher from frequent epidemics starting from Black Death to influ-
enza pandemic than from battle, all governments started looking for inexpensive
measures for keeping their army safe.
The first pandemic that attracted the notice of the health care policy makers
was that of Black Death in fourteenth-century Europe, which included cases of
three major strains of plague, namely, bubonic plague, pneumonic plague and
septicemic plague. The congested towns of late medieval Europe were typified
by poor sanitation and hygiene in comparison to some contemporary civiliza-
tions elsewhere, and thus became loci of epidemic disease. Despite the confusion
about exactly what was making people sick, people in medieval Europe identified
that they needed both preventative and curative health care to fight the disease.
In addition to using herbal treatments, religious rituals, such as bloodletting and
boil lancing, quarantine was a common preventative method. One of the earliest
quarantine methods was developed in Venice where the ships that came in and
out of the city were monitored and the travellers were put on forced isolation
for 40 days under the name of quarantine. As these efforts partially succeeded in
managing transmission of the disease, many other cities (like Marseilles, Genoa,
Naples, Rome) adopted similar protocols, which marked the beginning of public
health care policy. Governments, in association with urban municipalities of
bigger cities, closed ports, banned infected and suspicious travellers, and either
sent the sick or the healthy away in efforts to minimize contagion. Sanitation
was also a huge part of governmental health care efforts to combat the Black
Policy failure and options for government 325
Death. Homes of the deceased were sanitized, and their personal belongings
were burned, while the municipalities designated safe homes to house the sick.
While the Black Death pandemic marked the beginning of the welfare state’s
intervention to protect citizens’ health care needs, and civil society too came
up strongly to join the cause as well by educating the public about how to live a
healthy life was a concerted effort (Nevils-Karakeci 2020). Prompted by more
stringent sanitary enforcement during plague years, concepts of cleanliness and
sanitation gradually took hold in Europe’s cities.
In the first decade of the nineteenth century, scientists started to identify the
germs that cause the epidemics in the form of communicative diseases. This led
to sanitary movements in Europe and America, forcing radical changes in regula-
tions and health behaviours of the citizens, particularly in urban areas. Following
the Meat Inspection Act in the US in 1896 and prohibiting urban citizens to keep
livestock, large-scale protests hit several cities, including an irate butcher chasing
the sanitary inspector with knifes (Duffy 1990).
At the beginning of the twentieth century, when the developed countries
were confident that their battle against the communicative infectious diseases
is almost won, the nations were forced to implement emergency measures in
response to a tremendous health challenge in the form of influenza pandemic
in 1918. With the world divided by war, the multilateral health surveillance
system was not enough to control the pandemic among the soldiers, travel-
ling across the world and living in unhealthy conditions. Lot of municipalities
implemented disease-containment strategies in the form of closure of schools,
churches, theatres and suspension of public gatherings. Yale University can-
celled all on-campus public meetings. However, the measures were uncoordi-
nated in war-torn areas. Additionally, media played a confusing role, vouching
for conflicting health measures, creating panic, stigma and great anxiety among
the citizens (Tognotti 2013). During the second wave of influenza epidemic
in 1957–58, it was confirmed that quarantine methods and closure of schools/
asylums/churches, etc., can best postpone the onset of diseases by a few weeks
(Zhdanov 1959). Also, quarantine methods have always been criticized as often
being coercive, violating basic rights. The people belonging to lower economic
and social strata and marginalized sections have commonly been stigmatized
and faced discrimination (Rothstein et al. 2003). Availability of vaccines and
anti-microbial drugs played a very positive role in reducing the load of the
disease. The large-scale investment in public health on the one hand, and in
drug discovery on the other, helped to reduce the duration and severity of the
epidemic.
By the mid-twentieth century, the institutionalization of public health was
thought to be crucial for maintaining health and hygiene of countries (primar-
ily in Europe and Northern America) and preventing pandemics. Japan studied
European institutions in their preparation of becoming the world super power.
They also followed similar policies in their colonies in Korea, Taiwan and other
East Asian countries.
Across the countries, the neglect of public health policies and overall prepar-
edness for resisting pandemic waves jolted after the discovery of antibiotics and
widespread availability of antibiotics to cure the people from communicative
diseases. The elite capture of health care policy saw a shift away from prevention
and public health in general towards curative services, including tertiary health
care facilities and highly subsidized medical training. They did not bother about
the living conditions or behaviours of the citizens at the lower rung of the pop-
ulation, because they believed that even if their hygienic and sanitized living
conditions fail to prevent such diseases for them and their family members, they
would easily be cured in institutionalized medical facilities ( Jeffery 1988). Addi-
tionally, the spread of democratic institutions also affected public health services,
because the electorate typically prefer public funds to be used to provide private
goods (like medical care), rather than public goods like social sanitary measures
(Khaleghian and Gupta 2005). This led to neglect of public health in almost all
democratic developing countries, compared to the non-democratic regimes of
East Asia, USSR and Cuba (Das Gupta 2005).
This gross neglect of public health to contain the spread of infectious diseases
in the form of epidemic or pandemic has actually resulted to a lack of prepar-
edness to tackle pandemic even in the new millennium. Just before the cur-
rent Covid-19 pandemic hit the world, a report on Global Health Index (2019)
by Nuclear Threat Initiative (NTI), Johns Hopkins Center for Health Security
( JHU) and The Economist Intelligence Unit (EIU) clearly pointed out that none
of the countries across the globe were actually ready to fight back an epidemic
(www.ghsindex.org). The report identifies six dimensions of health security in
connection of preparedness to fight an epidemic. They are prevention, detection,
rapid response, health system, compliance with International norms and risk en-
vironment. The report found that the national health security is fundamentally
weak around the world. No country is fully prepared for epidemics or pandem-
ics, and every country has important gaps to address. The overall mean GHS
Index score among all the 195 countries assessed was 40.2 out of a possible score
of 100. Among the 60 high-income countries, the average GHS Index score is
51.9. In addition, 116 high- and middle-income countries did not score above
50. Interestingly, lowest scores were found in prevention and rapid response of
epidemic spread.
19.3 History of public health and pandemic

preparedness in India
In colonial India, the British were more interested in protecting British civilians
and army cantonments from the outbreak of epidemic. They never attempted
to alter the health behaviours of the native Indians, may be due to fear of hos-
tility. The policies of segregation and quarantine back home outlined their pol-
icy of keeping the Army and the British administrators in complete segregation
from general people in India. Those areas enjoyed far better sanitation, disease
surveillance and solid waste management. Institutions like All India School of
Public Health and Hygiene and the Calcutta School of Tropical Medicine were
created for public health training and research. There are reasons to believe that
the prime reason behind these efforts is to prepare a public health cadre to serve
the British army.
The spare but systematic colonial approach to public health service provision
is reflected in its successes and failures (Das Gupta 2005). Though during the first
half of the twentieth century, the mortality spikes from epidemics were fairly
handled, but diseases such as malaria and gastro-enteric infections continued to
take heavy tolls. Independent India’s First Five-Year Plan document found that
only 3% of households in India had toilets, and that much of the population lacked
basic water, drainage and waste disposal services (Planning Commission 1951).
However, with epidemiologic transition taking place in developed coun-
tries, the post-war era saw significant investment in medical research of non-
communicable diseases. This trend seriously undermined the efforts of global
research and development on existing communicable diseases, and also the newer
possible epidemics like Avian flu. Though multilateral agencies and donors in-
vested in vertical strands of disease control and care for separate diseases in de-
veloping countries like India, the horizontal integration among them remained
a far cry. Within public health, two sections that got attraction from the policy
makers are immunization programme and family planning programme. With
their problems and domains lying in complete segregation to each other and also
vis-à-vis the rest of the health policy, the success or partial success of these verti-
cal programmes remained isolated and toned down. This discouraged the build-
ing of an integrated health system identifying the outcomes of public health. As a
result, two things followed, which tended to change the contour of health policy
framework in India. First, there was a serious cut in public health budget at the
state levels, and second, the doctors and nurses, trained in curative care, became
far more lucrative professions compared to public health workers. The cadre of
sanitary inspectors and male health workers started vanishing from India, with
more stress on building public and private medical hospitals.
Introducing architectural correction to the health policy in India, National
Rural Health Mission initiated central government’s support plans to the states
for preventive and curative health care in 2005. Though public health and san-
itation still remain under the aegis of state list, central governments continued
support to states for building toilets, particularly for the poor. The central health
ministry has a major influence on the states’ public health policies because of
its fiscal leverage, convening power, technical resources, and ability to draw
and disseminate lessons across states. However, this has counter-productively
de-emphasized public health services, through a series of policy decisions since
the 1950s (Das Gupta et al. 2009). As a consequence, states spent only 8.9% of
their health budget on public health programmes in 2001–02 (Government of
India 2019), and this includes the states’ significant budgetary contribution to the
single-issue programmes. By contrast, they spent over 33.9% of their budget on
tertiary and secondary care services, illustrating the importance of the curative
sector to capture public budgets for health (GoI, 2019).
In this context, it must be mentioned that India spends an abysmally low
share of its GDP on health care at only 1.3% of its GDP in 2018 (GoI, 2019),
which made the health infrastructure highly inadequate. Additionally, the cap-
ital expenditure on health is abysmally low at 7% of total health expenditure
during 2016–17 (NHA 2016–17). It contrasts with other developing countries
such as Brazil, which spends 7.5% of its annual GDP on health, Bhutan, which
has allocated 3.6%, and Bangladesh, which dedicates 2.2% (World Bank 2019).
Among developed nations, South Korea has kept its health care expenditure at
8.1%, Japan 10.9% and the US at 8.5% of GDP. Subsequently, India has a severe
shortage of health care workers. According to the Health Ministry data released
in October 2019, there is one doctor for every 11,082 people, which is more than
ten times the doctor-patient ratio that the World Health Organization (WHO)
promulgated. India has 0.53 beds for 1,000 people compared with 0.87 in Bang-
ladesh, 1.1 in Indonesia, 2.11 in Chile, 2.73 in Turkey, 1.38 in Mexico, 4.34 in
China and 8.05 in Russia (OECD 2017). Low investment, in health sector in
general and public health in particular, created a real tight-fisted health infra-
structure, which has been proved to be incapable to handle any pandemic of this
order.
19.4 Private demand for prevention

Given the above discussion, it is evident that the public, rather than private,
control of infectious diseases is one of the main achievements of modern public
health. Across the globe, the health policy measures to combat and prevent epi-
demic and pandemic situations have focussed over-reliance of public monitoring
and subsidized treatment. Using the standard arguments about the positive ex-
ternal effects of disease prevention, economists often echo such arguments for an
active public role in the prevention of infectious diseases, such as AIDS, SARS
and Avian flu.
However, the issue of individual person’s private response to any pandemic
has rarely been discussed and evaluated. With the concepts of economic epide-
miology, one must consider the crucial interaction between the extent of disease
and the demand for prevention itself, which is increased by the extent of disease.
The theory identifies that an increase in the prevalence of an infectious disease,
i.e. the share of the population infected, induces growth in private ex-ante de-
mand for prevention. Central to the study of economic rationality under epi-
demics remains the prevalence-elasticity of private demand for prevention against
disease. It represents the degree to which prevention rises in response to a disease
outbreak and may differ in its form across diseases. For example, the elasticity for
vaccine-preventable diseases may represent the number of additional vaccina-
tions induced by each new infection, while that for sexually transmitted diseases,
it may represent the increase in the matching of sex partners who have the same
infection status. For Covid-19, the demand for using masks and alcohol-based
sanitizer could not be increased in developing countries before the infection
affected a significant share of population.
This type of prevalence elastic behaviour has two major, but somewhat con-
tradicting, implications. On the one hand, the growth of infectious disease is
self-limiting because it induces preventive behaviour, while on the other hand,
initially successful public health interventions actually make it progressively
harder to eradicate infectious diseases. This happens irrespective of type of mar-
kets under which the vaccines are produced. In the background of current media
explosion of information and the world-wide stringent patent laws in pharma-
ceutical market, a strongly communicative disease like Covid-19 seems impos-
sible to be eradicated all together. On the demand side, the private demand for
prevention would fall as soon as the wave of new cases declines, while on the
supply side the patent-protected producer of vaccines has a special dynamic in-
centive to increase mark-ups. Put simply, if there were economic fortunes to be made in
disease eradication, we would have more of them (Philipson 2000).
In epidemiological analysis, this hazard rate is supposed to be an increasing
function of prevalence. This means that the larger the proportion of infected
people in the population, the larger the fraction of uninfected people who be-
come infected in the next period. A higher prevalence increases the chance that
a susceptible individual will meet an infected individual. However, economic
incentives imply that the hazard rate into infection should be a decreasing function
of the prevalence of the disease, when private demand for prevention is preva-
lence elastic. As the number of infected individuals grows, uninfected individ-
uals face a larger risk of infection and hence raise their demand for prevention.
In essence, the issue boils down to the fact that with an epidemic or pandemic
outburst of highly communicative disease, there are two different kinds of re-
sponses that might be found to control further spread. There is response from
state to monitor and offer curative service, while the private demand for preven-
tive services rises. Given the fact that most of the developing countries, including
India, currently have a fragile structure of preventive service provision and pub-
lic health facilities, juxtaposed with non-availability of vaccines for Covid-19,
the government’s response seems to be concentrating again on curative services
with detection and treatment of patients already infected. The private demand
for prevention rises with the pandemic spreading its prevalence, in the form of an
attempt to minimize the touch points. The prime way to do that has been usage
of masks and sanitizers, along with some uptake of medicines from alternative
medical systems (homeopathy, Ayurveda and naturopathy) which is yet to prove
any direct connection with this specific disease. As discussed earlier, the pre-
ventive and public health service mechanism been long neglected and vaccines
unavailable, two things remained crucial: Number of positive cases and number
of deaths, representing morbidity and mortality, respectively.
19.5 Morbidity and mortality from Covid-19: a state-wise

analysis of morbidity and mortality
The discussion on spread and case fatality from Covid-19 remains at the heart of
the discussion on pandemic response by states. While the number of cases nor-
malized by population size represents the morbidity trend from the disease, that
is, the share of people affected from the disease, the case fatality (deaths per hun-
dred cases) denotes the mortality from the disease. Though incidence and death
from a disease seem to be highly correlated (because ill health is a pre-condition
of death), available literature suggests that the trend and nature of morbidity and
mortality often go in different directions (Murray 1996). It is evident from all
health rounds of National Sample Survey (NSS health rounds) that the morbidity
trend identified by self-reported cases of illness is by far the highest in states with
best health infrastructure (Kerala, Tamil Nadu), while the infant and child mor-
tality rates are the lowest in these states. However, the reported morbidity is the
lowest in states like Bihar and Uttar Pradesh where the health care infrastructure
(HI) is arguably the worst and mortality figures are the highest. This is explained
by a number of factors like usage of aggressive medical technology of postponing
death, which increases the frailty of the population and hence disease occurrences.
But the availability of HI and awareness to seek treatment are two most important
causes of increased disease reporting. It remains imperative to identify whether
this hypothesis of differential trend of incidence and death from Covid-19 and
their relationship with HI hold in a situation of widespread pandemic or not. The
observation related to this might help us to identify the right mix of health care
policy for the near future. The data sources used for the analysis are as follows:
1 All Covid-19-related data have been taken from https://www.covid19india.

org/ (searched on 31st October 2020 at 9.00 hrs).
2 Health infrastructure data has been collected from National Health Profile
2019 (14th Issue).
3 Gross State Domestic Product (GSDP) data has been collected from the
Ministry of Statistics and Programme Implementation; GoI, 2020. (MOSPI
Gross State Domestic Product. Ministry of Statistics and Programme Im-
plementation, Government of India. Searched on 10th November 2020 at
10.30 hrs).
4 Elderly population data has been collected from the Census of India (https://
www.censusindia.gov.in/vital_statistics/SRS_Report_2015/6.Chap%202-
Population%20Composition-2015.pdf, searched on 15th November 2020
at 20.30 hrs) and Elderly in India Report (CSO; http://mospi.nic.in/sites/
default/files/publication_reports/ElderlyinIndia_2016.pdf, searched on
15th November 2020 at 20.30 hrs).
5 Prevalence of NCDs has been estimated from NSS 75th round unit level
data on Social Consumption: Health, 2017–18.
Figure 19.1 represents the cases of Covid-19 reported per thousand population
across Indian states with HI rank and per capita GSDP rank. Clearly, the bigger
bubbles of cases per thousand of population lie on the top right corner (Delhi,
Sikkim, Chandigarh, Goa, etc.) and smallest bubbles lie on the lower left corner
(Bihar, UP, Jharkhand, MP). Majority of the big bubbles are actually related to
either better HI rank or higher GSDP per capita rank or both. Though Mahar-
ashtra and Karnataka have modest health infrastructure, they are richer states of
the country lying above the mean. They both have big bubbles corresponding
to higher incidence of Covid-19 disease. The only noteworthy exception seems
to be Andhra Pradesh, which records pretty high incidence but it has moderate
health infrastructure and GSDP per capita. The overall trend actually hints that
reporting of Covid-19 positive cases seems to be more in states with better in-
frastructure and income, thus supporting the original hypothesis. However, in
not all states the correlation is evident between tests performed and cases per
thousand of population (Figure 19.2). For example, Delhi has tested 23.1% of
its population, while the other richer and better equipped states (Chandigarh,
Goa, Sikkim, etc.) have tested far lower.1 Also, the states on the lower left corner
(except Madhya Pradesh) do not have smallest bubbles for tests per hundred of
population. West Bengal, lying in the median area of HI rank, but lower than
average in GSDP per capita, has the second lowest test coverage after Madhya
Pradesh. States with less than 10% of test coverage are Bihar, Uttar Pradesh,
Relationship
FIGURE 19.1 among cases per thousand of population, health infra-
structure and per capita Gross State Domestic Product (GSDP).
Relationship
FIGURE 19.2 among test coverage (tests per hundred of population),
health infrastructure and per capita Gross State Domestic Product
(GSDP).
Jharkhand, Chhattisgarh, Madhya Pradesh, Rajasthan (poor in terms of econ-

omy and health infrastructure), Gujarat, Maharashtra, Haryana (rich but with
poor health infrastructure), West Bengal and Punjab (in medium group in terms
of both indicators).
Thus, the positivity does not necessarily depend on the test coverage in the
case of Covid. It might depend on the population density and other issues related
to epidemiology. The fact that morbidity, like other diseases, depends positively
on HI and not exactly on test coverage brings in uses of elasticity of private de-
mand for prevention as discussed earlier. In the case of non-vaccine-preventable
disease like Covid-19, the other preventive practices might have reduced the
spread in some states. In this connection, it must be remembered that in most of
the countries, including India, tests have been done under contact tracing and
hence the test positivity rate (TPR), that is, the share of the positive cases out
of all tests, needs to be looked at too. Where the spread is low, the need for tests
may be low, at least initially. Thus, some states which report low coverage of tests
(tests per thousand of population) might not actually need more and hence their
TPR might be low. Figure 19.3 shows the TPR across the states with HI and per
capita GSDP rank. Interestingly, those states which possess better infrastructure
tend to have done better contact tracing with TPR values high for Delhi, Goa
and Chandigarh. Maharashtra has the highest TPR value, in spite of low health
infrastructure.
Here, an added angle is that decision to test in the later phase of the pandemic
has shifted at least partially to private individuals in all states in India. With a
lot of private laboratories receiving the permission to do Covid-19 tests, the
individuals themselves are opting for tests and the test coverage is supposed to
increase if more and more people are found to be positive out of those tested.
Thus, low test coverage in Madhya Pradesh can be justified by low need of tests
Test
FIGURE 19.3 positivity rate vis-à-vis health infrastructure rank and GSDP PC
rank.
as depicted by lower than average TPR in the states. However, TPR in West
Bengal is higher than Indian average, which fails to justify the low test coverage
in the Eastern state.
According to the existing theory, case fatality is inversely related to health
infrastructure and status of economy. However, when plotted against those two
indicators as in Figures 19.1 and 19.2, no pattern emerges. While states like Delhi
and Punjab enjoy better HI, their case fatality is high. Exactly opposite picture
emerges for Bihar: With very low HI, it records one of the lowest case fatality
rate (CFR). This essentially means that CFR is not exactly correlated with health
infrastructure or per capita GSDP and this definitely marks a departure from
usual mortality trends across the Indian states.
In order to capture the other plausible causes, we plot CFR vis-à-vis HI and a
composite index comprising ranks in share of aged population and share of hos-
pitalization from non-communicable diseases. These two are considered because
of the growing medical literature on high fatality among aged population and
those with high co-morbidities in the form of NCD chronic diseases. Figure 19.4
clearly identifies that there is a clear downward trend for CFR with the compos-
ite index, but no such relationship with health infrastructure. Bihar’s low CFR
and Punjab’s exceptionally high CFR can be, at least partially, explained by high
and low ranks in composite index, respectively.
If we look at the summary table (Table 19.1), in Maharashtra and West Bengal,
the test positivity is high, coupled with low tests per hundred population, clearly
indicating that the states needed more coverage in disease surveillance. How-
ever, Delhi tested almost one-quarter of its population as the need represented
by case fatality is high. Bihar’s low test coverage is not unjustifiable owning to
low incidence of the disease in terms of indicators of positive cases per thousand
Case
FIGURE 19.4 fatality rate per 100 positive cases vis-à-vis health infrastructure
rank and composite index of aged population share-wise rank and
NCD-related hospitalization-wise rank.
TABLE 19.1 Summary table for selected states and their indicators
States Positive Tests Positive Case Health Composite

Cases per per 100 Cases per Fatality Infra- Index
1,000 Population 100 Tests Rate per structure
Population 100 Positive
Cases
Maharashtra 13.7 7.3 18.78 26.2 Low High

Gujarat 2.5 8.8 2.87 21.6 Low High
Delhi 19.1 23.2 8.23 17.0 High High
Punjab 4.4 8.6 5.16 31.4 Medium High
West Bengal 3.8 4.6 8.20 18.4 Medium Medium
Bihar 1.8 9.2 1.96 5.0 Low Low
All India 2.1 3.5 5.87 17.2 Average Average
Source: Authors’ estimation.
population and positive cases per hundred tests. Again, the CFR is high in all
these states, except Bihar, thus indicating no correlation with HI. High CFR
is reported from states with higher old-age population and higher incidence of
non-communicable diseases.
19.6 Conclusions and way forward

The above discussion strongly confirms that there was and there is a policy void
in most of the countries across the globe to tackle an outburst of highly com-
municative disease. While recognizing the fact that restrictive policies of quar-
antine, lockdown and isolation can at best postpone the disease, majority of the
countries took resort to them, as they did in earlier centuries. The postponement
of the spread through these policies actually could have reduced the burden of
epidemic only if the health infrastructure and public health policy matrix were
robust enough for inserting correct dose of disease surveillance and treatment
of diseases. With the colossal neglect of public health, juxtaposed with extra-
ordinary burden of non-communicable diseases in most of the developed coun-
tries following the epidemiologic transition, there was huge loss of life. The
story of Indian states tells us a similar story. The disease surveillance mechanism
had hit a floor so that the public-funded test coverage was low and inadequate
in most of the states. The private demand for prevention (in the form of behav-
ioural change and going for detection) could not supplement the overall preven-
tive services. The death rates are very high in some states primarily owning to
demographic pattern and prevalence of non-communicable diseases. Thus, the
mortality from this highly communicable disease appears to have strong co-
occurrences of non-communicable diseases. This indicates that what we need at
this hour is not compartmentalized and segregated vertical programmes for spe-
cific diseases, not even for communicable and non-communicable diseases, but
a holistic approach towards health system strengthening, with renewed focus on
public health comprising strict disease surveillances and preventive mechanism.
The focus on investment in curative treatment might appear to be insufficient in
the face of low priority of prevention mechanism. In an integrated health care
policy, curative and preventive strands would have to be recognized as comple-
mentary sectors, which can deliver the outcome efficiently, rather than pick and
choose one as a substitute of the other.
Note
1 We have considered both rapid antigen test and RT-PCR test for Covid-19, as sep-
arate data for each kind of test is not available. It has to be noted here that as per
WHO guidelines, rapid antigen test is less reliable than RT-PCR test in detection
the infection (https://www.who.int/emergencies/diseases/novel-coronavirus-2019/
question-and-answers-hub/q-a-detail/coronavirus-disease-covid-19).
References
Coronavirus Outbreak in India (2020). Available at https://www.covid19india.org.
Searched on 31st October 2020 at 9.00 hrs.
Das Gupta, M. (2005). Public Health in India: A Dangerous neglect. Economic & Political
Weekly. Vol. 40. No. 49. pp. 5159–5165.
Das Gupta, M., Rajendra, S., Somanath, T.V. and Datta, K.K. (2009). How Might India’s
Public Health Systems Be Strengthened? The World Bank Policy Research Working
Paper 5140. Washington DC. Available at http://go.worldbank.org/ BW6NXMA570.
Duffy, J. (1990). The Sanitarians: A History of American Public Health. Urbana and Chicago:
University of Chicago Press.
Elderly in India Report (2020). Available at http://mospi.nic.in/sites/default/files/
publication_reports/ElderlyinIndia_2016.pdf, searched on 15th November 2020 at
20.30 hrs.
Global Health Security Index (2019). Available at www.ghsindex.org.

Government of India (2019). National Health Accounts Estimates for India 2016–17.
National Health Systems Resource Centre: Ministry of Health and Family Welfare,
New Delhi.
Government of India (2020). Economic Survey 2020–21. Ministry of Finance. GoI avail-
able at https://www.indiabudget.gov.in/economicsurvey.
Gross State Domestic Product (2020). Ministry of Statistics and Programme Implemen-
tation. GoI available at http://mospi.nic.in/data. Searched on 10th November 2020
at 10.30 hrs.
Jeffery, R. (1988). The Politics of Health in India. Berkeley: University of California Press.
Khaleghian, P. and Gupta, M.D. (2005). Public Management and the Essential Public
Health Functions. World Development. Vol. 33. No. 7. pp. 1083–1099.
Murray, C. (1996). Epidemiologic and Morbidity Transitions in India. In Monica Das-
Gupta, Lincon Chen & T. Srinivasan (eds). Health, Poverty and Development in India.
New Delhi: Oxford University Press , pp. 122–147.
National Health Profile 2019-14th Issue (2020). Central Bureau of Health Intelligence. Delhi:
Ministry of Health and Family Welfare. Government of India.
National Health Systems Resource Centre (2019). National Health Accounts Estimates for
India 2016–17. New Delhi: Ministry of Health and Family Welfare, Government of
India.
National Sample Survey (2019). NSS 75th round unit level data. Social Consumption:
Health, 2017–18. New Delhi.
Nevils-Karakeci, R. (2020). The Plague and the History of Healthcare Policy. Available at
https://interestingengineering.com/the-plague-and-the-history-of-healthcare-policy.
Organization for Economic Co-operation and Development (OECD) (2017). Data on
hospital beds per thousand. Available at https://data.oecd.org/healtheqt/hospital-
beds.htm. Searched on 10th November 2020 at 8.30 hrs.
Philipson, T. (2000). Economic Epidemiology and Infectious Diseases. In A. Cuyler &
J. P. Newhouse (eds). Handbook of Health Economics. Amsterdam, NY: Elsevier, pp.
1762–1797.
Planning Commission of India (1951). First Five Year Plan Document. Government
of India. Available at http://14.139.60.153/bitstream/123456789/7205/1/The%20
Fi r st %2 0 f ive%2 0 yea r %2 0 pl a n%2 0 a%2 0 d r a f t %2 0 out l i ne%2 0Pl a n n i n g %2 0
commission%20July%201951%20CSL-IO016212.pdf.
Population Census of India (2020). Available at https://www.censusindia.gov.in/
vital_statistics/SRS_Report_2015/6.Chap%202-Population%20Composition-2015.
pdf, searched on 15th November 2020 at 20.30 hrs.Rothstein, M.A., Alcalde, M.G.,
Elster, N.R., Majumder, M.A., Palmer, L.I., Stoone, T.H. et al., (2003). Quarantine &
Isolation: Lessons Learned from SARS. Louisville: Institute of Bioethics Health Policy
& Law. pp. 1–160.
Tognotti, E. (2013). Lessons from the History of Quarantine, From Palgue to Influenza.
Emerging Infectious Diseases. Vol. 19. No. 2. pp. 254–259.
World Bank (2019). www.worldbank.org, accessed on 20.10.2020.
Zhdanov, V.M. (1959). The 1957 Influenza Pandemic in the USSR. Bulletin of the World
Health Organization. Vol. 20. No. 2–3. p. 489.
20
HIGHER EDUCATION IN THE
POST-COVID ERA
India and the world
Saumen Chattopadhyay
20.1 Introduction
The universities remain one of the oldest surviving institutions after the church.
However, the universities have been undergoing transformation for ages but
nothing compared to the transformative changes we have been witnessing in
the recent decades. Imposition of Lockdown to ensure physical (or social) dis-
tancing during the Covid-19 Pandemic disrupted functioning of universities
with teaching-learning in the classroom coming to a halt. As the universities
strived to restore normalcy by shifting towards online academic interactions
and even administration, the question we need to address is whether there is
a case for reimagining the future of higher education in view of the recent
spurt in acceleration of higher education reforms triggered by adoption of digital
technology albeit perforce to a large extent. With the rapid intrusion of digital
technology, often called the ‘disruptive technology’ in almost every sphere of
our life, and increasing importance of knowledge to secure and sustain human
existence, universities are poised to seize new opportunities for economic and
non-economic reasons. At the same time, universities are faced with challenges
emanating from decline in public funding, changing demand for skills the uni-
versity graduates are embodied with, rising social aspirations to pursue higher
education, to generate relevant knowledge and connect with the society effec-
tively and meaningfully.
This chapter seeks to trace out the possible implications of increasing reliance
on online teaching and other related changes for the university and its contri-
bution towards sustainability of the society as a whole. We then examine India’s
transition to the ‘new normal’ after the disruption against the backdrop of the
National Education Policy (GoI 2020A). We end this paper with the question
how ready India is to join the global system of higher education. This needs to be
DOI: 10.4324/9781003220145-24
338 Saumen Chattopadhyay
juxtaposed with three changes. One, the increased internet-based connectivity

which has formed a global network of students and researchers contributing to-
wards higher education to evolve as a global public good; two, the continuation
of the neoliberal approach to corporatize universities; three, tendencies inherent
in the university system to accentuate differentiation both nationally and glob-
ally. This chapter looks at the way the evolving Indian universities are respond-
ing to the changes in the Post-Covid world.
20.2 Higher education matters

While it is understandable why unlocking the university gates took considerable
time, it is important for the sector and the country that higher education does not
get relegated as the economy limps back to normalcy because higher education
matters.
The higher education institutions (HEIs) perform multiple functions and
university output is, therefore, multi-dimensional comprising mainly teaching-
learning, doing research and knowledge generation, education as consumption
as a transformative experience and as generators of information by the process
screening and grading (Stiglitz 1975) in addition to the outreach activities.
Churning out skilled graduates and contributing towards knowledge generation
have long been recognized as the major factors behind economic growth. As stu-
dents experience university life in the process of their socialization, they evolve
with the formation of their preferences which are supposedly imbued with moral
values and responsible citizenship.
Rendering education as a screening device and credentials as signals is an
important function the universities perform. Credentials or certificates emit sig-
nals to the employers regarding the quality of human capital embodied in the
graduates and their suitability for the job market skill requirements. No other
institution can replace universities in the future for the two most important
functions that the universities perform at the present scale: one, contribution
towards human capital formation and two, screening or signalling (Palfreyman
and Temple 2017).
In this period of uncertainty and gloom associated with the loss of jobs, higher
education can offer immunity to the graduates to (re)-skill themselves and pre-
pare for the new normal in the job market. Higher education assumes critical
importance in the Post-Covid for being the most prominent site for knowledge
generation and by strengthening university–industry linkage for doing, argua-
bly, impactful research. Competition among the higher education institutions
in a market-based system in the USA and UK along with a tendency to foster
individualist culture can explain rapid spread of Covid-19 and high death tolls
observed this part of the world in sharp contrast to East Asia and parts of Europe
where values inculcated are aligned to higher education as a common good and
conduct of socially responsible behaviour (Marginson 2020).
Higher education in the post-Covid era 339
20.3 Transformation in higher education

Can the future of higher education be extrapolated from the past and the pres-
ent trends? There are certain features common to the transformation that we
witness in the academia world over. These changes are occurring at three levels,
the institutional, the national and the global. These transformations are mostly
being informed by neoliberal approach to higher education policymaking. At the
institutional level, university governance is being reformed based on the princi-
ples of new public management which recommends accounting and auditing of
teachers’ performances and incentivization to reward their good performances.
The neoliberal approach advocates a transition from the traditional input-based
funding of the universities, where teachers are considered to be trust-worthy and
are bestowed with academic freedom with accountability being restricted to self
and the peer groups towards output-based funding. In the process, accountability
is shifted to the market, quality assurance agencies and ranking parameters de-
signed by the national and global agencies. In this ‘utilitarian drift’, knowledge
produced is also more instrumental than intrinsic (McCowan 2017; Teichler
2015). At the national level, the approach of the policymakers has been to cate-
gorize universities and construction of a quasi-market to foster competitiveness
to achieve allocation efficiency and deliver quality though the relationship be-
tween the two is not only weak but uncertain too (Chattopadhyay and Sharma
2019). The inevitable outcome of this market-based restructuring is accentuation
of vertical differentiation within the higher education system which undermines
horizontal diversity (Teichler 2015). There is a tendency among the universities
towards filiopietism as universities emulate the elite world-class universities and at
the same time, there is a tendency to converge and look alike too (Palfrayman
and Temple 2017). At the global level, mobility of embodied human capital,
teachers and students witnessed a rapid rise mainly from the South to the North.
Knowledge being intrinsically in the nature of a public good (Stiglitz 1999; Mar-
ginson 2016) has developed features of a global public good facilitated by open
access system and university knowledge despite tendencies towards knowledge
capitalism. However, the global arena of knowledge remains polar and unequal
due to the domination of English and sustained stellar performances by some
Western nations like the USA, a major part of Europe and China (ibid.)
Amid this ongoing transformation in the wake of Pandemic, we witness ac-
celeration in the process of adoption of digital technology which is contributing
to the process of university transformation as evident from the exploration of
new innovative models of higher education to deal with the emerging challenges
(Economist Intelligence Unit 2020).
20.3.1 Transition to online teaching

Online teaching is in fact turning out to be a game changer as it has the poten-
tial to trigger significant changes in the way a university functions particularly
with regard to teaching-learning and academic collaboration. The Pandemic has

accelerated the process by causing sudden disruption which lasted for long. Irre-
spective of the Pandemic, the transformation can be understood in terms of time
and space with implications for scale and quality. The global online platforms
were approached by the universities to provide them quick assistance and helped
them tide over the crisis.
20.3.2 Changing notions of time and space

Classroom teaching is a typical example of a service in contrast to a product man-
ufactured by a firm with the help of a defined technology. Typically, a service
delivered lasts only during the period of its delivery. If teaching is recorded and
stored digitally, the service of teaching would not get extinguished once the class
gets over because the service acquires the status of a digital product which can be
used umpteen number of times the way we watch movies and enjoy music online
either free or at a price as the case may be. However, in the case of teaching in the
context of growth of knowledge and understanding, revision of content is under-
taken as and when it is felt to be necessary. This conversion of ephemeral service
to a lasting product is an important dimension of digitization of universities.
The other dimension is that in the conventional sense, delivery of service
desires physical presence of the intended beneficiaries in the proximity of the
deliverer as it happens in the case of teaching in a classroom. However, if teach-
ing is conducted online, the concept of space in the classroom loses its signifi-
cance as students from all parts of the world can join in the virtual space which
can be both synchronous and asynchronous (McClure 2015). This change is a
crucial one with far-reaching implications. The physical space in the classroom
transcends the boundaries of the four walls and can potentially encompass the
virtual global space. This mode over time will get integrated into the main-
stream delivery system of the university and in all possibility, the university will
gradually move towards a blended mode of delivery which would require both
‘in-classes’ teaching and online participation from the students. The other mode
which came into prominence is opting for courses from e-platforms called mas-
sive open online courses (MOOCs) which can be construed as an extension of
ODL.1 The teachers would now be expected to be structured and prepared in
their teaching as well as to be compatible and acceptable to the rest of the world
in terms of content and curricular design.
20.3.3 Implications for scale and quality

Online education and MOOCs in particular provide an immense scope for scal-
ing up to garner more revenue as physical capacity of the classroom and the ad-
mission policy cease to matter. The issue is whether quality of learning outcome
suffers with the expansion in scale. Quality of service depends on the quality of
human capital or expertise embodied which is not replicable or reproducible and
its delivery with passion and motivation. This is in fact true for any service. This
non-replicability of human capital embodied can explain how do the quality
differences persist in the world of entertainment, games and sports as well as
in the world of scholarship and knowledge. As top universities attract the top
minds, both students and teachers, good-quality human capital tends to get clus-
tered among the good-quality universities ranked at the top of the ranking table.
The competition in education market is based on what is called selection-based
competition as opposed to exchange-based competition in the case of a market
for products (Glennerster 1991; Winston 1999). For example, concentration of
Nobel Laureates among the best of the universities in the USA has rendered the
league table stable as top universities retain their positions at the top with minor
year-wise fluctuations, while those from below find it difficult to climb up. The
good-quality institutions can embark on massive expansion as online delivery
is scalable albeit up to a limit with the possibility of replication of content and
online delivery. It is not beyond dispute that quality can be maintained as it is in
the case of listening to music and watching a movie, where there is neither any
scope nor any need for interaction. With the rise to the dominance of MOOCs
as being promoted by many universities, reputed as well as the not so reputed
ones, students exercise their freedom to choose courses across a range of options
transcending the boundaries of the university campus. India is the second with
7.1 million global learners with the USA having 12.1 million and China 3.2 mil-
lion, indicating hybrid learning in the campus (IFC 2020). The students begin
to learn in a different way as teachers begin to teach differently in the absence of
physical presence of the students in the classrooms.
20.4 Unbundling and commodification

This phenomenon of online teaching has led to what is commonly called unbun-
dling (McCowan 2017) furthering the process of commodification. Bundling
and unbundling of various types are common in the market for consumption
goods. Different products of varying degrees of complementarity are bundled
together which is promoted in different ways with varying degrees of conveni-
ence to the consumers. Delivery of teaching can now be offered to the students
in an unbundled form to allow the students to exercise discretion over the choice
of courses over a flexible period of time devoid of the pleasure arising out of ac-
ademic and non-academic interactions which accrue to the students from living
in the university campus. This unbundling has implications for what a university
does and how does a university connect to the society (ibid.).
The institutions can now deploy teachers with the objective of cost minimi-
zation which has acquired new significance in the emerging era marked by com-
petition and diversified funding base. With the option making recorded lectures
available, the tasks of developing course contents and tutoring of the students can
become two separate activities to be performed by two categories of teachers, the
former with the high status and the latter with the low status. This would lead
to an overall diminution in the demand for teachers. This bundling and com-
modification of the courses as pointed out by McCowan (2017) undermines the
sanctity of university space, university mandate and holistic learning.
20.5 Internationalization
The development of digital capability and the spread of internet have now made
possible what is called ‘internationalization at home’ in contrast to internation-
alization abroad associated with global mobility of embodied human capital as
indicated by the rise in student and teacher mobility and increasing collaboration
among the universities across the international borders. The class for online de-
livery can be truly global as students can opt for courses being delivered by the
universities located abroad. This is what was being followed by many top uni-
versities which attract students from other countries during 2020 and even in the
first half of 2021. Even with the easing of travel restrictions in the post-Pandemic
era, the tendency for internationalization at home will remain popular (de Wit
and Altbach 2020).
To understand the global knowledge economy, Marginson (2011) has catego-
rized knowledge into three categories. First, knowledge as a product which has a
price and it is useful for the industry in the form of patents within the framework
of intellectual property rights (IPR); second, the open access system and third, the
knowledge as a ‘status good’ which creates hierarchy within the system triggered
by the world ranking of universities. These three constitute an open-ended sys-
tem which is heterogeneous with overlapping regions. Knowledge has long been
recognized as a global public good normatively speaking (Stiglitz 1999) because
of its intrinsic characteristic of non-rivalry and non-excludability as fundamental
knowledge is of use to all cutting across the international borders. Excludability
of knowledge is socio-economic and a construct of policy (Marginson 2016).
The global flows which emanate out of these three are conditioned not only by
economic principles, but also by the culture and social spheres. Open access sys-
tem and the knowledge produced by the universities which are invoked by the
world ranking agencies are partly relational and hence it remains outside the pur-
view of the economic relationships as a part of it overlaps with the financial and
industrial economy. Only a part of the knowledge produced by the universities
as evident in the entire gamut of research a university carries out is amenable to
economic valuation (Marginson ibid.). The growth of internet has facilitated the
global flows in the knowledge economy. The production of knowledge responds
to the need of the capitalist economy, the need of the ‘status competition’ and
those of the open source exchange. The economic conceptualization of knowl-
edge as a public good can highlight market failure ‘but does not really capture
the scale, fertility and disorder of the open source regime; nor does it pinpoint
the zero-sum logic of status competition’ (ibid. p. 6). Given the massive flow of
knowledge across the border, application of shadow pricing at this scale is just
impossible, he argues for another method of valuation and that is of social status
which is served by the growing importance of world ranking of universities.
The other way of looking at the ongoing transformation of the university
has to be understood against the backdrop of the crisis faced by capitalism as a
sustainable system. The emergent form of capitalism referred to as informational
capitalism is now altering the nature of knowledge and human beings for its sur-
vival (Ouellet and Martin 2018). There are three manifestations of informational
capitalism, globalization, financialization and commodification. Amid all these,
universities are turning into organizations. Economic agents are best viewed as
information processors, and market as a cybernetic mechanism for transforma-
tion of information. Universities are now serving as the sites for reproduction of
advanced capitalist system – as techno-science hub of technological innovation
useful for the valuation of the capital. In the new globalized system of privatized
knowledge and internet revolution, there has been a shift towards applied knowl-
edge referred to as Mode 2 which requires establishment of linkage between the
industry and the university. With the decline in public funding, the industry is
outsourcing risks of R&D to the university to generate value for their intangible
assets. This is being made possible because of the adoption of digital technology
which is a specific form of globalized neoliberal informational capitalism.
20.6 Indian higher education during the Pandemic

The Pandemic disrupted teaching-learning and research in the HEIs in a major
way because of inadequate digital capabilities, lack of preparedness and hesitancy
on the part of teachers to transit to online against the backdrop of poor access to
internet among a large section of the students community (Reddy et al. 2020).
The centre and the UGC insisted that without examinations, degrees cannot
be awarded. Six states ruled by the parties in opposition to the ruling party at
the centre approached the Supreme Court. The Court supported the UGC and
deadline for the conduct of examinations was extended up to the end of Oc-
tober 2020. The academic calendar was revised and the new academic session
was scheduled to begin latest by 1 November 2020. In the case of University of
Delhi, the Delhi High Court monitored the conduct of open book examina-
tion in two phases with allowance for offline in the second phase to ensure that
the students’ interests are adequately served and grievances are addressed. In a
study conducted by Bhushan (2020) based on a sample of 543 HEIs, 63% of the
HEIs admitted that less than 50% of the online classes could be held till the end
of June 2020, while for 76–80% of the HEIs, examinations could not be held
before the end of June 2020, and 27% of government HEIs disclosed that they
lacked wi-fi facilities. A significant proportion of the students back home often
equipped with deficient gadgets struggled on their sides to join the online classes
with poor access to internet with bandwidth issues. Overtime, a semblance of
teaching-learning was restored to many of the campuses. Expectedly, a complete
transition to online covering all the courses and inclusive of all the students did
not materialize.2
To help the universities tide over the unprecedented crisis, the UGC issued
circulars urging the HEIs to move towards MOOCs (GoI 2020B, 2020C). This
unveiling of the National Education Policy 2020 (GoI 2020A) in the midst of
unlocking seeks to give a further push for online teaching and popularization of
MOOCs.
20.6.1 NEP 2020 and the future of Indian higher

education in the global world
The NEP 2020 unveiled on 29 July 2020 gave a push to the ongoing process of
popularization of e-platforms and a move towards online teaching. The NEP
2020, arguably, is futuristic in its approach as it seeks to empower India to meet
the emerging challenges emanating from both within the system at the national
level and outside the system at the global level. The process of reconfiguration
of the Indian higher education system as envisaged in the NEP may lead to a
bipolar market as some institutions will find it difficult to survive in the scenario
engineered by the process of binary-accreditation, competitive research funding,
students’ freedom augmented by the increased use of technology (Chattopadhyay
2020).
20.6.2 Policy push for the adoption of digital technology

To restore teaching during the Pandemic, the UGC began to promote MOOCs
which was launched in 2016. It is most likely that this adoption of technology
will get entrenched in the process, the extent of which remains unknown.
The NEP in continuation states clearly that:
‘The recent rise in epidemics and pandemics necessitates that we are ready
with alternative modes of quality education whenever and wherever tradi-
tional and in-person modes of education are not possible….’ the NEP ‘rec-
ognizes the importance of leveraging the advantages of technology while
acknowledging its potential risks and dangers’.
(GoI 2020A; NEP 24.1)
A major change was introduced in the choice of courses on the online platform.
In a public notice issued on 31 August 2020, the UGC urged the HEIs to al-
low the students to opt to the extent of 40% of the courses from the SWAYAM
platform (GoI 2020B). The public notice solicited ‘Expression of interest’ for
171 UG level MOOCs in humanities and social sciences from the university
teachers. This requires that the credit transfers to be allowed and equivalences
to be worked out. The NEP has suggested setting up of National Higher Educa-
tion Qualification Framework (NHEQF) under the General Education Council
(GEC) to address this issue. This NEP recommends the creation of an Academic
Bank of Credits (ABC) for the students to store their credits earned digitally
emphasizing on students’ sovereignty as an important element of reform strategy
for making of the market. There is a clear indication for promotion of online ed-
ucation and popularization of online courses in the wake of the wider acceptance
of online education, albeit perforce, as the letter issued by the UGC (GoI 2020C)
unequivocally states that there is a “…need to sustain the momentum generated
for online courses during the pandemic” cutting across the rural–urban divide.
20.7 Implications expansion, inclusion, excellence and

employability
Policy thrust on the formation of digital capabilities and internationalization
in the context of overall approach to Indian higher education reform will have
implications for all the major objectives of higher education, expansion, inclu-
sion, excellence and employability. These may be understood with reference to
the formation of the higher education market the NEP seeks to construct with
active support and interventions from four institutions entrusted with the tasks
of regulation, funding, accreditation and setting standards for quality learning
outcomes. If the students are allowed to exercise their freedom in their choices
of courses which need not be confined within the university but across the uni-
versities, national and international possibly in due course of time and from
popularization of e-platforms designated for MOOCs with flexibility over the
duration of the study period, there appears to be a major policy push for the even-
tual construction of a global market for Indian higher education (GoI, 2020A,
2020B, 2020C). Veering towards MOOCs and adoption of online transaction
provides a scope for expansion and increase in enrolment in higher education.
The issue is whether it will be meaningful to say that this kind of expansion will
be assured of inclusiveness and quality.
The NEP recommends building up of world-class digital infrastructure
and world-class content for India to eventually evolve as a prominent interna-
tional player. As far as inclusion is concerned, two issues are important. One,
the increased stratification of the higher education system with the possibility
of the market being bipolar marked by segregation into two segments, one of
good-quality HEIs and the remaining ones, will undermine the benefits of in-
clusion. Two, for meaningful inclusion, the digital deprivation has to be over-
come. Reddy et al. (2020) present before us the gross inequality that exists in
access to internet and ownership of suitable digital infrastructure, computers or
tablets among the enrolled students across the states, rural and urban, and socio-
economic categories based on the data gathered from the NSSO 75th Round
held in 2017–18. Only 25% of the students have access to internet through any
devices, including smartphones, at All-India level with corresponding figures of
17% and 44% for rural and urban, respectively. For computer or digital devices
with access to internet, access falls dramatically to 9% at All-India level with 4%
and 11% for rural and urban, respectively. Among the top 10% incomes, two-
thirds have access to internet with 45% ownership of computers.
It is possible that the universities will move towards a blended mode of teach-
ing with a differentiated fee structure adding one more layer to the extent of dif-
ferentiation that exists. Unless the elite and top class research universities posture
as inclusive, the value differences among the graduates will reflect cost differ-
ences. Since expansion beyond a point by the reputed institution will dilute the
brand and reputation of the institution, and therefore mitigate the positionality
of university-specific credentials, driving for expansionism would have its limit.
The fees structure will vary depending on the branding of the institution which
is a deterrent for true inclusion in an increasingly stratifying system.
Quality will be ensured by the accreditation with the National Accreditation
Council (NAC) being given a prominent role in the scheme designed by the
NEP. Excellence means world-class research and overall performance par ex-
cellence at the global level. The Institutions of Eminence (IoEs) are expected to
gain strength over time to face international competition and feature in the list
of global ranking (GoI 2017).
Given the digital divide that exists in India, online access would remain a
problem for vast sections of the student population in the near future. With the
greater possibility of survival of the good-quality-accredited HEIs, both national
and international as mentioned above, the higher education system will get fur-
ther stratified.
With the growth of online teaching and e-platforms, the socialization in the
campus will remain limited which is not desirable for a country like India with
bewildering diversity. Students, who would prefer to be on campus, will have
to shell out more for non-academic benefits that a campus life offers. It will be
exclusionary for the students who will find it unaffordable.
20.7.1 Employability and job market

Often the rationale for raising enrolment ratio in higher education is contested
in view of the declining capacity of the labour market to absorb the graduates
because of the increasing deployment of capital-intensive technology. Marginson
(2016) admits that the returns may fall and hence there is a need to treat higher
education and the realm of job market as two independent spheres (ibid.). There
is already a proliferation of micro courses suitably designed to upgrade skills to
meet the changing job market requirements; universities will also be focusing
more on general skills, problem-solving skills which would train students to
quickly adapt to the imperatives of the dynamic job market. The global online
platforms like Coursera are trying to bridge the skill gaps by offering courses with
immediate market demand in partnership with the universities and enterprises.
The market for global online degree and micro-credentials is expected to rise
to $117 Billion by 2025 as per the Holoniq website. Teichler (2015) emphasizes
on enrichment and status associated with earning degrees which will make the
graduates employable in the face-changing job market imperatives.
20.7.2 Policy push for internationalization

IoE (GoI 2017) and the Category-I institutions have been given the man-
date for the purpose of featuring in the list of global ranking (GoI 2018). The
Research-Intensive universities, the IoE and the Category-I HEIs have been
given the freedom to enter into collaborations with foreign universities.
The Finance Minister in her Union Budget 2020–21 announced that exter-
nal commercial borrowing (ECB) and FDI in Indian higher education are to be
permitted. The NEP has suggested an enactment of new legislation to invite the
top 100 globally ranked universities to operate in India. It is to be seen whether
this would stem the migration of the Indian students to study abroad. At the same
time, the Indian higher education has the potential to develop into a hub which
can attract foreign students to study in India as claimed by the NEP. Though
previously there was a meek response from the foreign universities, with the
advent of online teaching, the foreign universities can now explore the possi-
bility of entering the Indian market at a relatively low investment. Because of
the AI and digital technology, the possibility of an increase in the involvement
and participation of the EdTech companies has risen dramatically as the univer-
sities have started looking for outsourcing of some crucial activities like holding
of examinations and job-oriented niche courses from platforms like Coursera.3
India is the second sending country after China in the world as outbound stu-
dents before Pandemic hit the world reached almost 280,000 (Wadhwa 2019).4 In
the absence of reputation and good-quality infrastructure, India attracts students
mainly from the Asian countries to the extent of only one-tenth of outbound
students. For India to emerge truly as a partner and active participant in the
global knowledge system, India has to produce good-quality research and feature
in the top 100–200 in global ranking. In view of the advent of digitization of
education transaction and possible entry of foreign universities, ‘internationali-
zation at home’ will become a preferred option.
20.7.3 Indian universities in global ranking

India has been an exporter of human capital and an importer of technology and
knowledge. India spends inadequate amount, 0.69% of GDP on R&D as noted
by the NEP (GoI 2020A, p. 45). In the world of knowledge, ranking of India’s
universities will give an indication how is India placed comparatively speaking.
World ranking of Indian universities in terms of three most used world rank-
ing templates which are the Academic Ranking of World University (ARWU)
(Shanghai Jiao Tong University Ranking 2020), the Times Higher Education
(THE World University Ranking 2020) and the QS Ranking (QS Ranking
2020), the performances of the Indian higher education institutions are not yet
satisfactory.5
The policymakers have given recognition to only THE and QS for the
Category-I universities under the scheme of Graded Autonomy (GoI 2018) possi-
bly because of the high weightage given to research, including presence of Nobel
Laureates and field medallists in ARWU may not be appropriate for the Indian
universities. In terms of ARWU, the US universities dominate the top 1,000
with 8 universities among the top 10, 41 among the 100, and 133 among the top
500. China has 168 ranked in the top 1,000, among which 81 are in the top 500,
6 are in the top 100; the UK has 85 in the top 1,000, 36 in 500 and 8 in 100.
With regard to ARWU, the ranking of the Indian institutes begins after 500;
for the THE and the QS, it begins with 300+ and 172, respectively. India stood
fourth with China, the USA and the UK being at the top in terms of research
output in 2018–19 (all types as per SciVal as on April 2020) but India ranks
eighth in terms of shares in internationally co-authored publications (Interna-
tional Facts and Figures 2020).
If we consider one indicator given the importance of research, per capita pro-
ductivity, the scores awarded to Harvard University as per ARWU is 79.3 as
compared to 16.4 to IISC and 9.0 to JNU. It can be safely concluded, in the field
of knowledge, that Indian universities have to traverse long distances. The rank-
ing of the Indian higher education institutions as per the National Institutional
Ranking Framework (GoI 2020D) are shown in the Table 20.1. Though ranking
as a proxy of excellence has been subject to severe criticism, the quantitative as-
sessment of research can be considered to be a reliable indicator of both quality
and quantity of knowledge production. There is a dire need for improving the
culture of scholarship in the Indian universities, but at the same time, there is
also a need for selecting a few universities with potential and nurturing them
with resources and autonomy to feature in the league table within a decade or so.
There is no technology we can import to revamp select Indian universities and
compete in the global level in the near future.

Disruption caused by the Pandemic has raised important questions regarding
the future role of the universities in imparting necessary skill and in production
of knowledge. Financial sustainability of the universities has become a big issue
in the face of diminution in public funding when cost recovery is proving to be
challenging in a highly competitive world. Adoption of digital technology brings
with it opportunities to mobilize revenue and cut costs of operation. The global
online higher education market, including MOOCs, degrees and Online Pro-
gramme Management providers, is expected to grow to $74 billion by 2025 from
$30 billion in 2018 (IFC 2020). Some of the changes which are being ushered
in as the universities are embracing online education and opting for the courses
being offered by the national and global online platforms will remain embed-
ded in the university governance structure. The envisioning in the NEP 2020
TABLE 20.1 Global and national ranking of Indian universities and top world universities as
per world ranking (2020)
Universities ARWU Universities THE Ranks Universities QS

Ranks Ranks
1 Harvard 1 Oxford 1 MIT 1

2 Stanford 2 California Institute 2 Stanford 2
of Technology
3 Cambridge 3 Cambridge 3 Harvard 3
4 MIT 4 Stanford 4 Oxford 4
5 University of 5 MIT 5 California 5
California Institute of
Berkeley Technology
1 IISC 501–600 (2) IISC 301–350 (2) IIT Bombay 152 (4)
2 IIT Madras 601–700 (1) IIT Ropar 301–350 (39) IIT Delhi 182 (3)
3 CU, DU 601–700 IIT Indore 351–400 (23) IISC 184 (2)
(11), (18)
4 IIT Delhi 701–800 (3) IIT Bombay 401–500 (4) IIT Madras 271 (1)
5 IIT 701–800 (5) IIT Delhi 401–500 (3) IIT KGP 281 (5)
Kharagpur
6 JNU 701–800 (8) IIT Kharagpur 401–500 (5) IIT Kanpur 291 (6)
7 AMU, VIT 801–900 Indian Institute 501–600 IIT Roorkee 383 (9)
(31), (28) of Chemical (34), (35),
Technology, IIT (9)
Gandhinagar,
IIT Roorkee
Source: Websites of ARWU, THE and QS as given in the references.

Note: Within parentheses, rankings of the Indian higher education institutes are as per the National In-
stitutional ranking Framework 2020.
imagines such a transformation. But this transformation may prove to be costly

for many universities. Unless there is a substantial increase in the allocation of
higher education budget and digital capabilities are enhanced and digital divide
is effectively addressed, the future of higher education looks promising for many
good universities and it is uncertain for the rest. Engagement at the global level
can be more rewarding and meaningful if the standard of Indian higher edu-
cation improves considerably. In terms of the number of students going abroad
for higher studies, India stands next to China. Unless India’s ranking improves,
India’s ability to attract students and enter into research collaboration will remain
limited. The entry of the foreign universities along with the students’ sovereignty
to choose courses will make the Indian market a bipolar one. This will result in
further stratification of the Indian higher education market which will affect the
inclusiveness of the expanding higher education system. However, despite the
vulnerabilities of the higher education sector, it is in the interest of the nation
that this sector is assigned adequate priority to negotiate well with the changing
landscape of the world of higher education.
Notes
1 McClure (2015) provides three narratives in favour of MOOCs. These are elite narra-
tive, democratic narrative and economic narrative. The elite narrative views MOOC
as an obligation of the rich and the privileged who study in the elite institutions to
open up their classrooms to the masses of the world. Democratic narrative argues for
empowering of the masses necessary for self-governance and economic narrative is
based on privatization of the market place and consumer choice.
2 In a survey conducted by Prof Sudhanshu Bhushan of NIEPA, 61% of the teachers
admitted that less than 50% of the courses could be covered by end June.
3 Coursera, arguably, is the largest online learning platform in the world. Universities
have partnered with Coursera to deliver courses and degrees.
4 India had 3,75,055 students studying abroad as compared to only 47,424 foreign stu-
dents studying in India (http://uis.unesco.org/en/uis-student-flow (accessed on 6
March 2021).
5 The three ranking agencies use different parameters with different weights. The
THE World University Rankings use 13 indicators classified into five core perfor-
mance indicators such as teaching, research, citations, international outlook and in-
dustry income which are assigned 30, 30, 30, 7.5 and 2.5% of weights, respectively.
Six metrics are used in the QS Ranking. The components are academic reputation
(40%), faculty to student ratio (20%), citations per faculty from SCOPUS (20%),
employer reputation from global survey (10%), proportion of international students
and proportion of international faculty (5% each). For Shanghai’s ARWU, the in-
dicators are quality of education which is proxied by the alumni of the institution
winning Nobel prizes and Field Medals (10% weightage) and 20% to Nobel prizes
and Field Medals by the existing faculty, quality of faculty is assessed by giving 20%
weightage to the ‘most cited’ researchers in 21 broad subject categories compiled
by Thomson Reuters, and 20% to papers published in Nature and Science. Out of
the three, the QS is the least research metric dependent and the methodology being
survey dependent to the extent of 50%; arguably, it suffers from subjectivities and
instability.
References
Bhushan, S. 2020. Covid 19 and Higher Education in India: A Survey Report. New Delhi:
NIEPA. https://theprint.in/india/education/90-lakh-govt-college-students-cant-
access-online-lessons-report-states-urges-aid/469631/ accessed on 6.10.2020.
Chattopadhyay, S. 2020. NEP 2020: An Uncertain Future for Indian higher Education.
Economic and Political Weekly, Vol. LV, No. 46, 21 November 2020, 23–27.
Chattopadhyay, S. and A. Sharma. 2019. A Neoliberal Approach to the Policy Making
in Pradip Kumar Biswas and Panchanan Das (eds.) Indian Higher Education during the
Post-Liberalisation Era. Singapore: Springer, 289–314.
de Wit Hans and Altbach, Philip G. 2020. Internationalization in Higher Education:
Global Trends and Recommendations for Its Future, Policy Reviews in Higher Educa-
tion, DOI: 10.1080/23322969.2020.1820898
Economist Intelligence Unit (The). 2020. New Schools of Thought: Innovative Models for De-
livering Higher Education, A Report by the Economist Intelligence Unit. https://www.
qf.org.qa/eiu#:~:text=In%20’New%20Schools%20of %20Thought,institutions%20
in%20today’s%20changing%20world; https://www.fimpes.org.mx/covid19/images/
banners/doctos/HEQatar.pdf
Glennerster, H. 1991. Quasi-Markets for Education?, Economic Journal, Vol. 101, No. 408,
September, 1268–1276.
Government of India. 2017. University Grants Commission (Declaration of Govern-

ment Educational Institutions as Institutions of Eminence) Guidelines 2017. https://
www.ugc.ac.in/pdfnews/2170800_Guidelines-for-Educational-lnstitutions-as-
lnstitutions-of-Eminence-2017.pdf (accessed on 12.08.2020).
Government of India. 2018. Ministry of Human Resource Development UGC (Catego-
risation of Universities (only) for Grant of Graded Autonomy) Regulations 2018. The
Gazette of India. 12th February 2018. (Part III, Section 4) https://www.ugc.ac.in/
pdfnews/1435338_182728.pdf (accessed on 13.08.2020)
Government of India. 2020A. The National Education Policy 2020, MHRD, New
Delhi. https://www.mhrd.gov.in/sites/upload_files/mhrd/files/NEP_Final_English.
pdf
Government of India. 2020B. Public Notice: Expression of Interest (UG Course Map-
ping/ SWAYAM) dated 31.07.2020, New Delhi: UGC. https:// www.ugc.ac.in/
pdfnews/ 4981711_ Public-Notice-EOI.pdf (accessed on 11.08.2020).
Government of India. 2020C. Office Memorandum-Popularisation of Courses-reg,
Ministry of Education, Department of Higher Education. https://www.ugc.ac.in/
pdfnews/ 0049565_SWAYAM_letter_final1.pdf accessed on 5.10.2020
Government of India. 2020D. National Institutional Ranking Framework (NIRF), web-
site: https://www.nirfindia.org/2020/OverallRanking.html.
International Facts and Figures. 2020. Universities UK International. Available online:
https://www.universitiesuk.ac.uk/policy-and-analysis/reports/Pages/international-
facts-figures-2020.aspx (accessed 1 Dec 2020).
International Finance Corporation (IFC). 2020. Toward Ending Education Privilege (Case
Study), Washington, DC: World Bank Group.
Marginson, S. 2011. The New World Order in Higher Education Research Rankings,
Outcomes Measures and Institutional Classifications in Michele Rostan and Mas-
similiano Vaira (eds.) Questioning Excellence in Higher Education: Policies, Experiences
and https://www.timeshighereducation.com/world-university-rankings/2020/world-ranking#!/
page/0/length/25/locations/IN/sort_by/rank/sort_order/asc/cols/stats Challenges in National
and Comparative Perspectives. Rotterdam: Sense Publishers.
Marginson, S. 2016. The Global Construction of Higher Education Reform in Karen
Mundy, Andy Green, Bob Lingard and Antoni Verger (eds.) Handbook of Global Edu-
cation Policy. 291–311. West Sussex: Wiley Blackwell.
Marginson, S. 2020. The Relentless Price of High Individualism in the Pandemic. Higher
Education Research & Development, DOI: 10.1080/07294360.2020.1822297
McClure, M. W. 2015. MOOCs: Hype or Hope? Conflicting Narratives in Higher Ed-
ucation Policy in Pavel Zgaga, Ulrich Teichler Hans G. Schuetze and Andrä Wolter
(eds.) Higher Education Reform : Looking Back-Looking Forward, Higher Education Re-
search and Policy, Vol. 8, 385–400. Frankfurt am Main: Peter Lang Edition.
McCowan, T. 2017. Higher Education, Unbundling and the End of the University as We
Know It. Oxford Review of Education, Vol. 4396, 733–748.
Ouellet, M. and E. Martin. 2018. University Transformations and the New Knowledge
Production Regime in Informational Capitalism, triple-C, Vol. 16, No. 1, 78–96.
https://www.triple-c.at/index.php/tripleC/article/download/875/1091?inline=1
Palfreyman, D. and P. Temple. 2017. Universities and Colleges: A Very Short Introduction.
Oxford: Oxford University Press.
QS Ranking. 2020. QS World Ranking. https://www.topuniversities.com/university-
rankings/world-university-rankings/2020.
Reddy A., S. Jose, and R. Vaidehi. 2020. Of Access and Inclusivity: Digital Divide in
Online Education in Economic and Political Weekly, Vol. 55, No. 36, 23–26.
Shanghai Jiao Tong University Ranking. 2020. http://www.shanghairanking.com/

Academic-Ranking-of-World-Universities-2020-Press-Release.html
Stiglitz, J. E. 1975. The Theory of Screening, Education, and the Distribution of Income.
The American Economic Review, Vol. 65, No. 3, 283–300.
Stiglitz, J. E. 1999. Knowledge as a Global Public Good in Inge Kaul, Isabelle Grunberg,
and Marc A. Stern (eds.) Global Public Goods: International Cooperation in the 21st Cen-
tury. New York: Oxford University Press, 308–325.
Teichler, U. 2015. The Future of Higher Education: A View Reflecting the State and
the tasks of Higher Education Research in P. Zgaga, U. Teichler, H. G. Schuetze and
A. Wolter (eds.) Higher Education Reform : Looking Back-Looking Forward, Higher Ed-
ucation Research and Policy, Vol.8, Frankfurt am Main: Peter Lang Edition. Times
Higher Education World University Ranking (THE) (2020). Available online:
https://www.timeshighereducation.com/world-university-rankings/2020/world-
ranking#!/page/0/length/25/locations/IN/sort_by/rank/sort_order/asc/cols/stats
(accessed on 4 March 2021).
Wadhwa, R. 2019. Pathways to Internationalization in Indian Higher Education: Re-
flections on Policy Options in S. Bhushan (ed.) The Future of Higher Education in India.
Singapore: Springer, 307–328.
Winston, G. C. 1999. Subsidies, Hierarchy and Peers: The Awkward Economics of
Higher Education. The Journal of Economic Perspectives, Vol. 13, No. 1 (Winter), 13–36.
21
LOCKDOWNS, MIGRANTS, AND THE
SPATIAL DISTRIBUTION OF COVID-19
CASES IN INDIA
Consequences of a “tough and timely decision”
Zakir Husain and Richa Kothari
21.1 Introduction
Lockdowns have been a major instrument in containing the spread of Covid-19
in many countries. It has been recognised that lockdown is a drastic public health
measure (Habibi et al., 2020; Sands et al., 2016), with far-reaching consequences
and adverse health outcomes (Poletto et al., 2014; Stone, 2020). The adoption
of such an extreme step requires a careful analysis. In India, however, a national
lockdown was introduced from 25th March giving a prior notice of a few hours,
and without adequate planning. As a result, there have been several unintended
adverse consequences on Indian society and economy. The crisis relating to mi-
grant workers, in particular, has been extensively documented in the media. In
a belated attempt to resolve this crisis, the Indian Railways introduced special
trains intended to transport the migrant workers and their families, stranded at
their places of work, back to their original residence. In this chapter, we argue
that the lack of preparedness and poor handling of the process of reverse migra-
tion drastically modified the spatial distribution of Covid-19. While Covid-19
was initially restricted to a few districts in the western side of the sub-continent,
the special trains led to a change in the spatial distribution of Covid-19 cases,
with the pandemic spreading to eastern and north-eastern states.
Migration is a phenomenon where people move from one place to another in
search for better opportunities like better education, better jobs, better medical
facilities, or for family and marriage. In India, there are an estimated 45.58 crore
migrants. While women comprise the majority of such migrants (68%), work
(9%) and education (1%) also motivate the movement of population between
regions. Most of the migrant workers are employed as daily wage labourers in
manufacturing and construction, in the unorganised sector. The sudden an-
nouncement of the lockdown (MHA Order No. 40-3/2020-D dated 24/3/2020)
DOI: 10.4324/9781003220145-25
354 Zakir Husain and Richa Kothari
led to the closure of factories and workplaces, leading to a loss of income, un-
certainty about the future, and, in many cases, homelessness. As these migrant
workers remit the greater part of their income back to the places of origin, they
had little cash reserves. The Government of India claimed to have sufficient food
grains with Food Corporation of India to meet demand for the next one and a
half year. It announced distribution of extra rations, part of which was free. The
ration cards of the migrant workers – entitling them to subsidised entitlements
of rice, wheat, and pulses – were generally left back home. Unable to access the
public distribution system, the migrant workers faced food shortages. In Delhi,
for instance, migrant workers were shown feeding on the rotten fruit dumped by
the Yamuna (Archana, 2020). There are also media reports of migrant workers
facing police brutality in their attempts to procure food from the market infring-
ing the lockdown (Nathanael, 2020).
From April, an exodus from the places of work in cities like Mumbai, Delhi,
and Surat started. In the hope that they would get income under the Mahatma
Gandhi National Rural Employment Guarantee Scheme (MNREGA) scheme,
the migrant workers sought to return to the safety of their homes. As public
transport was not available, they had to walk the huge distance back to their
origin – sometimes carrying their children on their backs or shoulders – through
forests, and over rivers, travelling in groups. During their long march, the mi-
grant workers faced the hostility of the state apparatus. Many were arrested for
violating the lockdown; in Gujarat, 120 workers were beaten up by the Gujarat
police and dumped in Maharashtra on 31st March (Shantha, 2020). The Ut-
tar Pradesh government formed special police teams to prevent migrants from
crossing state borders; those who somehow entered into Uttar Pradesh were
sprayed with disinfections or confined in small rooms (Gupta, 2020). Some of
them died of exhaustion, dehydration, or starvation. A heart-rending instance
is that of Jeeta Madkami, a 12-year-old girl who walked over 100 kilometres in
three days from Peruru village in Telangana in an attempt to return to her native
home in Bijapur district, Chhattisgarh. She died of dehydration and electrolyte
imbalance within 11 kilometres of her home (Mishra, 2020; Rao, 2020). Others
died of accidents. The case of 16 workers who were run over by a freight train
in Aurangabad on 8th May, when they had fallen asleep on the railway tracks,
exhausted from their journey made national headlines (Banerjee & Mahale,
2020). But many others also died when the vehicles carrying them had accidents.
Turning a Nelson’s eye to this humanitarian crisis, the Supreme Court accepted
the assurance of the Solicitor General on 31st March that “there is no person
walking on the roads in an attempt to reach his/her hometowns/villages”, and
a three-member bench commented that “it was not possible for the court to
monitor who is walking and who is not walking”. Another bench commented
on the Aurangabad tragedy, “how can anybody stop this when they sleep on the
railway tracks?” (Chhokar, 2020).
In face of persistent criticism, the Government of India finally took cogni-
sance of the plight of the migrant workers. Special trains (dubbed Shramik trains)
Spatial distribution of Covid in India 355
were announced to transport the migrant workers. Between 1st May and 9th
July, 4,165 special trains were operated, ferrying about 63 lakh migrant work-
ers. While Indians stuck abroad had been brought back without charging any
airfare, the migrants – most of whom had by now reached the bottom of their
meagre savings – were charged the normal sleeper class fare, plus an additional
Covid-19 surcharge of Rs. 50 (L. Agarwal, 2020). Although Mr. Lav Agarwal,
Joint Secretary of the Union Health Ministry, cryptically referred to a 85% sub-
sidy on 4th May, in reality, cash-strapped state governments either had to pay
this price or charge the migrants. The trains were over-packed, passengers were
deprived of water, and were given meagre allotments of food, which was often
rotten. Even though all passenger traffic was closed, inefficient spacing and lack
of co-ordination resulted in congestion on routes, and inordinate delays; some of
the trains often took circuitous routes travelling for several days before reaching
their destinations (Ameen, 2020). Some trains did not reach their destinations
at all, but dumped their passengers in a totally different place.1 Not surprisingly,
several deaths occurred on the Shramik trains (Dhingra, 2020). The chaos also led
to another disaster, with serious public health implications.
Historically, population movements have been known to spread disease from
one geographical area to another. For instance, the outbreak of yellow fever in
Philadelphia in the summer of 1793 is attributed to the influx of refugees from
the sugar cane-growing French colony of Haiti (Kim, 2014). Migration also
explains the recent spread of malaria to new areas (Martens & Hall, 2000). Re-
cent studies have also linked migration with the transmission of Covid-19 across
geographical regions. For instance, the spread of Covid-19 from Wuhan to other
areas has been linked to migrant workers returning to their native places (Chen
et al., 2020; Jia et al., 2020; Li & Ma, 2020; Shen, 2020; Shi et al., 2020).
The government had announced that adequate checks at the points of origin
would ensure that only asymptomatic passengers were travelling, while similar
screening would take place at the destination to identify Covid-19 cases and
quarantine them. In reality, such checks were not properly undertaken. The di-
version of trains to unscheduled destinations, unexpected stops, and late arrivals
also contributed to unmonitored disembarkation. Consequently, these migrant
labourers became the unwitting carriers of Covid-19 from hotspots like Maha-
rashtra, Gujarat, and Delhi to create new epicentres in states like West Bengal,
Bihar, Orissa, Chhattisgarh, and Assam from where the migrant workers had
originated. Studies undertaken by Indian researchers have also tried to link rising
incidence of Covid-19 to in-migration using econometric models (Kumar, 2020;
Lee et al., 2021). However, such studies were undertaken in the early phases of
Covid-19 transmission; consequently, the results are not very robust. Moreo-
ver, such studies have simply presented results of econometric models, without
undertaking any exploratory study of the spatial pattern in Covid-19 in India.
The present study applies spatial statistical methods to explore spatio-temporal
variations in the geographical patterns in Covid-19 using district-level data. The
study period is from 1st May to 2nd August.
21.2 Methodology
21.2.1 Research objective and hypotheses

The study investigates the association between Shramik trains and the change in
spatial distribution of Covid-19 active cases in India. Starting with an analysis
of the spatio-temporal distribution of active cases in India, we will consider
whether the change may be attributed to the migrant policy of the government.
We will examine whether the increase in active cases in eastern states starts with
the launching of the Shramik trains, and whether the incidence of active cases is
higher in states with high out-migration rates. Specifically, we will assess the
validity of three hypotheses:
H1: There has been a change in spatial distribution in Covid-19 cases, with a
sharp rise in active cases in eastern and north-eastern states.
H 2: The increase in Covid-19 cases in eastern and north-eastern parts of India is
associated with the introduction of Shramik trains.
H3: Incidence of Covid-19 active cases is higher in districts with high out-
migration rates.
21.2.2 Database
Data on district- and state-wise Covid-19 cases is publicly available on the
site https://www.COVID19india.org/. District- and state-level data were
downloaded from https://bit.ly/3gUIHwv and https://bit.ly/3kHEbnu, re-
spectively. The Government of India emphasises on case fatality rates and
recovery rates as indicators of success in controlling the pandemic. However,
both these two indicators have limitations. The authenticity of the reported
death rates is doubtful in a country where only 22% of deaths are certi-
fied; further, the death of Covid-19 patients with co-morbidities is often
not included within Covid-19 deaths (Biswas, 2020). For the same reason,
the recovery rate – confirmed cases less deaths – is also not a suitable marker
of progress of the pandemic. This study uses the number of active cases as a
measure of the spread of Covid-19. The justification of using active cases is
that it indicates the pressure on health infrastructure. In case of comparison
between regions, however, the number of active cases has to be adjusted for
variations in population. Hence, we have estimated active cases per million,
when comparing across regions.
Migration data from the National Sample Survey Office survey on migra-
tion (National Sample Survey Office, 2010) was taken from Chinmay Tumbe’s
webpage (https://sites.google.com/site/chinmaytumbe/home/migration-links).
In addition, we have used data for the control variables used in regression mod-
els from the District Census Handbooks (https://censusindia.gov.in/2011census/
dchb/DCHB.html).
21.2.3 Methodology
To answer the first question, viz. is there evidence of a change in the geograph-
ical pattern of the pandemic in India, we have used choropleth maps. These are
maps which use differences in shading or colouring within predefined areas to
indicate the average values of a particular quantity in those areas. The breaks can
be defined using quintiles, maximising within-group homogeneity, or custom-
ised divisions. As we want to observe the change in spatial pattern over time,
we have used the latter variety, applying the same categories in different time
periods. Such maps are called natural break maps.
In the next step, we estimated Global and Local Moran’s indices. The nature
of spatial distribution – whether it is random, or diversified, or concentrated – is
given by the value of Global Moran’s I (Moran, 1950), while Anselin’s Local Mo-
ran’s I (the most widely used Local Indicator of Spatial Association, LISA) iden-
tifies districts exhibiting spatial autocorrelation with their neighbouring districts
(Anselin, 2010). Statistical inference is based on Monte Carlo randomisation test
at 999 permutations. A limitation of the Local Moran’s I is that it is based on
a decomposition of the Global index. The Getis-Ord gi*(d) statistic, however,
identifies hotspots and the cold spots by looking at each characteristic feature of
the dataset within the context of its neighbouring features (Getis & Ord, 2010;
Ord & Getis, 2010). For a hotspot to be statistically significant, it has to be sur-
rounded by neighbours that are hotspots as well.
To answer the second question, viz. to ascertain whether there is an increase
in Covid-19 cases in eastern states after the starting of Shramik trains, we have
used Interrupted time series (ITS) models (Soumerai et al., 2015; Wagner et al.,
2002). It is a tool suitable for evaluating the effectiveness of large-scale public
health interventions (or shock) at a population level, in the absence or presence
of suitable counterfactuals. It has been widely used in public health interventions
such as vaccination, evaluation of health impacts of unplanned global financial
crisis, effect of smoking ban in public places on hospital admissions for acute
coronary events, etc.
The ITS model can be presented as follows:
Yt = β0 + β1T + β 2 X t + β 3TX t + εt (21.1)

where
Yt : aggregated outcome at time t,

T: time since the start of the study,
X t : a dummy variable indicating the pre-intervention period (coded as 0) or
post-intervention period (coded as 1),
β0 represents the intercept or the initial value of the outcome at T = 0,
β1 represents the underlying pre-intervention rate of temporal change,
β2 represents the level change following the intervention,
β3 indicates the slope change following the intervention, and
εt is the random error term, following a first order autoregressive [AR(q)] process,
where q is the appropriate number of lags.
In the present case, the starting of the Shramik trains (1st May) and the unlock
dates (1st June and 1st July) with a 14-day lag (i.e. 15th June and 15th July) are
taken as shocks. The regression model is, therefore,
Y = β1 + β 2 TIME + β 3TRAIN + β4 TRAIN × TIME + β5UL1

+ β6 UL1 × TIME + β7 UL 2 + β8 UL 2 × TIME (21.1a)
when
Y: active cases per million,

TIME: time since the start of study,
TRAINS: dummy variable representing the intervention period (pre-
intervention periods 0, otherwise 1); in our case, it represents the start of
Shramik trains on 1st May,
TRAINS × TIME: interaction of TRAINS and TIME variables,
UL1: dummy variable representing the intervention period (pre-intervention
periods 0, otherwise 1); in our case, it represents the date of unlock 1 with a
lag of 14 days i.e. on 15th June,
UL1 × TIME: interaction of UL1 and TIME variables,
UL2: dummy variable representing the intervention period (pre-intervention
periods 0, otherwise 1); in our case, it represents the date of unlock 2 with a
lag of 14 days i.e. on 15th July,
UL2 × TIME: interaction of UL2 and TIME variables.
The model is estimated for West Bengal, Bihar, Jharkhand, Orissa, Chhattisgarh,
and Assam.
Finally, to confirm the causal relationship between reverse migration and ac-
tive cases per million, we have estimated regression models. The datasets used in
this study have a hierarchical structure such that the Covid-19 cases are clustered
within districts, which, in turn, are clustered within states. The states may also
be clustered within the geographical regions. Hence, using normal econometric
models like linear regression would not give robust results – multi-level models
are appropriate in such cases. The multi-level models are estimated for three
dates, viz. beginning of May, beginning of July, and beginning of August. The
model estimated in the beginning of May gives an overview of the Covid-19
situation in the pre-reverse migratory phase, whereas the model estimated in the
latter dates reflects the post-reverse migration situation. The Shramik trains had
started on 1st May and ended on 9th July.
The multi-level regression can be presented as follows:
Yt = β0 + β1X t + ρU t + εt (21.2)
where
Yt: the aggregated outcome at time t; in our case, it is active cases per million,
Xt: is a covariate matrix for the fixed effects β; in our case, it is
L_ACTIVEM: lagged values of active cases per million,
MIGRANT: percentage of households with at least one migrant for eco-
nomic reason,
PLIT: percentage of literates to total population,
ASHA: percentage of villages with Accredited Social Health Activists in the
district,
DISTANCE: mean distance of villages from town weighted by female
population,
ρ: the covariate matrix for the random effects u.
The error vector (ε) is assumed to be a multi-variate normal with mean zero and
variance matrix σ 2 ε.
Initially, we estimated a two-level model, assuming that districts are clustered
within states. To check for robustness of our results, we added Region as another
level to create a three-level model. In this structure, districts are nested within
states, which are nested in regions.2
Finally, to take into account the varying impact of these variables across dis-
tricts, Geographically Weighted Regression (GWR) models are used (Ried et al.,
2017). The coefficients in this model vary with the geographical coordinates of
the locality, facilitating a spatial analysis of the coefficients. Assuming that there
are n observations, for observation i ∈ {1, 2,…, n} at location (ui, vi), the model is
yi = β0 (ui , vi ) + ∑ j β j (ui , vi ) xij + εi ,
where βj indicates the jth coefficient (Fotheringham et al., 2017). We have re-
gressed active cases per million on MIGRANT, PLIT (literacy level), ASHA
(percentage of districts covered by Accredited Social Health Activists), and DIS-
TANCE (mean distance between villages and nearest town). The model was
estimated using data for May, July, and August.
The exploratory spatial analysis was undertaken using Luc Anselin’s GeoDa
Version 1.14, the econometric analysis was carried out in Stata 14, and the GWR
model was estimated using the GWR Version 4.0.90 package (Nakaya et al.,
2009).
21.3 Results and discussion
21.3.1 Has regional distribution of active Covid-19 cases changed?

The number of active cases in India has increased over time. As on 16th Septem-
ber, the confirmed number of Covid-19 cases crossed the 5.1 million mark, and
the number of new daily cases fell slightly below the one lakh mark. The growth
rate, too, has increased. Although India took 167 days to reach the mark of 1
million cases, which is the slowest among the worst hit countries, the spread of
the disease has been racing ever since. In the next 21 days, India’s count reached 2
million. The pace accelerated even more, with the next 3, 4, and 5 million cases
being reported after 23 days, 16 days, and 13 days, respectively. After the US,
India has become the second country to cross the mark of 5 million cases. The
active cases in India are the second highest in the world with almost 1 million ac-
tive cases of Covid-19 on 16th September. Given the rate at which India is adding
numbers, it is even expected to overtake the US in terms of confirmed number
of Covid-19 cases in the first week of October. What is more important is the
geographical distribution of Covid-19 cases in India – with almost 80% of active
cases being reported from only ten states viz. Maharashtra, Andhra Pradesh,
Tamil Nadu, Karnataka, Uttar Pradesh, Delhi, West Bengal, Telangana, Bihar,
and Odisha. Although initial epicentres of the novel coronavirus spread were at
Tamil Nadu, Gujarat, and Delhi, now the disease is widespread throughout the
country with Maharashtra, Karnataka, Andhra Pradesh, and Chhattisgarh grow-
ing faster than the national average.
However, if we adjust active cases by population, then a pattern is revealed
(Figure 21.1). In May and June, the eastern and north-eastern states were rela-
tively Covid-19 free. But, from June onwards, the number of cases per million
persons started to rise in these two regions, particularly in the north-eastern
1200
1000
Active cases per million
800
600
400
200
0
May June July August
Central North West South East North-East
FIGURE 21.1 Regional distribution of active cases per million over time.
states. In contrast, the incidence of Covid-19 has gone down in North and Cen-
tral India. The southern states also experienced a sharp increase in cases, and have
overtaken western India since July.
The change in geographical pattern becomes clearer when we use natural
break choropleth maps (Figure 21.2). Initially, most of the districts are in the
green category (with active cases below 0.25 per million); only a few districts on
the left-hand side of the sub-continent had a relatively higher incidence of active
cases per million persons. By the end of May, however, we see a larger number
of districts in the blue category; further, quite a few are in the eastern side of
the country. The number of districts in the green category reduces in June and
July. By August, we find that the western and southern states have become hot-
spots; more importantly, the pandemic has spread extensively in the eastern and
north-eastern states, with a few hotspots in these states also.
FIGURE 21.2 District-wise distribution of active cases per million population.

FIGURE 21.3 LISA maps for active cases per million population.
Estimates for Global Moran’s I do not indicate the presence of distinct spa-
tial clusters. It is statistically significant and high only in July (0.29, z = 13.90)
and August (0.19, z = 8.26), indicating a concentrated spatial distribution of
Covid-19. Anselin’s local I’s for May shows a large cold spot (low-low pattern)
in the north-eastern part of India, and in Chhattisgarh (Figure 21.3). In August,
however, the low-low cluster in the north-eastern region disappears, indicating
the spreading of pandemic to this region. Cold spots are now observed in districts
located in northern, north-western, and central parts of India, while hotspots
(High-high) are observed in Mumbai-Pune, Andhra Pradesh-Telangana, and
Orissa belts. Most of the cold spots in the eastern and north-eastern regions are
also observed to have disappeared.
A more detailed analysis is provided by Getis-Ord g i*(d) statistics (Figure 21.4).
The number of cold spots increases over the study period by about 31%; in con-
trast, the number of hotspots is estimated to rise from 76 districts to 116 districts
(i.e. by 53%).
The spatial location of the hotspots too has changed over time (Figure 21.5).
At the beginning of May, hotspots defined over a radius of 500 kilometres are
restricted to Maharashtra. By August, however, the hotspots have covered the
entire southern part of India. The eastern and north-eastern parts of India have
become cold spots. It implies that in-migration has, if at all, a highly localised
effect – a result that we will examine later.
21.3.2 Out-migration and incidence of Covid-19

Using ITS regression models, we examine whether the start of the Shramik trains
is associated with an increase in Covid-19 cases in these states. The spikes from
reverse migration inflow spurts are quite visible in the graphs for the states bar-
ring Chhattisgarh and Orissa (Figure 21.6). The effect is apparent, but less pro-
nounced, in Bihar; it may reflect the low testing in Bihar in May–June.
Regression results, given in Table 21.1, provide specific information about the
impact of shocks and whether the resultant impact is statistically significant or
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
April May June July August
ColdSpot Neutral Hotspot
FIGURE 21.4 Change in hotspots and cold spots between May and August.
FIGURE 21.5 Choropleth maps for Getis-Ord gi*(d) statistics.
not. The values of β3, β5, and β7 denote changes in the intercept due to the shock,
while values of β4, β6, and β8 indicate changes in slope due to the shock. The for-
mer (β3, β5, and β7) captures the immediate effect of the shock, while the latter
(β4, β6, and β8) reflects the change in trajectory of the pandemic, or the long-run
impact of the pandemic. We find that β3 is negative and significant, except for
FIGURE 21.6 Graphs for ITS model – selected states.
Odisha, while the slope (β4) increases for all the six states, indicating a clear break
from the temporal trend and the start of a new time trend. Specifically, active
Covid-19 cases are observed to increase at a faster pace after the launching of the
Shramik trains. The first unlocking of the economy has a rather ambiguous effect.
Although the intercept (β5) decreases significantly in Bihar, Orissa, and Assam,
the change is insignificant in the remaining states. The trajectory (β6) remains
the same in West Bengal, Bihar, and Jharkhand, decreases in Chhattisgarh, and
TABLE 21.1 Results of ITS model for selected states
Variable West Bengal Bihar Jharkhand Orissa Chhattisgarh Assam
TIME (β2) 35.55*** 31.55*** 6.47*** 5.48*** −0.63*** −0.78***

TRAINS (β3) −576.58* −392.87** −207.02* 16.90 −220.55** −543.37**
TRAINS×TIME (β4) 75.88*** 36.65*** 12.96*** 18.77*** 24.02*** 58.30***
UL1 (β5) −1,297.44 −1,650.75*** −320.85 −558.84* −119.50 −917.07*
UL1×TIME (β6) 80.21 26.74 6.71 93.28*** −21.41*** 105.77***
UL2 (β7) 4,898.39*** 2,018.38*** 401.24 −880.44 389.54** 1,103.62*
UL2×TIME (β8) 220.74** 530.80*** 273.73*** 359.57*** 111.87*** −33.94
INTERCEPT (β1) 300.55*** 74.89*** 34.00*** 44.35*** 8.64*** 16.02***
Statistics
N 103 103 103 103 103 103
F 643.76*** 478.24*** 218.92*** 756.89*** 283.23*** 579.23***
Probability 0.00 0.00 0.00 0.00 0.00 0.00
Note: * denotes p < 0.05, ** denotes p < 0.01, *** denotes p < 0.001.
increases in Orissa and Jharkhand. After the second unlock, the growth rate of
Covid-19 increases in all the six states. ITS models, therefore, reveal a spurt in
Covid-19 cases coinciding with the launching of Shramik trains; this trend was
aggravated by the second unlock. In the next step, we examine whether there
is an association between reverse migration and the spreading of the pandemic.
21.3.3 Is migration the cause?

The reverse migration of migrant workers back to their origin generated a
potential health hazard in the form of a surge in the number of Covid-19
cases (Maji et al., 2020). The study attempts to estimate the possible surge in
Covid-19 cases due to return of migrant workers in the states of Maharashtra,
Uttar Pradesh, Madhya Pradesh, Rajasthan, and Bihar using an extended SEIR
model. The study predicts that the surge of cases will be minimal in Mahar-
ashtra (about 1,000 active cases under the worst case scenario), Uttar Pradesh
(less than 5,000 active cases under the worst case scenario), Bihar (less than
3,000 active cases under the worst case scenario), and Madhya Pradesh (less
than 3,000 active cases under the worst case scenario) and was expected to see
a high surge of Covid-19 cases. However, the peak would be reached in mid-
or end of May in the states of Uttar Pradesh and Rajasthan. The surge may be
controlled by conducting real-time PCR (RT-PCR)-based pooled testing on
the labour force as they reached their destinations. The study does not consider
the impact of migration on eastern and north-eastern states (except on Bihar);
in fact, it does not examine the impact of migration on the spatial distribution
of Covid-19 cases. Moreover, it does not anticipate the sudden spiralling of
cases from June onwards.
In contrast, our analysis reveals a substantial spike in Covid-19 cases after the
launching of Shramik trains. We next examine whether the spurt in cases may be
attributed to the return of migrant workers. As states have not revealed the num-
ber of migrants affected by Covid-19, we had to rely on media reports. While
some reports are available for May, the press soon lost interest in this issue. But
even for May, there is circumstantial evidence of a causal relationship between
reverse migration and spurt in Covid-19 cases.3
In reality, media reports the flouting of social distancing norms during the
journey and the lack of thermal screening upon arrival. Moizee (2020) cites a
migrant worker who claims, “We got our train tickets from Ambala and we were
tested there. But when we returned to Muzaffarpur neither were we thermally
screened nor were we asked if we showed Covid-19 symptoms”. Such lapses were
reported in other parts of Bihar, where passengers were allowed to go straight
home from the station without any testing or quarantine instructions. Till 31st
May, out of the 3,800 people who had tested positive in the state, 2,500 persons
(66%) were migrants (Moizee, 2020).
In West Bengal, Covid-19 cases were initially restricted to Kolkata, and the
urban areas of Howrah, Hooghly, North 24 Parganas, and South 24 Parganas.
The flow of migrant workers has reportedly led to Covid-19 spreading in rural
TABLE 21.2 Share of migrants among affected persons – West Bengal
District Affected Migrants Migrants (%)
Birbhum 254 212 83.46

Bankura 180 128 71.11
Maldah 321 280 87.23
Uttar Dinajpur 227 197 86.78
East Burdwan 139 128 92.09
Koch Bihar 264 223 84.47
West Mednipore 300 234 78.00
Purulia 85 82 96.47
Source: West Bengal Government Health Portal, https://www.wbhealth.gov.in/.
areas of these districts, and to green districts like Birbhum, Murshidabad, Uttar
Dinajpur, Purulia, Cooch Bihar, and Maldah (Bhattacharya, 2020). Officials of
the West Bengal Health Department said that 56% of the people testing positive
for the novel coronavirus in May were migrants who returned from other states;
one out of ten migrant workers who were tested turned out to be Covid-19
positive (Moneycontrol News, 2020). Media reports that the District Magistrate
of Maldah had claimed, “In my district, the spike is due to the migrant workers
testing positive. All of them are asymptomatic. … Tracing such asymptomatic
patients and keeping them isolated is primary on our agenda” (cited in Bhat-
tacharya (2020)). Data reported by the West Bengal Health Department in June
(Table 21.2) suggests that these claims cannot be brushed aside.
In the north-east, as on 20th May, 70% of new Covid-19 cases in Assam were
migrants (Chowdhury, 2020). Kamrup Metropolitan Area, a district in Assam
that had experienced a high influx of in-migrants, became a hotspot with the
number of cases rising from 4 (in May) to over 16,000 in August (Sitlhou, 2020).
A similar situation was reported in Tripura (Leivon et al., 2020).
21.3.4 A statistical analysis of reverse migration and Covid-19 cases

Correlations between migration indicators during the operation of Shramik trains
indicate a higher incidence of Covid-19 in districts with out-migrants, compared
to districts with a low number of out-migrants (Table 21.3). Further, the corre-
lation increases over time.
We next estimate multi-level regression models to confirm the causal rela-
tionship between reverse migration and active cases per million (Table 21.4).
Transmission of Covid-19 within the locality is captured by the lagged value of
cases; they are strictly not comparable as the lag period varies between the first
model and the subsequent models. However, at all points of time, past incidence
of active Covid-19 cases is an important determinant of current active cases. It
may indicate the presence of community transmission – as warned by health
experts in a recent statement (Press Trust of India 2020). The incidence of ac-
tive Covid-19 cases per million persons is positively related to literacy levels; it
TABLE 21.3 Correlation between migration indicators and incidence of Covid-19
Migration Links 1st May 31st May 2nd July
Households with at least one migrant for 0.1975 0.2958 0.3402

economic reason (%)
Migrants as per cent of population 0.1803 0.3075 0.3195
Male migrants as per cent of population 0.2076 0.3310 0.3509
Note: All correlations are statistically significant at 5% level.
TABLE 21.4 Results of multi-level regression models
Variable Two-Level Model Three-Level Model
May July August May July August
MIGRANT 0.01 4.81 7.04 0.01 4.81 6.35

0.09 3.79 2.06 0.09 3.79 1.82
L_ACTIVEM 1.19 3.03 1.53 1.19 3.03 1.53
37.60 7.82 15.41 37.60 7.82 15.35
LITP 0.00 0.00 0.00 0.00 0.00 0.00
4.94 2.77 0.32 4.94 2.77 0.27
ASHA −0.07 −3.27 0.54 −0.07 −3.27 0.63
2.33 5.51 0.33 2.33 5.51 0.38
DISTANCE 0.02 −0.74 −2.11 0.02 −0.74 −3.37
0.75 1.27 1.33 0.75 1.27 1.30
INTERCEPT 0.69 263.81 154.00 0.69 263.82 169.51
0.27 4.91 1.02 0.27 4.91 1.09
Intercept 1.18 4.76 5.93 0.84 4.42 5.58

3.28 22.95 34.06 0.04 0.37 0.12
Var(District) 2.73 5.58 6.53 0.84 4.42 5.58
93.55 191.36 225.11 0.04 0.37 0.12
Var(Region) – – – 2.73 5.58 6.53
– – – 93.55 191.36 224.78
Statistics
N 633 633 633 633 633 629
χ2 1,651.51 190.22 305.53 1,651.51 190.22 302.77
Probability 0.00 0.00 0.00 0.00 0.00 0.00
F 5.44 33.90 70.10 5.44 33.90 69.91
Probability 0.01 0.00 0.00 0.07 0.00 0.00
Inter-class
correlations
District 0.05 0.11 0.23 – – –
Region 0.02 0.08 0.11 0.05 0.16 0.23
Note: t-ratios are reported below coefficients.

may imply that greater awareness of the population leads to greater reporting of
Covid-19 cases. In August, however, the coefficient of PLIT is insignificant at
10% level. The coefficients of ASHA are negative and significant at 1% level in
May and July, indicating that grass root health workers may have spread informa-
tion, helping to contain Covid-19. Moreover, the absolute value of the coefficient
is increasing over time. The coefficient of DISTANCE is negative, but is statis-
tically insignificant at 10% level, in all three cases. The coefficients of migration
are positive in all three models. In the model for May, however, the coefficient
is statistically insignificant. In subsequent models, the coefficient is significant
at 5% level. Further, the migration effect seems to become stronger over time,
increasing from 4.81 in July to 7.04 in August.
FIGURE 21.7 Spatial variation of coefficient of MIGRANT.

To check robustness of the results, we have added another level to create a

three-level structure. In this structure, districts are nested within states, which
are nested in regions. Results are identical – with the impact of migration be-
coming stronger over time.
Table 21.4 reveals temporal variations in the impact of MIGRANT on active
cases. It is also possible to analyse spatial variations in the causal relationship between
MIGRANT and active cases across districts using a GWR model of active cases
upon MIGRANT, controlling for LITP, DISTANCE, and ASHA (Figure 21.7).
Results for May reveal that the value of the coefficient for MIGRANT is high
in the northern and southern regions – in the states of Punjab, Haryana, Gujarat,
Rajasthan, western Uttar Pradesh, Tamil Nadu, and Karnataka. There is also a
belt in southern Bengal and Jharkhand where values of the coefficient of MI-
GRANT are high. The coefficient ranges from −9.66 to 15.02, with a mean of
0.12 and median of 0.03. In July, migration exerts a stronger effect on active cases.
The range is −42.52 to 81.98, with mean of 39.46 and median of 0.84. High coef-
ficients are observed in almost the entire southern part of India – barring Kerala –
and in Gujarat-Rajasthan belt. In the Bengal belt, the cluster shrinks, but high
coefficient values are observed in almost the entire north-eastern region. In Au-
gust, high coefficients of MIGRANT are observed in the eastern coast, starting
from West Bengal and continuing up to Tamil Nadu. It is also high in Maharash-
tra, Karnataka, Telangana, Andhra Pradesh, western Uttar Pradesh, and Madhya
Pradesh. The reopening of factories in Maharashtra, southern states, and Uttar
Pradesh in the unlocking phase led to a third round of migration – with migrants
returning to their places of work. This may have contributed to the change in
spatial distribution of the coefficients of MIGRANT in August.
21.4 Conclusion
The analysis undertaken in this study shows that there has been a change in the
spatial distribution of active Covid-19 cases over time. Initially, Covid-19 flared
up in northern and western parts of India. After May, however, the epicentres
have gradually changed. While the southern states of Tamil Nadu, Karnataka,
Andhra Pradesh, Telangana, and Kerala saw a spike in Covid-19 cases, there was
also a rise in the infections in eastern and north-eastern states. Though low testing
rates are possibly hiding the magnitude of Covid-19 cases in these states, there are
indications that the epicentre may shift from the south to eastern India in the fu-
ture. Statistical analysis of trends in active Covid-19 cases in the eastern region in-
dicates that there is a break in the trend after the reverse flow of migration started.
We also show that states with high out-migration rates exhibit high incidence of
Covid-19 active cases; further, the strength of the linkage between migration and
Covid-19 cases has increased over time, particularly in states like Bihar, Orissa,
Assam, and Jharkhand. The evidence brought out by our analysis suggests that
there are sufficient grounds for accepting all the three research hypotheses of the
study. This calls for an assessment of the migrant policy to see what went wrong,
and suggests an alternative policy that could have been adopted.
When Covid-19 was raging in Europe and the US, air connectivity was
maintained with such hotspots – without proper screening of passengers,4
thereby allowing the entry of Covid-19 patients into India without screening.
As pointed out by Abhijit Banerjee, lockdown was announced as a knee jerk
reaction, without considering whether the economy and society of the country
would permit it to continue for a sufficiently long time to contain the pandemic
(Business Today & Today, 2020). Simulations using a Susceptible-Infected-
Recovered model (Kermack & McKendrick, 1927) reveal that if the lockdown
had been postponed by a few months, there would have been a modest increase
in cases. The time gained to plan for the lockdown may have reduced the sub-
stantial human costs on Indian society. In particular, the migrants could have
been conveyed in a phased manner back to their origin. Instead, after giving
time for Covid-19 to spread among migrant workers, they were transported in
sub-human conditions, without proper screening at the origin. While the state
governments also failed in their duty in screening and monitoring the incom-
ing passengers, delayed arrival at designated checkpoints and unscheduled stops
allowed the migrants to disperse without undergoing the planned testing. Inter
alia, this changed the spatial distribution of Covid-19 and laid the foundations
for the generation of new epicentres. It has already increased the case load for
the country, fanned the rise in cases after unlocking started, and led to infections
spreading to smaller towns and rural areas (Press Trust of India, 2020). The
latter development will challenge the weak health infrastructure in small towns
and rural areas:
As COVID-19 case load increases in the other 364 districts that are largely
tier 4 or rural, the critical patients may not be able to find the needed
healthcare. The other risk that exists is that as COVID-19 spreads in rural
India and many patients visit nearby towns and big cities for treatment, it
could worsen the situation in these places with what people are calling as
the second wave of COVID-19 cases.
(LiveCircles study, cited by Vora (2020))5
In addition, multiple regional peaks may be generated, so that cross-infection

between regions can set up a self-reinforcing vicious circle. And, an example of
black humour at its best (!), after the migrant workers reached home, factories
were reopened so that the migrant workers had to return to their places of work.
Whether this will lead to a second wave in Maharashtra, Karnataka, Andhra
Pradesh, and other northern states remains to be seen.
Critics have argued that the National Democratic Alliance government, un-
der Narendra Modi, has focussed more on image building and on dramatic an-
nouncements, and less on efficient implementation. Implementation of major
economic, social, and political policy changes is normally based on extensive
planning and widespread consultative processes. The implementation of policy
changes like demonetisation, introduction of General Sales tax, abrogation of
Article 370, banning of the triple talaq, and amendment of the Citizenship Act,
however, demonstrates an identical ruthless implementation of top-down deci-

sions. An identical lack of micro-level planning, lack of transparency, and dis-
regard of stakeholders and vulnerable groups marked the migrant policy of the
Central government. It remains to be seen in the future how costly will be India’s
attempt to demonstrate its capacity for taking “timely and tough” decisions.6
Notes
1 A train to Basti, Uttar Pradesh landed up in Ghaziabad, while a Pune bound ended
up in Purulia, West Bengal ( Joshi, 2020). A train bound for Gorakhpur from Vasai
Road, Mumbai travelled through West Bengal, Orissa, Bihar, and Jharkhand; during
this 60-hour journey, the passengers got food and water only twice (Ameen, 2020;
Dhingra, 2020).
2 The regions are formed as follows: North (Chandigarh, Delhi, Haryana, Himachal
Pradesh, Jammu and Kashmir, Ladakh, Punjab, Rajasthan, Uttarakhand, and Uttar
Pradesh), North-east (Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram,
Nagaland, Sikkim, and Tripura), Centre (Madhya Pradesh and Chhattisgarh), East
(Bihar, Jharkhand, Odisha, and West Bengal), West (Dadra and Nagar Haveli and
Daman and Diu, Goa, Gujarat, and Maharashtra), and South (Andhra Pradesh, Kar-
nataka, Kerala, Puducherry, Tamil Nadu, and Telangana).
3 The sole exception is Uttar Pradesh, where the positivity rate is about 3%. During
discussion, however, NGOs commented on the highly unreliable nature of Covid
data in Uttar Pradesh.
4 In an interview to Karan Thapar, Prashant Bhushan revealed that, while the Prime
Minister had claimed that screening of air passengers had started in January, his RTI
revealed that it was started only from 4th March. Moreover, between 4th and 23rd
March, only 19% of passengers were screened (Accessed from https://thewire.in/
government/prashant-bhushan-karan-thapar-COVID-19-govt-mismanagement on
3 September 2020).
5 In this context, it should be noted that the latest round of District Level Household &
Facility Survey reported that only 25% of rural Indians have access to public outpa-
tient (OPD) healthcare. The presence of undiagnosed and untreated co-morbidities
will aggravate the situation. Also of concern is the fact that about 70% of the aged
population in India resides in rural areas.
6 The national lockdown was hailed by the Regional Director of World Health Or-
ganization in South-east Asia as a “timely and tough” decision in April (N. Agarwal,
2020).
References
Agarwal, L. (2020). Fact Check: No, the Centre Isn’t Paying for Migrant Workers’ Train Journeys
Home. The Wire. https://thewire.in/government/indian-railways-migrant-workers-
fare. Accessed 5 February 2021.
Agarwal, N. (2020, April 14). WHO praises PM Modi’s “timely and tough” decision on corona
lockdown 2.0. Livemint. https://www.livemint.com/news/india/who-praises-pm-
modi-s-timely-and-tough-decision-on-corona-lockdown-2-0-11586851502720.
html. Accessed 5 February 2021.v
Ameen, F. (2020). Coronavirus: No food and water aboard Shramik Special trains. The Telegraph.
https://www.telegraphindia.com/india/coronavirus-no-food-and-water-aboard-
shramik-special-trains/cid/1775256. Accessed 5 February 2021.
Anselin, L. (2010). Local Indicators of Spatial Association-LISA. Geographical Analysis,
27(2), 93–115. https://doi.org/10.1111/j.1538-4632.1995.tb00338.x
Archana, K. C. (2020). Migrant Workers Scavenge For Bananas Near Delhi Cremation Ground.
India Times. https://www.indiatimes.com/trending/social-relevance/lockdown-
hunger-forces-migrant-workers-to-scavenge-for-bananas-near-delhi-cremation-
ground-510986.html. Accessed 5 February 2021.
Banerjee, S., & Mahale, A. (2020). 16 Migrant Workers Run Over by Goods Train Near
Aurangabad in Maharashtra. The Hindu. https://www.thehindu.com/news/national/
other-states/16-migrant-workers-run-over-by-goods-train-near-aurangabad-in-
maharashtra/article31531352.ece. Accessed 5 February 2021.
Bhattacharya, S. (2020, May 27). Returning Migrants Cause Surge of Covid-19 Cases in
Rural Bengal. Hindustan Times. https://www.hindustantimes.com/india-news/
ret ur n ing-m ig rants-cause-surge-of-cov id-19-cases-in-r ura l-benga l/stor y-
2wXPrIdPWNIpAXBLbRs3eM.html. Accessed 5 February 2021.
Biswas, S. (2020). India Coronavirus: The “Mystery” of Low Covid-19 Death Rates. BBC
News. https://www.bbc.com/news/world-asia-india-52435463. Accessed 5 February
2021.
Business Today, & Today, B. (2020, August). India Erred in Imposing Lockdown Restric-
tions; Entered, Exited Too Soon, Says Abhijit Banerjee. Business Today. https://www.
businesstod ay.i n/cur rent/economy-pol it ics/i nd ia- coronav i r us-lockdow n-
restrictions-entered-exited-too-soon-abhijit-banerjee/story/413101.html. Accessed 5
February 2021.
Chen, H., Chen, Y., Lian, Z., Wen, L., Sun, B., Wang, P., Li, X., Liu, Q., Yu, X., Lu, Y., Qi,
Y., Zhao, S., Zhang, L., Yi, X., Liu, F., & Pan, G. (2020). Correlation between the Migra-
tion Scale Index and the Number of New Confirmed Coronavirus Disease 2019 Cases in
China. Epidemiology and Infection, 148, e99. https://doi.org/10.1017/S0950268820001119
Chhokar, J. S. (2020). Migrant Worker Crisis: The Supreme Court Has Abdicated All Respon-
sibility. The Wire. https://thewire.in/law/supreme-court-migrant-workers-crisis-
abdicate-responsibility. Accessed 5 February 2021.
Chowdhury, R. (2020). Coronavirus Assam: COVID-19 Cases Rise In Assam As Migrants
Return, New Cases In Quarantine Centres. NDTV. https://www.ndtv.com/india-news/
coronavirus-assam-covid-19-cases-rise-in-assam-as-migrants-return-new-cases-in-
quarantine-centres-2231982. Accessed 5 February 2021.
Dhingra, S. (2020). “Lost” Trains, Dead Passengers — Railways Blames “Extraordinary
Times” for Shramik Train Mess. The Print. https://theprint.in/india/lost-trains-dead-
passengers-railways-blames-extraordinary-times-for-shramik-train-mess/430698/.
Accessed 5 February 2021.
Fotheringham, A. S., Yang, W., & Kang, W. (2017). Multiscale Geographically Weighted
Regression (MGWR). Annals of the American Association of Geographers, 107(6), 1247–
1265. https://doi.org/10.1080/24694452.2017.1352480
Getis, A., & Ord, J. K. (2010). The Analysis of Spatial Association by Use of Distance Statis-
tics. Geographical Analysis, 24(3), 189–206. https://doi.org/10.1111/j.1538-4632.1992.
tb00261.x
Gupta, A. (2020). Coronavirus: Migrant Workers Sprayed with “Disinfectants” on Return to
UP’s Bareilly, Oppn Attacks Yogi Govt. Jagran English. https://english.jagran.com/india/
coronavirus-migrant-workers-sprayed-with-disinfectants-on-return-to-ups-bareilly-
oppn-attacks-yogi-govt-10010186?utm_expid=.W6HdjhiBQ-ml0nTAajwI9g.0&utm_
referrer=https%3A%2F%2Fwww.google.com%2F. Accessed 5 February 2021.
Habibi, R., Burci, G. L., de Campos, T. C., Chirwa, D., Cinà, M., Dagron, S., Eccleston-
Turner, M., Forman, L., Gostin, L. O., Meier, B. M., Negri, S., Ooms, G., Sekalala,
S., Taylor, A., Yamin, A. E., & Hoffman, S. J. (2020). Do not violate the Interna-
tional Health Regulations during the COVID-19 outbreak. The Lancet, 395(10225),
664–666. https://doi.org/10.1016/S0140-6736(20)30373-1
Jia, J. S., Lu, X., Yuan, Y., Xu, G., Jia, J., & Christakis, N. A. (2020). Population Flow
Drives Spatio-Temporal Distribution of COVID-19 in China. Nature, 582(7812),
389–394. https://doi.org/10.1038/s41586-020-2284-y
Joshi, T. (2020). 40 Shramik Trains Lost Their Way, Railways Busy Providing Clarification. HW
English. https://hwnews.in/news/national-news/40-shramik-trains-lost-way-railways-
busy-providing-clarification/137847. Accessed 5 February 2021.
Kermack, W. O., & McKendrick, A. G. (1927). A Contribution to the Mathematical The-
ory of Epidemics. Proceedings of the Royal Society of London. Series A, Containing Papers
of a Mathematical and Physical Character, 115(772), 700–721. https://doi.org/10.1098/
rspa.1927.0118
Kim, S. (2014). Control Discourses and Power Relations of Yellow Fever: Philadelphia
in 1793. Korean Journal of Medical History, 23(3), 513–541. https://doi.org/10.13081/
kjmh.2014.23.513
Kumar, A. (2020). Modeling Geographical Spread of COVID-19 in India Using Network-based
Approach (27 April; Preprint). https://doi.org/10.1101/2020.04.23.20076489
Lee, J. N., Mahmud, M., Morduch, J., Ravindran, S., & Shonchoy, A. S. (2021). Migra-
tion, Externalities, and the Diffusion of COVID-19 in South Asia*. Journal of Public
Economics, 193, 104312. https://doi.org/10.1016/j.jpubeco.2020.104312
Leivon, J., Deb, D., & Agarwala, T. (2020). Behind Northeast’s Covid-19 Surge, Stranded
Migrants Returning Home. The Indian Express. https://indianexpress.com/article/
north-east-india/behind-northeasts-covid-19-surge-stranded-migrants-returning-
home-6457181/
Li, B., & Ma, L. (2020). Migration, Transportation Infrastructure, and the Spatial Transmission
of COVID-19 in China. http://qianxi.baidu.com/
Maji, A., Choudhari, T., & Sushma, M. B. (2020). Implication of Repatriating Migrant
Workers on COVID-19 Spread and Transportation Requirements. Transportation Re-
search Interdisciplinary Perspectives, 7. https://doi.org/10.1016/j.trip.2020.100187
Martens, P., & Hall, L. (2000). Malaria on the Move: Human Population Movement
and Malaria Transmission. Emerging Infectious Diseases, 6(2), 103–109. https://doi.
org/10.3201/eid0602.000202
Mishra, R. (2020). 12-Yr-Old Migrant Worker Dies After Walking 100km. Hindustan Times.
https://www.hindustantimes.com/india-news/12-yr-old-migrant-worker-dies-
after-walking-100km/story-8LHJHm2F61mDJZXqGd4SGP.html. Accessed 5
February 2021.
Moizee, S. (2020, June 2). No Testing for Coronavirus or Quarantine: Returning Migrants
Decry Bihar’s Apathy. The Quint. https://www.thequint.com/videos/news-videos/is-
bihar-turning-a-deaf-ear-to-the-covid-19-pandemic-returning-migrant-workers-
not-tested-coronavirus. Accessed 5 February 2021.
Moneycontrol News. (2020, June 17). Coronavirus Pandemic: 56% COVID-19 Cases in
West Bengal Are Migrant Workers. https://www.moneycontrol.com/news/trends/
coronavirus-pandemic-56-covid-19-cases-in-west-bengal-are-migrant-workers-
5419251.html. Accessed 5 February 2021.
Moran, P. A. P. (1950). Notes on Continuous Stochastic Phenomena. Biometrika, 37(1/2),
17. https://doi.org/10.2307/2332142
Nakaya, T., Fotheringham, A. S., Charlton, M., & Brunsdon, C. (2009). Semiparametric
Geographically Weighted Generalised Linear Modelling in GWR4.0, in B. G. Lees
and S. W. Laffaan (eds), Proceedings of 10th International Conference on Geocomputation:
1–5. http://www.geog.leeds.ac.uk/groups/geocomp/2009/PDF/Nakaya_et_al.pdf.
Accessed 19 June 2021.
Nathanael, M. P. (2020). Licensed to Beat, Abuse and Kill. The Hindu. https://www.
thehindu.com/opinion/op-ed/licensed-to-beat-abuse-and-kill/article31609626.ece
National Sample Survey Office. (2010). Migration in India: 2007–08. Ministry of Statistics
and program Implementation, Govt. of India: New Delhi.
Ord, J. K., & Getis, A. (2010). Local Spatial Autocorrelation Statistics: Distribu-
tional Issues and an Application. Geographical Analysis, 27(4), 286–306. https://doi.
org/10.1111/j.1538-4632.1995.tb00912.x
Poletto, C., Gomes, M. F., y Piontti, A. P., Rossi, L., Bioglio, L., Chao, D. L., Jr, I. M. L.,
Halloran, M. E., Colizza, V., & Vespignani, A. (2014). Assessing the Impact of Travel
Restrictions on International Spread of the 2014 West African Ebola Epidemic. Euro-
surveillance, 19(42). https://doi.org/10.2807/1560-7917.ES2014.19.42.20936
Press Trust of India. (2020, August 1). Covid-19 Cases Being Increasingly Reported from Small
Towns, Rural Areas. The Economic Times. https://timesofindia.indiatimes.com/india/
covid-19-cases-being-increasingly-reported-from-small-towns-rural-areas-experts/
articleshow/77858226.cms. Accessed 5 February 2021.
Rao, S. (2020). Tribal Girl, 12, Treks 100 km in 3 Days from Telangana, Dies before Reaching
Home. Times Of India. https://timesofindia.indiatimes.com/city/hyderabad/tribal-
girl-12-treks-100-km-in-3-days-from-telangana-dies-before-reaching-home/
articleshow/75283506.cms. Accessed 5 February 2021.
Ried, M., Eicher, M.-M., Neu, R., Sziklavari, Z., & Hofmann, H.-S. (2017). Evaluation
of the New TNM-Staging System for Thymic Malignancies: Impact on Indication
and Survival. World Journal of Surgical Oncology, 15(1), 214. https://doi.org/10.1186/
s12957-017-1283-4
Sands, P., Mundaca-Shah, C., & Dzau, V. J. (2016). The Neglected Dimension of Global
Security — A Framework for Countering Infectious-Disease Crises. New England
Journal of Medicine, 374(13), 1281–1287. https://doi.org/10.1056/NEJMsr1600236
Shantha, S. (2020). Gujarat Police to Probe Allegation That Migrant Workers Were Forced Into
Container Trucks. The Wire. https://thewire.in/government/gujarat-police-inquiry-
migrant-workers-container-trucks. Accessed 5 February 2021.
Shen, J. (2020). Analyzing the Determinants of the Spread of Covid-19 among the Provincial
Regions in China. https://doi.org/10.21203/rs.3.rs-26465/v1
Shi, Q., Dorling, D., Cao, G., & Liu, T. (2020). Changes in Population Movement Make
COVID-19 Spread Differently from SARS. Social Science & Medicine, 255, 113036.
https://doi.org/10.1016/j.socscimed.2020.113036
Sitlhou, M. (2020). In just two weeks, coronavirus has surged 59% among Assam’s
healthw orkers. Scroll.in, 16 August. https://scroll.in/article/970358/in-just-two-
weeks-coronavirus-has-surged-by-59-among-assams-health-workers. Accessed 19
June 2021.
Soumerai, S. B., Starr, D., & Majumdar, S. R. (2015). How Do You Know Which Health
Care Effectiveness Research You Can Trust? A Guide to Study Design for the Per-
plexed. Preventing Chronic Disease, 12, 150187. https://doi.org/10.5888/pcd12.150187
Stone, J. (2020). Why Travel Bans Don’t Work During An Outbreak Like Coronavirus. Forbes.
https://www.forbes.com/sites/judystone/2020/02/01/why-travel-bans-dont-
work-during-an-outbreak-like-coronavirus/#531610b853ea
Vora, R. (2020, August 19). Survey Finds Smaller Towns Unprepared to Handle Fast-
Spreading Covid-19. The Hindu BusinessLine. https://www.thehindubusinessline.
com/news/national/sur vey-f inds-smaller-towns-unprepared-to-handle-fast-
spreading-covid-19/article32392618.ece. Accessed 5 February 2021.
Wagner, A. K., Soumerai, S. B., Zhang, F., & Ross-Degnan, D. (2002). Segmented
regression analysis of interrupted time series studies in medication use research.
Journal of Clinical Pharmacy and Therapeutics, 27(4), 299–309. https://doi.org/
10.1046/j.1365-2710.2002.00430.x
22
LABOUR AND THE PANDEMIC
A study on work, employment, and work
situation
Lopamudra Banerjee and Snehashish Bhattacharya
22.1 Introduction
In this expository essay, we look at the uneven patterns of outcomes for working
populations during the Covid-19 pandemic and posit that the crisis has fore-
grounded some critical developments in the world of work. Using the analytical
lens of work situation, we examine these developments in the Global North and
the Global South. At first glance, the crisis reveals what Dodds et al. (2020) have
described as structural inequities, which produce adverse effects against some
more than others. On a closer look, however, the crisis brings out in sharp relief
certain particulars about the current situations of work and employment that of-
ten remain substratal. We hope to bring in this chapter a certain understanding of
what these particulars tell us about the conditions of labour in the contemporary
world of work.
We develop our arguments in the following manner. In the next section, we
briefly review how differences in Covid outcomes are linked to occupational
differences in current understanding. This perspective highlights the structural
asymmetries that drive risk distributions among various strata of working popu-
lations. While recognising this phenomenon, we, however, seek to emphasise a
separate dimension of this risk asymmetry. Over the past few decades, there has
been an increasing disjuncture between the notions of ‘work’ and ‘employment’,
with certain fundamental and wide-ranging changes taking place in the nature
of work and relations of employment. These categories are being reconfigured in
different ways in Global North and Global South. The changes, we argue, have
significant bearings on how the impacts of the pandemic are experienced by dif-
ferent segments of working populations. Drawing on this understanding, in the
third section of the chapter, we posit our notion of ‘work situation’ as an alterna-
tive vantage point to study the emerging forms of segmentations, precarisation,
DOI: 10.4324/9781003220145-26
Labour and the pandemic 377
and informalisation in the world of work.1 We employ this lens to illustrate, in

the light of Covid-19 pandemic, the specificities in the conditions of precarity
in Global North and Global South at the current juncture of global capitalism.
We summarise our main observations in the concluding section of the chapter.
22.2 Work and the pandemic
22.2.1 Occupational exposure, employment relations,

and unequal outcomes
Of the various ways in which Covid-19 shocks have adversely affected the work-
ing populations engaged in remunerative work, we consider two: Loss in liveli-
hood from loss of work and/or loss of steady flow of income, and loss in health
from disease exposure.2 Work shapes relative risks of losses through two chan-
nels. First, work arrangements determine how economic impacts of the pan-
demic and enforced virus-containment measures (like lockdown and quarantine)
affect retention and earning prospects of individuals; second, the nature of work
determines the possibilities of voluntary physical distancing in workplaces and
reduced interpersonal contacts.
We are, by now, familiar with these details. Indeed, in many studies, the link
between work conditions and Covid-19 outcomes is so emphasised that ‘occu-
pational exposure’ is used as a metric to assess the likelihood of disease exposure
and economic losses (Koh, 2020). In our essay, we recognise this reality. But we
posit that while a straightforward attribution of outcome inequalities to occupa-
tional differences is useful for descriptive purposes, a more nuanced understand-
ing of the empirical findings is possible if we go beyond the typography of risks
and outcomes in simple terms of occupational categories. For that, a study of the
underlying drivers of deprivation and advantage is called for.
A traditional approach in this regard has been to trace the conditions of un-
equal outcomes back to the positions that working individuals hold in the hi-
erarchical ranking within economic relations of production and exchange. The
approach is premised on a particular idea of employment relations that became
standard in the industrialised economies of the Global North in much of the
twentieth century following the Fordist model of production and accumulation.
In this approach, the structural positions of individuals are defined in terms of two
parameters: Endowments of productive resources and control over labour power.
The parameters are variously explored in Wright (1978) and Goldthorpe (1987),
among others.3 In this understanding, the twin parameters generate a structure
of stratification in the workforce. It is reasoned that those that are advantaged by
their greater access to ‘resource’ and ‘control’ have better standards of living and
more secure livelihood conditions. These conditions enable individuals to ward
off risks in times of crises. In contrast, the working individuals disadvantaged by
their deficits in ‘resource’ and ‘control’ have poorer material conditions of living
under normal conditions, and are more vulnerable to covariate shocks.
378 Lopamudra Banerjee and Snehashish Bhattacharya
Invoking this traditional approach, studies on uneven patterns of Covid-19

outcomes have been carried out for working populations across various countries
(see, for example, Holst et al., 2020; Jain et al., 2020; St-Denis, 2020). The com-
mon findings of these studies are that structurally disadvantaged individuals are
much more exposed to Covid-19-induced economic hardship and health losses.
Empirical evidence further shows a disturbing overlay between economic condi-
tions of vulnerability and social conditions of marginalisation. A disproportion-
ately higher number of individuals engaged in low-paying insecure work belong
to historically excluded identity groups.
These evidences force us to acknowledge the distributional asymmetries and
social inequities that predate the pandemic. On first blush, much of the discus-
sions on uneven outcomes of Covid-19 thus appear as discussions on structural
inequalities and discriminations. But as Covid-19 throws light on the institutions
and practices in which the subjects of the pandemic are embedded, several details
about contemporary conditions of work and employment also become clear to us.
22.2.2 Disjuncture between work and employment

There has been an increasing rupture between the notions of ‘work’ and ‘em-
ployment’ both in Global North and in Global South over the past few decades.
The Covid crisis illuminates how the ensuing changes in the nature of work and
employment relations bring about risk differences across different segments of
working populations. For instance, evidences from Global North suggest that the
health and livelihood losses are significantly higher for segments of working pop-
ulations with insecure and informal work arrangements, i.e. with work arrange-
ments outside traditional Fordist employment relations. Further, the proportion
of this segment of population in the total labour force is rather large and is grow-
ing over time. In the US, for example, an increasing number of workers are pri-
marily employed under ‘nonstandard’ – i.e. subcontracted, temporary, on-call,
on-demand, or freelance – work arrangements (BLS, 2017). In Global South,
where vast majority of working populations operate outside the formal (Ford-
ist) conditions of employment relations, the traditional ‘Northern’ perspective
did not have much perch to begin with. While most agrarian activities in these
economies are informal in nature, majority of the working population even in
non-agricultural sector operate under informal work arrangements (ILO, 2013).
To make sense of how the pandemic affected different segments of working pop-
ulations under these conditions, it is, therefore, not sufficient to talk in terms of
occupational exposure or even the relative deficits in people’s access to ‘resource’
and ‘control’. Our argument may become clearer if we take a critical look at how
the world of work has been changing under contemporary capitalism.
In its traditional characterisation in the context of a capitalist economy, re-
munerative work in the sphere of market completely overlaps with the concept
of employment. In the Global North, employment relations are usually marked
by formal (legally enforceable) employment contracts, direct relation between
productivity and wages, decent work conditions, job security, and binding la-
bour regulations. This specific nature of work/employment, a principal feature
of the Fordist economic structure since the post-War period, emerged as the
universal prototype for ‘standard’ work relations under capitalism. It has been
implicit that ‘authentic’ capitalist development makes it feasible for major sections
of working population to attain higher standards of living through expansion
and generalisation of such work/employment relations.
The economies of the Global South, in contrast, are traditionally theorised as
dual economies, with a fundamental fracture in their economic landscape. The
capitalist sector, where wage employment is the basic category through which
the notion of remunerative work is articulated, is a relatively small segment of
the economy, surrounded by a vast ‘traditional’/‘precapitalist’ segment, where
such notion of employment is absent. The process of capitalist development in
such ‘backward’ economies was expected to break down this dualism, leading to
the spread of formal wage labour relations across the economy and, thereby, the
emergence of ‘standard’ capitalist employment relations as the universal category
of work.
The ‘standard’ features of capitalist employment relations have, however, un-
dergone fundamental changes over the past few decades, with a fracture between
‘work’ and ‘employment’ emerging in the Global North, and the dualism be-
coming more entrenched in the Global South. The onset of globalisation, in-
ternational outsourcing and subcontracting, and large-scale expansion of digital
economy have brought in their wake a breakdown of Fordist structures in the
Global North, along with a weakening of labour unions and collective bargain-
ing. The relative decline in availability of stable manufacturing jobs and corre-
sponding expansion of low-end service jobs, along with rising informalisation
of employment relations, have subverted the traditional norms of employment.
This has increasingly ruptured the link between labour productivity and wages.
In the Global South, on the other hand, the anticipated transition of economic
structure did not take place. As a strand of recent work shows, there has, instead,
been a reconstitution of dualism with the process of capitalist development. A
direct implication of this ‘absence’ of transition is that the separation between
work and employment persists (Sanyal, 2007; Bhattacharya, 2019; Kesar & Bhat-
tacharya, 2020).
Thus, ‘standard’ employment relations have not emerged as the universal de-
scriptor of the world of work. Instead, even in Global North, there has been
an emergence of economic dualism, which was traditionally considered to be
a hallmark of underdevelopment (Temin, 2018). A corollary of this develop-
ment has been the growing pervasiveness of ‘wageless work’ (Denning, 2010;
Shaw & Waterstone, 2020). A related phenomenon has been the emergence of
platform or gig economy across the world. Individuals labouring and earning
livelihoods in such economies are not in ‘standard’ employment relations; nor are
they entrepreneurs in the traditional sense of the term. This workforce includes
freelancers, agency workers, independent contractors, contingent and temporary
workers (Woodcock & Graham, 2020). In the US, as of 2017, approximately

34% of workforce was employed as gig workers (BLS, 2017). The ‘non-standard’
workers at the upper tier, especially the so-called cloud/crowd workers, may
have formal ‘control’ over their direct means of labour and their own labour
power, but face increasingly precarious livelihood conditions. At the lower tier,
the ‘non-standard’ workers lack both ‘resource’ and ‘control’. But these individu-
als are not covered by the customary employment protection legislation extended
to the category of ‘worker’.
In the Global South, segmentations in working populations have become even
more entrenched. Cleavages have deepened between workers with ‘standard’
employment in the formal capitalist sector, and those with either ‘non-standard’
employment relations or working outside the space of employment relations
as non-capitalist self-employed petty producers/traders (Bhattacharya & Kesar,
2020). These petty producers/traders, who organise production or exchange
without employing any wage labour and often with the help of unpaid fam-
ily labour, continue to comprise a large proportion of the working population
(Ibid.). While they have ownership over their meagre resources and control over
their labour power, their livelihood conditions remain vulnerable. They are
neither capitalist entrepreneurs wielding control over labour power of others,
nor are they transformed into employees (wage labour) with formal employment
contracts through the anticipated transition process of capitalist development.4
Further, a majority – and an increasing proportion – of wage workers, both in
formal and in informal sectors, comprise a largely informalised workforce with
‘non-standard’ employment relations. In India, for example, as of 2017–18, only
23% of the workforce was employed in regular salaried work, of which 71% did
not have a written job contract (NSSO, 2019). Moreover, between 2004–05
and 2011–12, such informalised salaried employment grew by more than 9%,
while regular formal employment rose only by 3% (ILO, 2018). Such workforce,
while participating in wage employment, does not bear the same signifiers of
capitalist employment as imagined on the basis of ‘standard’ Fordist employment
relations.
There is also a disquieting pattern of overlap between ‘work’ and ‘identity’.
The more vulnerable work positions tend to be more frequently occupied by
people with marginalised social identities. This stylised fact is observed in both
Global North and Global South. In the US, for example, individuals from mar-
ginalised identity groups (Black, Latinx, immigrants, and others) are signifi-
cantly more likely to occupy low-end and precarious work positions, with this
likelihood increasing over time (BLS, 2017). This has importantly contributed to
the increasing inequality among the population in terms of both income and pos-
sibilities for social and economic mobilities. In India, similarly, there is a strong
correspondence between the precarious conditions of livelihoods and marginal-
ised identities. Marginalised groups (in terms of caste, ethnicity, gender, etc.) are
largely over-represented in low-paying occupations and under-represented in
high-paying ones (CSE, 2018).
Thus, we observe a new dualism in the world of work in Global North, even
as the existing dualism is being reconfigured and re-established in Global South.
What is then, perhaps, called for is a ‘Southern’ perspective, which is now be-
coming increasingly relevant even in the ‘Northern’ context (Bhattacharya &
Kesar, 2020). Covid-19 crisis brings out in sharp relief these developments that
often remain substratal. Our proposition in this essay is that much of the uneven
patterns of outcomes we observe for working populations in this Covid moment
are manifestations of these developments.
22.3 Work situations, precarity, and the pandemic
22.3.1 Work situation

A granular understanding of these uneven patterns of outcomes, and the risk
asymmetries among working individuals, is possible if we examine their work
situations. We use ‘work situation’ as theoretical shorthand to indicate the location
of a person within social processes of labour. Of the vectors that might indicate
this location, we consider three: The nature of work performed by an individual,
the circumstances under which work is performed, and the position of the performer
of work within the economic processes of production and distribution.
By ‘nature of work’, we refer to the set of technical tasks and duties that
one performs to facilitate production and exchange of commodities. The dif-
ferent tasks require different skills and specialisations, involving different fields
of knowledge and know-how. They entail the use of different sets of tools, ma-
chines, and material inputs. Depending on the nature, technological dependence
of work varies, and so do the levels of interpersonal contact required at work-
place. These differences are particularly relevant in the Covid-19 context. The
differences generate differential possibilities for people to participate in work
from remote locations and to maintain social distance in workplace.
‘Circumstances of work’ indicate how work is organised in terms of for-
mal and informal agreements. Circumstances reflect the norms of work and the
mutual expectations of individuals participating in work. They may also re-
flect the nature and levels of remunerations, benefits and entitlements, and the
extent of autonomy and flexibility at workplace. All these factors define the
degree of work stability, income sustainability, legal or social protections, and
extent of empowerment at work. Oftentimes, work circumstances may tally up
with endowments of productive resources. Differences in work circumstances
have important implications for disparities in abilities of individuals to cope with
livelihood and health risks posed by Covid-19.
By ‘position’ of an individual in the process of labour, we refer to their class
position, i.e. their specific location in the process of production, appropriation,
and distribution of surplus (Resnick & Wolff, 1987). Class position can be de-
termined depending on whether a working person is a direct producer of sur-
plus, or its appropriator, or a participant in the process of distribution of surplus.
For instance, in the space of commodity production, an individual could be a

worker who produces surplus, or a capitalist who appropriates and distributes this
surplus, or a self-employed who produces and appropriates their own surplus, and
so on. The class position of an individual has bearing on the specific ways in which
risks generated by Covid-19 affect them and their ability to endure such risks.
The intersections and specificities in ‘nature’, ‘circumstances’, and ‘positions’
define work situations of individuals. These dimensions are shaped by – and, in
turn, shape – the various economic, political, and cultural processes that affect
the world of work. Work situations may differ across, as well as within, the
following groups, given the extant occupational structure for non-agricultural
workforce in both Global North and Global South5: (1) Corporate owners and
managers, (2) small business owners employing wage labour (including informal
wage labour, particularly in the Global South), (3) salaried or independent pro-
fessionals, (4) long-term or ‘permanent’ workers participating in the formal sec-
tor (including unionised workers), (5) independent or self-employed individuals
participating in platform/gig economy, (6) informalised, contractual, short-term,
or outsourced labour (including wage workers in platform/gig economy), and (7)
petty producers or traders (particularly in the context of Global South).6 Catego-
ries (1)–(4) entail ‘standard’ employment arrangements, with obvious hierarchic
differences between employers and employees (although small business owners
employing informal wage labour may be characterised as informal employers as
well). Categories (5)–(7), in contrast, entail ‘non-standard’ work arrangements,
mostly comprising the informal economy.
We posit that work situations give rise to a particular configuration of out-
comes for individuals in the context of their work. The ‘positions’ of individuals,
for example, determine their control over surplus, which, in turn, affects their
chances of avoiding unfavourable outcomes in periods of crisis. However, ‘cir-
cumstances’ may vary within each position, determining whether individuals are
able to retain the surplus and use it in their favour. For example, while categories
(5) and (7) may appropriate their own surplus, they may not be able to determine
its retention and distribution, given their work circumstances. This will compro-
mise their ability to cope with livelihood shocks. Individuals in categories (1) and
(2), and the independent professionals in (3) may, in contrast, be better situated
in this regard, given their work circumstances. Similarly, workers in category
(4) may be better-off, in general, than those in (6), given the differences in their
work circumstances. On the other hand, variations in ‘nature’ of work determine
the differential ability of individuals with similar ‘positions’ and ‘circumstances’
to ward off adverse health outcomes. For example, among those with otherwise
‘better’ work situations (in terms of position and circumstance), there may be
differential possibilities for technologically mediated remote work. Furthermore,
for those with ‘non-standard’ and/or precarious work situations (in terms of posi-
tion and circumstance), access to employment and livelihoods, however tenuous,
may be possible even during crisis depending on the nature of work (e.g. in the
platform/gig economy).
Our proposition is that the differences in the outcomes of Covid-19 shocks we

are observing across the different segments of working populations are an articu-
lation of this underlying multiformity in work situations. This can be illustrated
by drawing from the vast number of case studies, primary surveys, and other
empirical work that have emerged on the impact of the pandemic on labour.
22.3.2 Work situations, uneven outcomes, and precarity

The experiences of individuals working outside ‘standard’ employment relations,
both in Global North and in Global South, provide relevant instances. In the
upper tier, we have the independent self-employed, many of them participating
in the platform/gig economy. Given their nature of work, these individuals of-
ten experience lower disease exposure (Rogers et al., 2020). Given their work
circumstances and class positions, however, precarity in livelihood conditions
has extended into this stratum as well (Schwander, 2019; Vlandas, 2020). These
livelihood conditions are subjected to two opposing forces during the pandemic.
Work and income flows for these individuals have increased as many enterprises
moved operations online with Covid outbreak. Contrastingly, the sharp eco-
nomic downturn has reduced the use of online labour platforms alongside other
types of non-standard work. The terms ‘distancing bonus’ and ‘downscaling loss’
are used to capture these opposing effects (Stephany et al., 2020). In the initial
phases of the pandemic, the former outweighed the latter; over time, however,
downscaling loss dominated. Simultaneously, these individuals started facing a
tighter market as the number of workers seeking income through online labour
platforms grew with company closures and business shutdowns.
The pandemic, however, has unmasked the lore of entrepreneurial culture at
the upper tier of non-standard work. Recent scholarship has drawn attention to
how high-skilled individuals are forced to turn to entrepreneurship as their last
recourse in situations of joblessness and redundancies (Serviere, 2010). Individu-
als with this work situation do not employ wage workers, are often home-based
and engaged in work based on low capital outlay, and have unstable income flows
arising from uncertain demand for their products and services (Block & Sandner,
2009). These are classic features of petty commodity production in the Global
South. While these forms of self-employment have always been the recourse of
the working poor in the Global South, the emerging dualism in the world of
work in the Global North has necessitated these new types of self-employment,
and thereby unleashed a process of precarisation of even high-skilled individu-
als operating at frontiers of modern technology. Even as such self-employment
is valorised as a sign of resilience and creative risk-taking, the Covid situation
points, rather, to increased economic vulnerability and distress for these individ-
uals (Comunian & England, 2020).
In the lower tier of non-standard and informalised work, the livelihood losses
from unemployment or reduced hours in the Global North have been substan-
tially higher for those that work in brick-and-mortar establishments, with nurses
and medical workers being notable exceptions due to the specific nature of their
work (Adams-Prassl et al., 2020). In the US, for example, in the initial phase
of the pandemic, unemployment in brick-and-mortar establishments rose by as
high as 12 percentage-points for women and 9 percentage-points for men (Ibid.).
Individuals working through online platforms as on-demand gig workers (espe-
cially in frontline delivery services), in comparison, experienced relatively steady,
and sometimes increased, work flows (Apouey et al., 2020). A major flux has
been observed in this tier. Those with flexible skills moved from non-essential to
essential services for subsistence. While this survival-strategy ensured livelihood,
it automatically increased the risk of disease exposure in both Global North and
Global South (Ibid.; Nair, 2020).
The possibility of remote work, however, does not exist for all. In the US, for
example, telework is possible for only 7% of workers in accommodation and food
services, as compared to 86% of workers in professional, scientific, and technical
services (Dingel & Neiman, 2020). Furthermore, while 30% of non-Hispanic
white workers report being able to work remotely, only 20% of non-Hispanic
black and 16% of Hispanic workers are able to do so (BLS, 2019).
Much of the work in the lower tier entails face-to-face interactions and is
‘geographically-tethered’ (viz. location-dependent) (Spurk & Straub, 2020). Per-
versely, there has been an attempt to revalorise these so-called ‘low-status’/‘low-
skill’ work at the Covid moment. However, given the nature of such work, risk
of disease exposure tends to be higher in this tier, even though the increased
risk is not matched by ‘hazard pay’. While these individuals are ‘forced to have
the freedom’ to take risks even without monetary fall-back options or commu-
nitarian/social security networks, their remunerations do not reflect the returns
to risk. Furthermore, given the over-representation of historically marginalised
identity groups engaged in work of this nature, an identity-driven pattern in
adverse health outcomes is observed in both Global North and Global South. In
the US, for example, Rogers et al. (2020) find that the prevalence of deaths from
Covid among non-Hispanic black (21%) is disproportionately higher than their
share in total population (12%). We can speculate that a similar pattern would be
observed in the Global South as well, given the high correspondence between
marginalised identities and lower tier jobs (for India, see CSE, 2018). The condi-
tions of vulnerability of such low tier informal workers have, therefore, further
exacerbated during the pandemic (FAO, 2020).
Further, given their work circumstances, individuals engaged in non-standard
work neither qualify as employees entitled to employment benefits, nor do they
have effective control over their own surplus. This gives rise to a situation of
endemic precarity and lack of ability to cope with livelihood and health shocks.
In the US, for example, one-in-ten full-time workers does not have access to an
employer-provided paid sick leave plan, while six-in-ten of the part-time workers
do not have such access (Causa & Cavalleri, 2020). Those in such non-standard
work arrangements in the Global North also may not sufficiently benefit from
government support programmes launched to promote labour market recovery
from the pandemic, as neither are they officially classified as employees nor do
many of them have sufficient tax-obligations to be classified as self-employed
(Apouey et al., 2020).7 In the Global South, absence of health coverage has left
this segment financially vulnerable against health shocks. Further, in the absence
of social protection or adequate direct transfer programmes, there has been an
unprecedented increase in inequality across various segments of informal work-
ing populations, with severe deepening of poverty at the lower tier. In India, for
example, a survey of vulnerable populations across 12 major states during the
period of lockdown find that about 50% of such households did not receive any
form of cash transfers, with the exclusion rates being particularly high for mar-
ginalised communities (Kesar et al., 2020).
For a vast segment of self-employed petty producers/traders in the informal
economy in the Global South, there have been severe livelihood losses in the
initial phases of enforced lockdowns, irrespective of the nature of their work. In
India, for example, about 90% of respondents in the survey cited above reported
loss of employment during this period, along with a 54% overall drop in earnings.
Those in the lower quintiles of income distribution experienced higher drop in
earnings compared to those in the higher quintiles, exacerbating the already
stark pre-Covid income inequality (Ibid.). This economic distress forced them to
resume work at the risk of health loss as soon as lockdowns were lifted. For the
informalised contractual daily wage workers, the pandemic brought widespread
job losses and furloughs, with chances of call-back varying according to the nature
of their work. If one looks through the lens of identity, this phenomenon appears
starker, given the gender and caste asymmetry across works of different nature.
In the context of India, while about 35% of men were affected in terms of their
employment status during Covid lockdown and in immediate post-lockdown
period (April–September 2020), the corresponding figure for women was 70%.
Further, while 37% of women who lost jobs during the lockdown remained un-
employed post-lockdown, among the men the same was true only for about 7%
(Abraham et al., 2021b). In terms of caste, the magnitude of employment loss for
the lower-ranked caste groups was three times higher than that for the higher-
ranked groups (Deshpande & Ramachandran, 2020).
The workers with relatively better economic and social backgrounds, espe-
cially in urban areas, were impelled into frontline delivery work of the gig econ-
omy or onto petty trading. For those without these backgrounds, and especially
those who had migrated out of rural-hinterlands in search of livelihood, the
pandemic generated a humanitarian crisis. The migrant workers, who have their
living conditions tethered to their work in urban areas, found their very daily
existence threatened when pushed out of employment (Adhikari et al., 2020).
For sheer survival, the workers have been forced to re-join the workforce by
reducing their reservation wages to a distressingly low level. While in terms of
official macro-level data these churnings may show up as spurts of economic ac-
tivities and celebrated as indicators of recovery, household- and individual-level
surveys highlight the push-factors rather than pull-factors of distress-led work
force participation. In India, for example, while 95% of the December 2019 level
workforce was back in employment by August 2020, almost one in every three of
those who were formally employed in April 2020 became (or, rather, compelled
to become) self-employed in the informal economy by August 2020 (World
Bank, 2020). Other studies have reported about 42% of those who were daily
wage workers in December 2019 becoming self-employed by August, most likely
due to lack of regular work opportunities (Abraham et al., 2021a).
In comparison, the experiences of working populations operating inside
‘standard’ employment relations as long-term or ‘permanent’ workers, both in
Global North and in Global South, are observed to be somewhat less precarious.
While their work circumstances reduced relative economic risks, the nature of
work reduced health risks. For the salaried professionals, the more common phe-
nomenon has been furloughing rather than outright loss of livelihood. In many
cases, however, there has been a decline in salary and benefits, including cost-
of-living adjustments or retirement contributions from employers. In the UK,
for example, 15% of salaried employees had been furloughed from their main
job, while only 3% lost their main job. There was, however, a clear gradient in
patterns of furloughing across the pre-Covid earnings distribution. Almost 30%
of the bottom quintile suffered furloughs compared to less than 8% of the top
quintile (Gardiner & Slaughter, 2020). Further, conditions of work for many of
these individuals have changed, along with changes in how workloads are spread
over the day. Working hours have increased, on average, without a commensu-
rate increase in compensations. While work-from-home allows the flexibility
not available earlier, labour hour is encroaching on what would have previously
been leisure time, with the burden on women being disproportionately high,
given the gendered norms of unpaid household and care work (Alon et al., 2020;
Deshpande, 2020).
While it is possible to provide several other illustrations, our purpose here has
been to explicate the following key argument. The heterogeneous outcomes for
different segments of working populations in this Covid moment are a manifes-
tation of disparate nature of work, work circumstances, and class positions. These
dimensions reinforce and contradict each other to bring forth a configuration of
outcomes that reveal the entrenchment of dualism in the world of work both in
Global North and in Global South, which has deepened the conditions of pre-
carity for a vast majority of working population.
22.4 Conclusion
Our purpose in this essay has been to explore what the Covid crisis reveals about
the conditions of labour under contemporary capitalism. While the dual shocks
of pandemic and lockdown have affected working populations everywhere, the
effects are not the same for everyone. A growing body of literature shows that
differences in occupation can explain the differences in individuals’ risks of eco-
nomic and health losses from the shocks. This approach provides a valuable way
to think about how the Covid outcomes differ across different segments of work-
ing populations. Yet, to appreciate why these differences exist, we need to look
deeper into the conditions of precarity.
The causal roots of vulnerability have often been explained in terms of po-
sitions people hold within employment relations that have come to be consid-
ered as the ‘standard’ in the post-War period in Global North. This approach,
however, does not fully reveal the precarities experienced by a rising number
of working individuals in Global North engaged in wageless work and with
non-standard work arrangements. Nor can they say much about the experiences
of vast majority of working populations in Global South operating outside stand-
ard employment relations.
Our argument in this essay has been that Covid crisis makes this disjuncture
between ‘work’ and ‘employment’ vividly visible. The unevenness in Covid-19
risk, and the observed inequality in economic and health outcomes, can be seen
as articulations of this phenomenon. To explore this, we have invoked the notion
of work situation, as understood in terms of the nature of work, the circum-
stances under which work is performed, and the position of the performer of
work within the economic processes of production and distribution. We have
sought to illustrate that work situations have important bearings on economic
and health outcomes for working individuals in both Global North and Global
South. This analytical lens provides us with a broader perspective on conditions
of labour under global capitalism at the present juncture.
Notes
1 The expression ‘work situation’, as developed in the stratification literature, captures
social relations in workplace (Lockwood, 1958). We look beyond this benchmark
definition in this essay.
2 In this essay, we do not focus on the important realm of unpaid work. Unpaid work,
often carried out in the form of household labour, has significant gender dimensions.
The relative risks associated with such work are often distinctly different from those
related to remunerative work.
3 Critical theoretical distinctions, of course, exist in the approaches in Wright and
Goldthorpe. While Wright locates the issue of employment relations in the context
of production structures, Goldthorpe is attentive to market structures. Here, we rhe-
torically swipe the approaches together to make our larger points.
4 The unpaid family labour in these non-capitalist enterprises, however, neither have
control over resources or their own labour power, nor are they wage workers em-
ployed through formal contractual agreements.
5 The agricultural sector has certain specificities that we are not exploring in this essay,
given space constraints.
6 Categories (1) and (2) appropriate the surplus produced by their workers, whereas
independent professionals in (3), and those in (5) and (7) appropriate their own sur-
plus. Categories (4) and (6), however, produce surplus that is appropriated by their
employers.
7 The recently legislated Families First Coronavirus Response Act in the US aim to
provide some support to non-standard workers by allowing them access to paid sick
leave for up to two weeks (Causa & Cavalleri, 2020).
References
Abraham, R., Basole, A., & Kesar, S. (2021a). Tracking employment trajectories in the Covid-19
pandemic: Evidence from Indian panel data. (Centre for Sustainable Employment Working
Paper No. 35). Bengaluru: Azim Premji University.
Abraham, R., Basole, A., & Kesar, S. (2021b). Down and out: The gendered impact of the
Covid-19 pandemic on India’s labour market. (Centre for Sustainable Employment Work-
ing Paper No. 40). Bengaluru: Azim Premji University.
Adams-Prassl, A., Boneva, T., Golin, M., & Rauh, C. (2020). Inequality in the impact of
the coronavirus shock: Evidence from real time surveys (IZA Discussion Paper No. 13183).
Bonn: Institute of Labour Economics (IZA).
Adhikari, A., Goregaonkar, N., Narayanan, R., Panicker, N., & Ramamoorthy, N.
(2020). Manufactured maladies: Lives and livelihoods of migrant workers during
COVID-19 lockdown in India. The Indian Journal of Labour Economics. https://doi.org/
10.1007/s41027-020-00282-x.
Alon, T., Doepke, M., Rumsey, J. & Tertile, M. (2020). The impact of COVID-19 on
gender equality. (National Bureau of Economic Research Working Paper Series No.
26947). Cambridge, MA: National Bureau of Economic Research (NBER).
Apouey, B., Roulet, A., Solal, I., & Stabile, M. (2020). Gig workers during the COVID-19
crisis in France: Financial precarity and mental well-being. Journal of Urban Health.
https://doi.org/ 10.1007/s11524-020-00480-4.
Bhattacharya, R. (2019). Labour beyond the labour market: Interrogating marginality. In
Workers and margins: Grasping erasures and opportunities, ed. N. Jammulamadaka, 45–62.
Singapore: Palgrave Macmillan.
Bhattacharya, S., & Kesar, S. (2020). Precarity and development: Production and labour
processes in the informal economy in India. Review of Radical Political Economics, 52(3),
387–408.
Block, J., & Sandner, P. (2009). Necessity and opportunity entrepreneurs and their duration in
self-employment: Evidence from German micro data (SOEP Papers on Multidisciplinary
Panel Data Research. No. 191). Berlin: Deutsches Institut für Wirtschaftsforschung
(DIW).
Bureau of Labor Statistics (BLS). (2017). Contingent and alternative employment arrangements.
Washington, DC: United States Department of Labor.
Bureau of Labor Statistics (BLS). (2019). Job flexibilities and work schedules – 2017–2018 data
from the American time use survey. Washington, DC: United States Department of Labor.
Causa, O., & Cavalleri, M. C. (2020). How non-standard workers are affected and protected dur-
ing the Covid-19 crisis: Stylised facts and policy considerations. Centre for Economic Policy
Research. Available at: https://voxeu.org/article/how-non-standard-workers-are-
affected-and-protected-during-covid-19-crisis. Accessed on: 28 November 2020.
Centre for Sustainable Employment (CSE). (2018). State of working India 2018. Bengaluru:
CSE, Azim Premji University.
Comunian, R., & England, L. (2020). Creative and cultural work without filters:
Covid-19 and exposed precarity in the creative economy. Cultural Trends, 29(2),
112–128.
Denning, M. (2010). Wageless life. New Left Review. 66 (November–December), 79–97.
Deshpande, A. (2020). The Covid-19 pandemic and lockdown: First order effects on gender gaps
in employment and domestic time use in India (GLO Discussion Paper No. 607). Essen:
Global Labour Organization (GLO).
Deshpande, A., & Ramachandran, R. (2020). Is COVID-19 “the great leveler”? The critical
role of social identity in lockdown-induced job losses (Ashoka University Discussion Paper
No. 34). Sonipat: Ashoka University.
Dingel, J. I., & Neiman, B. (2020). How many jobs can be done at home? (National Bureau of
Economic Research Working Papers No. 26948). Cambridge, MA: National Bureau
of Economic Research (NBER).
Dodds, K., Castan Broto, V., Detterbeck, K., Jones, M., Mamadouh, V., Ramutsindela,
M., Varsanyi, M., Wachsmuth, D., & Yuan Woon, C. (2020). The COVID-19 pan-
demic: Territorial, political and governance dimensions of the crisis. Territory, Politics,
Governance, 8(3), 289–298.
Food and Agriculture Organization (FAO). (2020). Impact of COVID-19 on informal work-
ers. Rome: FAO. Available at: http://www.fao.org/3/ca8560en/CA8560EN.pdf. Ac-
cessed on: 28 November 2020.
Gardiner, L., & Slaughter, H. (2020). The effects of the coronavirus crisis on workers: Flash find-
ings from the resolution foundation’s coronavirus survey. London: Resolution Foundation.
Available at: https://www.resolutionfoundation.org/app/uploads/2020/05/The-
effect-of-the-coronavirus-crisis-on-workers.pdf. Accessed on: 22 February 2021.
Goldthorpe, J. H. (1987). Social mobility and class structure in modern Britain. Oxford: Clar-
endon Press.
Holst, H., Fessler, A., & Niehoff, S. (2020). Covid-19, social class and work experience in
Germany: Inequalities in work-related health and economic risks. European Societies,
23(sup1), S495–S512.
International Labour Organization (ILO). (2013). Measuring informality: A statistical manual
on the informal sector and informal employment. Geneva: International Labour Office.
International Labour Organization (ILO). (2018). India wage report: Wage policies for decent
work and inclusive growth. Geneva: International Labour Office.
Jain, R., Budlender, J., Zizzamia, R., & Bassier, I. (2020). The labour market and poverty
impacts of Covid-19 in South Africa (SALDRU Working Paper No. 264). Cape Town:
SALDRU, University of Cape Town.
Kesar, S., Abraham, R., Lahoti, R., Nath, P., & Basole, A. (2020). Pandemic, informality,
and vulnerability: Impact of COVID-19 on livelihoods in India (Centre for Sustainable Em-
ployment Working Paper No. 27). Bengaluru: Azim Premji University.
Kesar, S., & Bhattacharya, S. (2020). Dualism and structural transformation: The infor-
mal manufacturing sector in India. The European Journal of Development Research, 32,
560–586.
Koh, D. (2020). Occupational risks for COVID-19 infection. Occupational Medicine, 70(1),
3–5.
Lockwood, D. (1958). The blackcoated worker: A study in class consciousness. London: George
Allen and Unwin.
Nair, G. (2020). The gig economy in the pandemic. The India Forum. Issue: June 5, 2020.
Available at: https://www.theindiaforum.in/sites/default/files/pdf/2020/06/05/the-
gig-economy-in-the-pandemic.pdf. Accessed on: 28 November 2020.
National Sample Survey Organisation (NSSO). (2019). Annual report - periodic labour force
survey 2017–18. Ministry of Statistics and Programme Implementation, New Delhi:
National Statistical Office.
Resnick, S., & Wolff, R. (1987). Knowledge and class: A Marxian critique of political economy.
Chicago and London: University of Chicago Press.
Rogers, T. N., Rogers, C. R., VanSant-Webb, E., Gu, L. Y., Yan, B., & Qeadan, F.
(2020). Racial disparities in COVID-19 mortality among essential workers in the
United States. World Medical & Health Policy, 12(3), 311–327.
Sanyal, K. (2007). Rethinking capitalist development: Primitive accumulation, governmentality
and post-colonial capitalism. New Delhi: Routledge.
Schwander, H. (2019). Labour market dualization and insider–outsider divides: Why this
new conflict matters. Political Studies Review, 17(1), 14–29.
Serviere, L. (2010). Forced to entrepreneurship: Modeling the factors behind necessity
entrepreneurship. Journal of Business and Entrepreneurship, 22(1), 37–53.
Shaw, I. G. R., & Waterstone, M. (2020). Wageless life: A manifesto for a future beyond capi-
talism. Minneapolis: University of Minnesota Press.
Spurk, D., & Straub, C. (2020). Flexible employment relationships and careers in times of
the COVID-19 pandemic. Journal of Vocational Behavior, 119, 103435. https://doi.org/
10.1016/j.jvb.2020.103435.
St-Denis, X. (2020). Sociodemographic determinants of occupational risks of exposure
to COVID-19 in Canada. Canadian Review of Sociology, 57, 399–452.
Stephany, F., Dunn, M., Sawyer, S., & Lehdonvirta, V. (2020). Distancing bonus or down-
scaling loss? The changing livelihood of us online workers in times of COVID-19.
Tijdschrift voor Economische en Sociale Geografie, 111, 561–573. https://doi.org/10.1111/
tesg.12455.
Temin, P. (2018). The vanishing middle class: Prejudice and power in a dual economy. Cam-
bridge, MA: MIT Press.
Vlandas, T. (2020). The political consequences of labour market dualization: Labour
market status, occupational unemployment and policy preferences. Political Science Re-
search and Methods, 8(2), 362–368.
Woodcock, J., & Graham, M. (2020). The gig economy: A critical introduction. Cambridge,
MA: Polity Press.
World Bank. (2020). Beaten or broken? Informality and COVID-19. South Asia Economic
Focus (October). Washington, DC: World Bank.
Wright, E. O. (1978). Race, class, and income inequality. American Journal of Sociology,
83(6), 1368–1397.
23
COVID-19 PANDEMIC AND
MIGRANT WORKERS
India and the world
Srobonti Chattopadhyay
23.1 Introduction
Migrants, especially the ones migrating from less developed or developing coun-
tries to developed nations have been in discussion for quite a few years. How-
ever, the term “Migrant Workers” has acquired fresh significance in the context
of the worldwide Covid-19 pandemic during 2019–20.1 International migrants
not being the citizens of their destination countries are quite obviously deprived
of several social security measures that the nationals of the concerned countries
are entitled to. The Covid-19 pandemic created an unprecedented situation in
world history and took almost the whole world by surprise. It hardly left any-
one any time to cope with the unexpected shocks that the pandemic generated.
Of course, the primary one among these relates to health issues. What kinds of
healthcare services are required, whether it is feasible to provide everyone the
required healthcare support, how to deal with an unknown virus that mutates
in an unprecedented and unexpectedly fast pace, left everyone in a state of total
confusion. The fast escalating number of deaths due to the Covid-19 conta-
gion created a huge panic in the public psyche across the globe. But these have
been just the immediate impact. There are also equally serious issues beyond
the health-related ones, those that have long-run consequences for the world
economy. The Covid-19 pandemic invited total or partial lockdown in many
countries that led to disruption of economic activities in various sectors which,
in turn, has created huge uncertainties regarding livelihood. Many activities had
to be totally suspended, while many others had to substantially scale down. The
obvious fallout of all these has been large-scale laying-off of workers. Different
parts of the world witnessed sudden surges in unemployment. In many cases, the
recovery prospect also seems very bleak and therefore the prospective way out of
this unemployment appears quite remote. However, the extent of unemployment
DOI: 10.4324/9781003220145-27
392 Srobonti Chattopadhyay
and the possibility of recovery vary from sector to sector depending on the na-
ture of the economic activities associated with them, the kinds of goods and ser-
vices they produce, etc. The kind of labour requirement in terms of their levels
and types of skill in the various activities is also a decisive factor here. Therefore,
the obvious question that comes up is what awaits the migrant workers during
and after the pandemic?
The next crucial question to ponder is whether the situation of the migrant
workers within national boundaries is any different from their international
counterparts. Intra-country migrants are in a somewhat different situation from
the international migrant workers in terms of the fact that they are the citizens
even in their destinations. But still we must note that the international migrant
workers are not the only ones facing the plight as described above. Even the
intra-country migrant workers are having a hard time. To understand the sit-
uation of migrant workers in the backdrop of the Covid-19 pandemic better,
we need to analyse a few facts related to these people. We analyse the cases
of international and intra-country migrant workers separately. First, we elabo-
rate the case of international migrant workers and then move on to the case of
intra-country migrant workers with specific reference to India.
The plan of the chapter is as follows: Section 23.2 analyses the case of migrant
workers in the backdrop of the Covid-19 pandemic and the consequent disrup-
tion of economic activities across the globe, Section 23.3 studies the situation of
Indian migrant workers, both within country and abroad, along with its long-
run implication for the Indian economy, Section 23.4 discusses the policy options
and finally, Section 23.5 concludes the chapter.
23.2 Migrant workers: the international scenario

To proceed further in the discussion, first of all we need to define who the mi-
grant workers are. Fact Sheet No. 24(Rev.1)2 of the OFFICE OF THE UNITED
NATIONS HIGH COMMISSIONER FOR HUMAN RIGHTS defines a mi-
grant worker as “a person who is to be engaged, is engaged or has been engaged
in a remunerated activity in a State of which he or she is not a national”. As per
Article 3, this definition excludes the employees of international organizations, Gov-
ernment officials, persons sent or employed by a State or on its behalf outside its territory
who participate in development programmes and other cooperation programmes, investors,
refugees and stateless persons, students and trainees, non-national non-resident seafarers
and workers on an offshore installation. Of course, this definition applies to interna-
tional migrant workers only. The International Labour Organization (ILO) sug-
gests that both the countries of origin and destination gain from the contribution
of the migrant workers though in different ways. In destination countries, the
migrant workers definitely contribute to growth and development. At the same
time, the countries of their origin largely benefit not only from their remittances
but also from the skills they acquire during their migration experience.3
Covid-19 pandemic and migrant workers 393
Next, we look at the proportion of migrant workers among the international

migrants along with their major destination countries as well as the profiles of
migrant workers. According to the ILO global estimates on migrant work-
ers,4 out of approximately 258 million international migrants, 164 million were
migrant workers in 2017, i.e. in the year 2017, the migrant workers comprised
nearly 64% of all international migrants. Of these migrant workers, a little less
than half (46.9%) were located in two broad sub-regions – North America
and Northern, Southern and Western Europe (destination-wise distribution of
international migrant workers in 2017 is shown in Figure 23.1), where 55.4%
of all female workers and 40.9% of all male workers were located. 40.8% of all
workers in the Arab States and 1.0% of all workers in Southern Asia were mi-
grants, making them the regions with the highest and lowest shares of migrants
in the labour force, respectively (ILO, 2018).5 Among all migrant workers world-
wide in 2017, 68.1 million or approximately 41.6% were female. Male migrant
workers made up 95.7 million or 58.4% of the total (ibid.) and outnumbered
female migrant workers globally, but with significant regional variations. Ap-
proximately 8% of all migrant workers in 2017 were between 15 and 24 years of
age (ibid.). Of the estimated 150.3 million migrant workers in 2013, 11.5 million
were migrant domestic workers (ILO, 2015).6 The region-wise distribution of
migrant workers in 2017 in terms of destination countries is illustrated in Table
23.1 and Figure 23.1.
According to the World Migration Report 2020,7 in 2017, high-income
countries hosted 68% of migrant workers, an estimated 111 million people,
while middle-income countries hosted an additional 47 million migrant workers
(29%), and 5.6 million (3.4%) were residing in low-income countries. Examining
the changes in proportional distribution, it is observed that in 2017, a noticea-
ble change in destination country category was observed – from 2013 to 2017,
high-income countries witnessed a 7 percentage point fall in migrant work-
ers (from 75% to 68%), while upper-middle-income countries experienced a 7
percentage point rise (from 12% to 19%). Economic growth in middle-income
countries and/or changes to labour immigration regulations in high-income
TABLE 23.1 Region-wise distribution of migrant workers in 2017
Regions Number of Migrant Workers in Million

Northern America 37.72
Europe 52.48
Latin America and the Caribbean 4.43
Africa 12.96
SE Asia & Pacific 11.64
Arab States 22.80
Asia 21.81
Total 164
Percentage of Migrant Workers
13.30%
Northern America
23%
Europe
Latin America and the

13.90% Caribbean
Africa
SE Asia & Pacific

7.10%
Arab States
7.90%
32%
Asia
2.70%
FIGURE 23.1 Region-wise distribution of migrant workers.

Source: https://migrationdataportal.org/themes/labour-migration.
countries may have resulted in this apparent shift. “The share of migrant workers
in the total workforce across country income groups was quite small in low-
income (1.9%), lower-income (1.4%) and upper-middle-income countries (2.2%),
but much greater for high-income countries (18.5%)”.8
Figure 23.1 suggests that North America and Europe constitute the most
sought after destinations for migrant workers and being in the group of high-
income countries, whose total labour force have the highest proportion of mi-
grant workers, the number of migrant workers is quite high. However, we have
to now figure out in what sectors most of these migrant workers are engaged.
According to the Bureau of Labor Statistics data, in 2019, there were
28.4 million foreign-born people in the U.S. labour force, comprising 17.4% of
the total labour force.9 In the USA and Canada, the farming industry is largely
dependent on the migrant workers, and that is why these workers constitute “the
lifeblood of farming industry in USA and Canada”. Mostly manual, tedious and
unskilled jobs at both small- and large-sized farms are performed by the migrant
workers across these countries, mainly involving tending of vegetables, fruits and
farm animals. A large number of workers migrate from faraway lands, majorly
from Caribbean islands and Central and South American countries, for these
mundane jobs each year. Some of them are seasonal workers, while some others
get employed on a temporary basis round the year. These workers include both
the legal ones with proper visas, as well as illegal ones without a permit to stay or
work. The situation of the migrant workers can be described following Kumar
(2017) as follows:
Migrant workers undertake menial jobs that others shun. They are silent
and invisible and fear of repercussions due to the lack of laws protecting
them. They come, mostly men, under-educated, and poor, do their rou-
tine jobs that no one else wants, they collect their wages, they do a meagre
existence, send money home to support their families, and, when the sea-
son ends they go back. The cycle repeats during every planting to harvest
season. Their wages are often low. They have no social life. And they have
no labour union privileges to protect them.10
According to Moyce and Schenker (2018):
the majority (71.1%) of migrant workers worldwide are engaged in the ser-
vice industry, including domestic work, food services, and administrative
or professional work. Other occupational sectors include industry (manu-
facturing and construction) and agricultural work. The ILO estimates that
women make up 44.3% of all migrant workers. Women are six times more
likely than men to be engaged in domestic work abroad (48).11
According to Fasani (2020), around 31% of employed working-age individuals

constitute the key workers in the EU. The key workers are the ones providing
vital services. In EU, the largest five categories of key workers include teaching
professionals (14.5%), skilled agricultural workers (11.9%), science and engineer-
ing associate professionals (11.1%), personal care workers (10.3%) and cleaners
and helpers (9.9%) and 13% of key workers are immigrants in EU,12 though this
share varies widely across Europe, from over 40% in Denmark and France to just
above 10% in Bulgaria and Slovenia. 13% of key workers, on average, are immi-
grants in the EU. The wide variation across member states of the EU is observed
in Figure 23.2. In Eastern European countries like Romania, Bulgaria, Poland
and Slovakia, the share of immigrant key workers is quite close to zero, while in
countries like Italy, Belgium, Germany, Sweden and Austria, it fluctuates around
20%. Ireland (26%), Cyprus (29%) and Luxembourg (53%) mark the largest fig-
ures. In most countries, the share of Extra-EU migrant key workers is larger than
their EU-migrant13 counterparts.14
Among the key workers in the EU, the largest five categories consist of
teaching professionals (14.5%), skilled agricultural workers (11.9%), science and
engineering associate professionals (11.1%), personal care workers (10.3%) and
cleaners and helpers (9.9%). As natives by far form the largest proportion of the
labour force, their distribution across key occupations is quite similar to the
overall distribution. However, the story of migrant workers is very different.
For the two groups of migrant workers, EU and extra-EU, important differ-
ences can be observed – these people seem to be more concentrated in low-skill
FIGURE 23.2 Country-wise distribution of remittances.

Source: https://www.rbi.org.in/scripts/BS_ViewBulletin.aspx?Id=17882#CH1.
occupations. To be precise, the first three key occupational categories for mi-
grant workers from EU itself are cleaners and helpers (20.9%), personal care
workers (12.5%) and teaching professionals (11.1%). The top two occupations for
Extra-EU migrant workers are just the same with cleaners and helpers consti-
tuting 27.8% and personal care workers constituting 16.6%. The third position,
however, most frequently adopted occupation being that of drivers and mobile
plant operators (9%).
The levels of skills and educational attainments of migrant workers, in com-
bination with other existing hurdles to the recognition of foreign qualifications
and to the access to certain professions in the EU region, are likely to determine
how well represented in each occupation the foreign-born workers are going to
be. In most member states, EU-mobile key workers15 are predominantly middle
or highly educated, while extra-EU key workers tend to have lower education
(especially in countries such as Italy, Spain, Portugal and Greece). This is partly
due to the original skill distribution of migrants in each country, as well as to the
process of selection of individuals into migration. If we consider the concentra-
tion of migrants in key occupations by educational level (low, middle, high), we
see that, irrespective of their level of education, EU-mobile workers’ concentra-
tion in key occupations closely follows that of the native population.
Fasani (2020) comments that many EU member states were induced to tem-
porarily shut down large sections of their economies following the rapid spread of
the Covid-19 contagion aiming at slowing down the virus propagation rate and
letting national health systems provide adequate care to all seriously affected citi-
zens. According to him, while the forced shutdown has confined big proportions
of the workforce at home, some essential functions are still required to be per-
formed for keeping the European citizens healthy and safe and protecting them
from starvation during the pandemic. “Key workers” are fighting in the front
line through performing these crucial tasks of Europe’s Covid-19 response. The
understanding of who these workers are and under which conditions they can
effectively continue the provision of their essential services is crucial for the
formulation of any informed strategy to cope with the pandemic. An important
concern here is whether these “key workers” are part of those groups of indi-
viduals who may be particularly vulnerable during the crisis. “The government-
imposed heterogeneity in the lockdown across occupations and industries will
obviously lead to differential impact of the recession on different groups of
workers based on their pre-COVID-19 employment”. Moreover, the economic
impacts of the pandemic may be expected to further vary by age, gender, house-
hold structure, type of employment contract, firm size, etc. (Bell et al., 2020,
Adams-Prassl et al., 2020, Hupkau and Petrongolo, 2020). The nationality of
workers is also an important variable to take note of. “Since immigrants are
typically more exposed to economic downturns than natives (Dustmann et al.,
2010), assessing their involvement in key occupations is the first step to conceive
interventions that may reduce their vulnerability”.16
According to a report in UN News dated 8th April, 2020:
Workers in four sectors that have experienced the most “drastic” effects
of the disease and falling production are: food and accommodation (144
million workers), retail and wholesale (482 million); business services and
administration (157 million); and manufacturing (463 million).
Together, they add up to 37.5 per cent of global employment and this is
where the “sharp end” of the impact of the pandemic is being felt now, the
ILO chief added.17
From the analysis here, we can figure out that the economic activities hard hit
by the Covid-19 pandemic and the consequent lockdown in various countries
involve medium-skilled workers. The migrant workers in these sectors are defi-
nitely going to suffer a lot as during recovery, the local non-migrant workers will
have better access to all types of jobs due to their better familiarity among and
with the locals and hence better access to social networks.
23.3 Indian migrant workers: within country and abroad

The plight of the migrant workers due to the lockdown in India as a measure to
prevent the spread of the novel-coronavirus made headlines worldwide. These
migrant workers were the ones who migrated within the country. However, a
huge number of workers also migrate to other destination countries from India
and even these workers had difficult times as a consequence of the Covid-19
pandemic. To analyse the situation of these migrant workers, let us first figure
out the major destinations of such workers.
According to the India Labour Migration Update 2018,18
India is a major country of origin and transit, as well as a popular destina-

tion, for workers across international borders. As per official figures, there
are over 30 million Indians overseas, with over 9 million of the Indian
diaspora concentrated in the GCC region (now known as the Cooperation
Council for the Arab States of the Gulf ). Over 90% of Indian migrant
workers, most of who are low- and semi-skilled workers, work in the Gulf
region and South-East Asia.
How the Covid-19 pandemic is affecting the GCC region is well elaborated
by John (2020).19 According to John (2020), with the continued sharp rise in
Covid-19 in the Gulf countries, the governments are adopting drastic measures
to enforce social distancing by announcing curfews, travel bans and the shut-
down of large segments of their economies. As observed by most economists,
these steps are definitely leading to severe economic disruption that will con-
tinue further during the coming days. Two factors are decisive about how severe
the economic impact on the GCC will be; they are, first, the duration of restric-
tions on the movement of people and economic activities; and second, the size
and efficacy of fiscal responses to the crisis. As most governments have adopted
and implemented well-designed fiscal stimulus packages prioritizing health ex-
penditure for containing the spread of the virus, expectations are high that the
likelihood of a deep economic recession can be reduced to a minimum.
According to the World Migration Report 2020, in 2019, India had the larg-
est number of migrants living abroad, the number being 17.5 million; Mexico
and China held the second and the third ranks, respectively, with the respective
number of migrants from these countries being 11.8 million and 10.7 million.
The same report suggests that in 2018, India was the highest remittance-recipient
country, earning 78.6 million USD. China and Mexico followed India, with
China earning 67.4 million USD and Mexico, 37.5 million USD. Among the
remittance-sending countries, the USA topped the list, sending 68.0 billion
USD, while the United Arab Emirates ranked second, sending 44.4 billion USD
and Saudi Arabia ranked third, sending 36.1 billion USD. Thus, it is quite evi-
dent that India’s foreign exchange earnings are quite dependent on remittances
from abroad. A Reserve Bank of India (RBI) report in 2018 suggests that,
82 per cent of the total remittances received by India originated from

seven countries, viz., the United Arab Emirates (UAE), the United States
(US), Saudi Arabia, Qatar, Kuwait, the United Kingdom (UK) and Oman
(Figure 23.2). With over 90 per cent of overseas Indians working in the
Gulf region and South-East Asia (ILO, 2018) – mostly semi-skilled and
unskilled workers – the Gulf Cooperation Council (GCC) countries
accounted for more than 50 per cent of total remittances received in 2016–
17, notwithstanding a sharp decline in oil prices and fiscal tightening in
these countries. The Indian diaspora in the US, characterised by high skills
and high earnings, is the second largest contributor.
To understand how this big pool of migrant workers has been and is going
to be affected by the Covid-19 pandemic, two different categories need to be
considered – first, the case of those migrant workers who lost jobs due to suspen-
sion of economic activities and had to stay back in their countries of destination
and second, the case of those migrant workers who were repatriated by the Indian
Government after losing job. About the first category, Dilip Ratha, lead econo-
mist of the World Bank, in an interview with the Economic Times, suggests that,
The coronavirus crisis is going to produce unprecedented levels of effects

on the migrant economy. The first is pretty direct: Because of social dis-
tancing and the lockdown, businesses are mostly shut and only a few people
are working. That means a lot of migrant workers are now either unem-
ployed or waiting for something (positive) to happen. It also means a lot of
people are stranded. In addition, there is the effect of oil prices. There has
been a significant decline in demand for oil, which has been an additional
pressure on the economy of the Gulf countries. This means more fiscal re-
straint, reducing the government’s ability to provide protection to foreign
workers.20
Yadav (2020) observes that, as mentioned by the lead economist of the World
Bank, millions of Indian migrant workers in Gulf countries are in serious trouble
as a result of the coronavirus pandemic and the reduction in oil price. The coro-
navirus pandemic will generate unprecedented effects on the migrant economy.
A huge number of workers who are living under unhygienic conditions are also
unable to return home as a consequence of travel restrictions imposed by the
Gulf countries. Amnesty International as well as some other organizations has
expressed concerns about the public health risk of migrant workers by mention-
ing very common issues such as overcrowded accommodation that makes social
distancing impossible. Just not this much, since most of the businesses are shut
down due to social distancing and countrywide lockdowns, most of the migrant
workers are jobless or on lookout for a job or stuck in places. They have no
income and hence cannot afford food and basic amenities. Payments of salaries
have been stopped by some companies raising serious concerns about the basic
survival of the migrant workers in a foreign land. Several companies in the oil
and gas sector have put many migrant workers’ lives at risk amidst the pandemic
by running the production at the normal rate through defiance of the strict gov-
ernment guidelines of staying at home and following the quarantine protocols.
Though these countries have announced a financial package for protecting their
economies, it has been brought to the fore by the activists that very little has been
done for migrant workers in terms of financial help. This may be attributable to
the corrupt system that is driven by the discrimination against foreign workers.
The narratives above depict the miserable conditions of the migrant workers
who are stranded in their destination countries and cannot return home. The
obvious question that arises here is what is happening to those migrant workers
who were repatriated? Most of the repatriated migrant workers had been in the
laid off category. There seems little chance that all of them will be able to go
back to their erstwhile destination countries during the post-Covid-19 situation.
It will also be difficult for them to find jobs locally as they have to compete with
the existing non-migrant local labour force, particularly when the possibility of
fresh job creation seems to be a distant or may be an impossible dream. Moreo-
ver, most of these migrant workers hail from poor or very poor families. Return-
ing to their home countries, they are back in their poverty-ridden residences, to
stay in dingy environments sans any means for social distancing. Their misery is
deepened by the uncertainty regarding the recovery of their livelihood.
The consequences of these large-scale job losses of Indian migrant workers
abroad are not limited just to the misery of the migrant workers; they are quite
far-reaching. The job losses abroad have a direct impact on remittances. The
World Bank predicts that remittances to India from various countries will drop
by 23% in 2020.21 The drop in remittances on the one hand will adversely affect
GDP and on the other hand it will put India in a very difficult position in terms
of meeting international payment obligations. These consequences are likely to
be of long-term nature. How and when the damage can be compensated for
remains a big question.
We next come to the issue of migrant workers within the country. As ob-
served by Khanna (2020)22, the internal migrants form a crucial part of India’s
economy; the Economic Survey 2016–17 estimates suggest that almost 100 mil-
lion internal migrants are there in India, who constitute about 20% of the total
workforce and an estimated 10% of India’s economic output is attributable to
their contributions (Al-Jazeera, 2020)23; thus, they play an important role in the
economy of the country. Due to the sudden announcement of the lockdown in
India from 24th March, 2020, many migrant workers in different parts of the
country suddenly turned jobless. They did not have enough savings to fall back
on, were unable to pay house rents and therefore were forced to walk miles, not
having any alternative arrangements for staying, half-starved on the road. Many
of them lost lives on the way due to the fatigue of such a long and tedious jour-
ney, especially during the scorching summer. A number of them also died due
to road accidents or being run over by trains on railway tracks. Many flocked
beneath the Yamuna river bridge, somehow surviving on one meal a day under
very unhygienic conditions, with no provisions for any social distancing (Ghosh
(2020a)).24 As observed by Yadav (2020)25:
In India, most of the lowest paying jobs are in sectors like construction,
hospitality, textiles, and domestic workers. 90% of the workforce comprises
the informal or ‘unorganised’ sector of the economy, and migrant workers

are a big part of it. According to the World Bank, amidst the COVID-19
pandemic and lockdown, India’s 40 million migrant workers are facing the
major brunt of the situation. This pandemic has created major issues like
starvation, being stranded away from the family, and loss of jobs which in
turn will increase the risk of modern slavery a thousand-fold. First, there
will be an increased risk of enslavement due to the increased supply of
workers who are at risk of exploitation. Second, not to forget the disrup-
tion of education due to COVID-19, most of the children will be forced
out of school to make a living to sustain their families. A huge number of
children will be trafficked across borders where they will get paid a meagre
salary and might also suffer physical, emotional, and sexual violence. Not
only are the migrants at high risk of the above issues due to the COVID-19
pandemic but also are victims of social discrimination.
According to a report published in India Today on 16th June, 2020,26 Global

Consultants made a forecast of 130 million job losses in India due to the Covid-19
pandemic, 40% of it being blue-collared jobs. The report further suggests that the
industries to be worst hit due to the Covid-19 pandemic include Aviation, Au-
tomobile, Travel and Tourism, Food, Entertainment and Manufacturing. While
industries like Hospitality industry which includes hotels and restaurants and
malls, Jewellery industry, and Retail industry were likely to be able to relax a lit-
tle following unlock 1.0, Healthcare, Pharmaceuticals, e-commerce, IT services,
and essential retail were the ones that did well despite the Covid-19 pandemic.
Khanna (2020) suggests that, “According to the early estimates, tens of mil-
lions of migrant workers were left unemployed in India by the end of March
2020 due to the lockdown”. Among various industries, manufacturing, mining,
retail and wholesale trade and hospitality are said to have the highest dependence
on migrant workers from other states, according to an analysis by India Ratings
& Research.27 Therefore, the migrant workforce associated with manufacturing
is likely to be the hardest hit with a grim recovery prospect, while the ones
associated with retail trade still have some hope. However, it is not just the
migrant workers who faced ordeal due to the lockdown, small businesses also
suffered badly as the migrant workforce left the cities. In fact, after the unlock
started, the recovery process of these small businesses got adversely affected as
the migrant workers had left the cities during the lockdown. This is evident
from a report published in Hindustan Times on 2nd June, 2020 as reported by
Shrangi (2020)28 which suggests that according to various trade associations,
with a large number of migrant workers going back to their hometowns in the
midst of the ongoing nationwide lockdown, a massive workforce shortage hit
the operations in industrial, transport and other sectors, which reopened after
an almost two-month interruption. As per Traders’ bodies’ estimates, nearly
60–70% of the labour force employed in Delhi left the city during the lockdown
imposed to halt the spread of the coronavirus disease (Covid-19). The workers
who went back to their hometowns include both skilled and unskilled categories
and consist of those engaged as machine operators in factories, loaders, drivers,
housekeeping staff, as well as those involved in packaging, construction workers,
hand-embroidery workers, salespersons and security guards, among others. The
business owners from the industrial, wholesale and retail, and transport sectors,
and those running small eateries and restaurants reported that with a large chunk
of workforce gone, the lockdown-hit business that was slowly picking up got
severely adversely affected.
It is difficult to estimate exactly how many migrant workers are affected by the
lockdown. However, Singh (2020),29 in a report published in The Indian Express
on 29th April, 2020, observes that no official data about the inter-state migrants
in the country is available; only the estimates for 2020 made by Professor Am-
itabh Kundu of Research and Information System for Developing countries are
somewhat suggestive. His estimates, based on the 2011 Census, NSSO surveys
and economic survey, show that a total of about 65 million inter-state migrants
are there, and 33% of these migrants are workers. Conservative estimates suggest
that 30% of them are casual workers, while another 30% work on a regular basis
but in the informal sector. If street vendors are added, another vulnerable com-
munity which is not captured by the worker data can be identified; this would
imply that there are 12–18 million people residing in states other than that of
their origin who have been placed at a risk of losing their income. According to a
study conducted by the Centre for the Study of Developing Societies (CSDS) and
Azim Premji University in 2019, an estimated 29% of the population in India’s
big cities survives on daily wages. This, therefore, is the number of people who
would logically want to move back to their states.
It is not difficult to understand that the way this mass exodus of migrant work-
ers from cities during lockdown has affected the recovery of small businesses in
the post-lockdown phase, more serious long-run consequences await the Indian
economy. Due to lack of workforce, many of the small business activities may
be forced to shut down and if they do, that will further worsen the already grim
employment situation. It is going to act like a vicious circle that the lockdown
initiated – suspension of economic activities leading to loss of employment for
the migrant workers, thus forcing them to leave cities in mass scale leading to
dearth of labour during the post-lockdown situation forcing some small business
activities to shut down resulting in further loss of employment opportunities in
the long run which would again compel a part of the migrant workforce to leave
and the downward spiral will continue. Not just small businesses, even the small
house owners who used to survive on the rent from migrant workers are also
badly affected. Their earnings are also not to recover until the migrant workers
are back in cities. So the externality stretches quite far.
However, the migrant workers who left the cities to go back to their village
homes during lockdown have started returning to cities due to lack of livelihood
opportunities in villages. Ghosh (2020b) in a report published in The Hindu on
17th August, 202030 observes: While a steady stream of daily wagers are returning in
search of livelihood, the ones who “survived” lockdown in the Capital are not too hopeful
of the current scenario. For the reasons noted above, the prospective employment
opportunities for the returning migrant workers do not seem too bright; more-
over, they have to compete with those who somehow managed to stay back in
cities during the lockdown and survive on whatever little opportunities they
could access. This is actually making things all the more difficult for both types
of migrant workers – those who left and those who did not.
23.4 Policy options: national level and multilateral

The discussion on the plight of migrant workers clearly suggests that left to itself
the world economy is not in a position to auto-correct and create conducive
situation for the huge mass of migrant workers. Government intervention will
definitely be required to cure the economic scars that the Covid-19 pandemic
leaves behind. Moreover since this pandemic has been an unprecedented global
scenario, government policies across the world need to be well coordinated to
address the problems properly and effectively. This calls for multilateral initia-
tives in policy making along with the ones at the national levels. This is espe-
cially important in the context of the migrant workers.
As Fasani (2020) observes:
There are two major lessons for policymaking arising from this evidence.
First, migrant workers are playing a critical role in performing basic func-
tions in EU societies hit by the COVID-19 epidemic. This calls for inter-
ventions in the short run that may allow foreign-born workers to better
cope with the crisis and keep contributing to its solution. Second, low-
educated migrants – and not just high-skilled ones – are employed in occu-
pations that are key for hosting societies. This latter fact suggests the need
for reconsidering – in the aftermath of the pandemic crisis – a migration
policy debate which is currently almost entirely focused on the importance
of attracting high skilled migrants to the EU.31
For dealing with the issue of international migrant workers, it is important for
all countries to formulate well-coordinated migration policies which can ensure
safety and security for the migrant workers in their destination countries. For
ensuring better job prospects for international migrant workers, the governments
in countries like India need to ensure better training facilities in acquiring of
skills as per needs of the global markets.
The policy of skill enhancement activities is not just limited to the case of
international migrants.
The analysis of which kinds of business activities fared well despite the pan-
demic and which ones have a better recovery prospect during the post-lockdown
situation clearly indicates a drive towards computerization and digitization. On-
line retail trading has become quite popular these days and has been showing a
rising trend for some time. The pandemic and the consequent lockdown accel-
erated its pace. This trend is undoubtedly going to continue. Many other forms
of services like payment methods, processing of official papers, etc., have also
become digitized and they are less likely to revert to manual mode. Thus, easing
the skill generation process needs to be policy priority for the governments in
the long run too. However, to deal with the immediate ordeal and uncertainty
of the migrant workers, the government needs to incentivize economic activi-
ties intensive in unskilled, semi-skilled and medium-skilled labour; the small-
and medium-scale industries in both village and urban areas therefore need to
be given a targeted well-designed policy support that can make these activities
self-sustaining in the long run and at the same time absorbing a substantial part
of the migrant workers to help them overcome their immediate ordeal.
23.5 Conclusion
The issue of migrant workers is definitely intertwined with the world economic
scenario. Thus, the policies to ensure the smooth functioning of global trade are
going to help them, but that will be only to an extent. The problems related to
migrant workers, both at international and national levels, call for separate and
targeted treatment of the issue. It must be noted that the problems that the migrant
workers face have been there; the Covid-19 pandemic made these problems more
visible to the public eye while aggravating some of the problems. As the analysis
in this chapter has established, the migrant workers contribute substantially to the
world economic activities. Therefore, more care and attention need to be dedi-
cated in formulating policies to address the problems of migrant workers.
Notes
1 The immigrants and migrant workers are differentiated on the basis of their duration
of stay in the destination countries; the immigrants have a long-term plan to stay or
may be settle down permanently in the destination country, while the migrant work-
ers do not have such long-term plans, and their stay in destination countries therefore
is of a shorter duration.
2 https://www.ohchr.org/Documents/Publications/FactSheet24rev.1en.pdf.
3 https://www.ilo.org/wcmsp5/groups/public/---ed_protect/---protrav/---migrant/
documents/publication/wcms_208594.pdf.
4 https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/
documents/publication/wcms_652001.pdf.
5 Ibid.
6 https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/
publication/wcms_436343.pdf.
7 https://www.un.org/sites/un2.un.org/files/wmr_2020.pdf.
8 https://publications.iom.int/system/files/pdf/wmr_2020_en_ch_2.pdf.
9 https://www.bls.gov/opub/ted/2020/mobile/foreign-born-workers-made-up-17-
point-4-percent-of-labor-force-in-2019.htm#:~:text=In%202019%2C%20there%20
were%2028.4,17.4%20percent%20of%20the%20total.
10 https://english.mathr ubhum i.com/features/social-issues/m ig rant-workers-

of-north-america-1.1928654.
11 ht t p s://w w w. a n nu a l r ev iew s .or g /doi /pd f /10.114 6/a n nu r ev- publ he a lt h -
040617-013714.
12 https://knowledge4policy.ec.europa.eu/publication/immigrant-key-workers-
their-contribution-europes-covid-19-response_en.
13 The migrant workers in Europe from non-European countries are termed as Extra-
EU migrants, while the migrants from European countries are being referred to as
EU migrants.
14 https://voxeu.org/article/covid-19-immigrant-workers-europe.
15 The key workers who move freely within the EU are referred to as EU-mobile key
workers.
17 https://news.un.org/en/story/2020/04/1061322.
18 https://ww w.ilo.org/wcmsp5/groups/public/---asia/---ro-bangkok/---sro-
new_delhi/documents/publication/wcms_631532.pdf.
19 https://www.orfonline.org/research/the-pandemic-in-the-gulf-64882/.
20 https://econom ictimes.indiatimes.com/opinion/inter views/covid-19-w il l-
have-unprecedented-ef fect-on-m igrant-economy-dilip-ratha-world-bank/
articleshow/75648695.cms?from=mdr.
21 Ibid.
22 https://journals.sagepub.com/doi/full/10.1177/0972063420935542
23 https://www.aljazeera.com/ajimpact/hungr y-desperate-india-virus-controls-
trap-migrant-workers-200402031123631.html
24 https://w w w.thehindu.com/news/national/coronavir us-workers-stranded-
on-yamuna-floodplain-without-basic-amenities/article31349018.ece.
25 https://criticaledges.com/2020/07/12/covid-19-on-indian-migrant-workers/.
26 https://www.indiatoday.in/education-today/latest-studies/story/india-to-lose-130-
million-jobs-due-to-covid-19-pandemic-report-1689590-2020-06-16.
27 h t t p s : // w w w . b l o o m b e r g q u i n t . c o m /e c o n o m y - f i n a n c e / i f - i n d i a s -
migrant-workers-dont-return-it-spells-trouble-for-these-sectors.
28 https://www.hindustantimes.com/delhi-news/as-migrant-workers-return-home-
lack-of-workforce-hits-businesses-in-delhi/story-5I90E2zIUOSiWkpvIQaq6M.
html.
29 h t t p s : // i n d i a n e x p r e s s . c o m /a r t i c l e /e x p l a i n e d /c o r o n a v i r u s - i n d i a -
lockdown-migran-workers-mass-exodus-6348834/.
30 https://www.thehindu.com/news/cities/Delhi/they-are-back/article32370322.ece.
References
Articles
Adams-Prassl, Abigail, Teodora Boneva, Marta Golin and Christopher Rauh (2020),
“Inequality in the impact of the coronavirus shock: Evidence from real-time sur-
veys.” https://www.iza.org/publications/dp/13183/inequality-in-the-impact-of-the-
coronavirus-shock-evidence-from-real-time-surveys (Last accesseed on April 18,
2021).
Bell, Brian, Nicholas Bloom, Jack Blundell and Lucy Pistaferri (2020), “Prepare for large
wage cuts if you are younger and work in a small firm”. https://voxeu.org/article/
prepare-large-wage-cuts-if-you-are-younger-and-work-small-firm, 6 April. (Last
accesseed on April 18, 2021).
Fasani, Francesco (2020), “Immigrant key workers in Europe: The COVID-19 response
that comes from abroad.” https://voxeu.org/article/covid-19-immigrant-workers-
europe (May 5, 2020) (Last accessed on April 18, 2021)
Ghosh, Shinjini (2020a), “Coronavirus | Workers stranded on Yamuna floodplain
without basic amenities”, The Hindu. https://www.thehindu.com/news/national/
coronavirus-workers-stranded-on-yamuna-f loodplain-without-basic-amenities/
article31349018.ece (April 16, 2020) (Last accessed on April 18, 2021)
Ghosh, Shinjini (2020b), “Lack of jobs force migrant labourers to return to cities they
left in March”, The Hindu. https://www.thehindu.com/news/cities/Delhi/they-are-
back/article32370322.ece (August 17, 2020) (Last accessed on April 18, 2021)
Hupkau, Claudia, and Barbara Petrongolo (2020), “Work, care and gender during the
Covid‐19 crisis.” Fiscal studies 41.3, p.623–651.
John, Issac (2020), “The pandemic in the Gulf: What it portends for NRIs in the region.”
https://www.orfonline.org/research/the-pandemic-in-the-gulf-64882/ (April, 2020)
(Last accessed on March 20, 2021)
Khanna, Anoop (2020), “Impact of Migration of Labour Force due to Global COVID-19
Pandemic with Reference to India.” Journal of Health Management 22.2: 181–191.
https://journals.sagepub.com/doi/full/10.1177/0972063420935542
Kumar, Sasi (2017), “Migrant workers of North America.” https://english.mathrubhumi.
com/features/social-issues/migrant-workers-of-north-america-1.1928654 (May 17,
2020) (Last accessed on April 18, 2021)
Moyce, Sally C. and Marc Schenker (2018), “Migrant workers and their occupational
health and safety”, Annual Review of Public Health. https://www.annualreviews.org/
doi/pdf/10.1146/annurev-publhealth-040617-013714 (Last accessed on April 18,
2021)
Shrangi, Vatsala (2020), “As migrant workers return home, lack of workforce hits busi-
nesses in Delhi”, Hindustan Times. https://www.hindustantimes.com/delhi-news/
as-migrant-workers-return-home-lack-of-workforce-hits-businesses-in-delhi/
story-5I90E2zIUOSiWkpvIQaq6M.html ( June 2, 2020) (Last accessed on April 18,
2021)
Singh, Sushant (2020), “Explained: Indian migrants, across India”, The Indian Express.
https://ind ianexpress.com/ar ticle/explained/coronav ir us-ind ia-lockdown-
migran-workers-mass-exodus-6348834/ (April 29, 2020) (Last accessed on April 18,
2021)
Yadav, Renu (2020), “Impact of COVID-19 on indian migrant workers”, Critical Edges.
https://criticaledges.com/2020/07/12/covid-19-on-indian-migrant-workers/ ( July
12, 2020) (Last accessed on April 18, 2021)
Reports
COVID-19: Impact Could Cause Equivalent of 195 Million Job Losses, Says ILO Chief,
UN News, April 8, 2020 https://news.un.org/en/story/2020/04/1061322 (Last
accessed on April 18, 2021)
“Covid-19 Will Have Unprecedented Effect on Migrant Economy: Dilip Ratha, World
Bank” (Interview with Indulekha Arvind, ET Bureau), The Economic Times. May
10, 2020 https://economictimes.indiatimes.com/opinion/interviews/covid-19-
will-have-unprecedented-effect-on-migrant-economy-dilip-ratha-world-bank/
articleshow/75648695.cms?from=mdr (Last accessed on April 18, 2021)
Foreign-Born Workers Made Up 17.4 Percent of Labor Force in 2019, The Economics
Daily, May 29, 2020. https://www.bls.gov/opub/ted/2020/mobile/foreign-born-
workers-made-up-17-point-4-percent-of-labor-force-in-2019.htm#:~:text=In%20
2019%2C%20there%20were%2028.4,17.4%20percent%20of %20the%20total (Last
Globalising People: India’s Inward Remittances, RBI Bulletin, November 14, 2018.
https://www.rbi.org.in/scripts/BS_ViewBulletin.aspx?Id=17882#CH1 (Last accessed
on April 18, 2021)
Hungry, desperate: India virus controls trap its migrant workers. Al-Jazeera. Time. April 2,
2020 https://www.aljazeera.com/ajimpact/hungry-desperate-india-virus-controls-
trap-migrant-workers-200402031123631.html (Last accessed on April 18, 2021)
If India’s Migrant Workers Don’t Return, It Spells Trouble for These Sectors, Bloomb-
erg Quint, August 8, 2020. https://www.bloombergquint.com/economy-finance/
if-indias-migrant-workers-dont-return-it-spells-trouble-for-these-sectors (Last
ILO Global Estimates on International Migrant Workers: Results and Methodology.
https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/
documents/publication/wcms_652001.pdf (Last accessed on April 18, 2021)
ILO Global Estimates on Migrant Workers: Results and Methodology – Special Fo-
cus on Migrant Domestic Workers. https://www.ilo.org/wcmsp5/groups/public/
---dgreports/---dcomm/documents/publication/wcms_436343.pdf (Last accessed
on April 18, 2021)
Immigrant Key Workers: Their Contribution to Europe’s COVID-19 Response. April
24, 2020. https://knowledge4policy.ec.europa.eu/publication/immigrant-key-
workers-their-contribution-europes-covid-19-response_en (Last accessed on April
18, 2021)
India Labour Migration Update 2018. https://www.ilo.org/wcmsp5/groups/public/
---asia/---ro-bangkok/---sro-new_delhi/documents/publication/wcms_631532.pdf
(Last accessed on April 18, 2021)
India to Lose 130 Million Jobs due to Covid-19 Pandemic: Report, India Today, June
16, 2020
https://www.indiatoday.in/education-today/latest-studies/story/india-to-lose-130-
million-jobs-due-to-covid-19-pandemic-report-1689590-2020-06-16 (Last accessed
on April 18, 2021)
Labour Migration, Migration Data Portal. https://migrationdataportal.org/themes/
labour-migration (Last accessed on April 18, 2021)
Migration and Migrants: A Global Overview, Chapter 2, World Migration Report 2020,
p. 33. https://publications.iom.int/system/files/pdf/wmr_2020_en_ch_2.pdf (Last
World Migration Report 2020. https://www.un.org/sites/un2.un.org/files/wmr_2020.
pdf (Last accessed on April 18, 2021)
The International Convention on Migrant Workers and its Committee, Fact Sheet No.
24 (Rev.1), Office of the United Nations High Commissioner for Human Rights.
https://www.ohchr.org/Documents/Publications/FactSheet24rev.1en.pdf (Last
24
THE FUTURE OF LABOUR
SEGMENTATION AFTER COVID-19
Mélika Ben Salem
24.1 Introduction
Job creation has been highly challenging since ICT innovations have led to the
new era of digitalization, threatening many economic tasks considered as routine
ones and allowing easy replacement of human beings by machines. Moreover,
this digitalized world relies on highly global supply chains, determining the al-
location of labour across the world. The pandemic could reshuffle the cards by
first modifying significantly the logic of the global value chains, and then the
labour cost and its allocation. However, it could also dampen the digitalization
process observed before. The result of the pandemic is then highly unclear on
the medium- or long-term trend of labour, and especially its division between
formal and informal employment on the one hand and salaried employment
and self-employment on the other one. Indeed, job opportunities are an issue in
most developing countries, partially depending on outsourcing from developed
ones. Then, the prevalence of self-employment is an answer to the job short-
age, self-employment being likely informal and preventing economies of scale
achievement, and then job creations. Does then the pandemic reduce total em-
ployment, without changing the share of informal employment, but increasing
self-employment by reducing available tasks already threatened by digitalization?
On the contrary, as the advent of the Covid-19 has been associated with environ-
ment issues, green policies could generate new job opportunities but the nature
of these jobs, informal or not, self-employed or not, is not so easy to forecast.
This chapter is an attempt to answer these questions by first assessing the
direction taken by the labour market aggregates. This amounts to comparing
trends in developing countries, illustrated here by India, to the worldwide ones
before the pandemic. Thereafter, predictions of these trends after the pandemic
are formulated relying on the literature initiated by Acemoglu and Autor (2011).
DOI: 10.4324/9781003220145-28
Labour segmentation’s future after Covid-19 409
Indeed, the development of automation linked to innovation, associated with

artificial intelligence, is at the heart of the job creation process in this century,
provoking mixed vision on the future of work, in either developed or devel-
oping countries. Using machines and computers to substitute for human la-
bour is an old story beginning at the end of the nineteenth century, resulting in
the past by some increase in demand for labour and wages. If this consequence
seems expected for the fourth industrial revolution, the economic adjustment
to its full realization appears slow and costly for the labour market to adapt,
at least in the short and medium runs. Indeed, technological progress encour-
ages capital-intensive activities that translate in job destruction, magnified for
the unskilled, in the developed countries. For developing countries, it implies
labour shortage and skill mismatch as their exporting companies adopt these
innovations. Indeed, two opposite movements in developing countries’ labour
demand characterize globalization: Outsourcing that exploits the lower labour
cost in developing countries and jobless economic growth. The occurrence of
the pandemic changes the balance between these forces as it caused a violent
and sudden slowdown in international trade that strongly affected global value
chains and then countries with high dependence to a limited range of export
and import goods. The persistence of the pandemic itself but also of its conse-
quences threatens normalization, by creating a new trend in labour markets, as
working from home, which is biased towards some type of employment, rather
formal and skilled, magnifying the main effect of the actual technologies. The
discussion below will develop this issue by taking into account the main feature
of developing countries labour markets that is segmentation, along the lines of
self-employment and informality. The following section is dedicated to a brief
comparison between Indian and worldwide labour markets providing evidence
on the labour market segmentation and the third one attempts to outline the
analysis of the future of labour segmentation.
24.2 A brief overview of labour outcomes: India vs. the world1

According to the standard view, labour segmentation is defined by the quality
of the job, described by different inputs, as working conditions.2 In developing
countries, the initial focus was on access to social protection – like unemploy-
ment and health insurance, minimum wage or retirement rights and pensions,
defining formal workers. This section then documents the dominance of infor-
mal employment in India, given that a majority of workers are self-employed and
that the service sector becomes the new job opportunity pool.
24.2.1 Self-employment and informality

Labour force in developing countries is growing under demographic pressures,
which are especially acute in India whose population will reach 1.66 billion in
410 Mélika Ben Salem
Self and Vulnerable Employment, Unemployment Labor force participation rate vs GDP growth rate
6.5
90
65
8
80
60
6
% of total employment
% of population 15+
70
% of labor force
4
5.5
%
55
60
2
50
5
50
0
−2
4.5
40
45
1991 1995 1999 2003 2007 2011 2015 2019 1991 1995 1999 2003 2007 2011 2015 2019
India: Self E. World: Self E. India: Vuln. India: LFPR World: LFPR India: Emp.
World: Vuln. India: Unemp. World: Unemp. World: Emp. India: GDP World: GDP
Source: WDI 2020
FIGURE 24.1 Employment and growth – India vs. the world.
2050. Nevertheless, the evolution of labour force participation rate, here ex-
pressed as the ratio of the labour force over the population aged 15 and more,
shown in Figure 24.1 – right panel, is decreasing monotonically since 1991 and
India shares this feature with the rest of the world, although the decrease is
stronger in India since the mid-2000. Indeed, at the end of the period, barely
one Indian over two participates in the labour market, whereas the worldwide
average is over 60%. Employment (as a percentage of the population aged 15
and more) follows the same evolution, with a last value of 46%. The puzzling
fact is that economic growth in India remains high, around 6%, compared to
the worldwide average less than 3% and seems not to translate into job creation
or more job opportunities for India workers. Indeed, the Indian employment
elasticity3 is equal to 0.23 over the period, which is lower than the worldwide
value (0.39).
Moreover, close values between employment and the labour force partici-
pation rate are not surprising as Indian unemployment (as a percentage of the
labour force) remains stable around 5.5% for the whole period, 1991–2019. The
world average of unemployment evolves less smoothly as in developed countries
unemployment aligns more with economic conditions. A possible explanation of
this divergence relies on the agent’s choice between unemployment and inactiv-
ity according to her educational attainment. Indeed, labour force participation in
India is higher among the advanced education group, suggesting that unemploy-
ment rate should be higher also, as shown in Table 24.1. For the basic education
TABLE 24.1 Education and labour market outcomes in India
Education, % 1994 2000 2005 2010 2012 2018

Labour force (% of the working age population group with the same education)
Advanced 72.3 69.2 67.3 62.7 63.2 59.9
Basic 56.5 55.8 56.9 54.3 51.6 48.9
Intermediate 51,6 50,7 52,7 45,5 43,7 40,8
Unemployment (% labour force group with the same education)
Advanced 10.0 10.3 8.7 7.4 8.4 15.3
Basic 3.4 3.9 2.8 2.4 2.4 4.2
Intermediate 9.4 9.2 7.9 6.9 6.3 10.8
Source: WDI (2020).
TABLE 24.2 Informal employment in India (2012 Indian surveys)
Percentage Shares In the Informal Outside Total

(Excluding Agriculture) Sector
Share in total employment 80.9 (64.3) 6.5 (12.2) 88.2 (78.1)a

Total employment 81.1 (64.7) 18 (33.7) 100.0a
Urban 58.6 (64.3) 15.2 (12.2) 75.9 (74.1)
Own-account workers 94.3 0 94.3
Contributing family workers 98.8 1.2 100
Source: ILO (2018).

a The difference is due to households. In brackets the corresponding percentages excluding agri-
cultural employment.
group, only one worker over two participates in the labour market, and in that
case, they rarely are unemployed. The intermediate education group behaves like
the basic (resp. advanced) one in terms of labour force participation (respectively,
unemployment rate). Education is one of the pieces of labour segmentation, the
more educated having better working conditions, and as expected offering more
protection against the current of the Covid-19.
Another distinctive feature of labour markets in developing countries is the
overwhelming weight of self-employment, having decreased over the period
but remaining high at 86% in 2019, whereas the worldwide average is less than
50%. Figure 24.1 also shows that vulnerable employment, which includes own-
account and contributing family workers, is very close to self-employment. As
can be seen from Table 24.2, the nature of vulnerable employment and hence
self-employment itself in India is largely informal – e.g. without access to social
security benefits, as 94.3% own-account and 100% contributing family workers
are informal. Formal employment applies to less than 20% Indian workers, as
the share of informal employment in total employment is 88.2%.4 Again, ed-
ucation is key to determine the allocation between formal and informal jobs,
highlighting the trade-off between unemployment and inactivity according to

informality.
The last decades testify an increase in the informal employment in the formal
sector – including registered incorporated firms, largely offsetting the corre-
sponding decrease in the informal sector, with the share of total employment
in and outside the informal sector, respectively, equal to 81.1% and 18.8%, the
remainder working in households.
Consequently, looking at labour segmentation along the lines of self-
employment and informality provides a very similar picture in India. Reducing
informality in the labour market then raises the question of job destruction, as a
majority of Indian workers are self-employed, or the issue of decent work. When
discussing the impact of the pandemic on the future of work, it is then necessary
to have in mind its impact on the nature and quality of jobs.
24.2.2 Productive structure and informality

If we go further in the analysis of the labour market segmentation, economic sec-
tors add information as they are more or less correlated with working conditions,
as written contracts and/or access to social security benefits. Indeed, all around
the world, agriculture experiments in average a large decrease in employment
over the period, the share of agricultural employment in total employment is
equal to 60% in the nineties and ends at 40%. The value added of agriculture
as a percentage of GDP is rather low, declining slowly beyond and below 20%.
In India, agriculture is then the biggest employer with the lowest productiv-
ity. Furthermore, agriculture provides the largest share of informal jobs, 48.6%,
compared to the other sectors that contribute one-quarter each to informal em-
ployment as can be seen in Table 24.3.
It does not necessarily imply that rural-urban migration in India helps es-
caping from informal employment, as the share of informal employment in ur-
ban areas remains high, at 75%. Indeed when excluding agricultural activities
from employment, the share of informal employment in total employment is
only barely different, at 78.1%. The change proves to be remarkable in the com-
position of informal employment across production units: 64% in the informal
sector vs. 33.7% outside (the remainder working in households). We can draw
a first conclusion at this point of the analysis: If reducing agricultural share in
TABLE 24.3 Informal employment by economic sector in India
Percentage Shares Agriculture Industry Services
Share in total employment 99.7 81.4 77.6

Share in informal employment 48.6 25.1 26.4
Share in formal employment 1.1 42.6 56.3
Source: ILO (2018).

GDP and employment is a necessary condition to improve job quality, it is not

a sufficient one, which implies that urbanization maintains the actual profile of
labour segmentation. Second, the mutation of the productive structure, gener-
ally defined as the first step of the economic modernization, seems skipping the
stage of industrialization in favour of tertiarization. This could be dramatic as
the industrial development, characterized by large-scale production, is associated
with the emergence of trade union and the implementation of employment law,
ensuring more formal job opportunities. On the contrary, the switch to a service
economy tends to threaten the bargaining power of workers in the employment
relationship by encouraging self-employment and small production scales, and
then deterring working conditions for more workers.
The economic survey 2019–20 produced by the Government of India 5 pro-
duces a detailed description of the recent evolution of the services sector. Indeed,
the hierarchy of industry and services in terms of employment (as a share in
total employment) and value added (as a percentage of GDP) is approximately
the same: The service sector is the most productive, reaching 50% of Indian
GDP, which is lower than the worldwide average, whereas the industrial values
added are similar. When looking at employment from the World Development
Indicators, the picture is similar as the Indian service sector accounts for 30% in
2019, and the worldwide average being 50%. Growth rates are higher in India,
reflecting the traditional catch-up phenomenon. The main driver of Indian GDP
growth rate is clearly the services sector.
The key issue at this point is that the content of jobs in services is decisive to
understand the impact of the Covid-19, now and in the future. Indeed, the cur-
rent response to the pandemic in the absence of a vaccine was the development
of “working from home” strategy, which requires a high degree of digitalization
in the production process, not encountered by all occupations.
24.3 Covid-19: changing labour segmentation?

The comparison of Indian labour outcomes with their worldwide counterpart
helps to identify the nature of the market segmentation. If formality, e.g. access
to a safety net protecting against negative shocks as income or job losses, remains
a distinctive element of this segmentation when comparing developing and de-
veloped countries, it is more and more difficult to distinguish it from employ-
ment status. Indeed, informal employment and self-employment seem to overlap
to the point of being confounded, and India is a good example of this. Indeed,
salaried employment and formalization grow with economic development and
even if India has experimented higher growth rates in the last three decades com-
pared to the worldwide average, a majority of her workers are self-employed and
informal employment has not decreased. However, there is evidence of hetero-
geneity within the informal segment,6 and hence among self-employed workers.
The informal segment is composed of two tiers: The low tier includes a majority
of wage-earners waiting for a formal job, the up tier a majority of self-employed
workers having chosen voluntarily to be informal. This voluntary choice is ex-

plained by several factors: High human and financial capital, low quality of in-
stitutions and weak public good provision. Education and economic sectors will
add on the design of the labour market segmentation: Skilled workers are likely
formal as economic sectors with capital-intensive production, requiring high-
skilled workers.
Finally, the borders between formal and informal jobs on the one hand, and
standard and non-standard jobs on the other hand tend to fade away: This evo-
lution of labour segmentation could be associated with automation, leading to
digitalization.7 As the current state of the labour market in developing countries,
where labour is abundant and capital scarce, is mainly driven by labour demand,
we first argue that labour segmentation in its current form remains with stronger
relevance, due to the substitution effect at work with digitalization. Second, the
outbreak of the pandemic has frozen or even deteriorated the position of the most
vulnerable in the labour market, expanding segmentation.
24.3.1 Recent background: automation and the replacement effect

One of the major stylized facts on labour markets in developed countries is job
polarization, which describes the decline in the share of employment in middle-
wage occupations, compared to the rise or stability of this share in the two other
occupations (low-wage and high-wage). Automation and off shoring due to glo-
balization are the main explanations of polarization. Autor and Dorn (2013) assign
it to automation and distinguish two forms of it: Employment polarization –
describing the sectoral change in occupations for low-skilled workers (from in-
dustrial to service ones), and wage polarization, corresponding to the rising of
earnings at the top and bottom of the distribution. For our purpose, we will
focus on employment polarization. For Autor (2015), the substitution mechanism
at play in automation, especially for employment polarization, leads definitively
to net job destruction. Automation also generates complementarities between
capital and labour. This author predicts that the continuous thread of new tech-
nologies in AI will reinforce the substitution in favour of computers for routine
tasks, while magnifying the need of labour for innovations. Acemoglu and Re-
strepo (2019) build on these substitute/complementary components a theoretical
framework for understanding the effects of automation on labour demand. The
re-allocation of production tasks to capital and labour as a result of automation
is determined by two mechanisms. The first one, called displacement effect,
deprives labour of some productive activities or tasks to assign them to capital.
This substitution mechanism is well-known to increase labour productivity by
reducing the labour share in value added, with a possible negative impact on
labour demand. The second one, called reinstatement effect, reflects the job cre-
ation process associated with automation, given that the new productive tasks
are assigned to labour because of its comparative advantage. Contrary to the
displacement effect, it raises labour share and labour demand. The final impact
of automation on the composition of production (tasks assigned, respectively, to

labour and capital) and then on labour demand is unclear. From a welfare per-
spective, this result is secondary compared to the result on productivity. Produc-
tivity gains, that are expected in the case of automation, will improve wealth but
the final wealth sharing and the inequalities issues depend on social and political
institutions.
Acemoglu and Restrepo (2019) apply their framework to the US labour mar-
ket over the period 1947–2017 and explain the growth of employment from the
joint action of the displacement and reinstatement effects. They found that the
substitution effect accelerates during the past three decades whereas the reinstate-
ment effects weaken, leading to a slowdown of employment growth.
As Indian and US labour markets have in common that they are flexible,
and Indian economy being largely open to international markets, the prevalence
of the displacement effect over the reinstatement one could explain the jobless
growth in India as shown in the previous section. Moreover, in order for the job
creation mechanism to outweigh the substitution mechanism, it requires rapid
adjustment of workers skills to the new tasks as well as an efficient matching
process in the labour market. Indian labour market does not complete these con-
ditions as underlined by Chowdhury (2014). Indeed, she shows that the source
of skill shortages in the Indian labour market could be explained by skill mis-
matches or allocative inefficiency. Skill mismatch has two components: The lack
of skills in the labour supply, and over-education, which corresponds, in employ-
ment, to the gap between the education attainment and the requirements of the
job. The last component of skill mismatch reinforces the displacement effect: It
increases the competition with high-skilled workers for jobs originally intended
for less-skilled workers. If over-education is high in India, it could explain why
the labour force participation rate of people with basic education is lower than
the one of people with advanced education, as seen in the previous section.
Calvão and Thara (2019) provide a picture of digital work in India. Their goal
is to assess the impact of the Fourth industrial revolution, on the Indian labour
market, by looking at the changes in activities like driving, and the IT industry.
Their main finding is the rising job insecurity caused by the digitalization pro-
cess, even for the more educated workers in the IT industry, for whom formality
is regular. The new threat on job opportunities is the reverse outsourcing as au-
tomation goes on replacing labour for an always larger range of tasks and induces
foreign companies to go back home. This issue would worsen the over-education
phenomena by increasing the surplus of unemployed high-skilled labour.
Rising economic development does not keep all its promises in terms of job
opportunities and improving living standards under the combined effects of glo-
balization and automation. International competition among exporting firms
and digitalization lead first to an increase in international and domestic labour
demand but biased in favour of high-skilled workers, who could either migrate
or stay home. Nevertheless, this increase in job offers, especially obvious in the
sector service, is far from satisfying the population needs. It changes the domestic
wage distribution by magnifying income inequalities. As digitalization deepens,

employment polarization becomes more pronounced and decreases the upper tail
of the occupation distribution, which cannot be replaced by capital. The ability
of the Indian labour market to implement the reinstatement effect by creating
new tasks and also solving the skill mismatch issue determines the future trajec-
tory of employment along segmentation.
24.3.2 The outbreak of the Covid-19

To understand the impact of the pandemic on labour market segmentation, we
look at how it interferes with the logic of automation driving currently the em-
ployment path. The Covid-19 is an exogenous and common shock to all coun-
tries, resulting then in circular answers from countries according to their social,
economic and political singular conditions, as they are all related by global value
chains. This productive dependence, as every country could differ in the number
and intensity of tasks she performs along the global value chains, also explains the
choice of policy measures to deal with the pandemic. For developing countries
involved deeply in the global value chains, it is a matter of concern as developed
countries could accelerate the end of offshore insourcing and outsourcing. We
choose to focus on the direct adjustment made by the labour markets following
the outbreak of the pandemic, the predictions being conditional to some invar-
iance in these global value chains. In the case of India, this assumption is not so
strong as, according to the Government of India (2020),8 India is characterized
by lower participation in global value chains compared to the major exporting
nations in East and Southeast Asia.
In most countries, the primary effect is a huge increase in unemployment.
Some countries have resorted to short-time working or temporary lay-offs that
were publicly funded – in whole and in part, as a buffer against the health crisis.
Indeed, these policy measures allow a faster economic recovery by sustaining
purchasing power and then consumption. For countries that decided to imple-
ment lockdowns, working from home, when the occupation allows it, is adopted,
in addition or not to the unemployment measures. Then, identifying workers
that are likely being unemployed provides a picture on how the pandemic could
interact with the current state of the labour market as driven by digitalization.
Kaplan et al. (2020) classify occupations in order to depict the heterogeneity in
exposure to Covid-19 due to work. Social interaction, which captures the risk
of having the Covid-19 due to employment, and flexibility, defined as the pos-
sibility to working from home, determine this sensitivity. Social interaction, as
flexibility, could be high or low. The combination of these modes provides four
groups of occupations. An additional group includes occupations considered as
essential by local authorities, and they are characterized by low flexibility.9 Ac-
cording to this classification, we can see that the pandemic magnifies the transi-
tion to unemployment or to informal jobs or inactivity in developing countries
along the same lines than dualism; the standard result that education protects
from unemployment is also valid for the pandemic.10 The typical profile of the
worker whose activity is threatened by the pandemic combines low earnings,
youth, low skills, temporary contract, service job like in the tourism sector.
Furthermore, the pandemic fosters digitalization because a way to avoid
disrupting business is working from home, which is a form of automation.
Service jobs around and within the workplaces are then threatened. Here, a
huge difference between developed and developing countries could be ob-
served as working from home depends highly on IT infrastructure quality. In-
deed, Garrote Sanchez et al. (2020) report that 1 job in 5 across the world offers
telecommuting possibilities, this number decreasing to 1 in 26 in low-income
countries. And this estimation is certainly an upper bound as it does not take
into account internet access. Consequently, the impact of the pandemic on dig-
italization by this way is mitigated in developing countries, at least in the short
run. In India, the service sector is characterized by a large spectrum of occupa-
tions and skills, from own-account workers, mainly informal to regular high-
skilled workers, largely implied in IT occupations, meaning that the pandemic
has heterogeneous effects on employment in service sector. Nevertheless, on
average, the most popular worker profile in the service sector remains the first
one. Indeed, for the Government of India (2020), job creation’s strategy, due to
India’s comparative advantage in labour-intensive activities, relies on economic
sectors exploiting rather unskilled labor, like textiles, clothing, footwear, toys
and network products in the case India specializes in final assembly, again
requiring less-skilled workers. Firms could also react to the casual shortage
of labour supply provoking the pandemic by fostering. This investment may
prove to be irreversible, even with the end of the pandemic, again challenging
job opportunities.

Predictions on the future of work and labour segmentation in developing coun-
tries could be made even if the uncertainty on the duration of the pandemic
remains high. The trajectory of employment will not be changed structurally,
given that global value chains become resilient to the pandemic, the current
main drivers staying on move. Demographic pressures, employment shortages
and automation will remain relevant, at least in the short run, to explain the
dynamics of labour supply and labour demand. Even if working from home will
not spread as fast as in developed countries, it will be adopted questioning the
relevance of service jobs depending on workplaces and their relocation. Autor
and Reynolds (2020) point out that their confidence on automation to create
more jobs than workers to fill them has been shaken by the outbreak of the
Covid-19 in the US labour market. For developing countries, the global picture
of labour segmentation before and after the pandemic remains oddly stable, pri-
marily governed by self-employment. The recovery from the pandemic could be
then designed on well-known policy programmes as the one for decent work.
Nevertheless, any sizable change in the global value chains could render null and
void the invariance of employment trajectory in the long run.
At the moment of chapter’s publication, uncertainty remains high about the
future of work in the world and in India in particular. ILO (2021) develops two
scenarios, one optimistic and another one more pessimistic, highlighting the
heterogeneity of the recovery among countries. The pandemic results in di-
vergences in the evolution in total employment and by economic sectors across
countries. Common thread, however, is that low-paid employment experienced
greater job destruction, and self-employment paid the heaviest price to the pan-
demic, which again questions the future of informality. Some Indian economists,
such as Ashwini Deshpande,11 worry that India’s annual budget misses the point
on income support, especially to reduce the huge gap between paid and self-
employed workers. This kind of policy measures, according to ILO (2021), ex-
plains most of the differences of the pandemic impact on national labour markets
and the success of their recovery.
Notes
1 Figures in this section are based on the World Bank Indicators dataset and tables
extracted from the appendices of ILO publication in 2018, relying on Indian surveys
carried out in 2012.
2 ILO in 1999 defines a multidimensional concept, decent work, that covers four
dimensions as employment creation, rights at work, social protection and social
dialogue.
3 This raw calculation is obtained by estimating the OLS regression of the logarithm of
the labour force on the logarithm of the GDP (PPP, constant 2017 US dollars).
4 The corresponding shares for the World and developed countries, reported in ILO
(2018), figure 5, panel C, are, respectively, 61.2% (50.5%) and 18.3% (17.1%).
5 See chapter 9 of volume 2, https://www.indiabudget.gov.in/economicsurvey/doc/
vol2chapter/echap09_vol2.pdf.
6 See the seminal paper of Fields, G. (2011). Labour market analysis for developing
countries. Labour Economics, 18, S16–S22.
7 Digitalization refers to labour transformation by automation-based technology.
8 See chapter 5 of volume 1, https://www.indiabudget.gov.in/economicsurvey/doc/
vol1chapter/echap05_vol1.pdf.
9 Examples of occupations are, respectively, nurse (essential), retailers (high social in-
teraction but low telecommuting), engineers (low social interaction and telecommut-
ing), teachers (high social interaction and telecommuting) and software developers
(low social interaction and high telecommuting).
10 For example, for the US labour market, Kaplan et al. (2020) find that workers with
high social interaction occupations and low possibility of working from home are also
the more vulnerable to negative income risks.
11 Missing in Budget: Push for jobs, welfare. The Indian Express, February, 2, 2021.
References
Acemoglu, D. & Restrepo, P. (2019). Automation and new tasks: How technology dis-
places and reinstates labor. The Journal of Economic Perspectives, 33(2), 3–30.
Acemoglu, D. & Autor, D. (2011). Skills, tasks and technologies: Implications for employ-
ment and earnings. In O. Ashenfelter & D. Card (Eds), Handbook of Labor Economics,
Volume 4b (pp. 1043–1171). Elsevier.
Autor, D. & Reynolds, E. (2020). The nature of work after the COVID crisis: Two few
low-wages jobs. The Hamilton Project – Brookings, Essay 2020-14.
Autor, D. (2015). Why are there still so many jobs? The history and future of workplace
automation. Journal of Economic Perspectives, 29(3), 3–30.
Autor, D. & Dorn, D. (2013). The growth of low-skill service jobs and the polarization of
the US labour market. American Economic Review, 103(5), 1553–1597.
Calvão, F. & Thara, K. (2019). Working futures: The ILO, Automation and digital work
in India. International Development Policy, 11, 223–246.
Chowdhury, S. R. (2014). Skill mismatches in Indian labour market: Policy priorities &
challenges ahead. The Indian Journal of Industrial Relations, 49(3), 422–438.
Government of India (2020). Economic Survey 2019–2020. New Delhi: Ministry of
Finance, Department of Economic Affairs, Economic Division. https://www.
indiabudget.gov.in/economicsurvey.
Garrote Sanchez, D., Gomez Parra, N., Ozden, C., Rijkers, B., Viollaz, M. & Winkler,
H. (2020). Who on earth can work from home. World Bank Policy Research Working
Paper 9347.
ILO (2021). ILO Monitor: COVID-19 and the world of work. https://www.ilo.org/wcmsp5/
groups/public/@dgreports/@dcomm/documents/briefingnote/wcms_767028.pdf
ILO (2018). Women and men in the informal economy: A statistical picture. https://www.ilo.
org/global/publications/books/WCMS_626831/lang--en/index.htm
Kaplan, G., Moll, B. & Violante, G. (2020). The great lockdown and the big stimulus: Tracking
the pandemic possibility frontier for the US. NBER Working Paper 27794.
INDEX
Note: Bold page numbers refer to tables; italic page numbers refer to figures and page
numbers followed by “n” denote notes.
Aadhaar-linked bank accounts 159 aggregate supply (AS) shocks: aggregate

Abe, Messrs 53 production function 140; domestic
Abiad, A. 3 capital 141; foreign capital 141; macro
Abraham, A. 290 theory 138; non-food price 142;
Abraham, R. 290 NSO 139; post-Covid shift 138, 139;
accelerated digitalisation 38 Solow-type one-sector growth model
Acemoglu, D. 262, 408, 414–415 140; WPI≈139
active cases, regional distribution 360; air quality index (AQI) 199
district-wise distribution 361, 361; alcohol-based sanitizer 329
Getis-Ord gi*(d) statistics 362, 363; Alfaro, L. 232
growth rate 359–360; hotspots and cold Alon, T. M. 310
spots 362, 363; LISA maps 362, 362 Al-Othaimeen,Yousef A. 84
adult mortality risk 111 al-Qaeda 81, 83, 89n24
Afghanistan conflict: al-Qaeda 81, 83, Anderson, S. 310, 311
89n24; coronavirus case 84; international Andrés Manuel López Obrador 53
support 83; Ministry of Interior Affairs Ang, E. 239–240, 249
85; pandemic amidst 83–85; Taliban animal-based foods 195
attacks 81–82; towards peace 82–83; animal labour 283–284
US-Taliban talks, Qatar 82–83; war Annual Surveys of Industry (ASI) 7, 262
89n23; Watson Institute at Brown Ansar Allah movement 78
University 82 antagonism 44
Agarwal, L. 355 anti-Asian racism 45
aggregate demand-aggregate supply (AD- anticipated transition process 380
AS) model 130; ASI 7; demand-side antimicrobial resistance 195
contraction 5; firm fragility 7; IIP 7; apocalyptic pandemic 76
interaction of 5, 5; international trade 6; AQI see air quality index (AQI)
lay-offs and unemployment 6; liquidity Ardern, Jacinda 44
crisis 6; non-fragile firms 7; OECD Arumugam, Dr. U. 279
6; social distancing 5–6; supply chain asset ownership, widowhood and
disruption, India 6–7; supply chain deprivation 314–316, 315
networks 7 ‘Atmanirbhar Bharat’ package 13–14, 22,
aggregate production function 140 25, 131, 260
422 Index
atmanirbharta 76 capital-intensive import-competing

Autor, D. 408, 414, 417 sector 189
Ayushman Bharat scheme 21 capital-intensive production 122
Azoulay, Audrey 15–16 Carberry, P. 280
carbon emissions reduction 40
Baekgaard, M. 52 case fatality rate (CFR) 333, 334
Bag, S. 290 caste-and gender-based disparities, India's
BAI see Business Assessment Index (BAI) lockdown 68
balanced budget 174n1 casual labour 286, 287
balanced growth process 299 Center for the Monitoring of the Indian
balance of payment equilibrium: and BP Economy’s (CMIE) unemployment data
curve 134; modified standard model 215, 216
135–137 Central Statistical Organisation
Baldwin, J. R. 262 (CSO) 278
Baldwin, R. 289 Centre for the Study of Developing
Bandyopadhyay, S. 15, 26, 312–314 Societies (CSDS) 402
Banerjee, Abhijit 371 Chakrabarti, S. 10
Banerjee, Lopamudra 9 Chakraborty, Chandrima 11
Banker, R. D. 282 Chakraborty, L. 164, 165
Baser (Colonel) 84 Chakraborty, S. 18, 110
Basole, A. 290 Charnes, A. 282
Basu, S. 215 Chatterjee, P. 6–7, 13, 23, 269
Bhattacharyya, R. 14 Chattopadhyay, Saumen 16, 25
Beck, T. 233–234 Chattopadhyay, Srobonti 9
behavioural patterns 1–2 Chetty, R. 150
BEI see Business Expectations Index (BEI) China: Covid-19 cases 109, 110; life
Ben Salem, Melika 10 expectancies 110, 110–111
Bhattacharjee, Manojit 10 “Chinese virus” 45
Bhattacharya, R. 14, 26 Choi, T.Y. 239–240
Bhattacharya, S. 9 Chowdhury, S. R. 415
Bin Laden, Osama 81 climate change 195, 196–198, 207; “banner
biodiversity 194–195 year” 197; greener economic growth
Birla, B. 290 203–205; health and environmental
Black Death 324 policy 198; low-carbon economy 198;
Black (bubonic) Plague 75 Paris Agreement 197; root causes 198;
‘black swan’ event 233 socioeconomic impact 197; World
Boehm, J. 262 Health Organisation 196; zero-carbon
Bol, D. 52 economy 197
Bolsonaro, J. 53 CMIE unemployment data see Center for
Bose, Montu 18 the Monitoring of the Indian Economy’s
Bottazzi, G. 232 (CMIE) unemployment data
Brazil 1; Covid-19 cases 109, 110; life coefficient of variation (CoV) 217, 223
expectancies 110, 110–111 cohort-level mortality 127n4
Bros, Catherine 26 Collins, C. 310
brown environmental consequences 196 concomitant policy uncertainty 41
brown projects 203 concussion 40
Brown, R. 232 conflicts 24–25; in Afghanistan 81–85;
Business Assessment Index (BAI) 23 de-escalation 87; India-China border
“business as usual” scenario 32 79–81; violence amidst virus 75–88; in
Business Expectations Index (BEI) 23 Yemen 77–79
constant returns to scale (CRS) 282–283
Cainelli, G. 239 consumer behaviour 146
Calvão, F. 415 Consumer Pyramids Household Survey
Cantore, N. 109 (CPHS) 216, 219–220
Index 423
Consumption Expenditure Survey and migrant workers 391–404, 393, 394,

Debt-Investment Survey 292 396; migration 353–356; MNREGA
consumption, widowhood and deprivation scheme 354; morbidity and mortality
312–313 18; multi-level regression 358–359; out-
Conte, Giuseppe 44 migration and incidence 362–364, 364,
Cooper, W. W. 282 365, 366; policy actions, India 163–173,
coordinated supply shock 235 173; policy failure before 323–335,
coronavirus pandemic: de-globalization (see 331–334, 334; population movements
de-globalization); digital technologies 355; reverse migration 366, 367, 368,
2; economic disruption 1; economic 369, 369–370; RT-PCR test 335n1,
impact 2, 2 (see also economic impact); 366; Shramik trains 355, 356, 358, 366;
economic revival policy, post-corona spatio-temporal variations 355; special
period 16, 17, 18–26, 19, 20, 22; global trains 354–355; statistical inference 357;
economy 2; macroeconomic front “timely and tough” decisions 372; trust
lockdown 1; manufacturing activities in governments 26; West Bengal Health
1; production and transport systems 1; Department report 367, 367; widowed
social impact 2, 2 (see also social impact); and elderly women, India 309–317,
unemployment 1; unorganized informal 315, 317; women workers 9; see also
sector 1; work from home 1; see also coronavirus pandemic
Covid-19 pandemic Covishield 14
corporate world: business-model CPHS see Consumer Pyramids Household
disruptions 202; environmental footprint Survey (CPHS)
202; environmental policies 203; green credit markets 148
projects 203; pandemic and post- CRS see constant returns to scale (CRS)
pandemic responses 202; physical climate CSO see Central Statistical Organisation
risk 201; post-pandemic world 202; (CSO)
social acceptability 203; supply chains
201; videoconferencing 202 Daniele, G. 52
Covaxin 14 Dasgupta, D. 6, 215, 289
Covid19India.org (C19I) 95 Das, M. 18, 110
Covid-19 pandemic: active cases, regional data envelopment approach (DEA) 282
distribution 359–362, 360–363; acute Datta, P. K. 10
coronary events 357; cases, lockdown Deaton, A. 312
India 4–5; climate change 196–198, debt trap diplomacy 25
207; corporate world 201–203; cross- decarbonisation 208
infection 371; database 356; econometric de-globalization 13–14; bargaining model
models 358; economic impact 149–152, 191; capital-intensive import-competing
151–152; firm fragility 231–253, 236– sector 189; Financial Crisis 178; Financial
237, 238, 242; fiscal stimulus 22; gender Meltdown 179; global trading 178, 179;
inequalities 9; global annual growth Great Depression 178; hard-earned
deceleration 3; global political economy “flatness” 180; long-run tariffs 188–189;
31–47, 33, 34, 35, 39, 41; governments market-driven model 182; MNREGA
trust 50–62, 51, 53, 54–57, 58–61; wage 183; NTMs 180; OECD 179;
GWR model 359; Hindu-Muslim pre-pandemic protectionism 181–182;
conflict, India 92–103, 98, 100; pro-social welfare bias 189; public
household adjustments, India 213–228, distribution system 184; RBI policies
214, 217, 219–222, 226; income 184, 185; RTAs 180; self-reliance 182,
distribution shock 23; Indian agricultural 184, 185, 189; short-term stabilization
sector 278–288, 281, 285–286; India’s policies 185, 186; tax concessions 180–
lockdowns 198–201; lockdowns 353; 181; theoretical implications 186–189;
macroeconomic impact, Indian economy trade and investment 179; unilateral
129–142, 137, 139; manufacturing tariffs 187–188; US-China trade war
industries, India 260–276, 265–270, 192–193; world exports 179, 179; world
266, 268, 271, 272, 272–275, 277; FDI 179, 180
424 Index
degree of religious polarisation 104n3 economic incentives 23, 147

demand-deficient economy 215 economic relief packages 24
demand management policies: ‘Atmanirbhar economic resilience 146
Bharat’ package 22; fiscal stimulus Economic Resurgence Plans 24
package, Asian countries 21, 22; economic revival policy, post-corona
‘helicopter money’ 22; Keynesian fiscal period: debt trap diplomacy 25; demand
response 23; macroeconomic analysis 21; management policies 21–23, 22;
policy declarations 21–22 economic relief packages 24; education
Demirguc-Kunt, A. 234 quality 25; fiscal and debt sustainability
Demographic and Health Surveys 18, 19; gender parity 26; ‘green’
(DHS) 314 component 25; health and educational
Deshpande, A. 310 infrastructure 18; health-care system
de-unionisation 43 20–21; Hindu-Muslim conflicts 24–25;
developing countries: Aadhaar-linked bank human development indicators 16,
accounts 159; fiscal stimulus 158–159; 17; labour market measures 24; open
household incomes 158; MNREGA 158; economy 19; strategies and policies 20,
MSME 159; myriad welfare schemes 20; supply-side impact 23–24; tourism
159; PDS 158; political accountability and hospitality 25–26; traditional vs.
159; tax-transfer scheme 158 online education 25; with and without
Dev, M. S. 3, 279 social security 18, 19
Dey, S. 269 Economist Intelligence Unit (EIU) 326
DHS see Demographic and Health Surveys economy recession 33, 33
(DHS) education: categories 220, 220; gradient,
DiCarlo, Rosemary 76 heterogeneity 222–223; labour market
difference-in-difference strategy 263, 264, outcomes, India 410, 411; quality 25; see
272–273, 273, 274 also higher education, post-Covid era
disease-containment strategies 325 effect of pandemic on firms and households
disruptive technology: higher education, see firm fragility; household adjustments,
post-Covid era 337 India
“distortionary” welfare schemes 43 elderly protection, India's lockdown 70, 71
Dodds, K. 376 e-learning 16
domestic violence 77; for women 85–86, emotional response 50, 62
90n36, 91n39 employment: categories 220, 221, 222;
Dorn, D. 414 demand-side effects 9; formal 411; GDP
Dubey, J. D. 317 growth 7–8; global unemployment 7–8;
Dutta, Arijita 18 high-skilled workers 9; ILO report 9;
impact on workers 8, 8; informal, India
e-commerce entities 235 410, 411; informal workers, vulnerabilities
Economic Census (EC) 303 10; labour force participation rate 410;
economic contraction 160 labour market 8–9, 11; labour-saving
economic growth 194; Covid-19 cases, technologies 10; labour surplus states
India 4–5; financial markets 3; GDP and 11–12; ‘non-standard’ workers 380;
GVA growth predictions, India 3–4, NSSO data 10; polarization 414; pre-
4; global demand and supply chain 3; and post-corona unemployment rates,
greener 203–207; OECD 3; WEO 3 India 10–11, 11; share 38, 39; ‘standard’
economic impact: AD analysis 5, 5–7; AS capitalist 379, 383, 386; vulnerable 411;
analysis 5, 5–7; cross-sectional volatility with and without social security 18, 19;
152; distributive effect 150; economic vs. work situation 378–381; youth 10
growth 3–5, 4; employment 7–12, 8, 11; environmental quality 1–2
GDP 149, 150; Great Depression 150; Epidemic Diseases Act 224
income distribution data 150; international epidemic evolution 53–57
trade 12, 12–14, 13; mortality risk 149; epidemics: adult mortality risk 111;
share prices 150, 151; standard deviation aggressive policy measures 109;
151, 152;World Wars 150 balanced growth path 115–117, 116;
Index 425
Covid-19 cases 109, 110; economic Gallegati, M. 232

cost 109; human capital-dependent Galwan valley control 79–80
sector 114–115; insurance mechanism Gangopadhyay, Anupriya 14
126; intergenerational wealth dynamics Garrote Sanchez, D. 417
113–114; life expectancies 110, 110–111; gender-based violence 85
management 58–59; OLG economy gender-day scheme 87
111–112; pandemic-related mortality gender inequalities 9, 88, 317
effect 111; production function 114–115; gender parity 26
raw labour life 112–113; short-and long- Geographically Weighted Regression
run effects 117–126 (GWR) models 359
epidemiological impact 150 GFCF see gross fixed capital formation
Esteban, J. 97 (GFCF)
Estupinan, X. 290 Ghose, Arpita 11
European Center for Disease Prevention Ghosh, P. 23, 148
and Control 58 Ghosh, Shinjini 402
evidence-based policy 44 Gini Coefficient 219, 219, 227
exchange rate 164 global demand and supply chain 3
global economy 2
Fasani, Francesco 395, 396, 403 Global Health Index 326
fertilizer, rice production 283–284 globalisation: political economy 36, 40–42;
financial crisis 178, 179 social entitlements 42
financial markets 3, 36 global ranking, Indian universities 347–348,
Financial Meltdown 179 349
financial sustainability, higher education, global trade: de-globalization 178, 179;
post-Covid era 348 GVC 40, 41; political economy 34, 35
firm fragility: anecdotal evidence 251; global value chain (GVC): carbon emissions
‘Bare-Bones Survival’ 250, 252; reduction 40; concomitant policy
‘Blessing in Disguise’ 250, 251; capital uncertainty 41; concussion 40; domestic
value margin 237; CARES Act 252; economies 42; domestic production
‘Chokehold’ 250, 252; definition 239; 40; fragility 36; global trade 40, 41;
e-commerce firms 232; in economic manufacturing industries, India 261;
systems 235, 236–237; economy-wide protectionism 41; re-nationalising 40;
fragility 232; empirical import 240–241; re-organisation 31; stagnation/reversal
‘Exit with Contagion’ 250, 252; ex post 32; supply-side shock 40
measurements 253; financial fragility Goldthorpe, J. H. 377
index 237; generalizability 239; industrial Goods and Services Tax (GST) 159
ecosystems 233–235; inter-firm links Great Depression 1, 33, 34, 42;
238; macroeconomics 232, 237, 238; de-globalization 178; economic impact
MERS 231; micro-economic foundation 150; informal sector, India 290; short-run
233; modeling 241–251; network-based demand-supply management 133
measures 250; ‘non-fragile’ balance sheet ‘green’ component 25
233; non-pharmaceutical interventions greener economic growth: climate change
231, 232; policy implications 252–253; 203–205; green fiscal recovery packages
risk-sharing networks 239; SARS 204; India 206–207; international 205–
231; supply/demand side 231; supply 206; Kyoto protocol 204; low-carbon
network 239; supply network outcomes, investments 203; sustainable recovery
limitations 251; vulnerability 240 efforts 204
fiscal and debt sustainability 18, 19 green fiscal recovery packages 204
fiscal austerity 42–43 green projects 203
fiscal stimulus package, Asian countries “green” recovery 205, 206
21, 22 Green Stimulus Index 206
Fordist economic structure 379, 380 Gross Capital Formation 163
formal employment 411 Gross Domestic Product (GDP): fiscal
fossil fuel-based industry 206 policy 142; and GVA growth predictions,
426 Index
India 3–4, 4; Indian agricultural sector, Hindu-Muslim conflict, India 24–25;

efficiency loss 278; infection rate and BJP effect 96; Casualties 98, 98–100;
economic activity 146; long-run growth, C19I 95–96; civic engagement 93;
post-Covid-19 139; manufacturing communal violence/harmony 94,
industries, India 260; policy actions, 102; Covid-related outcomes 94, 95,
India 173; political economy 33–34, 34, 101–103; degree of religious polarisation
35, 36 104n3; empirical specification 97–98;
gross fixed capital formation (GFCF) 3 endogeneity 95; ethnic violence 92;
Gross State Domestic Product (GSDP) 330, “externalities” 94; “fractured” growth,
331, 331, 332 social capital 102; “Hindu-rashtra” 95;
growth rate effect 119–122, 121 inter-group conflict 93; microeconomic
GST see Goods and Services Tax (GST) data 103; OLS district-level regressions
Gupta, S. 290 101, 100; party’s ideology/stance/
Guterres, Antonio 83–84 policy 97; Recovery rate 101; statistical
significance 101;Varshney–Wilkinson
Hadi, Mansour 78–79 dataset 96; violence effects 93
Hale, T. 58 homogenous regions (HRs) 216
Halpern, L. 263 household adjustments, India 11; C19
harmonized unemployment rates 27n1 incidence 213; CMIE 215; CPHS 216,
health: climate change 198; economic 219–220; demand-deficient economy
factors 129; economic revival policy, 215; economic activity 227; educational
post-corona period 18, 20–21; policy categories 220, 220; education gradient
matrix 323; public health risk 324–326, heterogeneity 222–223; employment
399; widowed and elderly women, India categories 220, 221, 222; expenditure
316, 316–317 218; financial assets 295, 296; GDP
health care infrastructure (HI) 330–331 228; Gini Coefficient 219, 219, 227;
health policy matrix 323 HRs 216; income 217–218, 224–227,
‘helicopter money’ 22 226; labour-surplus 228; “lives lost”
Hepburn, C. 204, 207–208 215; long-term negative consequences
herd immunity: biological and economic 216; low probability events 228;
149; economic impact, Covid-19 MGNREGA 228; occupation gradient
149–152, 151–152; optimal taxation heterogeneity 223–224; Oxford C19
model 153–158 government response tracker 213, 214;
Hershbein, B. 38 PSU 216; public policy response 213,
heterogeneity: education gradient 222–223; 214; quarterly trends 217, 217, 219;
occupation gradient 223–224 residual 218; social security infrastructure
Hetherington, M. J. 51 213; twin E crisis 213; unanticipated
higher education institutions (HEIs) economic shock 215; uncertainty 228;
338, 345 welfare and poverty estimates 217
higher education, post-Covid era: digital household incomes 158
technology 337; disruptive technology household population, India 66–67
337; employability 346–347; financial household savings 36
sustainability 348; global ranking, Houthi, Mohammad Badr al-Din 78
Indian universities 347–348, 349; HEIs Hsieh, C.-T. 232
338, 345; in India, during pandemic human development indicators 16, 17
343–345; internationalization 342–343; human labour 283–284
internet-based connectivity 338; human resilience 88
job market 346–347; MOOCs 345; Hussain, Zakir 4–5, 8–9, 18
NAC 346; NEP 337, 345, 346; policy
push, internationalization 347; policy IHDS see India Human Development
thrust 345; transformation 339–341; Survey (IHDS)
unbundling and commodification IIP see Index of Industrial Production (IIP)
341–342; vulnerabilities 349 ILO see International Labour Organization
high-skilled jobs 38 (ILO)
Index 427
IMF see International Monetary Fund India Human Development Survey

(IMF) (IHDS) 314
income assessment 148 Indian agricultural sector, efficiency loss:
income shocks: digital payments 224; agricultural market reforms 280; budget
Epidemic Diseases Act 224; ex-post allocation 280; CSO 278; food supply
risk-mitigating mechanisms 225; greater chains 280; GDP 278; input-and output-
employment 226, 226–227; household oriented measures 281, 281; NSSO
adjustments, India 217–218; hunger 225; 279; PLFS 278; rice 280–281, 283–284;
migration 225–226 surplus labour 280; TE 281, 281–283; see
income taxation 147–148; ahistoric and also rice production
static analysis 158; Mirrlees mechanism Indian economy, macroeconomic impact:
156; Pigouvian taxes 156; revenue aggregate supply shocks 138–139, 139;
neutrality 157; risk averse agents 158; tax fiscal packages 129; health economics
liability 155–157; tax-transfer policies factors 129; long-run growth, post
157; truth-telling direct mechanisms Covid-19 139–142, 145; monetary
155–156 stimulus packages 129; post Covid-19
Index of Industrial Production (IIP) 7, 262, 129–133; short-run demand-supply
264; difference-in-difference strategy management 129, 133–138
272–273, 273, 274; GDP 271; import Indian migrant workers: CSDS 402;
share and dependency 272; Indian destinations of 397–398; GCC 398–399;
economy 270–271; tobacco industries India Labour Migration Update (2018)
274–275; vulnerability 274–275, 275; 398; post-lockdown phase 402; poverty-
YoY change 271, 271–272 ridden residences 400; public health
India 1; Ayushman Bharat scheme 21; risk 399; quarantine protocols 399;
Covid-19 cases 109, 110; economic RBI report 398; skilled and unskilled
cost 109; economic revival policy, post- categories 401–402; social distancing
corona period 16, 17, 18–26, 19, 20, 398; World Migration Report 398
22; employment and growth 410, 410; India’s lockdowns 198–201; AQI 199; cash
fiscal and debt sustainability 18, 19; transfers 66; caste-and gender-based
fiscal deficit 22–23; foreign trade 13, disparities 68; Covid-19 incidence
13; GDP and GVA growth predictions 71–72; disinfection campaigns 200;
3–4, 4; green/brown projects 194–208; economic impact 201; economic
greener economic growth 206–207; production loss 64; vs. economy 72–73;
Hindu-Muslim conflict 92–103, 98, 100; educational institutions 64; elderly
household adjustments 213–228, 214, protection 70, 71; inter-household
217, 219–222, 226; human development altruism 70; intra-household motivation
of 16, 17; informal sector 289–305; life 70; “macro” perspective 68; “micro”
expectancies 110, 110–111; lockdown perspective 68, 70; MNREGA 67;
4–5, 53, 55–57, 64–73; migrant workers morbidity and mortality risks 65; MSW
397–403; pandemic package 182–186, 200; occupational structure 64, 65;
183, 185; pandemic preparedness pandemic control measures 198; relief
326–328; policy actions 163–173, 173; package 67; self-isolation 67; self-
pre-and post-corona unemployment selection outcomes 70; serological testing
rates 10–11, 11; relief package 21, 70; side effects 199; social distancing
22; ‘shramik’ trains 5; social security 67, 69; social protection 69; tradeoff
measures 18; supply chain disruption 67–69; transport services 64, 67; universal
6–7; widowed and elderly women transfers 66; vulnerability to 65–66
309–317, 315, 317 indirect tax effect 172–173, 173
India-China conflict: Aksai Chin 80; industrial ecosystem, firm fragility
diplomacy 81; Galwan valley control modeling: agent firms 241; balance-
79–80; Ladakh standoff: 79–80, 89n22; sheet-based financials 248; model
Line of Actual Control 80; post- specification 241–249; supply network,
pandemic levels 80–81; transgressions modeling shocks 245–249
border 80 “inefficient” welfare schemes 43
428 Index
influenza pandemics 75 Johnson, B. 59

informal employment, India 410, 411; by Jonas, O. B. 289
economic sector 412, 412
informal occupational structure 65–66 Kahn, L. B. 38
informal sector, India: “entitlement failure” Käki, A. 239
304; food transfer 290; Great Depression Kanagavalli, G. 279
290; income impact 308; institutional Kango, Ujjwal 14, 26
credit 296; labour income 289; long Kaplan, G. 416
run 299, 301, 301–304; market-linked Kapoor, R. 317
economic relief package 296; medium Kesar, S. 290
run 298–299, 300; MSMEs 296; Keynesian fiscal response 23
OAE 290; short run 290, 297–298; Keynesian macro framework 133
unidirectional war 304; vulnerability 290, Keynesian recession 160
291, 292, 293–297, 296 Khan, F. 290
informal workers, vulnerabilities 10 Khan, M. I. 290
in-person service jobs 38 Khanna, Anoop 400, 401
institutional credit 296 Khera, R. 67
intergenerational wealth dynamics 113–114 Klein, N. 45
Intergovernmental Panel on Climate knowledge economy 16
Change 196 Korean new deal 205
international commodity markets 36 Kothari, Richa 4–5, 8–9, 18
International Labour Organization (ILO) Krause, D. R. 239–240
37, 392 Kumar, A. 280
international migrant workers 391; Bureau Kumar, S. 280, 395
of Labor Statistics data 394; heterogeneity Kundu, A. 10, 18
397; ILO 392–393; pre-COVID-19 Kushti, N. 91n41
employment 397; region-wise distribution Kyoto protocol 204
393, 393, 394; skills and educational
attainments 396; teaching professionals labour: income 289; marginalisation 43;
395, 396;World Migration Report 393 surplus states 11–12; vulnerability 387;
International Monetary Fund (IMF) 33 work 376 (see also work situation)
international trade: de-globalization labour market 8–9, 11, 38; changes 31, 37;
13–14; foreign trade, India 13, 13; inequalities 45; loss in working hours
labour-intensive production sectors 27n1; measures 24; post Covid-19,
14; US-China trade war 13–14; world Indian economy 132; recovery 384–385
merchandise trade volume 12, 12–13 labour-saving technologies 10
internet-based connectivity 338 labour segmentation: automation and
intra-country migrants 392 replacement effect 414–416; capital-
Iran-Saudi rivalry 78 intensive activities 409; Covid-19
ITS regression models 362–364, 364, outbreak 416–417; definition 409;
365, 366 human and financial capital 414;
Iyer, S. 95 productive structure and informality
Izzan, Muhammad 78 412, 412–413; salaried employment 408,
413; self-employment and informality
Jaimovich, N. 38 409–412, 410, 411; social protection 409
Jain, S. 6–7, 13, 23, 269 Ladakh standoff: 79–80, 89n22
Jan Dhan Yojana (JDY) 158, 164 Lahiri, H. 18, 165, 166
Jeffries, F. L. 240 Lahoti, R. 290
Jha, S. 93 laissez-faire 154
job polarisation 38–39, 42, 43, 46 Line of Actual Control, India-China
Joe, W. 311 conflict 80
John, Issac 398 lockdown, Europe 9
Johns Hopkins Center for Health Security Lopez-Obrador, A. M. 53
(JHU) 326 Los, B. 262
Index 429
low-carbon economy 198 level and multilateral policy 403–404;

low-skilled jobs 38, 43 unemployment 391–392
Milgrom, P. 240
macroeconomics 117; analysis 21; demand Ministry of Statistics and Programme
and supply shock 146; fragility 237; Implementation (MOSPI) 263
open-economy 168 Minsky, H. P. 237
Macron, E. 51, 59 Mirrlees, J. 147, 153; income taxation
macro theory 138 155–158
Mahatma Gandhi National Rural Mitra, A. 24, 93, 94, 99
Employment Guarantee Act modified Keynesian standard model 136
(MNREGA) 158, 354; de-globalization Modi, Narendra 40, 51, 371
183; greater employment 226, 226–227 Molina, C. 39
Maity, B. 15, 26, 312–314 Mondal, B. 317
Maliszewska, M. 12 Monetary Policy Rate Corridor 164
Maloney, W. F. 39 money market equilibrium: and LM curve
Manida, M. 279 134; modified standard model 135
Mann, C. L. 36 Montalvo, J. 97
Mansoor, K. 290 Moradhvaj 317
manufacturing industries, India: ASI morbidity-mortality debate 324; analysis of
262, 263–264; Atmanirbhar Bharat 330–334, 331–334, 334
260; demand and supply shocks 261; Morning Consult Political Intelligence 58
dependency 261–262, 266–270, 267– Morrison, S. 59
270, 268, 277; difference-in-difference Moyce, Sally C. 395
strategy 263, 264; GDP 260; GVC 261, MSW see municipal solid waste (MSW)
262; heterogeneity 260, 262; IIP 262, Mueller, J. E. 50
264, 270–275, 271–272, 272–275; Mukherjee, T. 290
import share 264, 265, 266, 266, 266, Mukherji, A. 11, 22, 24, 215
267; international trade 261; MOSPI Mulligan, R. F. 237
263–264; NIC 262; OLS regression 264; multilateral health surveillance 325
output share 264, 265, 266, 266, 266, municipal solid waste (MSW) 200
267; production function 263; trade Munshi, Soumyanetra 15, 24
shock 262;VI 263 myriad welfare schemes 159
market-based system 338
market-driven model 182 NAS see National Accounts Statistics (NAS)
market-linked economic relief package 296 Nash bargaining model 188
Marshal-Lerner effect 172 Nath, P. 290
massive open online courses (MOOCs) 345 National Accounts Statistics (NAS) 130
Meat Inspection Act 325 National Accreditation Council (NAC) 346
medical heath infrastructure 21 National Commission for Women 85
Merkel, A. 44, 59 National Democratic Alliance government
Merkley, E. 52 371
Mexican tariffs 181–182 National Education Policy 16
Micklethwait, J. 36 National Education Policy 2020 (NEP
micro informal firm-based clusters 299 2020) 337, 344–346
micro, small and medium enterprises National Industrial Classification (NIC)
(MSME) 132, 159, 252; informal sector, 262
India 296 National Rural Health Mission 327
Middle East Respiratory Syndrome National Sample Survey (NSS) 312, 330
(MERS) 231 National Statistical Office (NSO) 139
middle-skilled jobs 38–39 national value-added tax 159
migrant labour, rice production 284 negative demand shock 233–234
migrant workers: Indian 397–403; Nellis, G. 95
international 391, 392–397, 393, 394, Nelson, M. 51
396; intra-country migrants 392; national Nelson, R. 125
430 Index
new institution, long run process: brand PFCE see private final consumption
value 304; broader value chain 304; data expenditure (PFCE)
bank, government’s direct intervention Phelps, E. 125
302–303; land transfer 303; market Pigouvian taxes 147–148, 160n3; income
access, government support 302; taxation 156; optimal taxation model
metal-based manufacturing 303; micro 154–155
informal firm-based clusters 299; region/ Pohit, S. 279
area mapping 301, 301–302; social policy actions, India: balanced budget
infrastructure, government support 302; 174n1; economic effects 165; exchange
survey-based research 302 rate 164; fiscal and monetary measures
New Keynesian model 164 163; GDP 173; Gross Capital Formation
Nishi, H. 237 163; indirect tax effect 172–173,
non-communicable diseases 335 173; investment propensities 177;
non-tariff measures (NTMs) 180 macroeconomic indicators 163, 175;
Nuclear Threat Initiative (NTI) 326 Monetary Policy Rate Corridor 164;
New Keynesian model 164; open-
Obama, Barack 81 economy-Keynesian model 165–169,
occupational exposure, work situation 169; rate cut effect 170, 170–172,
377–378 176–177; short-run-open-economy
occupational structure, work situation 382 model 163; trade-off 165
occupation gradient, heterogeneity policymakers 147
223–224 policy thrust 345
Olken, B. A. 232 political economy: automation and jobs
online education 38 37–39, 39; “business as usual” scenario
open-economy-Keynesian model 19; 32; demand-and supply-side shocks
financial sector 168–169, 169; real sector 36; economy recession 33, 33; financial
165–168 markets 36; GDP 33–34, 34, 35, 36;
“Opinion Leadership School” 51, 62 globalisation 36, 40–42; global trade
optimal taxation model: agent’s net utility 34, 35; Great Depression 33, 34; GVC
153; economic activity 153; incentive 31, 36, 40–42, 41; herd immunity 32;
effects 154; laissez-faire 154; Mirrlees household savings 36; international
income taxation 155–158; normative tax commodity markets 36; labour market
policy 153; Pigouvian taxes 154–155; changes 31, 37; populism 43–46;
social welfare function 153; tax liability public sentiment 36; self-reliance
153; tax policies 154 47; social entitlements 42–43; social
Orbán,Viktor 45 welfare regimes 31; socio-political
Organisation for Economic Co-operation emergency 32–33; techno-administrative
and Development (OECD) 179; infrastructure 47; vaccination progress 32
unemployment 27n1 political stability 26
overlapping-generations (OLG) economy populism: “Chinese virus” 45;
111–112 de-industrialisation 45; hyper-
own account enterprise (OAE) 290 representation 45; income inequalities
45; job polarisation 45; policy reforms
Padhee, Dr. A. K. 280 46; political cleavages 45; post-pandemic
Pal, Dipyaman 11 world 44; social demands 45; social space
pandemic see Covid-19 pandemic 43–44; trust in government 51–52
pandemic preparedness: in India 326–328; post-Covidian era, green/brown projects
public health policy 324–326 194–208
Pandey, Abhilasha 14 post Covid-19, Indian economy: AD-AS
Pangong Tso 79 130; Atmanirbhar Bharat package 131;
Patnaik, U. 165 Economic Survey 131; economy after
Periodic Labour Force Survey (PLFS) 65, lockdown 132; GDP 130, 132; income
132, 278 distribution 138; income inequality
personal protective equipment (PPE) 234 132; Indian economy, macroeconomic
Index 431
impact 129–133; labour market 132; revenue neutrality 157

long-run growth 139–142, 145; MSME revenue-neutral tax policy 154–155
132; NAS 130; National Sample Surveys reverse migration 366; migration indicators
133; PLFS 132; PLIS 132; WIR 133l vs. incidence 367, 368; multi-level
unemployment rates, India 10–11, 11 regression models 367, 368, 370; spatial
predatorprey models 146 variation 369, 370
pre-pandemic protectionism 181–182 Reynal-Querol, M. 97
pre-pandemic trade policy 181 Reynolds, E. 417
prevalence-elasticity, private demand 328 rice production: animal labour 283–284;
primary sampling units (PSU) 216 casual labour 286, 287; efficiency 280–
private demand, prevention 328–329 281, 284–286, 285; fertilizer 283–284;
private final consumption expenditure human labour 283–284; input slack
(PFCE) 3 change 286, 286; manure 283; migrant
productivity-linked incentive scheme labour 284; NSSO 286; percentage
(PLIS) 132 change 286, 286; surplus labour 287
product market equilibrium: and IS curve Rogers, T. N. 384
134; modified standard model 135 Rosenzweig, S. C. 95
profit-maximization 245 rural Chhattisgarh, vulnerability 66
pro-social welfare bias 189
PSU see primary sampling units (PSU) Saad-Filho, A. 36
psychological attitude 1–2 Sadhu, K. 10
public distribution system (PDS) 158, 184 Saha, Debdatta 7, 23
public health policy: matrix 335; pandemic Sahoo, P. P. 279
preparedness 324–326 Saikia, N. 317
purchasing managers’ index (PMI) Sarkar, Runa 25
manufacturing 23–24 SARS-CoV-2 67, 195
Saudi-Iran conflict 89n14
Rajeev, M. 6, 215, 289 Saudi-led Operation Decisive Storm 78
“rally-around-the-flag” effect 50, 51, 52; Schenker, Marc 395
approval rate, Morning Consult Political Schroeder, S. 237
Intelligence 58; epidemic management Security and Exchange Board of India
58–59; first phase correlations 59, 60; (SEBI) 235
“Opinion Leadership School” 62; second self-employment 43, 417–418; enterprises
phase correlations 59, 61, 62; severity 292; and informality 409–412, 410, 411;
measure 58, 58–59; threat and comfort- labour segmentation 408; work situation
seeking behaviour 57 383, 385
Ranjan, D. 165 self-fulfilling process 299
Ratha, Dilip 399 self-help groups (SHGs) 158
Rath, S. 279 self-insurance 116–117, 126
Ratnaakar, Bandit 305 self-isolation 67
Ravindran, S. 310 self-quarantine 21
Raychaudhuri, Ajitava 6 self-reliance 47; de-globalization 182, 184,
Ray, D. 3, 4, 18, 93, 94, 97, 99, 148, 310, 311 185, 189
Ray, N. 290 self-selection 147
real-time PCR (RT-PCR) 366 Sen, A. 150, 309
Reardon, T. 280 Sengupta, R. 3, 279
Reed, R. 240 Severe Acute Respiratory Syndrome
regional trading agreements (RTAs) 180 (SARS) 231
related party transactions (RPT) 241 sexual/physical violence 86
Reserve Bank of India (RBI) 235; shadow pandemic 15, 77; violence against
de-globalization 184, 185; Indian women 85–87
migrant workers 398; rate cut effect 170, Shah, A. 165
170–172 Shah, M. 310
Restrepo, P. 414–415 Sharma, M. 290
432 Index
Shatrunjay, Arjun 11, 22 Solow-type one-sector growth model 140,

Sheridan, A. 109 142
SHGs see self-help groups (SHGs) Sterner, T. 198
Shia-Sunni conflict 89n14 Stoltenberg, Jens 83
short-and long-run effects, pandemic: Stringency Index 53–54, 54–57, 58, 63n3
absorption rate 125; aggregate output Subramanian, S. 3
effect 118–119; asymmetric effect supply chain: disruption, India 6–7;
122–126; capital-intensive production networks 7
122; full-fledged analysis 125; growth supply-side impact 23–24
rate effect 119–122, 121; human capital surplus labour 287
accumulation 123; human capital and survey-based research 302
production organization 122–124; Susceptible-Infected-Recovered model 371
late-life mortality 117; morbidity effects
124, 125, 126, 126; mortality rate 124; Taliban attacks 81–82
productivity loss 122–123; sectoral tax concessions, de-globalization 180–181
output composition 124–125, 126; tax-GDP ratio 18
self-insurance mechanism 126; skilled tax liability 153
labour 117; symmetric impact 118–122; tax revenues 148
unskilled labour 117 tax-transfer policies 23, 157, 158
short-run demand-supply management 129; teaching professionals 395, 396
AD 138; balance of payment equilibrium technical efficiency (TE): measurement
134–137; distributional issue 134; first- 282–283; output-oriented measure 281,
order linear approximations 133; Great 281–282
Depression 133; income distribution telemedicine 38
post Covid 138; IS-LM-BP framework temporary unemployment 43
137, 137; Keynesian macro framework test positivity rate (TPR) 332, 333, 334
133; money market equilibrium 134, Thara, K. 415
135; product market equilibrium 134, Thomas, E. 164
135; “pump priming” 133; surplus 137; Thorgren, S. 232
textbook analysis 137; uncertainty 134, tobacco industries 274–275
135, 144–145 TPR see test positivity rate (TPR)
short-run-open-economy model 163 traditional vs. online education 25
Shramik trains 355, 356, 358, 366 Trudeau, J. 51, 59
Shrangi,Vatsala 401 Trump, Donald 40, 43–45, 79, 83, 87
Shrivastava, A. 95 trust in government: approval rates of
Singh, Manish Kumar 199 51, 51, 52–53, 53, 54–57; emotional
Singh, Sushant 402 response 50, 62; epidemic evolution
Siu, H. E. 38 53–57; opinion leadership school 51;
smallpox 75 populism 51–52; public policies 50;
social demands 44 “rally-around-the-flag” effect 50, 51, 52,
social distancing 5–6, 15, 31, 33, 37; India’s 57–62, 58–61; social distancing measures
lockdown 67, 69; trust in government 53–57; socio-political attitudes 52
53–57 truth-telling direct mechanisms 155–156
social entitlements 42 Tumbe, C. 228
social impact: Covaxin 14; Covishield
14; human interactions 14; knowledge Ujjawala Scheme 165
economy 16; social distancing 15; social uncertainty: household adjustments,
interactions 14–15; ‘traditional vs. online India 228; short-run demand-supply
education 15–16; work from home 15 management 134, 135, 144–145
social inequities, work situation 378 unemployment 1, 412; CMIE 215; migrant
social media 45 workers 391–392; pre-and post-corona
social welfare 31, 153, 186 unemployment rates, India 10–11, 11;
socio-cultural relationships 1–2 temporary 43
socio-political emergency 32–33 unilateral tariffs, de-globalization 187–188
Index 433
United Arab Emirates (UAE) 87, 89n16 “missing women” phenomenon

United Nations High Commissioner for 309–311; mortality and morbidity 309;
Refugees (UNHCR) 89n13 NSS 312–313; policy implications 317;
universal basic income (UBI) scheme 43 societal constraints and discrimination
unorganized informal sector 1 313; urban poverty 313; vulnerability
US-China trade war 13–14, 182; 310, 311
de-globalization 192–193 Wilkinson, S. 92
US-Taliban talks, Qatar 82–83 Williams, T. A. 232
WIR see World Inequality Report (WIR)
vaccine-preventable diseases 328–329 Wooldridge, A. 36
Van Bergeijk, P. A. 178 work from home 1, 15, 413
Vandenbussche, H. 262 work situation 387n1; circumstances
Vandewalle, L. 3 of work 381; ‘distancing bonus’ 383;
variable returns to scale (VRS) 282–283 distributional asymmetries 378;
Varshney, A. 93, 102 ‘downscaling loss’ 383; vs. employment
Varshney–Wilkinson dataset 96 378–381; employment relations
violence amidst virus 75–88 377–378; face-to-face interactions 384;
Vučić, Aleksandar 45 ‘geographically-tethered’ 384; hazard pay
vulnerabilities: acute 290–296; economic 384; individuals positions 382; labour
shocks, low savings and exposure 292, market recovery 384–385; location 381;
294, 296–297; higher education, post- macro-level data 385; nature of work
Covid era 349; IIP 274–275, 275; labour 381; occupational exposure 377–378;
387; rural-urban self-employment 290, occupational structure 382; precarity
291, 292, 293–295; widowed and elderly 384–385; ‘resource’ and ‘control’ 377;
women, India 310, 311 self-employment 383, 385; social
vulnerability index (VI) 263 inequities 378; unequal outcomes
377–378
wage-labour–based poverty alleviation World Development Indicators 413
schemes 22 World Economic Outlook (WEO) 3, 33
‘wageless work’ 379 World Inequality Report (WIR) 133
Wallace, Ben 83 World Migration Report 393, 398
Wang,Y. I. 240–241 World Trade Organization (WTO) 181, 186
waste management systems 195 Wright, E. O. 377
Weaver, S. 95
WEO see World Economic Outlook Yadav, Renu 399–401
(WEO) Yemen conflict: Ansar Allah movement 78;
wholesale price index (WPI) 139 civil war in 78; Covid-19 in 78, 79; in
widowed and elderly women, India 9; asset post-pandemic 77; poverty 78; Saudi-led
ownership 314–316, 315; consumption Operation Decisive Storm 78
312–313; crowdsourced data 311; DHS
314; economic activities 310; face zero-carbon economy 197
discrimination 309; health 316, 316–317; zoonotic pathogens 195

The COVID-19 Pandemic, India and The World - Economic and Social Policy Perspectives-Routledge India (2021)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The COVID-19 Pandemic, India and The World - Economic and Social Policy Perspectives-Routledge India (2021)

Uploaded by

Copyright:

Available Formats

The Covid-19 pandemic has had far-reaching social and economic consequences

Engagingly written, this comprehensive volume compiles original research by

Ananya Ghosh Dastidar is a professor in economics, Department of Finance

Soumyen Sikdar is a professor in economics, Indian Institute of Management

ISBN: 978-0-367-68886-8 (hbk)

1 Corona pandemic and the changing global economy 1

2 The Covid-19 pandemic and its impact on global

3 Riding the wave or going under? The Covid-19

5 Violence amidst virus: an essay on conflicts in the

6 Brothers in arms? Covid-19 and Hindu-Muslim

7 The long shadow of epidemics 109

8 The macroeconomic impact of Covid-19 on the

9 Developing economic herd immunity in the

10 Covid-19 and an assessment of some policy actions

11 Impact of the coronavirus pandemic on

12 Green or brown: global environment in the

13 Coping with Covid-19 and its consequences:

14 Firm fragility: a re-interpretation in the times of the

15 Impact of Covid-19 on manufacturing industries in

16 Measurement of efficiency loss in Indian

17 The pandemic and the informal sector of India:

18 Covid-19 pandemic: implications for widowed and

19 Policy failure before Covid-19 pandemic and options

20 Higher education in the Post-Covid era:

21 Lockdowns, migrants, and the spatial distribution of

22 Labour and the pandemic: a study on work,

23 Covid-19 pandemic and migrant workers: India and

24 The future of labour segmentation after Covid-19 408

1.1 Economic and social impact of corona pandemic 2

7.5 Asymmetric effect of morbidity shock 125

21.5 Choropleth maps for Getis-Ord gi*(d) statistics 363

13.5 Households by employment categories who experienced

18.2 If suffering from any chronic ailment (self-reported) among

Sutirtha Bandyopadhyay is currently an assistant professor in economics at

Lopamudra Banerjee teaches economics and on the environment at Benning-

Mélika Ben Salem is a full professor of economics at the Université Gustave

Manojit Bhattacharjee is presently working as an assistant professor at Ma-

Rajesh Bhattacharya is an associate professor in the Public Policy and Man-

Snehashish Bhattacharya teaches at the Faculty of Economics, South Asian

Rajib Bhattacharyya is an associate professor in the Department of Economics,

Ranajoy Bhattacharyya is a professor and head of Economics Division, Indian

Montu Bose is an assistant professor of TERI School of Advanced Studies,

Catherine Bros holds a master’s degree in international and development eco-

Saumya Chakrabarti is a professor and head of the Department of Economics

Chandrima Chakraborty is currently working as an assistant professor in the

Shankha Chakraborty is a professor of economics at the University of Oregon.

Partha Chatterjee is currently a professor and head of the Economics Depart-

Saumen Chattopadhyay is a faculty at the Zakir Husain Centre for Educa-

Srobonti Chattopadhyay is an assistant professor in the Department of Eco-

Mausumi Das is a professor at the Institute of Economic Growth, Delhi. Before

Ananya Ghosh Dastidar is a professor in the Department of Finance and Busi-

Pratip Kumar Datta teaches economics and mathematics at Rajatpur I.N.

Arijita Dutta is a professor and head in the Department of Economics at the

Anupriya Gangopadhyay is currently a research associate at the Centre for

Arpita Ghose is currently a professor in the Department of Economics, Ja-

Parikshit Ghosh is a professor of economics at the Delhi School of Economics.

Zakir Husain is a graduate from the Economics Department, Presidency Col-

Prime Minister’s High-Level Committee for preparing a status report on the

Richa Kothari is a research scholar at the Economics Department, Presidency

Anirban Kundu is an assistant professor at Presidency University, Bengaluru,

Bipasha Maity is an assistant professor of economics at Ashoka University in

economics of gender, labour economics and economic history. Prior to joining

Anirban Mitra is a senior lecturer (associate professor) in economics at the Uni-

TABLE.1.2 Differential impact on workers of various sectors

TABLE 2.1 Global GDP growth forecasts: percentage change in GDP

3.2 The evolution of the epidemic and the timing of the

TABLE 3.1 Change in approval ratings