Professional Documents
Culture Documents
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.
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
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
PART I
Conf licts, trust and society 29
4 India’s lockdown 64
Debraj Ray, S. Subramanian and Lore Vandewalle
viii Contents
PART II
Macroeconomic issues and environment 107
PART III
Sectoral perspectives 211
PART IV
Health, education and labour 321
Index 421
FIGURES
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
as Applied Economic Perspectives and Policy, Journal of Asian Economics, Margin: The
Journal of Applied Economic Research, and Indian Economic Journal.
book chapters with Routledge, Taylor and Francis, IGI Global, etc. The author
also presented research papers in different institutes and universities of India.
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.
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.
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.
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.
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.
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.
TABLE 1.1 India’s growth predictions in GDP, GVA and its components (%)
2020–21 2021–22
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
Corona pandemic & changing global economy 5
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.
AS2
AS1
Price
AS0
AD0
AD1
AD2
0 Y4 Y3 Y2 Y1 Y0 Output
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
8 Rajib Bhattacharyya et al.
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
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
10 Rajib Bhattacharyya et al.
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
12 Rajib Bhattacharyya et al.
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.
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.
Corona pandemic & changing global economy 13
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,
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.
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.
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.
Corona pandemic & changing global economy 17
18 Rajib Bhattacharyya et al.
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
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.
REMOVAL OF
DEVELOPMENT DEMAND SECTOR STRATEGIES OF
SUPPLY
OF HEALTH MANAGEMENT SPECIFIC INTERNATIONAL
BOTTLENECK COOPERATION
CARE SYSTEM POLICIES MEASURES
STRATEGY
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
Corona pandemic & changing global economy 21
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’.
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
Corona pandemic & changing global economy 23
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.
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.
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
26 Rajib Bhattacharyya et al.
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
Corona pandemic & changing global economy 27
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.
28 Rajib Bhattacharyya et al.
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.
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
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.
Covid-19 pandemic and its impact 35
FIGURE 2.2 Real GDP growth (%) for world, developed and emerging economies.
Source: BMI Research.
10
-5
-10
-15
2017
2018
2019
2020
2021
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)
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)
Covid-19 pandemic and its impact 37
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.
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
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.
40 Rajesh Bhattacharya and Ujjwal Kango
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
42 Rajesh Bhattacharya and Ujjwal Kango
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.
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
44 Rajesh Bhattacharya and Ujjwal Kango
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
46 Rajesh Bhattacharya and Ujjwal Kango
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.
Covid-19 pandemic and its impact 47
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.
48 Rajesh Bhattacharya and Ujjwal Kango
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_
755910/lang--en/index.htm. Last Accessed on: 12 March 2021.
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/.
Last Accessed on: 12 March 2021.
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/.
Last Accessed on: 12 March 2021.
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.
Covid-19 pandemic and its impact 49
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_
755910/lang--en/index.htm. Last Accessed on: 12 March 2021.
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.
Last Accessed on: 12 March 2021.
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
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
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.
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
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
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
b Brazil
.6
.6
80
200
new cases smoothed per million inhab.
.55
.55
60
150
Stringency index
Approval Rate
Approval Rate
.5
.5
40
100
.45
.45
20
50
.4
.4
0
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
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.
Riding the wave or going under? 55
c Canada
80
50
.7
.7
new cases smoothed per million inhab.
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
Day Number Day Number
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
d France
80
.4
.4
new cases smoothed per million inhab.
80
.35
.35
60
60
Stringency index
Approval Rate
Approval Rate
40
.3
.3
40
.25
.25
20
20
.2
.2
0
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
e Germany
80
80
.6
.6
new cases smoothed per million inhab.
.55
.55
60
60
Stringency index
Approval Rate
Approval Rate
.5
.5
40
40
.45
.45
20
20
.4
.4
0
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
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
new cases smoothed per million inhab.
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
Day Number Day Number
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
g Japan
50
15
.35
.35
new cases smoothed per million inhab.
40
10
Stringency index
Approval Rate
Approval Rate
.3
.3
20 30
.25
.25
10
.2
.2
0
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
h Mexico
60
.7
.7
80
.65
.65
40
Stringency index
Approval Rate
Approval Rate
40
.6
.6
20
20
.55
.55
0
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
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
Riding the wave or going under? 57
i USA
.46
.46
80
200
new cases smoothed per million inhab.
.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
Day Number Day Number
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
j United Kingdom
80
80
.65
.65
new cases smoothed per million inhab.
.6
.6
60
60
Stringency index
Approval Rate
Approval Rate
.55
.55
40
40
.5
.5
20
20
.45
.45
.4
.4
0
Stringency Index Approval Rate new cases smoothed per million inhab. Approval Rate
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
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
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.
Riding the wave or going under? 59
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 3.3 Correlates of the approval ratings of the heads of state between day 50 (February 19) and day 150 (May 29)
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)
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.
Riding the wave or going under? 61
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.
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.
Riding the wave or going under? 63
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?
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
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
India’s lockdown 69
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.
India’s lockdown 71
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?
72 Debraj Ray et al.
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.
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.
74 Debraj Ray et al.
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)
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
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
5.2.1.3 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
Violence amidst virus 79
Houthis and they have put up a strong defence, prolonging the conflict with a
continuous military stalemate in various provinces.14
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.
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.
5.2.3.1 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.
82 Soumyanetra Munshi
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.
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.
Violence amidst virus 83
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.
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.
Violence amidst virus 85
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.
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
86 Soumyanetra Munshi
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 amidst virus 87
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.’
(The Plague, Albert Camus)
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.
88 Soumyanetra Munshi
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.
(The Plague, Albert Camus)
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
September 16, 2020).
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.
Violence amidst virus 89
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).
Violence amidst virus 91
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.
Brothers in arms? 95
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
96 Anirban Mitra and Arnab Mukherji
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
Brothers in arms? 97
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:
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,
98 Anirban Mitra and Arnab Mukherji
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
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%.
Brothers in arms? 99
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
100 Anirban Mitra and Arnab Mukherji
TABLE 6.2 The effect of Hindu-Muslim conflict (Casualties, Killed and Outbreak) on Covid-19
deaths: OLS district-level regressions
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%.
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.
104 Anirban Mitra and Arnab Mukherji
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.
Brothers in arms? 105
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).
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
FIGURE 7.2 Life expectancies in Brazil, China, India and the USA.
Data: United Nations World Population Prospect, 2019.
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
112 Shankha Chakraborty and Mausumi Das
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
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
The long shadow of epidemics 113
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 )
114 Shankha Chakraborty and Mausumi Das
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:
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
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)
The long shadow of epidemics 115
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 .
b
Wt +1 R / 2 − 1 ≡ gW , with prob. p
gW ≡ −1=
Wt a
R − 1 ≡ gW , otherwise
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 < <
gW
gWa (w.p. 1 − p)
R
R gWb (w.p. p)
2
p
0 p p̄ 1
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
The long shadow of epidemics 117
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.
( )
Yt = (1 − α ) ργ 1−δ K t + α(ωµ H t ) (7.18)
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
120 Shankha Chakraborty and Mausumi Das
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
The long shadow of epidemics 121
growth rate
b
gW
gW
gW
a
gW
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.
122 Shankha Chakraborty and Mausumi Das
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
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
YtK = ργ 1−δ K t
and
YtH = 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
( )
Yt = (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 )
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
124 Shankha Chakraborty and Mausumi Das
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.
1 + (1 − α )ρ
αω + αωλ( p ) b
θ ( p ) − 1 ≡ gW , with prob. p
αω + [1 + (1 − α ) ρ ] λ ( p ) 2
W
gW ≡ t +1 − 1 =
Wt αω
[1 + (1 − α )ρ ] θ ( p ) − 1 ≡ gW
a
, otherwise.
αω + [1 + (1 − α )ρ ] λ( p )
growth rate
g̃ b
W
g W (ϵ=0)
g W (ϵ=0.3)
g̃ a
W
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
growth rate
gR
W (ϵ=0)
gR
W (ϵ=0.3)
gP
W (ϵ=0)
gP
W (ϵ=0.3)
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
The long shadow of epidemics 127
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 YtK decreases by a proportion 1 − γ 1−δ < 1 − γ , while YtH 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.
128 Shankha Chakraborty and Mausumi Das
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
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.
132 Ajitava Raychaudhuri
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
The macroeconomic impact on India 133
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.
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
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.
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.
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.
A LM0/
/
A
A//
A///
P AS/
AD
AS// AS
/
AD AD //
P0
P0// P0/
Y0
Y0/ Y1 Y0// Y
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//.
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
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.
The macroeconomic impact on India 143
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.
144 Ajitava Raychaudhuri
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
The macroeconomic impact on India 145
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.
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
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.
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.
Developing economic herd immunity 151
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.
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.
∫ t ( θ ) . f ( θ ) dθ = ∫ c (θ − c ) f (θ) dθ − T = 0
1
θ θ
Developing economic herd immunity 155
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 .
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
1y 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()
Developing economic herd immunity 157
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
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
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.
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.
Developing economic herd immunity 161
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
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
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
Assessment of some policy actions 167
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)
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 )
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)
e
YY BB
e
YY YY1 BB1 BB
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.
YY BB1
YY1 BB
A1
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)
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)
−E Y Be + E e BY
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 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 ) Pe − NX PX PX e = ((1 − mI ) (−I P ) Pe − (1 − m X ) X P PX e > 0 (10.A.7)
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 Pe is sufficiently close to zero, (10.A.11) will
be larger than (10.A.10). Since, Pe 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.
E t Be − E e Bt
dY = < 0 (10.A.12)
−E Y Be + E e BY
E Y Bt − E t BY
de = ≥< 0 (10.A.13)
−E Y Be + E e BY
−Bt −E t
Now, de > 0 ⇒ >
−BY −E Y
Or, Horizontal Shift of BB > Horizontal Shift of YY
⇒
−NX P Pt
>
( )
C ′ {(1 − t )} + {− (I p pt )} + −NX P Pt
(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
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
FDI
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
Coronavirus pandemic and de-globalization 181
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.
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.
TABLE 11.1 Policies undertaken by the Government of India (apart from international
trade and investment 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
184 Ranajoy Bhattacharyya et al.
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.
Coronavirus pandemic and de-globalization 185
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”.
186 Ranajoy Bhattacharyya et al.
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.
( )
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 Pi is the international price of good i.
7 tij, 0 is the tariff/subsidy on import good i by country j.
( )
8 P j = Pi 1 + t j is the price of good i in country j.
i , 0 i , 0
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 ( )
= Pi 1 + tij, 0 , where Pi 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 ) Pi 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.
( ( ) ( )) ( ( ) ( ))
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 , * = −
Pi (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.
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.
Coronavirus pandemic and de-globalization 191
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
∂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)
Likewise, we can derive the tariff profiles for partner countries f1 and f2.
192 Ranajoy Bhattacharyya et al.
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.
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.
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
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
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.
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
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
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.
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
216 Arnab Mukherji and Arjun Shatrunjay
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?
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.
218 Arnab Mukherji and Arjun Shatrunjay
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.
Coping with Covid-19 and its consequences 219
Source:
1 Author calculations from CMIE’s CPHS data.
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.
Source:
1 Author calculations from CMIE’s CPHS data.
Coping with Covid-19 and its consequences 221
TABLE 13.4 Income trends for households by employment categories who faced income
losses
Source:
1 Author calculations from CMIE’s CPHS data.
222 Arnab Mukherji and Arjun Shatrunjay
Source:
1 Author calculations from CMIE’s CPHS data.
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.
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.
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
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
228 Arnab Mukherji and Arjun Shatrunjay
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.
Coping with Covid-19 and its consequences 229
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.
230 Arnab Mukherji and Arjun Shatrunjay
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
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
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.
TABLE 14.1 Definition and expression of fragility in economic systems in the existing
literature
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)
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
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
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
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:
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.
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 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.
( )
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
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
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.
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
( )
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)
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 .
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
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.
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).
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
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.
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
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.
Impact of Covid-19 on manufacturing 263
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.
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
Impact of Covid-19 on manufacturing 267
TABLE 15.4 Panel regression between output share and import share
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
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.
268 Partha Chatterjee and Shweta Jain
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,
Year Correlation
2012 0.185
2013 0.177
2014 0.194
2015 0.161
2016 0.219
Impact of Covid-19 on manufacturing 269
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.
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
Year Correlation
2011–12 0.923
2012–13 0.812
2013–14 0.942
2014–15 0.899
2015–16 0.867
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
270 Partha Chatterjee and Shweta Jain
Year Correlation
2012 0.139
2013 0.107
2014 0.173
2015 0.102
2016 0.180
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
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
FIGURE 15.3 oY change in IIP for the months April–July 2020 sorted by import
Y
share.
272 Partha Chatterjee and Shweta Jain
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.
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
Impact of Covid-19 on manufacturing 273
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
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
274 Partha Chatterjee and Shweta Jain
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* p < 0.1.
Notes: Standard errors in brackets are clustered at industry-year level. *** p < 0.01, ** p < 0.05,
* 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
Impact of Covid-19 on manufacturing 275
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.
276 Partha Chatterjee and Shweta Jain
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.
Impact of Covid-19 on manufacturing 277
APPENDIX
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
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:
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.
Output(y)
y* f(x)
y0 A(x0,y0)
x* x0
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.
16.2.1.1 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.
Efficiency loss in agricultural sector 283
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
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.
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.
Efficiency loss in agricultural sector 285
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
Casual
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
288 Dipyaman Pal et al.
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
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.
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)
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
TABLE 17.3 Total employment and vulnerable employment across labour-size class
among self-employed during 2010–11
Manufacturing
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%).
The pandemic and the informal sector 295
TABLE 17.4 Total employment and vulnerable employment across labour-size class
among self-employed during 2015–16
Manufacturing
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
296 Anirban Kundu et al.
TABLE 17.5 Share (%) of households owning financial assets (in Rs.) other than shares
and debentures as of 30.06.2012 in rural areas
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
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
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.
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.
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 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.
302 Anirban Kundu et al.
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.
17.3.3.6 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).
304 Anirban Kundu et al.
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
306 Anirban Kundu et al.
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.
The pandemic and the informal sector 307
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/
308 Anirban Kundu et al.
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
Fall in output
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
312 Sutirtha Bandyopadhyay and Bipasha Maity
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.
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.
314 Sutirtha Bandyopadhyay and Bipasha Maity
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.
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
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
Observations 15,303 15,303
Panel B: Average Land Holding by Households (Acres)
2005 2011–12
Households where women transitioned from being 4.25 3.56
married in 2005 to being widowed in 2011–12
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%
married in 2005 to being widowed in 2011–12
Observations 1,694 1,694
Households where women continued to remain married 62% 64%
in 2005 and 2011–12
Observations 15,303 15,303
Panel D: Average Count of the Number of Animals Owned
2005 2011–12
Households where women transitioned from being 4.87 4.35
married in 2005 to being widowed in 2011–12
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.
Males:
Yes 28.34% 28.39%
Observations 6,627 1,011
Females:
Yes 30.85% 33.42%
Observations 3,011 3,326
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.
Covid-19 pandemic 317
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.
Acknowledgement
Maity acknowledges financial support from Ashoka Research Grant 2019–21,
Ashoka University. All remaining errors are our own.
318 Sutirtha Bandyopadhyay and Bipasha Maity
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.
Covid-19 pandemic 319
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.
320 Sutirtha Bandyopadhyay and Bipasha Maity
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.
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.
326 Arijita Dutta and Montu Bose
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.
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
328 Arijita Dutta and Montu Bose
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.
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.
330 Arijita Dutta and Montu Bose
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).
332 Arijita Dutta and Montu Bose
Relationship
FIGURE 19.2 among test coverage (tests per hundred of population),
health infrastructure and per capita Gross State Domestic Product
(GSDP).
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
334 Arijita Dutta and Montu Bose
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
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.
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.
336 Arijita Dutta and Montu Bose
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
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.
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
Higher education in the post-Covid era 343
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.
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.
‘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
Higher education in the post-Covid era 345
(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.
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.
on enrichment and status associated with earning degrees which will make the
graduates employable in the face-changing job market imperatives.
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.
TABLE 20.1 Global and national ranking of Indian universities and top world universities as
per world ranking (2020)
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
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.
Higher education in the post-Covid era 351
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.
356 Zakir Husain and Richa Kothari
21.2 Methodology
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).
Spatial distribution of Covid in India 357
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:
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,
when
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)
Spatial distribution of Covid in India 359
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).
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.
Spatial distribution of Covid in India 361
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 21.3 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.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
April May June July August
FIGURE 21.4 Change in hotspots and cold spots between May and August.
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
364 Zakir Husain and Richa Kothari
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
Note: * denotes p < 0.05, ** denotes p < 0.01, *** denotes p < 0.001.
Spatial distribution of Covid in India 365
366 Zakir Husain and Richa Kothari
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.
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).
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.
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.
Spatial distribution of Covid in India 371
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
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
Spatial distribution of Covid in India 373
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
374 Zakir Husain and Richa Kothari
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
Spatial distribution of Covid in India 375
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
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
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
380 Lopamudra Banerjee and Snehashish Bhattacharya
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.
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
Labour and the pandemic 385
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
386 Lopamudra Banerjee and Snehashish Bhattacharya
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
Labour and the pandemic 387
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).
388 Lopamudra Banerjee and Snehashish Bhattacharya
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).
Labour and the pandemic 389
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.
390 Lopamudra Banerjee and Snehashish Bhattacharya
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.
13.30%
Northern America
23%
Europe
Arab States
7.90%
32%
Asia
2.70%
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
Covid-19 pandemic and migrant workers 395
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
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
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
Covid-19 pandemic and migrant workers 397
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.
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
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,
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,
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
400 Srobonti Chattopadhyay
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
Covid-19 pandemic and migrant workers 401
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
Covid-19 pandemic and migrant workers 403
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.
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
404 Srobonti Chattopadhyay
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.
Covid-19 pandemic and migrant workers 405
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).
406 Srobonti Chattopadhyay
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)
Covid-19 pandemic and migrant workers 407
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
accessed on April 18, 2021)
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
accessed on April 18, 2021)
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
accessed on April 18, 2021)
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
accessed on April 18, 2021)
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
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
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
Labour segmentation’s future after Covid-19 411
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,
412 Mélika Ben Salem
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.
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.
Labour segmentation’s future after Covid-19 419
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.
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