You are on page 1of 172

Global Monitoring Report 2010

The MDGs after the Crisis


Global Monitoring Report 2010

The MDGs after the Crisis


Global Monitoring Report 2010

The MDGs after the Crisis


© 2010 The International Bank for Reconstruction and Development / The World Bank

1818 H Street NW
Washington DC 20433
Telephone: 202-473-1000
Internet: www.worldbank.org
E-mail: feedback@worldbank.org
All rights reserved

1 2 3 4 13 12 11 10

This volume is a product of the staffs of The World Bank and The International Monetary Fund. The
findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of
the Board of Executive Directors of The World Bank, the Board of Executive Directors of The Interna-
tional Monetary Fund, or the governments they represent.
The World Bank and The International Monetary Fund do not guarantee the accuracy of the data
included in this work. The boundaries, colors, denominations, and other information shown on any map
in this work do not imply any judgement on the part of The World Bank or The International Monetary
Fund concerning the legal status of any territory or the endorsement or acceptance of such boundaries.

Rights and Permissions


The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work
without permission may be a violation of applicable law. The International Bank for Reconstruction and
Development / The World Bank encourages dissemination of its work and will normally grant permission
to reproduce portions of the work promptly.
For permission to photocopy or reprint any part of this work, please send a request with complete
information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA;
telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com.
All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office
of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422;
e-mail: pubrights@worldbank.org.

ISBN: 978-0-8213-8316-2
eISBN: 978-0-8213-8424-4
DOI: 10.1596/978-0-8213-8316-2

Cover image: “Escape Route,” by Iyke Okenyi, 2006, courtesy of the World Bank Art Program.
Cover design: Debra Naylor of Naylor Design.
Interior photographs: Yosef Hadar / World Bank (10), Curt Carnemark / World Bank (28), Ray Witlin /
World Bank (68), Curt Carnemark / World Bank (96), Tran Thi Hoa / World Bank (120).
Contents

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

Abbreviations and Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv

Goals and Targets from the Millennium Declaration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi

Overview: MDGs after the Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 Millennium Development Goals: Significant Gains


before the Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2 Lessons from Past Crises—and How the Current


Crisis Differs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3 Growth Outlook and Macroeconomic Challenges in


Emerging and Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

4 Outlook for the Millennium Development Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

5 The International Community and Development—Trade,


Aid, and the International Financial Institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Appendix: Classification of Economies by Region and Income,


Fiscal 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

GLOBAL MONITORING REPORT 2010 v


vi CONTENTS GLOBAL MONITORING REPORT 2010

Boxes
2.1 Defining growth cycles in developing countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.2 Aggregate economic shocks and gender differences: A review of the evidence . . . . . 32
2.3 Crises as opportunities for reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.4 Using safety nets to lower the cost of reducing poverty . . . . . . . . . . . . . . . . . . . . . . 47
2.5 Are external shocks becoming more important than internal shocks for
developing countries? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.6 Human development suffered severely during crises in developing countries . . . . . . 53
2.7 Gender differences in impacts of the crisis: Evidence from East Asia . . . . . . . . . . . . 54
3.1 Quality of macroeconomic policies in low-income countries . . . . . . . . . . . . . . . . . . 83
3.2 Mobilizing additional revenue in developing countries: Key issues for tax policy
and revenue administration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.3 A fiscal rule for commodity exporters: The cases of Chile and Nigeria . . . . . . . . . . 87
4.1 Uncertainty and risk in projecting attainment of the MDGs . . . . . . . . . . . . . . . . . . 99
4.2 Estimating the impact of growth on human development indicators . . . . . . . . . . . 102
4.3 Assumptions for the archetype countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
4A.1 MAMS: A tool for country-level analysis of development strategies . . . . . . . . . . . 114
5.1 Facilitating trade through logistics reforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.2 The allocation of aid from private sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.3 The IMF’s engagement with low-income countries . . . . . . . . . . . . . . . . . . . . . . . . 138
5.4 Gender equality as smart economics: A World Bank Group action plan . . . . . . . . 143
5.5 Crisis-related initiatives of the International Finance Corporation . . . . . . . . . . . . 145
5.6 Action Plan for Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Figures
1 Serious global shortfalls loom for the human development MDGs . . . . . . . . . . . . . . 3
2 Key indicators plummet from their overall mean during growth decelerations,
all countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3 The long-run effect of slower growth on selected MDGs is worrisome . . . . . . . . . . . 7
1.1 But Africa’s poverty rate is falling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.2 At the global level, serious shortfalls loom for the human development MDGs . . . . 14
1.3 Since the 1990s growth in developing countries has accelerated . . . . . . . . . . . . . . . 14
1.4 Poverty reduction is substantial in all regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.5 Another view: Poverty rates and the number of poor people are falling rapidly. . . . 15
1.6 Net enrollment rates are rising in many countries . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.7 Gender parity is close in primary education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.8 More people have improved sources of water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.9 Progress lacking on ratio of employment to population . . . . . . . . . . . . . . . . . . . . . 19
1.10 Female enrollment in tertiary education lags in Sub-Saharan Africa
and South Asia, 2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
GLOBAL MONITORING REPORT 2010 CONTENTS vii

1.11 The contraceptive prevalence rate is low for low-income countries . . . . . . . . . . . . . 20


1.12 HIV prevalence rates and estimated deaths are showing signs of decline . . . . . . . . . 22
1.13 Improving access to antiretroviral treatment is still far from universal . . . . . . . . . . 22
1.14 Fragile states have made the least progress toward the MDGs . . . . . . . . . . . . . . . . 23
1.15 Progress in Sub-Saharan Africa is significant but still insufficient—partly
because of low starting points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.16 Many countries are falling short of most MDGs, 2009 . . . . . . . . . . . . . . . . . . . . . . 25
1.17 Poverty responds less to growth when the initial poverty rate is high . . . . . . . . . . . 25
2.1 Key human development and gender indicators plummet from their overall
mean during growth decelerations, all countries . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.2 Key human development and gender indicators also fall below their overall
means during growth decelerations in Sub-Saharan countries, if less so . . . . . . . . . 34
2.3 During growth decelerations, economic and institutional indicators diverge
far from the overall means . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.4 Health spending growth rate is more volatile than its per capita level or
GDP growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.5 Public education spending is less closely tied to GDP growth than is health
spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.6 Aid to education and health does not appear to be closely related to
GDP growth, 1998–2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.7 Despite intense fiscal pressures, Mexico’s federal funding for health and
education is set to rise in 2009–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.8 Average pharmaceutical expenditures fall in Eastern Europe, especially in the
Baltics, before beginning to rise again . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.9 Undisbursed HIV/AIDS grants from the Global Fund to Fight AIDS,
Tuberculosis, and Malaria, Rounds 1–7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.10 Food-related safety net programs are more common in Africa than elsewhere . . . . 45
2.11 Economic performance and MDG outcomes are better with good policy . . . . . . . . 52
2.12 Spending cutbacks in crisis-affected households are jeopardizing future
welfare in Armenia, Montenegro, and Turkey. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.13 The crisis sharply reduced wage earnings in middle-income countries . . . . . . . . . . . 58
2A.1 Projected Global Fund to Fight AIDS, Tuberculosis, and Malaria and U.S.
PEPFAR HIV/AIDS grants as of April 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2A.2 Projected Global Fund to Fight AIDS, Tuberculosis, and Malaria and U.S.
PEPFAR HIV/AIDS grants per AIDS patient as of April 2009 . . . . . . . . . . . . . . . . . 64
3.1 Short-term indicators of production and trade are recovering . . . . . . . . . . . . . . . . . 71
3.2 Commodity price indexes rebounded strongly in 2009 . . . . . . . . . . . . . . . . . . . . . . 71
3.3 Bond spreads have declined in emerging markets and developing countries . . . . . . 72
3.4 Share prices have recovered sharply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.5 Exchange rates have been less volatile: Daily spot exchange rates . . . . . . . . . . . . . . 73
3.6 The cost of external debt financing has come down . . . . . . . . . . . . . . . . . . . . . . . . 73
3.7 The share of nonperforming loans to total loans has been rising . . . . . . . . . . . . . . . 74
3.8 Bank financing to emerging markets dropped sharply in 2009 . . . . . . . . . . . . . . . . 75
viii CONTENTS GLOBAL MONITORING REPORT 2010

3.9 Changes in terms of trade have swung sharply since 2008 . . . . . . . . . . . . . . . . . . . 76


3.10 External imbalances have come down in emerging and developing countries . . . . . 76
3.11 Almost all countries rebuilt their international reserves in 2009 . . . . . . . . . . . . . . . 77
3.12 Deteriorating terms of trade sometimes reinforce contraction in economic
activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.13 Monetary policy conditions became more accommodating in 2009 . . . . . . . . . . . . 78
3.14 Average year-on-year growth in money and the money gap in emerging markets . . 78
3.15 Fiscal deficits expanded in 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.16 Growth in real primary spending, 2010 projections . . . . . . . . . . . . . . . . . . . . . . . . 79
3.17 Most countries responded with expansionary fiscal and monetary policy . . . . . . . . 82
3.18 After previous crises, low-income countries recovered more slowly than the
world economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.19 Growth of terms of trade and external demand in low-income countries in
past and current crises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.20 Output losses are highly persistent, especially under external demand shocks . . . . . 91
3.21 In Sub-Saharan Africa terms-of-trade shocks have larger and more
persistent effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.22 In low-income countries, growth downbreaks are more associated with
terms-of-trade shocks, giving hope for smoother recovery . . . . . . . . . . . . . . . . . . . 93
4.1 Framework linking policies and actions with development outcomes . . . . . . . . . . . 98
4.2 The long-run effect of slower growth on selected MDGs is worrisome . . . . . . . . . 104
4.3 The long-run effect of slower growth is especially worrisome in
Sub-Saharan Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.4 Annual GDP growth for LIRP under four cases . . . . . . . . . . . . . . . . . . . . . . . . . . 109
4.5 Simulated MDG outcomes for the LIRP archetype under alternative cases . . . . . . 112
4.6 Simulated MDG outcomes for the LIRR archetype under alternative cases . . . . . . 113
5.1 Trade has bottomed out and started to recover . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.2 Baltic Dry Index points to a fragile rebound in shipping by sea . . . . . . . . . . . . . . . 122
5.3 Short-term trade finance messages increased steadily from Jan. 2009 to
Feb. 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
5.4 Tariff rates fell except in upper-middle-income countries, 2008–09. . . . . . . . . . . . 124
5.5 About 350 trade-restrictive measures and 80 trade-liberalizing measures
have been implemented or initiated since the onset of the crisis, but some
have already been removed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
5.6 Net official development assistance rose in real terms in 2008 and 2009 . . . . . . . . 129
5.7 Significant amounts of official development assistance are in debt relief and
humanitarian assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.8 Trends in gross official development aid from bilateral donors, by type,
2000–08 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
5.9 Gross official development aid from bilateral donors, 2008 . . . . . . . . . . . . . . . . . 130
5.10 Fragile states received $21.3 billion net official development assistance in 2008 . . 131
5.11 Net official development assistance from all sources, by income group,
2000–08 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
GLOBAL MONITORING REPORT 2010 CONTENTS ix

5.12 Net ODA varies widely as a share of GNI in Sub-Saharan Africa . . . . . . . . . . . . . 136
5.13 Debt stock of heavily indebted poor countries is expected to come down by
80 percent in end-2009 NPV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.14 Multilateral development banks substantially increased their disbursements,
2000–09 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Maps
1.1 Africa is the only region with high extreme poverty . . . . . . . . . . . . . . . . . . . . . . . . 12
1.2 Proportion of population living with HIV is still high but declining in
Sub-Saharan Africa, 1990, 2001, and 2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.1 Around 9 million young children die before their fifth birthday . . . . . . . . . . . . . . . 40
2.2 An infant in a developing country is ten times more likely to die than a
newborn in a developed country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.1 How the crisis undermined GDP growth in 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.2 Across the world, 884 million people lack access to safe water—84 percent of
them in rural areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4.1 In 2007, 72 million children worldwide were denied access to education . . . . . . . 100
4.2 Tuberculosis kills around 1.3 million people a year, or 3,500 a day. . . . . . . . . . . . 110
5.1 Each year of a girl’s education reduces, by 10 percent, the risk of her
children dying before age five . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.2 Emissions in high-income countries overwhelm those in developing countries . . . 140

Tables
1 Global output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Poverty in developing countries, alternative scenarios, 1990–2020 . . . . . . . . . . . . . . 7
1.1 Poverty reduction is more difficult in poor countries . . . . . . . . . . . . . . . . . . . . . . . . 26
1.2 Poverty reduction is several times more difficult in Sub-Saharan Africa. . . . . . . . . . 26
1.3 Poverty gaps and ratio of mean income of the poor to the $1.25-a-day poverty
line are worse for low-income regions or countries, 2005 . . . . . . . . . . . . . . . . . . . . 26
2.1 Correlation coefficients between growth acceleration and deceleration and
human development indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.2 Frequency of growth acceleration and deceleration, growth rates, and GDP
per capita, 1980–2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.3 Correlation coefficients between economic cycles and economic and
institutional indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.4 World Bank lending for safety nets before and since the food, fuel, and
financial crises, 2006–11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.5 World Bank portfolio allocations to social safety nets, by region, 2009–10 . . . . . . . 46
2A.1 Differences between sample averages: Human development and gender
indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2A.2 Differences between sample averages: Sub-Saharan Africa . . . . . . . . . . . . . . . . . . . 60
2A.3 Differences between sample averages: Economic and institutional indicators . . . . . 61
x CONTENTS GLOBAL MONITORING REPORT 2010

3.1 Global output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70


3.2 Net financial flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3 Inflows of international remittances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.4 Growth regression results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.1 Poverty in developing countries, alternative scenarios, 1990–2020 . . . . . . . . . . . . 102
4.2 Trends for other MDG human development indicators by region and
alternative economic scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
4A.1 Alternate scenarios for poverty reduction, based on a poverty line of
$1.25 a day, by region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
4A.2 Alternate scenarios for poverty reduction, based on a poverty line of
$2.00 a day, by region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
4A.3 Detailed data for archetypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.1 World Bank Group trade-related activities, 2007 and 2008 . . . . . . . . . . . . . . . . . 128
5.2 Distribution of debt distress by country group, end-July 2009 . . . . . . . . . . . . . . . 137
5.3 Gross commitments by IFIs, 2007–09 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Foreword

T
he world is five years from the target responses adopted by many advanced, emerg-
date for the Millennium Development ing, and developing countries, as well as the
Goals (MDGs). We are still recover- swift and sizable assistance provided by inter-
ing from a historic financial and economic national financial institutions and multilat-
crisis. The recovery is uncertain and likely eral development banks. Policy responses and
to be uneven. We know from past crises that international cooperation have been better
the harms to human development during bad than in previous crises.
times cut far deeper than the gains during The postcrisis MDG scorecard is still being
upswings. tallied. Numbers can only be gathered with
Under these conditions, it is especially time-lags and are often incomplete. It is there-
important to consider actions to achieve the fore difficult to take a sharp snapshot of the
MDGs by the 2015 deadline. We need to learn developing world and to analyze the effective-
lessons from MDG experiences to date. This ness of international aid in real time.
2010 Global Monitoring Report can contrib- Despite these measurement challenges, we
ute to that assessment, as part of an MDG will certainly see significant harm to educa-
review led by the United Nations. tion, health, nutrition, and poverty indica-
How has the world performed in overcom- tors, especially in low-income countries. This
ing poverty and fostering human development is not a time for complacency. It is a time for
since the onset of the crisis? This year’s report, exceptional efforts. For example, timely and
The MDGs after the Crisis, aims to answer well-designed conditional cash transfer pro-
this and other critical questions. It highlights grams not only increase household incomes,
lessons from the crisis and presents forecasts but also help children—boys and girls—stay
about poverty and other key indicators. in school and learn. To beat major diseases
We learned from the 1990s crises that and reduce maternal mortality, we need to
macroeconomic stabilization is not enough. work on health systems in a holistic manner.
If strong safety nets are not in place when This means addressing issues ranging from
crises hit, malnutrition and school dropouts financing, service delivery systems, regulation,
increase, potentially leading to the loss of an to governance of the systems. To mitigate the
entire generation. damaging effects of the crisis, we must ensure
A key lesson from this financial crisis is that inclusive and sustainable global growth, main-
the economic and social impact of the down- tain and expand an open international trade
turn would have been far worse if not for the and fi nancial system, deliver on aid commit-
effective—and often extraordinary—policy ments, and encourage the private sector.

GLOBAL MONITORING REPORT 2010 xi


viii FOREWORD GLOBAL MONITORING REPORT 2010

To meet the MDGs, the developing world secondary education by 2015. More girls
must revive its growth and reinforce its resil- than ever in history complete primary school.
ience to shocks. Countries that sowed in times Almost two-thirds of developing countries
of plenty were able to reap in times of loss. reached gender parity at the primary school
Fiscal policy buffers must therefore be rebuilt level by 2005. However, at higher levels of
to allow for future countercyclical responses. schooling, female enrollment lags seriously.
Effective and efficient social safety nets—the And the quality of secondary and tertiary edu-
fi rst line of defense against adverse shocks to cation needs significant improvement.
the poor—must be strengthened. Progress in reducing maternal mortality
Progress on Goal 1—halving extreme pov- is advancing more quickly than we had esti-
erty and hunger—is advancing in fits and mated earlier. This report includes the new
starts. Poverty rates are forecast to continue findings just reported in The Lancet that the
falling in the wake of the crisis, but will do maternal death toll worldwide dropped from
so more slowly. The global rate for extreme 526,300 in 1980 to around 342,900 in 2008,
poverty is projected to be 15 percent in 2015,
far below the latest UN estimates of some
down significantly from 42 percent in 1990.
500,000 for the same year. These signs of
Much of the progress in reducing extreme pov-
erty has taken place in East Asia, where pov- improvement are encouraging. But the prog-
erty dropped from 55 percent in 1990 to 17 ress is fragile and we are still far from reach-
percent in 2005. If this report’s baseline projec- ing the global target of a 75-percent reduction
tion for a recovery holds, the developing world in maternal deaths by 2015 from the ratio that
will reach the poverty reduction goal by 2015. prevailed in 1990. As we emerge from the cri-
However, the crisis has harmed many sis, we must also renew our efforts to achieve
people. By the end of this year, we estimate universal access to reproductive health.
that an additional 64 million people will fall The World Bank Group and the Interna-
into extreme poverty due to the crisis. And tional Monetary Fund have stepped up to the
by 2015, 53 million fewer people will have challenge posed by the crisis. We have taken
escaped poverty. We estimate the poverty rate numerous initiatives to limit the slide in global
for Sub-Saharan Africa will be 38 percent by economic growth and avert the collapse of the
2015, rather than the 36 percent it would have banking and private sectors in many countries.
been without the crisis. The continent will We have also provided financing to govern-
therefore fall short of Goal 1. ments and the private sector, helping to soften
Goal 1 also encompasses the aim of halv- the impact of the crisis on the poor. And we
ing the proportion of people who suffer from have scaled up our support for social safety
hunger. The developing world is off track to nets.
meet this goal. Reducing malnutrition deserves With the deadline for the MDGs fast
more attention, because nutrition has a mul- approaching, we must recognize and over-
tiplier effect on the success of other MDGs, come obstacles in reaching the targets for
including infant mortality, maternal mortality, tackling extreme poverty, hunger, and disease.
and education. Child malnutrition accounts Business as usual will not work. At a time of
for more than a third of the disease burden of uncertainty, we need to extend our limited
children under five. And malnutrition during resources further. We must build upon the
pregnancy accounts for more than 20 percent progress made in improving gender equality,
of maternal mortality. education, and environmental sustainability.
We will likely meet the Goal 3 target The actions we take today will shape future
of achieving gender parity in primary and opportunities and challenges.

Robert B. Zoellick Dominique Strauss-Kahn


President Managing Director
The World Bank Group International Monetary Fund
Acknowledgments

T
his report has been prepared jointly by Ariel Fiszbein, Ann Harrison, Mohammad
the World Bank and the International Zia Qureshi, Martin Ravallion, Augusto de la
Monetary Fund (IMF). In preparing Torre, and Dominique van der Mensbrugghe.
the report, staff also consulted and collabo- Several staff members also made valuable
rated with the African Development Bank contributions, including the following from
(AfDB), the Asian Development Bank (ADB), the World Bank: Luca Bandiera, Uranbileg
the European Bank for Reconstruction and Batjargal, Shaohua Chen, Sie Chow, Lire
Development (EBRD), and the Inter-American Ersado, Elisa Gamberoni, Julien Gourdon,
Development Bank (IDB). The cooperation Hiau Looi Kee, Maria Hazel Macadangdang,
and support of staffs of these institutions are Andrew Mason, Aaditya Mattoo, Israel
gratefully acknowledged. Osorio-Rodarte, Claudio Enrique Raddatz
Delfin S. Go was the lead author and man- Kiefer, Prem Sangraula, Nistha Sinha, Stacey
ager of the report. Richard Harmsen led the Tai, Carolyn Turk, Marijn Verhoeven, and
team from the IMF. Principal authors of the var- Hassan Zaman.
ious parts of the report included Jorge Arbache, Other contributors from the IMF included
Jean-Pierre Christophe Chauffour, Stefano John Brondolo and Mario Mansour; research
Curto, John Elder, Vijdan Korman, Maureen assistance was provided by Emmanuel Hife
Lewis, Hans Lofgren, and Mariem Malouche and Ioana Niculcea.
(World Bank); Andrew Berg, Chris Papageor- Contributors from other institutions
giou, Catherine Pattillo, and Jarkko Turunen included Gaston Gohou and Timothy Turnere
(IMF); Malvina Pollock, Sherman Robinson, (AfDB); Indu Bhushan, Valerie Reppelin-Hill,
William Shaw, and Karen Thierfelder, (consul- Gina Marie Umali, and Edeena Pike (ADB);
tants). Sachin Shahria and Song Song were key Yannis Arvanitis, Gary Bond, and James Ear-
members of the core team and assisted with the wicker (EBRD); and Susana Sitja Rubio and
overall preparation and coordination of the Luis F. Diaz (IDB).
report. Guidance received from the Executive
The work was carried out under the gen- Directors of the World Bank and the IMF
eral guidance of Justin Lin, Senior Vice Presi- and their staffs during discussions of the draft
dent and Chief Economist, and Hans Timmer, report is gratefully acknowledged. The report
Director, Development Economics (DEC) also benefited from many useful comments
Prospects Group, both of the World Bank. and suggestions received from the Bank and
The circle of advisers included Shantayanan IMF management and staff in the course of its
Devarajan, Shahrokh Fardoust, Deon Filmer, preparation and review. Additional informa-

GLOBAL MONITORING REPORT 2010 xiii


xiv ACKNOWLEDGMENTS GLOBAL MONITORING REPORT 2010

tion and data, including background papers, Keane and Jorge del Rosario.
are available on the dedicated Web site, www Bruce Ross-Larson was the principal edi-
.worldbank.org/gmr2010. The multilingual tor. Martha Gottron did the final copyediting.
Web sites accompanying the report were pro- From the World Bank’s Office of the Publisher,
duced by Roula Yazigi, Rebecca Ong, Swati Stephen McGroarty, Susan Graham, Aziz
Priyadarshini Mishra, and Mohamed Has- Gökdemir, and Denise Bergeron managed the
san. Rebecca Ong and Merrell Tuck-Primdahl design, production, printing, and distribution
managed the dissemination activities. The of the report.
translation process was coordinated by Sheila
Abbreviations and Acronyms

ADB Asian Development Bank


AfDB African Development Bank
AIDS acquired immune deficiency syndrome
AfDF African Development Fund
AsDF Asian Development Fund
CIS Commonwealth of Independent States
CPIA Country Policy and Institutional Assessment
DAC Development Assistance Committee
EBRD European Bank for Reconstruction and Development
EU European Union
FDI foreign direct investment
G-8 Group of Eight
G-20 Group of Twenty
GDP gross domestic product
GNI gross national income
HIPC heavily indebted poor country/countries
HIV human immunodeficiency virus
IBRD International Bank for Reconstruction and Development
IDA International Development Association (World Bank Group)
IDB Inter-American Development Bank
IFC International Finance Corporation (World Bank Group)
IFI international financial institution
IMF International Monetary Fund
MCI Monetary Conditions Index
MDGs Millennium Development Goals
MIGA Multilateral Investment Guarantee Agency (World Bank Group)
NGO nongovernmental organization
ODA official development assistance
OECD Organisation for Economic Co-operation and Development
OPEC Organization of the Petroleum Exporting Countries
PEPFAR President’s Emergency Plan for AIDS Relief
PPP purchasing power parity
SDR special drawing rights
UN United Nations
WTO World Trade Organization

GLOBAL MONITORING REPORT 2010 xv


Goals and Targets from the Millennium Declaration
GOAL 1 ERADICATE EXTREME POVERTY AND HUNGER
TARGET 1.A Halve, between 1990 and 2015, the proportion of people whose income is less than $1.25 a day
TARGET 1.B Achieve full and productive employment and decent work for all, including women and young people
TARGET 1.C Halve, between 1990 and 2015, the proportion of people who suffer from hunger
GOAL 2 ACHIEVE UNIVERSAL PRIMARY EDUCATION
TARGET 2.A Ensure that by 2015, children everywhere, boys and girls alike, will be able to complete a full course of primary
schooling
GOAL 3 PROMOTE GENDER EQUALITY AND EMPOWER WOMEN
TARGET 3.A Eliminate gender disparity in primary and secondary education, preferably by 2005, and at all levels of education no
later than 2015
GOAL 4 REDUCE CHILD MORTALITY
TARGET 4.A Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate
GOAL 5 IMPROVE MATERNAL HEALTH
TARGET 5.A Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio
TARGET 5.B Achieve by 2015 universal access to reproductive health
GOAL 6 COMBAT HIV/AIDS, MALARIA, AND OTHER DISEASES
TARGET 6.A Have halted by 2015 and begun to reverse the spread of HIV/AIDS
TARGET 6.B Achieve by 2010 universal access to treatment for HIV/AIDS for all those who need it
TARGET 6.C Have halted by 2015 and begun to reverse the incidence of malaria and other major diseases
GOAL 7 ENSURE ENVIRONMENTAL SUSTAINABILITY
TARGET 7.A Integrate the principles of sustainable development into country policies and programs and reverse the loss of
environmental resources
TARGET 7.B Reduce biodiversity loss, achieving by 2010 a significant reduction in the rate of loss
TARGET 7.C Halve by 2015 the proportion of people without sustainable access to safe drinking water and basic sanitation
TARGET 7.D Have achieved a significant improvement by 2020 in the lives of at least 100 million slum dwellers
GOAL 8 DEVELOP A GLOBAL PARTNERSHIP FOR DEVELOPMENT
TARGET 8.A Develop further an open, rule-based, predictable, nondiscriminatory trading and financial system (including a
commitment to good governance, development, and poverty reduction, nationally and internationally)
TARGET 8.B Address the special needs of the least-developed countries (including tariff- and quota-free access for exports of
the least-developed countries; enhanced debt relief for heavily indebted poor countries and cancellation of official
bilateral debt; and more generous official development assistance for countries committed to reducing poverty)
TARGET 8.C Address the special needs of landlocked countries and small island developing states (through the Programme of
Action for the Sustainable Development of Small Island Developing States and the outcome of the 22nd special
session of the General Assembly)
TARGET 8.D Deal comprehensively with the debt problems of developing countries through national and international measures
to make debt sustainable in the long term
TARGET 8.E In cooperation with pharmaceutical companies, provide access to affordable, essential drugs in developing countries
TARGET 8.F In cooperation with the private sector, make available the benefits of new technologies, especially information and
communications

Source: United Nations. 2008. Report of the Secretary-General on the Indicators for Monitoring the Millennium Development Goals.
E/CN.3/2008/29. New York.
Note: The Millennium Development Goals and targets come from the Millennium Declaration, signed by 189 countries, including 147 heads of state and
government, in September 2000 (http://www.un.org/millennium/declaration/ares552e.htm) and from further agreement by member states at the 2005
World Summit (Resolution adopted by the General Assembly–A/RES/60/1). The goals and targets are interrelated and should be seen as a whole. They
represent a partnership between the developed countries and the developing countries “to create an environment—at the national and global levels
alike—which is conducive to development and the elimination of poverty.”
Overview: MDGs after the Crisis

W
hat is the human cost of the global mortality, the number of children who will be
economic crisis? How many people denied education, and the increase in discrim-
will the crisis prevent from escap- ination against women. Based on that assess-
ing poverty, and how many will remain hun- ment, the report identifies key policies nec-
gry? How many more infants will die? Are essary for the developing countries, donors,
children being pulled out of schools, not get- and the international financial institutions
ting the education they need to become more (IFIs) to reestablish progress toward the Mil-
productive adults and making it virtually lennium Development Goals (MDGs).
impossible to reach 100 percent completion The MDGs provide powerful benchmarks
in primary education by 2015? What are the for measuring global progress on key devel-
gender dimensions of the impacts? These are opment outcomes, calling attention to the
some of the questions as the global economy enormous challenges in low-income coun-
comes out of the worst recession since the tries. The goals have likely contributed to
Great Depression. the progress itself, galvanizing governments,
The questions do not have immediate donors, civil society, private agencies, and
answers—partly because the data to assess the media to support human development.
development outcomes are incomplete and But uniform goals—reducing poverty by
collected infrequently but also because half, infant mortality by two-thirds, mater-
impacts can take several years to emerge. For nal mortality by three-quarters—can under-
example, deteriorating health and nutrition estimate progress in poor countries. Why?
today could lead to higher mortality rates Because the greater the distance to the goals
in subsequent years. Lower investments will from low starting points in poor countries,
hamper future progress in sanitation and the greater the improvement needed to reach
water supply. Fewer children in school will the targets. While the extent to which coun-
lower completion rates in later years. And tries are on track to achieve the MDGs in
household incomes that fall far below the 2015 varies widely, recent improvements have
poverty line will delay escapes from poverty. been widespread, as have the losses caused by
This report uses indirect evidence to assess the crisis.
the impact of the crisis on several indica- From the 1990s until the onset of the crisis
tors, including the number of people who in 2008, developing countries, including low-
will not escape poverty, the increase in infant income countries, made significant progress

GLOBAL MONITORING REPORT 2010 1


2 OVERVIEW GLOBAL MONITORING REPORT 2010

in human development. However, the cri- be required to confront the challenges in the
sis attacked two critical drivers of progress new global economic environment.
toward the MDGs: faster growth and better Ten years after the adoption of the MDGs,
service delivery. The impact was undoubt- the international community is intensifying
edly negative because of the severity of the its monitoring of the progress toward these
recession and the tendency for indicators of goals. The United Nations has called on
human development to decline much more in member states to convene a formal summit
bad times than they improve in good times. on the MDGs in 2010 to review implementa-
But these asymmetric effects are estimated tion of the agreement, and the leaders of the
from past crises, which were often driven by Group of Eight (G-8), meeting in L’Aquila
internal shocks, such as domestic policy fail- in 2009 renewed their commitment to miti-
ures, conflict, and institutional breakdowns. gate the impact of the crisis on developing
By contrast, the current crisis was driven by countries.
an external shock, and policies and institu-
tions in developing countries have improved
considerably in the past 15 years. Moreover,
The MDG indicators showed
many countries have maintained social safety
significant progress before the
nets in the face of the income decline. That is
crisis
why the impact on the MDGs could be more When the crisis hit, many countries had
moderate than in past crises. already made considerable progress in
Even so, the analysis and projections dis- reducing extreme poverty. Globally, pov-
cussed in this report indicate that the dete- erty had fallen 40 percent since 1990, and
rioration in human development is severe, the developing world was well on track to
with effects likely to last for several years. reach the global target of cutting income
This grim outlook has been taken seriously poverty in half by 2015. Thanks to rapid
by the international community. The Inter- growth, especially in China, East Asia had
national Monetary Fund (IMF), the World already halved extreme poverty. Although
Bank, and the regional multilateral develop- Sub-Saharan Africa was unlikely to reach the
ment banks have sharply boosted their assis- target, poverty had been falling rapidly since
tance to developing countries. Despite some the late 1990s. The goal was more ambitious
increase in protectionist measures, develop- for Africa than for other regions, because the
ing countries have largely maintained their 1990 incomes of a large part of the African
access to markets, and the very real danger population were far below the poverty line.
of widespread beggar-thy-neighbor policies And Africa implemented reforms later than
has been avoided. Although aid expanded other regions and therefore benefited later
through 2008, it was at levels far below from accelerating income growth.
what is needed to meet donor commit- Progress on MDGs outside poverty was
ments for total aid and aid to Sub-Saharan uneven. Developing countries were on track
Africa. to achieve gender parity in primary and sec-
Policy responses to the crisis have repercus- ondary education and access to safe water,
sions that must be dealt with. The expansion although countries were falling behind on
of fiscal deficits—required to sustain demand gender parity in tertiary education and
in the depths of the recession—must be reined empowerment of women. Progress was
in by developing and advanced countries good on primary school completion, nutri-
alike. Additional resources will be required tion, maternal mortality and (less so) sanita-
so that the frontloading of concessional assis- tion, even if results at the global level were
tance and the rapid expansion in lending by expected to fall short of targets (figure 1). The
the multilateral development banks do not health goals appeared most challenging. Most
result in a sharp decline in multilateral flows regions were off track, with East Asia, Latin
in the coming years. And shifts in the orga- America, and Europe and Central Asia doing
nization and staff expertise of the IFIs may better than other regions.
GLOBAL MONITORING REPORT 2010 OVERVIEW 3

FIGURE 1 Serious global shortfalls loom for the human development MDGs

100

80

60
percent

40

20

0
MDG 1.a MDG 1.c MDG 2 MDG 3 MDG 3 MDG 4 MDG 5 MDG 7.c MDG 7.c
extreme hunger primary gender gender child maternal access to access to
poverty completion parity parity mortality mortality safe water sanitation
rate (primary) (secondary) under five

distance to goal achieved


distance to goal to be on track to achieve the target by 2015

Source: Staff calculations based on World Development Indicators database.


Note: Based on available data as of 2009, which can range from 2005 to 2009.

The crisis interrupted this progress, but Several factors produce the asymmetric
the effects will not be apparent for many response.
more years. Data needed to assess the degree
of deterioration in development indicators • Economic indicators and quality of insti-
will not be available for two or more years, tutions and policy, such as political stabil-
and some impacts—for example, on mortal- ity, voice and accountability, regulatory
ity rates and school completion rates—will framework, rule of law, and government
materialize only after several years. So this effectiveness, tend to decline sharply in
report uses historical examples and indirect downturns. Distinguishing cause and con-
evidence to assess the effects of the crisis on sequence is difficult, but vicious circles dur-
progress toward the MDGs. ing crises are stronger than virtuous ones
during prosperous times.
• Public and private spending on social ser-
Past crises generated vices can easily be disrupted during eco-
exceptionally poor outcomes nomic crises, just when people need them
The impact of economic cycles on MDG most.
indicators is highly asymmetric. The dete- • Safety nets were uncommon in developing
rioration in bad times is much greater than countries in previous crises.
the improvement during good times (figure • Donor funding also came under pressure if
2). Vulnerable groups—infants and children, the crisis was global or if aid effectiveness
especially girls, particularly in poor countries declined during crises. But there is some
of Sub-Saharan Africa—are disproportion- evidence that official development assis-
ately affected during crises. For example, tance has provided countercyclical support
during contractions, female enrollment in since 2003.
primary and secondary education drops more
than male enrollment. And the consequences
of this disproportionate impact persist long
Why this crisis may be different
into the future. Once children are taken out
for low-income countries
of school, future human capital is perma- Policies and institutions improved before
nently lowered. the crisis. The economic performance of
4 OVERVIEW GLOBAL MONITORING REPORT 2010

FIGURE 2 Key indicators plummet from their overall mean during growth decelerations, all countries

a. Life expectancy at birth b. Mortality rate


8
50
6 40
4 30

deaths, thousands
years gained/lost

20
2
10
0 0
–2 –10
–4 –20
–30
–6
–40
–8 –50
women men total
infant mortality child mortality under 5
growth acceleration growth deceleration (per 1,000 live births) (per 1,000)

growth acceleration growth deceleration

c. Primary completion rate d. Gender equality, ratio of girls to boys


30 45
35
percentage point change

percentage point change

20
25
10 15
5
0
–5
–10 –15
–25
–20
–35
–30 –45
girls boys total primary secondary tertiary
enrollment enrollment enrollment
growth acceleration growth deceleration
growth acceleration growth deceleration

Source: World Bank staff calculations based on data from World Development Indicators database. See chapter 2 for more details.
Note: The panels show differences of averages during growth accelerations and decelerations from overall averages.

developing countries is highly correlated in developing countries mean that they are
with the quality of policies. Many countries better prepared to cope with shocks. Thus
entered the crisis with better policies and fis- the impacts on human development outcomes
cal positions than they had in previous epi- may be less severe if conditions do not deteri-
sodes of contraction. orate and lead to widespread policy failures.
Unlike many previous crises, the current Spending on social safety nets has been
crisis was not caused by domestic policy relatively protected so far. Lower initial fis-
failure. Historically, almost 90 percent of cal deficits and higher priorities for social
the output volatility in low-income coun- spending have protected education and health
tries has been generated by internal condi- spending in most countries. Up-to-date infor-
tions and shocks, such as policy failures and mation is incomplete, but scattered informa-
conflicts. Since the 1990s output volatility tion provides some examples. For example, of
in low-income countries has lessened, and 19 programs initiated and monitored by the
the relative frequency of external shocks has IMF and implemented in collaboration with
increased. Stronger institutions and policies the World Bank in 2008–09, 16 budgeted
GLOBAL MONITORING REPORT 2010 OVERVIEW 5

higher social spending for 2009. Of these, conditionality frameworks. By the end of
9 were countries in Sub-Saharan Africa: February 2010, the IMF had committed a
Burundi, Republic of Congo, Côte d’Ivoire, record high $175 billion (including precau-
Liberia, Malawi, Mali, Niger, Togo, and tionary financing) to emerging and other
Zambia. Several African countries with pov- developing countries with balance of pay-
erty reduction strategies have protected their ments difficulties; the commitments included
funding for social sectors. And some coun- a sharp increase in concessional lending to
tries with fiscal space (Kenya and Nigeria) the world’s poorest countries. The IMF also
have protected capital expenditure, mainly implemented a general allocation of special
for infrastructure. But there are also exam- drawing rights equivalent to $250 billion,
ples of forced contractions in social spending. including almost $100 billion to emerging
Countries with precrisis fiscal and debt issues market economies and developing countries,
(such as Ethiopia and Ghana) had to under- $18 billion of it to low-income countries.
take fiscal tightening. HIV/AIDS (human Standard access to IMF fi nancing has been
immunodeficiency virus/acquired immune doubled, a new flexible credit line without ex
deficiency syndrome) funding has been post policy conditions for countries with very
largely sustained, but with a new concern for strong track records has been adopted, and
the efficiency of resource use. the provision of exceptionally large loans has
become easier, while safeguards have been
preserved.
The international community Responses by multilateral development
responded strongly to the crisis banks have sought to protect core develop-
Despite widespread fears, developing ment programs, strengthen the private sec-
countries’ market access was not signifi- tor, and assist poor households. More than
cantly reduced. At the end of 2009, 350 $150 billion has been committed since the
trade-restrictive measures had been put in beginning of the crisis (two-thirds from the
place around the world, some 20 percent of World Bank Group). International Bank for
them nontariff measures, such as quantita- Reconstruction and Development (IBRD)
tive restrictions, import licenses, standards lending almost tripled in fiscal 2009, and
requirements, and subsidies. Trade remedies the fi rst half of fiscal 2010 shows the stron-
were also on the rise. But in the aggregate, gest IBRD commitments in history ($19.2
protectionism has been contained. The trade- billion, up from $12.4 billion in the same
restricting or -distorting measures introduced period in fiscal 2009). Commitments by the
since October 2008 have amounted to only regional multilateral development banks also
about 0.5 percent of world merchandise increased sharply, by more than 50 percent
trade. Governments and multilateral develop- from 2007 to 2009. Low-income countries
ment institutions supported developing coun- tapped more deeply into multilateral con-
tries’ exports by bolstering trade finance. The cessional resources in 2009, in part through
Group of Twenty leaders pledged $250 bil- frontloading multiyear allocations.
lion in support of trade at their April 2009 Donors increased aid volumes in real
London Summit; the World Bank Group terms through 2009. Following an 11.7 per-
provided guarantees and liquidity for trade cent increase in 2008, total net official devel-
finance through the International Finance opment assistance (ODA) from the OECD’s
Corporation’s Global Trade Finance Program Development Assistance Committee (DAC)
and Global Trade Liquidity Program. And countries rose slightly by 0.7 percent in real
export credit agencies stepped in to prevent a terms in 2009. But in current U.S. dollars,
complete drying up of trade finance. it actually fell from $122.3 billion in 2008
A massive IMF rescue was designed to to $119.6 billion in 2009. The 2009 figure
limit economic contraction and contagion. represents 0.31 percent of members’ com-
The global nature of the crisis led the IMF bined gross national income (GNI). Aid from
to act swiftly to boost lending and modify non-DAC donors, led by Saudi Arabia, rose
6 OVERVIEW GLOBAL MONITORING REPORT 2010

63 percent in real terms in 2008 to $9.5 bil- supported commodity exporters, but com-
lion. Development assistance from China modity prices remain below their precrisis
will likely more than double in the next three levels.
years. Private aid, also substantial, is rising Trade is recovering unevenly across
rapidly. And progress continued in reduc- regions. World trade contracted by 12 per-
ing poor countries’ debt burden through the cent in 2009, and all regions experienced
Highly Indebted Poor Countries (HIPC) Ini- deep declines in imports. Signs of recovery are
tiative and the Multilateral Debt Relief Initia- evident, but trade remains fragile. At the end
tive. For 35 post-HIPC-decision-point coun- of 2009 global trade was still below its pre-
tries, the debt burden will be reduced by 80 crisis level. Almost a year into the recovery,
percent. the dollar value of global trade remains 20
percent lower than it was before the crisis.
The impact of the crisis on poverty will be
The recovery is stronger than long lasting. Poverty rates will continue to fall
expected, but the outlook for the after the crisis, but more slowly (table 2).1 By
MDGs remains worrisome 2015 the global poverty rate is projected to be
GDP growth in emerging market and devel- 15 percent, not the 14.1 percent it would have
oping economies is projected to accelerate to been without the crisis. The crisis will leave an
6.3 percent in 2010. Most economies show additional 64 million people in extreme pov-
signs of recovery (table 1), although many erty by the end of 2010. The recovery will not
countries remain dependent on exceptional make up all the lost ground. And as a result
policy stimulus, and in most countries growth of the crisis, 71 million fewer people will have
is not strong enough to undo the damage escaped poverty by 2020. For Sub-Saharan
caused by the sharp deceleration in incomes Africa, the poverty rate is expected to be 38
and social conditions in 2009. Fiscal deficits percent by 2015, rather than the 36 percent it
in emerging market and developing economies would have been without the crisis, lifting 20
rose by almost 3 percent of GDP in 2009 and million fewer people out of poverty.
are projected to remain high in 2010. Finan- The medium-term impact on other MDGs
cial market conditions for these economies may also be considerable. Illustrative and
are improving and capital flows are returning, indicative results from growth analyses2 sug-
although international bank financing and gest persistent gaps between precrisis and
foreign direct investment flows are projected postcrisis trends in 2015 (figure 3):
to remain weak in 2010. The rebound of
commodity prices in tandem with the global • An additional 55,000 infants might die in
recovery in manufacturing production has 2015. And 260,000 more children under

TABLE 1 Global output


percent change
Projections

Region 2007 2008 2009 2010 2011–13

World output 5.2 3.0 –0.6 4.2 4.4


Advanced economies 2.8 0.5 –3.2 2.3 2.4
Emerging and developing economies 8.3 6.1 2.4 6.3 6.6
Central and Eastern Europe 5.5 3.0 –3.7 2.8 3.8
Commonwealth of Independent States 8.6 5.5 –6.6 4.0 4.1
Developing Asia 10.6 7.9 6.6 8.7 8.6
Middle East and North Africa 5.6 5.1 2.4 4.5 4.8
Sub-Saharan Africa 6.9 5.5 2.1 4.7 5.7
Western Hemisphere 5.8 4.3 –1.8 4.0 4.2
Source: IMF World Economic Outlook. See chapter 3 for further discussions.
GLOBAL MONITORING REPORT 2010 OVERVIEW 7

TABLE 2 Poverty in developing countries, alternative scenarios, 1990 –2020


Scenario 1990 2005 2015 2020

Global level
Percentage of the population living on less than $1.25 a day
Postcrisis 41.7 25.2 15.0 12.8
Precrisis 41.7 25.2 14.1 11.7
Low-growth 41.7 25.2 18.5 16.3

Number of people living on less than $1.25 a day (millions)


Postcrisis 1,817 1,371 918 826
Precrisis 1,817 1,371 865 755
Low-growth 1,817 1,371 1132 1053
Source: World Bank staff calculations.

FIGURE 3 The long-run effect of slower growth on selected MDGs is worrisome

b. MDG 3: Gender parity in primary


a. MDG 2: Primary completion rate and secondary education
96 98.0
94
97.0
92
percent

percent

90 96.0
88
95.0
86
84 94.0
2007 2009 2011 2013 2015 2017 2019 2007 2009 2011 2013 2015 2017 2019
low-growth postcrisis precrisis low-growth postcrisis precrisis

c. MDG 4: Child mortality under five d. MDG 7c: Access to safe drinking water
75 17
73 15
deaths per 1,000

13
percent

71
11
69
9
67 7
65 5
2007 2009 2011 2013 2015 2017 2019 2007 2009 2011 2013 2015 2017 2019
low-growth postcrisis precrisis low-growth postcrisis precrisis

Source: World Bank staff calculations.

five could have been prevented from dying in Developing countries must
2015 in the absence of the crisis. The cumu- maintain good policies and
lative total from 2009 to 2015 could reach effective service delivery
265,000 and 1.2 million, respectively.
• An estimated 350,000 more students might Growth and institutional quality rein-
be unable to complete primary school in force each other. Before the crisis, policy
2015. reforms triggered an impressive acceleration
• Some 100 million more people might of growth in the developing world, which in
remain without access to an improved turn helped to strengthen institutions and
source of water. economic fundamentals. One of the dangers
8 OVERVIEW GLOBAL MONITORING REPORT 2010

of the crisis is that reforms might be aban- resist protectionist pressures and keep mar-
doned, leading to policy reversals and a kets open as expansionary policies unwind.
deteriorating economy. It is important that Beyond Doha, there is a need to broaden
all countries adopt credible medium-term cooperation on cross-border policy mat-
fiscal adjustment plans to bolster confidence ters that are not on the Doha Development
in macroeconomic policies and that they Agenda (climate change, and food and energy
undertake policy reforms to secure long- security). The crisis has also revealed the
term growth. importance of strengthening monitoring and
The effectiveness of safety nets should be public reporting of government measures to
enhanced given their importance in cush- increase transparency in the trading system
ioning the effects of crises and in reducing (Global Trade Alert, Global Antidumping
poverty. Safety net programs in low-income Database, World Trade Organization [WTO]
countries are often small and fragmented, monitoring reports).
covering only a small percentage of the poor Better monitoring of trade finance is
and vulnerable. There are real concerns about needed. Although recent data indicate that
whether they are affordable and administra- trade fi nance is recovering, a mechanism is
tively feasible in light of the various negative needed to collect data and monitor the mar-
incentives they might create. Understanding ket systematically and reliably—to assess
what kind of safety nets will serve social assis- how current interventions influence the sup-
tance best, what their implementation chal- ply of credit and trade flows, and to provide a
lenges are, and how to develop such programs useful early warning of stress in trade credit.
for maximum effectiveness should inform Developing countries’ trade logistics
policy reforms in developing countries. need further support. Lowering trade costs
If the global recovery remains weak, through better trade regulations, trade logis-
spending shifts, internal resource mobiliza- tics, and infrastructure can make a criti-
tion, and better service delivery can help, cal contribution toward development. The
but these tools have limits. In the face of Second Global Review of Aid for Trade in
declining external revenues, shifting expendi- Geneva in July 2009 found that donors were
tures to protect social services and increasing offering more and better aid for trade and
domestic tax collections can keep the MDG that cooperation between developing coun-
indicators from deteriorating to a worst-case tries is engaging new partners. Sustaining
level. But higher taxes can also retard prog- efforts to deliver on the commitments at the
ress on the poverty MDG by reducing house- 2005 WTO Ministerial Meeting (in Hong
hold income and spending. The negative Kong, China) to expand aid for trade should
effects of a tax increase need to be offset by continue to be a priority. And more such aid
better policies and service delivery. Even so, needs to be directed to low-income countries,
better development outcomes hinge critically which receive only about half the total.
not only on a better policy environment but Aid has to expand to meet previous com-
also on a rapid global recovery that improves mitments. The expected medium-term impact
the export conditions, terms of trade, and of the crisis on low-income countries has
capital flows for low-income countries. heightened the urgency to scale up aid. But
current donor spending plans leave a $14 bil-
lion shortfall in the commitments to increase
The global community must aid by $50 billion by 2010 (in 2004 dollars).
continue to support developing And the Group of Eight Gleneagles commit-
countries ment to double aid to Africa by 2010 has yet
Multilateral cooperation in trade must to be reflected in core development aid to the
be strengthened. Completion of the Doha region. Aid to Africa has grown 5 percent
Round is important in the aftermath of the annually since 2000, but much of it has been
crisis, because it would help governments in the form of debt relief or emergency and
GLOBAL MONITORING REPORT 2009 OVERVIEW 9

humanitarian assistance, not new finance. increase aid. All countries should adopt cred-
Reaching the 2010 target requires a further ible medium-term fiscal adjustment plans to
increase of $20 billion. Donor spending plans bolster confidence in macroeconomic policies
indicate that only an additional $2 billion is and undertake policy reforms to secure long-
programmed, leaving a gap of $18 billion. term growth.
Moreover, considerable scope remains for The international financial institutions
strengthening aid effectiveness by making aid need to adapt to the new global environ-
more predictable; rationalizing the division ment. In the absence of increased resources
of labor among donors; untying aid from the from donors, the crisis-induced frontloading
provision of goods and services in the donor of concessional resources by the International
country; increasing reliance on need and merit Development Association and other multilat-
to guide aid allocations; and addressing the eral agencies implies that concessional flows
problem of countries that receive too little aid. from these institutions must decline in the
near future. Similarly, the sharp rise in IBRD
commitments highlights the need for discuss-
Necessary reactions to the crisis ing a capital increase to avoid an eventual
raise further policy challenges falloff in lending. Changes in responsibilities
Developing countries’ fiscal positions dete- and organization of the IFIs are on the hori-
riorate. Several developing countries main- zon: increased demand for technical services
tained spending and expanded fiscal deficits will shift requirements for staff expertise;
to support domestic demand during the crisis. coordination among the IFIs will need to be
Indeed, more than one-third of these coun- strengthened; and proposals to improve the
tries introduced discretionary fiscal stimu- responsiveness of the multilateral develop-
lus plans in 2009. Absent such support, the ment banks (such as decentralization at the
impact on individual countries’ growth and World Bank) are under consideration. The
the shortfall in global demand would have rapid response of the global economic com-
been even greater than they were. But the munity to the downturn helped avoid a new
rapid expansion of fiscal deficits and greater Great Depression, but decisive leadership still
reliance on domestic fi nance in many coun- is required to ensure a rapid and sustainable
tries may not be sustainable. The deteriora- recovery.
tion in debt ratios in low-income countries is
particularly worrisome.
Optimal exit policies from policy support Notes
depend on country circumstances. Countries
1. This projection is based on household surveys
with weak private demand should continue in more than 100 countries and on the effect of
policy support if they have the fiscal space. growth on household consumption.
But countries facing financing constraints 2. These analyses are based on the estimated
cannot delay adjustment. Donors should relationships between GDP growth and the
assist them by meeting their commitments to MDGs, which can vary by country.
1
Millennium Development Goals:
Significant Gains before the Crisis

W
hat is the human cost of the global enormous challenges in low-income coun-
economic crisis? How many peo- tries. The goals have likely contributed to
ple will the crisis prevent from the progress itself, galvanizing governments,
escaping poverty, and how many will remain donors, civil society, private agencies, and
hungry? How many more infants will die? the media to support human development.1
Are children being pulled out of schools, But uniform goals—reducing poverty by half,
making it virtually impossible to reach 100 infant mortality by two-thirds, maternal mor-
percent completion in primary education by tality by three-quarters—can underestimate
2015? What are the gender dimensions of the progress in poor countries. Why? Because the
impacts? These are some of the questions as pace of progress is inversely related to initial
the global economy comes out of the worst conditions, particularly the greater distance
recession since the Great Depression. to the goals from low starting points in poor
The questions raised here are hard to countries.2 And although the extent to which
answer immediately, partly because the data countries are on track to achieve the MDGs
to assess development outcomes are incom- in 2015 varies widely, recent improvements
plete and collected infrequently but also have been widespread. So, too, are the losses
because impacts can take several years to caused by the interruption in progress.
emerge. For example, deteriorating health This chapter offers an overview of prog-
and nutrition now will lead to higher mortal- ress in the decade before the crisis. It serves as
ity rates later. Lower investments will ham- the starting point for a more forward-looking
per future progress in sanitation and water analysis and explains what is at stake when a
supply. Fewer children in school will lower period of strong growth in many developing
completion rates in later years. And house- countries, including the low-income countries
hold incomes that fall far below the poverty in Africa, is interrupted.3 Subsequent chapters
line will delay escapes from poverty. examine economic forecasts, how develop-
The Millennium Development Goals ment indicators responded to previous crises,
(MDGs) provide powerful benchmarks for and how the current crisis differs, thus pro-
measuring global progress on key devel- viding the building blocks to answer the ques-
opment outcomes, calling attention to the tions about the human costs of the crisis.

GLOBAL MONITORING REPORT 2010 11


12 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

MAP 1.1 Africa is the only region with high extreme poverty

Poverty rate:
Share of population living on less than $1.25 a day, % (2005) Green
(De
≥50
25–49.9
10–24.9
2–9.9
<2 Canada
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: PovcalNet, the World Bank.

Because the MDGs are more nutrition, and sanitation—was lagging at the
ambitious for poor countries, global level (figure 1.2).
global progress is mixed
Before the global economic crisis in 2008–09, But substantial progress is
overall progress on the MDG targets were
evident in many areas
particularly strong on poverty reduction, even Economic growth is a key driver in reducing
in Africa (figure 1.1). Progress was also made poverty and achieving other desired develop-
on gender parity in primary and secondary ment outcomes, and it is there that progress
education, maternal mortality, and on reli- has been most evident (figure 1.3). Economic
able access to improved water. Progress was growth in developing countries has acceler-
less encouraging on gender parity in tertiary ated, thanks to improved macroeconomic pol-
education and other targets for the empow- icies and a hospitable global environment—
erment of women. Of greatest concern were rapidly expanding world trade, favorable
the human development goals. Progress on commodity prices, more foreign aid and debt
most of them—especially for child mortal- relief, abundant low-cost capital, and large
ity but also for primary school completion, remittance flows. The 12 years before the
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 13

IBRD 37731
APRIL 2010

nland
en)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Portugal Spain Turkey Armenia baijan Turkmenistan of Korea
Monaco Greece Tajikistan Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire Cameroon
African Rep. Somalia Palau
Togo Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Kenya
São Tomé and Príncipe Gabon Congo Rwanda Singapore Nauru
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

Figure 1.1 But Africa’s poverty rate is falling

Source: World Development Indicators.


14 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 1.2 At the global level, serious shortfalls loom for the human development MDGs

100

80

60
percent

40

20

0
MDG 1.a MDG 1.c MDG 2 MDG 3 MDG 3 MDG 4 MDG 5 MDG 7.c MDG 7.c
extreme hunger primary gender gender child maternal access to access to
poverty completion parity parity mortality mortality safe water sanitation
rate (primary) (secondary) under five

distance to goal achieved


distance to goal to be on track to achieve the target by 2015

Source: Staff calculations based on latest available data as of 2009 from the World Development Indicators database.

FIGURE 1.3 Since the 1990s growth in developing countries has accelerated

a. Growth in developing countries has been b. Average incomes in Sub-Saharan Africa have
rapid relative to rich countries risen in tandem with those in other regions
10 8

8 6
6
% annual change
growth rate, %

4
4
2
2
0
0

–2 –2

–4 –4
1975 1980 1985 1990 1995 2000 2005 2010 1990 1992 1994 1996 1998 2000 2004 2006 2008 2010
developing countries GDP growth developing countries Sub-Saharan Africa
developing countries trend GDP growth developing countries high-income
high-income countries GDP growth (excluding China and India) countries
high-income countries trend GDP growth

Source: Staff calculations based on World Development Indicators database.

crisis capped a remarkable period of sustained been greater for the goals most influenced by
economic growth, technological advances, and economic growth, such as income poverty.
globalization that started in 1950, and spread Extreme poverty is falling rapidly. Despite
to a widening number of developing coun- growing populations, the number of poor
tries in Asia and Latin America and finally people living on less than $1.25 a day in
to low-income countries in Africa.4 Since the developing countries fell from about 1.8 bil-
mid-1990s growth in Sub-Saharan Africa lion in 1990 to 1.4 billion in 2005—from 42
has been comparable to that in other regions. percent of the population to 25 percent. The
As a result, progress toward the MDGs has global MDG target of 21 percent poverty is
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 15

FIGURE 1.4 Poverty reduction is substantial in all regions

Poverty headcount ratio at $1.25 a day


70

60

50
percentage

40

30

20

10

0
East Asia Europe and Latin America Middle East South Sub-Saharan
and Pacific Central Asia and the and Asia Africa
Caribbean North Africa

1990 2005

Source: PovcalNet, World Bank.


Note: Poverty rate is given as purchasing power parity (PPP) rate.

FIGURE 1.5 Another view: Poverty rates and the number of poor people are falling rapidly

80 2.0
1.8
1.6
proportion of people below

population, billions

60 1.4
1.2
poverty line, %

1.0
40 0.8
0.6
0.4
20 0.2
0
1981 1984 1987 1990 1993 1996 1999 2002 2005
0 Sub-Saharan Africa
1981 1984 1987 1990 1993 1996 1999 2002 2005
South Asia (excluding India)
Sub-Saharan South Asia East Asia and Pacific (excluding China)
East Asia and Pacific India China other regions

Source: PovcalNet, World Bank.

the one most likely to be exceeded, but the growth, the number of poor people actually
economic crisis adds new risks to prospects rose from 436 million to 456 million.
for reaching the goal. With the precrisis surge of growth in Sub-
Much of the progress is attributable to East Saharan Africa, the proportion of Africans
Asia, which reduced the incidence of poverty living on less than $1.25 a day fell from 58
from 55 percent in 1990 to 17 percent in 2005 percent in 1990 to 51 percent in 2005, but
(figure 1.4). China reduced its poverty rate the absolute number of poor people rose
from 60 percent to 16 percent, as the absolute from 296 million to 388 million (figure 1.5).
number of extremely poor fell from 683 mil- Despite Africa’s recent progress, the pace of
lion to 208 million. India reduced the share of economic growth is still not fast enough there
its population living in poverty from 51 per- to cut the 1990 poverty rate by half in 2015.
cent to 42 percent, but because of population In almost every other region, it is.
16 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 1.6 Net enrollment rates are rising in many countries Higher enrollments are shrinking the gen-
der gap in education. Gender equality and
Turkmenistan 76 99 female empowerment, the third MDG, are
São Tomé and Principe
important not only in themselves but also
84 94
because they improve progress on the other
Mongolia 79 97 MDG targets related to poverty, hunger, dis-
Rwanda 75 86 ease, and education. As more girls than ever
Nepal 66 84
complete primary school, many countries
have reached gender parity in primary educa-
Lao PDR 62 79
tion (figure 1.7). All told, almost two-thirds
Tanzania 49 73 of developing countries reached gender parity
Burundi 47 71 at the primary school level by 2005, and the
Sierre Leone 41 69 MDG 3 target of achieving gender parity in
Central African Republic
primary and secondary education can be met
43 59
by 2015. Sub-Saharan Africa is making good
Guinea-Bissau 39 54
progress but is far behind the global target.
Guinea 39 51

Burkina Faso 27 46 Access to safe drinking water is on track glob-


Ethiopia 31 45 ally and in most regions. Rapid expansions
Somalia 12 22
of infrastructure spending account for part of
this increased access. Progress on this part of
0 20 40 60 80 100
MDG 7 remains vital for child survival and
percentage
various health improvements. Between 1990
2000 2006 and 2006 more than 1.6 billion people gained
access to improved sources of drinking water,
Source: UNICEF 2007.
raising the proportion of the population with
Universal primar y education is within such access from 76 percent to 86 percent (fig-
reach. Many countries are close to provid- ure 1.8). As many as 76 developing countries
ing universal primary education. In more are on track to hit the target. But 23 develop-
than 60 developing countries, over 90 per- ing countries have made no progress, and 5
cent of primary-school-age children are in others have fallen behind.
school; the number of children out of school
fell from 115 million in 2002 to 72 million New findings suggest a significant drop
in 2007, even with growing populations. In worldwide in the maternal mortality. New
2007 the primary school completion rate analysis of maternal deaths in 181 countries
reached 86 percent for all developing coun- from better data found a significant decline
tries together—93 percent for middle-income globally.5 Aggregate maternal deaths decreased
countries but just 65 percent for low-income by over 35 percent from about 526,300 in
countries. Net enrollment rates are rising in 1980 to 342,900 in 2008. More than half of
several poor countries (figure 1.6). Because of all maternal deaths were concentrated in six
the substantial improvements, the world will countries—India, Nigeria, Pakistan, Afghani-
come close to the goal of universal primary stan, Ethiopia, and the Democratic Republic
school completion in 2015 (MDG 2) but still of Congo. All told, maternal deaths for every
fall short. For Sub-Saharan Africa and South 100,000 live births decreased markedly from
Asia the lower rates of 60 percent and 80 per- 422 in 1980 to 320 in 1990 and to 251 in
cent in 2007 still constitute advancement over 2008. The yearly rate of the decline in the
the 51 percent and 62 percent, respectively, in global maternal mortality ratio since 1990
1991. But with 41 million primary-school- was 1.3 percent (with an uncertainty range of
age children in Sub-Saharan Africa and 31.5 1.0–1.5). Progress is still varied. The Arab
million in South Asia out of school, the task Republic of Egypt, China, Ecuador, and
of meeting the target remains challenging. Bolivia have been achieving rapid gains, and
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 17

FIGURE 1.7 Gender parity is close in primary education

100
primary enrollment, %

80
ratio of girls to boys,

60

40

20

0
East Asia Europe and Latin America Middle East South Sub-Saharan World
and Pacific Central Asia and the and Asia Africa
Caribbean North Africa

100
secondary enrollment, %

80
ratio of girls to boys

60

40

20

0
East Asia Europe and Latin America Middle East South Sub-Saharan World
and Pacific Central Asia and the and Asia Africa
Caribbean North Africa

1991 2007

Source: World Development Indicators.

23 countries are on track with this MDG 5. FIGURE 1.8 More people have improved sources of water
In Sub-Saharan Africa, central and eastern
regions showed some improvement since
World
1990, but southern and western regions
showed deterioration because of the signifi-
Sub-Saharan Africa
cant number of pregnant women who died
from HIV infection. In southern Africa, the
South Asia
maternal mortality ratio increased from 171
in 1990 to 381 in 2008. Middle East and North Africa

Where progress has been mixed Latin America and the Caribbean
or lacking
Europe and Central Asia
The recent food crisis has complicated prog-
ress on fighting malnutrition and hunger. East Asia and Pacific
The developing world is not on track to halve
the proportion of people who suffer from 0 20 40 60 80 100

hunger. Because reducing malnutrition is population with access, %


essential to success on several other MDGs, 1990 2006
including infant mortality, maternal mortal-
ity, and education, it has a multiplier effect. Source: World Development Indicators.
18 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

Child malnutrition accounts for more than a adequate sanitation. In Sub-Saharan Africa
third of the disease burden of children under the proportion with access rose from 26 per-
age five. And malnutrition during pregnancy cent in 1990 to just 31 percent in 2006, and
accounts for more than 20 percent of mater- in South Asia, from 18 percent to 33 percent.
nal mortalities. MDG 7 also calls for integrating sustainable
The proportion of children under five who development into country policies and pro-
are underweight declined from 31 percent in grams and reversing the losses of environ-
developing countries in 1990 to 26 percent mental resources. Progress on this broader
in 2007, a much slower pace than needed to environmental agenda, although fairly slow, is
halve malnutrition by 2015. Progress has been picking up as the world focuses on environ-
slowest in Sub-Saharan Africa and South mental sustainability and climate change.6
Asia, with severe to moderate stunting affect-
ing as many as 46 percent of children under Prospects are worst for most MDGs relat-
five—more than 140 million children. ing to health, such as infant mortality. The
under-five mortality rate in developing coun-
Progress on full and productive employ- tries declined from 101 deaths per 1,000 to
ment, especially for women, was lack- 74 between 1990 and 2007, showing notable
ing even before the crisis. (figure 1.9). but insufficient progress to meet MDG 4 for
The employment-to-population ratio is reducing child mortality under five by two-
the proportion of a country’s working-age thirds. In 2006, 10 million children died
population (ages 15 years and older) that before age five from preventable diseases, com-
is employed. Considerable underemploy- pared with 13 million in 1990. The human
ment in informal sectors and subsistent immunodeficiency virus/acquired immune
activities of rural areas are of course hard deficiency syndrome (HIV/AIDS) epidemic
to account for in developing countries. The and civil conflicts have impeded Sub-Saharan
female employment ratios have consistently Africa’s progress. Its under-five mortality rate
been lower than male ratios, particularly in stood at 144 deaths per 1,000 in 2008, down
the Middle East and North Africa and in from 185 in 1990. Sub-Saharan Africa has 20
South Asia. Nonetheless, progress is noted percent of the world’s children under age five
in the female ratios for Latin America and but 50 percent of all child deaths. Progress in
the Caribbean and to a slight extent in the reducing infant mortality is also well short of
Middle East and North Africa (figure 1.9). the target in South Asia.
The situation is similar for universal
Gender parity is weak beyond primary access to reproductive health (MDG 5.b). For
education. MDG 3 also calls for gender par- example, the contraceptive prevalence rate
ity in tertiary education (figure 1.10), gender has increased for all income groups between
equality in employment, and greater political 1990 and 2007 but is still quite low at only
representation of women. Progress toward 33 percent for low-income countries in 2007
these targets has been slower and less even. (figure 1.11).
The gender targets face added risk from the
current crisis because evidence from past cri- Progress in halting the spread of major com-
ses suggests that women in general are more municable diseases has been mixed (MDG
vulnerable. 6). An estimated 33.4 million people were
living with HIV/AIDS in 2008; there were
Access to sanitation has been elusive. Sani- 2.7 million new infections and about 2 mil-
tation coverage, another important target lion AIDS-related deaths. The rapidly rising
of MDG 7 on environmental sustainability, trends of HIV spread and related deaths that
rose from 43 percent in 1990 to 55 percent in were recorded in the 1990s were halted in
2006, in low- and middle-income countries. the 2000s and were showing some signs of
But the global target will be missed. Almost decline in recent years (figure 1.12). However,
half the people in developing countries lack further actions are still necessary to achieve
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 19

FIGURE 1.9 Progress lacking on ratio of employment to population

a. Men
100
employment to population ratio, %

90
80
70
60
50
40
30
20
10
0
East Asia Europe and Latin America Middle East South Asia Sub-Saharan
and Pacific Central Asia and the and North Africa
Caribbean Africa

b. Women
100
employment to population ratio, %

90
80
70
60
50
40
30
20
10
0
East Asia Europe and Latin America Middle East South Asia Sub-Saharan
and Pacific Central Asia and the and North Africa
Caribbean Africa

1991 2007

Source: World Development Indicators.

significant reversals. Sub-Saharan Africa about 1 million people annually, 80 percent of


remains the region most heavily affected by them children in Sub-Saharan Africa.
HIV worldwide, accounting for over two-
thirds of all people living with HIV and for More attention should be given to achieving
nearly three-fourths of AIDS-related deaths in environmental goals. According to the World
2008. HIV prevalence has declined in recent Bank’s 2010 World Development Report on
years in Sub-Saharan Africa (map 1.2), but it development and climate change, develop-
has risen in other regions, albeit from much ing countries can shift to lower-carbon paths
lower levels. while promoting development and reducing
Antiretroviral treatment (ART) now poverty, as long as they receive financial and
reaches almost a third of people living with technical assistance from high-income coun-
HIV/AIDS in developing countries (figure tries. High-income countries, which produced
1.13). But few countries will meet the target of most of the greenhouse gas emissions of the
universal access to treatment anytime soon. past, must act to shape our climate future, tak-
The prevalence of tuberculosis, which killed ing actions quickly to reduce their own carbon
1.3 million people in 2008, has been declin- footprints and boost development of alterna-
ing in all regions except Sub-Saharan Africa. tive energy sources. The costs for getting there
Mortality from malaria remains high—at will be high but still manageable. A key way
20 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 1.10 Female enrollment in tertiary education lags in Sub-Saharan Africa and South Asia, 2007

140
enrollment, ratio of girls to boys, %

120

100

80

60

40

20

0
East Asia Europe and Latin America Middle East South Sub-Saharan
and Pacific Central Asia and the and Asia Africa
Caribbean North Africa

primary secondary tertiary

Source: World Development Indicators.

FIGURE 1.11 The contraceptive prevalence rate is rose during 2003–05 but fell in both 2006
low for low-income countries and 2007, dropping from 0.33 percent of
donor gross national income (GNI) in 2005
100
to 0.28 percent in 2007. The ratio of ODA
90
to GNI reached 0.31 percent in 2009, but to
80
married women ages 15–49

meet donors’ aid commitments, aid increases


using contraception, %

70
will have to be larger and sustained. Donors
60
need to shield aid budgets from the fiscal
50
impact of the financial crisis. In trade the
40
largest implementation gap is the failure to
30
conclude the Doha Round of negotiations.
20
Progress has been greater in providing debt
10
relief to poor countries, thanks to the Heav-
0
all developing low-income middle-income ily Indebted Poor Countries Initiative and the
countries countries countries Multilateral Debt Relief Initiative. The lack
1990 2007 of specific targets is hampering the transfer of
technology to developing countries and their
Source: World Development Indicators. access to essential drugs.8

to slow climate change is to ramp up funding


for mitigation in developing countries, where Progress varies by type of
most future growth in emissions will occur.7 country
Inside the global picture is considerable
More progress is needed on developing a variation across income groups, regions, and
global partnership for development. MDG 8 countries. Middle-income countries have
covers cooperation in aid, trade, debt relief, progressed fastest toward the MDGs. As a
and access to technology and essential drugs. group, they are on track to achieve the tar-
Net disbursements of official development get for poverty reduction. But many of them
assistance (ODA) from the Development still have large concentrations of poverty, in
Assistance Committee of the Organisation part reflecting great income inequality. These
for Economic Co-operation and Development concentrations of poverty, together with
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 21

large populations in some countries, mean MAP 1.2 Proportion of population living with
that middle-income countries remain home HIV is still high but declining in Sub-Saharan
Africa, 1990, 2001, and 2007
to a majority of the world’s poor in absolute
numbers. Many middle-income countries
also continue to face major challenges in 1990
Mauritania Mali
Niger
achieving the nonincome human develop- Cape Verde
The Gambia
Senegal
Burkina
Chad
Sudan
Eritrea
Faso
ment goals. Guinea-Bissau Guinea
Sierra Leone Ghana
Nigeria Central Ethiopia
Liberia Cameroon African Rep.
Progress has been weaker in low-income Côte D’Ivoire Togo Uganda
Somalia

countries, with considerable variability. It has Benin


Equatorial Guinea GabonCongo Rwanda Kenya
Dem. Rep. Burundi Seychelles
São Tomé and Príncipe of Congo
been slowest in fragile and confl ict-affected Tanzania Comoros

states (figure 1.14). Wracked by confl ict and Angola Zambia Malawi Mayotte
(Fr)
Mozambique
hampered by weak capacities, these states— Zimbabwe
Namibia Botswana
Madagascar
Mauritius
more than half of them in Sub-Saharan Swaziland
South Lesotho
Africa—present difficult political and gov- Africa

ernance contexts for the effective delivery


of development finance and services. Frag-
ile states account for close to a fi fth of the
2001
population of low-income countries but more Mauritania Mali
Cape Verde Niger
than a third of their poor people. Much of The Gambia
Senegal
Burkina
Chad
Sudan
Eritrea
Guinea-Bissau Guinea Faso
the challenge of achieving the MDGs will be Sierra Leone Ghana
Nigeria Central Ethiopia
Liberia Cameroon African Rep. Somalia
concentrated in low-income countries, espe- Côte D’Ivoire Togo Uganda
Benin GabonCongo Rwanda Kenya
cially fragile states. Equatorial Guinea
São Tomé and Príncipe
Dem. Rep. Burundi
of Congo
Seychelles
Tanzania Comoros

Angola Zambia Malawi Mayotte


Most regions are progressing rapidly toward Mozambique
(Fr)
Zimbabwe Madagascar
the goals. Thanks to rapid economic growth, Namibia Botswana
Mauritius
especially in China, East Asia has already South
Swaziland
Lesotho
Africa
halved extreme poverty. South Asia is on
track to do the same, but it is seriously off
track on most human development goals. For
2007
the health goals most regions are off track, Mauritania Mali
Niger
Cape Verde Chad Eritrea
although the rate of progress varies substan- The Gambia
Senegal
Burkina Sudan
Guinea-Bissau Guinea Faso
tially, with East Asia and the Pacific, Latin Sierra Leone Ghana
Nigeria Central
Cameroon African Rep.
Ethiopia
Liberia Somalia
America and the Caribbean, and Europe Côte D’Ivoire Togo Uganda
Benin
Equatorial Guinea GabonCongo Rwanda Kenya
Dem. Rep. Burundi
and Central Asia doing better than the other São Tomé and Príncipe of Congo
Seychelles
Tanzania Comoros
regions.
Angola Zambia Malawi Mayotte
(Fr)
Mozambique
Zimbabwe Madagascar
Sub-Saharan Africa is a special case. It is Namibia Botswana
Mauritius
easy to see that Sub-Saharan Africa lags on South
Swaziland
Lesotho
Africa
all the MDGs, including poverty reduction. IBRD 37740
APRIL 2010
But that is only half the story—because the
region has made progress. Practically all the Prevalence of HIV in Sub-Saharan Africa: % of population ages 15–49
MDG curves for Sub-Saharan Africa (figure
≥25 13–24.9 6.0–12.9
1.15) have been headed in the right direc-
1.0–5.9 less than 1 no data
tion for more than 10 years, but the progress
required looks steeper in direct compari- Source: United Nations 2009.
sons with other regions because of Africa’s
lower starting points. Achievements are more Many countries are likely to fall short of
apparent when viewed against the severe eco- most of the goals. Among countries for
nomic stagnation that afflicted the region which there are data, the proportion off
from the 1970s to the early 1990s. track exceeds that of countries on track for
22 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 1.12 HIV prevalence rates and estimated deaths are showing signs of decline

1.2 5

4
% of adults, ages 15–49

0.9

deaths, millions
3
0.6
2

0.3
1

0 0
1990 1993 1996 1999 2002 2005 2008 1990 1993 1996 1999 2002 2005 2008
estimates high and low estimates

Source: UNAIDS/WHO.
Note: The HIV adult prevalence rate is defined as the proportion of people ages 15 and above who are infected with HIV.

FIGURE 1.13 Improving access to antiretroviral with respect to economic growth depends on
treatment is still far from universal how the incomes of poor people move with
average growth; whether growth is in broad
10 45
9
terms reaching poor people; and how any
40
8 increase in incomes interacts with the income
35
7 distribution and the poverty line. How-
30
6 ever, when a common poverty goal like the
millions

percent

25
5 MDG 1 is applied to all developing countries
20
4 against a uniform poverty yardstick (such as
15
3 $1.25 a day), starting points or country cir-
2 10
cumstances matter greatly. Starting points
1 5
thus help explain why, despite the precrisis
0 0
people people ART progress in growth and poverty reduction,
receiving needing coverage low-income countries are still far from reach-
ART ART (right axis) ing the poverty MDG.
(left axis) (left axis)
Recent empirical analysis reveals that a
Dec. 2007 Dec. 2008 high initial poverty incidence slows progress
against poverty at any given growth rate.9
Source: UNAIDS/WHO.
Note: ART - antiretroviral treatment. Analysis of more than 600 household sur-
veys in 116 countries since 1980 confirms the
all MDGs except poverty reduction and gen- story: in two countries with the same eco-
der parity in school (figure 1.16). This result nomic growth rate, poverty reduction will
likely comes from low starting points in poor be slower in the country that begins with the
countries. higher poverty rate (figure 1.17). There are
some data and observation issues to consider.
At low poverty rates, there is considerable
How starting points affect results noise, and elasticity becomes volatile, espe-
Progress varies positively with income growth cially at the $1.25-a-day poverty threshold.
but is conditioned by starting points or coun- Some more advanced countries have no com-
try circumstances. This phenomenon is best putable observations in this range. Twelve
illustrated for poverty reduction. In general, countries with zero poverty rates in the initial
the elasticity, or responsiveness, of poverty period have zero or irregular elasticity; seven
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 23

FIGURE 1.14 Fragile states have made the least progress toward the MDGs

100
progress toward goal to date, %

80

60

40

20

0
MDG 1.a MDG 2 MDG 3 MDG 3 MDG 3 MDG 4 MDG 7.c MDG 7.c
extreme primary gender parity gender gender child mortality access to access to
poverty completion (primary and parity parity under five safe water sanitation
rate secondary) (primary) (secondary)

middle-income countries low-income countries fragile states

Source: Staff calculations based on World Development Indicators database.


Note: Most recent data as of 2009.

other countries with nonzero poverty rates line. One explanation is that the poverty gap,
have zero elasticity. Almost all these outliers which reflects the depth of poverty as well
belong to transitional economies in Eastern as its incidence, is greater in poor countries.
Europe and include the turbulent years in A related “distance” measure is the percent
the 1990s when their economies, incomes, ratio of the average income of the poor to the
and income distributions were undergoing poverty line; the distance between the aver-
fundamental changes. Even at a $2.00-a-day age income of the poor and the poverty line
poverty line, seven countries have zero elas- is greater in many low-income countries. In
ticity. When the initial poverty rate is greater 2005 Sub-Saharan Africa’s poverty gap and
than 10 percent, the pattern becomes broadly its mean ratio of the average income of the
more regular.10 poor to the $1.25-a-day poverty line are 20.1
At the aggregate or regional level, the percent and 59.7 percent, respectively. These
responsiveness (median elasticity) of poverty numbers are worse than comparable figures
to growth is generally lower in low-income for other regions (table 1.3). Middle-income
countries, which tend to have higher initial countries tend to have better numbers. The
poverty rates for both the $1.25-a-day and the two large countries, China and India, are
$2.00-a-day international thresholds (table exceptions in the sense that, despite their
1.1). Elasticity also depends on the choice starting points in income and poverty rates,
of a global poverty threshold, although the poverty reduction has been rapid not just
pattern holds in general. For middle-income because of the high growth rates but also
countries, particularly upper-middle-income because the poverty gap and the average
ones, the lower $1.25 threshold becomes less income distance from the poverty line have
meaningful as the median poverty rate falls been relatively low. For all these countries,
below 4 percent. The regional numbers are many of the poor are already close to the
summarized in table 1.2. $1.25-a-day threshold and growth can more
The pattern clearly reflects the more dif- easily raise their incomes over the poverty
ficult circumstances in poor countries and line. In contrast, in African countries like
fragile states. For example, in fragile states Liberia, Mozambique, Rwanda, and Tanza-
the median elasticity of extreme poverty to nia, which have very high poverty gaps and
growth is lowest at the $1.25 a day poverty greater distance to the thresholds, the same
24 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 1.15 Progress in Sub-Saharan Africa is significant but still insufficient—partly because of low
starting points

a. Extreme poverty b. Gender parity


70 110
58 100
60 100
51
50
% of population

90
88
40 37 82

percent
80
30
29 70
20

10 60

0 50
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015
actual $1.25/day path to goal actual goal
projected $1.25/day

c. Child mortality under five d. Access to water and sanitation


200 185
90
74
69
144
150
deaths per 1,000

% of population

60
51

100
42 37

30 26
50 62

0 0
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015
actual goal without water access actual
without sanitation access actual
without water access goal
without sanitation access goal

e. Primary completion rate


110
100
100

90

80
percent

70 67

60 57
57

50
47
40
1990 1995 2000 2005 2010 2015

actual girls goal girls


actual boys goal boys
Source: World Development Indicators.
Note: PPP is purchasing power parity.
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 25

FIGURE 1.16 Many countries are falling short of most MDGs, 2009

100
percentage of developing countries

50

0
MDG 1.a MDG 2 MDG 3 MDG 3 MDG 4 MDG 5 MDG 7.c MDG 7.c
extreme primary gender gender child births access to access to
poverty education parity parity mortality attended safe water sanitation
(primary) (secondary)

seriously off track off track on track achieved no data

Source: Staff calculations based on World Development Indicators database.


Note: The data cover 144 developing countries and the latest available information as of 2009.

FIGURE 1.17 Poverty responds less to growth when the initial poverty rate is high

1 1
0 0
elasticity to growth ($1.25/day)

elasticity to growth ($1.25/day)

–1 –1
–2
–2
–3
–3 –4
–4 –5
–5 –6
–7
–6
–8
–7 –9
–8 –10
–9 –11
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
headcount, % ($1.25/day) headcount, % ($2/day)

Source: Staff calculations from PovcalNet, the World Bank.


Note: There are data or observation issues especially when the poverty rate is low; see text for details.

level of growth would be unlikely to push than in a richer country of comparable popu-
many households above the poverty line in a lation size, the number of people to move out
short period of time. of poverty will be much larger in the poor
Another explanation is simply arithmetic, country.
having to do with the way the poverty MDG All this simply means that a long period
is defi ned.11 It is easier for a middle-income of sustained and shared growth will be cru-
country to halve its poverty rate from, say, 10 cial to meet the poverty MDG in many poor
percent to 5 percent than for a poor country countries. To really achieve rapid progress,
to cut its rate by even less than half, say, from growth may have to reach the rates in China
50 percent to 35 percent. Although the rate and India. Although a pro-poor growth rate
of reduction is lower in the poorer country will raise the income of the poor, the high
26 MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

TABLE 1.1 Poverty reduction is more difficult in poor countries


Median poverty headcount at initial
Median growth elasticity of poverty year with respect to poverty line
with respect to poverty line (%)

Country group $1.25 a day $2.00 a day $1.25 a day $2.00 a day
Low-income countries –1.01 –0.53 66.02 85.77
Fragile states –0.81 –0.54 51.46 74.99
Lower-middle-income countries –1.65 –0.88 15.11 29.54
Upper-middle-income countries –1.04 –1.41 3.60 11.32
Source: World Bank staff calculations from PovcalNet.
Note: The values in the table are the medians of {ln(H2/H1/ln(M2/M1} in each category.

TABLE 1.2 Poverty reduction is several times more difficult in Sub-Saharan Africa
Median poverty headcount at initial
Median growth elasticity of poverty year with respect to poverty line
with respect to poverty line (%)

Region $1.25 a day $2.00 a day $1.25 a day $2.00 a day


East Asia and Pacific –1.43 –0.79 48.6 77.9
Europe and Central Asia –2.00 –1.12 3.7 10.2
Latin America and the Caribbean –2.03 –1.44 8.8 20.9
Middle East and North Africa –2.89 –2.06 4.6 19.6
South Asia –1.05 –0.48 66.5 88.1
Sub-Saharan Africa –0.76 –0.36 66.0 83.9
Total –1.18 –0.81 18.4 39.4
Source: World Bank staff calculations from PovcalNet.
Note: The values in the table are the medians of {ln(H2/H1/ln(M2/M1} in each category.

TABLE 1.3 Poverty gaps and ratio of mean income of the poor to poverty gaps imply that it will take longer or
the $1.25-a-day poverty line are worse for low-income regions or more effort for the poor to cross the poverty
countries, 2005
line. In light of the fi nding that the poverty
Ratio of mean income rate itself may take more time to evolve, what
Poverty gap with of those below $1.25 a
respect to $1.25 a day day to the poverty line low-income countries achieved before the cri-
(%) (%) sis is indeed remarkable.
Sub-Saharan Africa 20.1 59.7
South Asia 9.2 75.9
East Asia and Pacific 5.6 71.3
Latin America and the The full impact of the global
Caribbean 4.9 65.1 economic crisis still lies ahead
Middle East and North Africa 1.4 84.1
Europe and Central Asia 1.1 67.8 It is likely that the world has not yet seen the
full impact of the crisis on the MDG indica-
Middle-income countries
tors. A slow global recovery could imply that
Albania 0.1 91.7
Brazil 2.8 65.6 progress on MDG indicators will stray fur-
Mexico 0.1 89.6 ther from the path it was on before the cri-
Thailand 0.01 96.4 sis. Will economic growth and development
Large countries
China: urban 0.2 90.2
deteriorate because of a less hospitable global
China: rural 6.5 75.2 economic environment brought about by the
India 10.0 76.1 global economic crisis? Chapters 2, 3, and 4
African countries look at lessons from past crises, the current
Tanzania 48.6 45.1
Mozambique 42.0 48.1 economic situation, the prospects for growth,
Liberia 40.7 50.7 and the outlook for the MDGs. Chapter 5
Rwanda 38.2 49.9 focuses on the actions and policies for attain-
Source: PovcalNet, World Bank. ing the MDGs—and beyond.
GLOBAL MONITORING REPORT 2010 MILLENNIUM DEVELOPMENT GOALS 27

Notes Expectations of Aid and Development Suc-


cess.” World Development 35 (5): 735–51.
1. See, for example, Sachs 2005. Easterly, W. 2009. “How the Millennium Devel-
2. A point raised by Clemens and Moss (2005); opment Goals Are Unfair to Africa.” World
Clemens, Kenny, and Moss (2007); Easterly Development 37 (1): 26–35.
(2009); and Vandemoortele (2009). Go, D., and J. Page, eds. 2008. Africa at a Turn-
3. This observation is documented in several ing Point? Washington, DC: World Bank.
recent studies. See, for example, World Bank Hogan, M., K. Foreman, M. Naghavi, S. Ahn, M.
(2008) on the decoupling of trend growth Wang, S. Makela, A. Lopez, R. Luzana, and
for developed and developing countries. For C. Murray. 2010. “Maternal Mortality for 181
Africa, see IMF (2008), Go and Page (2008), Countries, 1980–2008: A Sytematic Analysis
Ndulu (2008), and World Bank (2000). of Progress towards Millennium Development
4. Rodrik 2009. Goal 5.” Lancet. www.lancet.com, April 12,
5. Hogan and others 2010. Weak vital registra- 2010.
tion systems in developing countries make IMF (International Monetary Fund). 2008.
the maternal mortality ratio one of the hard- Regional Economic Outlook: Sub-Saharan
est things to measure, and previous estimates Africa. Washington, DC (October).
showed very little change over time. The new Ndulu, B. J. 2008. The Political Economy of Eco-
study supplemented national vital registra- nomic Growth in Africa, 1960–2000. Cam-
tion data with other sources, such as sibling bridge, U.K.: Cambridge University Press.
histories in household surveys, census data, Ravallion, M. 2009. “Why Don’t We See Poverty
and death surveys. It also carefully examined Convergence?” Policy Research Working Paper
uncertainty in the expected maternal death 4974. World Bank, Washington, DC.
rate from five sources—stochastic variance Rodrik, D. 2009. “Growth after the Crisis.” Paper
in the input data, nonsampling error in data prepared for the Commission on Growth and
systems, errors in the covariates (such as GDP Development, Harvard Kennedy School, Cam-
per capita, educational attainment, HIV sero- bridge, MA.
prevalence, and the like), and estimation error Sachs, J. 2005. The End of Poverty. New York:
from using simulation methods. Penguin Books.
6. Environmental sustainability and its links UNICEF 2007. Progress for Children, A World
to the MDGs were a major focus of Global Fit for Children, Statistical Review (6, Decem-
Monitoring Report 2008 (World Bank and ber). New York.
IMF 2008). The 2010 World Development United Nations. 2008. “Delivering on the Global
Report also focused on development and cli- Partnership for Achieving the Millennium
mate change (World Bank 2010). Development Goals.” MDG Gap Task Force
7. See World Bank (2010) for an in-depth discus- Report. New York.
sion and treatment of this issue. ———. 2009. The Millennium Development
8. United Nations 2008. Goals Report. New York.
9. Ravallion 2009. Vandemoortele, Jan. 2009. “The MDG Conun-
10. T he $1. 25 -a- day relat ionsh ip c a n be drum: Meeting the Targets without Missing
approximated by a logarithmic regression: the Point.” Development Policy Review 27 (4):
log(elasticity) = –1.0634 + 0.4725 (initial pov- 355–71.
erty rate), n = 101, which is significant at 1 per- World Bank. 2000. Can Africa Claim the 21st
cent (t = 5.13). See Ravallion (2009) for care- Century? Washington, DC: World Bank.
ful estimation and testing of convergence. ———. 2008. Global Development Finance
11. Easterly 2009. 2008: The Role of International Banking.
Vol. 1. Review, Analysis, and Outlook. Vol. 2.
Summary and Country Tables. Washington,
DC: World Bank.
References ———. 2010. World Development Report 2010:
Clemens, M. A., and T. J. Moss. 2005. “What’s Development and Climate Change. Washing-
Wrong with the Millennium Development ton, DC: World Bank.
Goals?” Brief. Center for Global Development, World Bank, and IMF. 2008. Global Monitoring
Washington, DC. Report 2008: MDGs and the Environment:
Clemens, M. A., C. J. Kenny, and T. J. Moss. 2007. Agenda for Inclusive and Sustainable Devel-
“The Trouble with the MDGs: Confronting opment. Washington, DC: World Bank.
2
Lessons from Past Crises—and
How the Current Crisis Differs

H
istorically, periods of sharp con- How growth volatility affects
traction have been extremely harm- human development and gender
ful for human development. Social indicators
indicators tend to deteriorate rapidly during
economic downturns and improve slowly dur- It is commonly observed that human devel-
ing economic booms. Moreover, vulnerable opment indicators deteriorate during growth
groups, such as children and women, are more downturns. Also true, but more difficult to
exposed to the effects of growth volatility. calculate, is that deteriorations in human
The asymmetric response of social indica- development indicators during downturns
tors to the growth cycle likely results from tend to exceed improvements during eco-
contractions associated with conflict or weak nomic booms (box 2.1 explains the definition
institutions that impair government services. of growth cycles used here). For example, life
With global crises, donor spending may also expectancy is 2 years longer during growth
come under pressure. accelerations than the overall average, but 6.5
There are several reasons, however, why years shorter during decelerations (figure 2.1).
this crisis may be different from previous cri- Infant mortality is 8 per 1,000 lower during
ses for low-income countries—social spending accelerations, and 24 per 1,000 higher during
has been largely protected so far; precrisis pol- decelerations. The primary school completion
icies and institutions were better; and external rate is 4 percent higher during accelerations
shocks, not domestic policy failures, were the but 25 percent lower during decelerations.
main cause of the current crisis for developing Further evidence for asymmetry is the size of
countries. correlation coefficients relating social indica-
Nonetheless, the impact on the Millennium tors with upturns and downturns (table 2.1).
Development Goals (MDGs) is worrisome. In general, the correlation between social indi-
In particular, several rapid and qualitative cators and periods of deceleration is stronger
assessments find that households are already than the correlation between social indicators
making painful adjustments, particularly in and periods of acceleration (for details, see
middle-income countries. annex 2A.1).

GLOBAL MONITORING REPORT 2010 29


30 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

BOX 2.1 Defining growth cycles in developing countries

The historical growth patterns considered in this that the growth acceleration episode is not a recov-
study are derived from a dataset for 163 countries ery from a recession.
covering 1980–2008. A growth acceleration episode
meets three conditions for at least three consecutive A growth deceleration episode meets these three con-
years: ditions in reverse. The framework is from Arbache
and Page (2007), which extends the methodology in
• The four-year forward-moving average growth Hausmann, Pritchett, and Rodrik (2005) by examin-
rate minus the four-year backward-moving aver- ing both accelerations and decelerations and by mak-
age growth rate exceeds zero for each year. ing each country’s long-run growth rate endogenous.
• The four-year forward-moving average growth rate Testing the sample means of development indicators
exceeds the country’s average growth rate, mean- for significant differences during periods of growth
ing that the pace of growth during accelerations is acceleration and deceleration can show whether
faster than the country’s trend. Thus the defi nition countries that experience more growth fluctuations
of episodes of growth acceleration (or deceleration) face slower progress on the MDGs and identify
is endogenous to each country’s long-run rate of how growth cycles affect changes in development
growth. indicators.
• Average GDP per capita during the four-year for-
ward-moving period exceeds the average during Source: Arbache and Page 2007; Arbache, Go, and Korman
the four-year backward-moving period, ensuring 2010.

Economic downturns also have a dispro- participation of women and men, with impor-
portionate impact on girls relative to boys. tant implications for how families adjust to
Life expectancy of girls and boys increases by economic crises (box 2.2).
two years during good times but decreases by Despite some commonalities, the relation-
about seven years for girls and six years for ship between growth volatility and develop-
boys during bad times. The primary educa- ment outcomes varies across countries and
tion completion rate rises 5 percent for girls regions. Initial conditions, regional spillovers,
and 3 percent for boys during good times but trade arrangements, economic geography, and
decreases 29 percent for girls and 22 percent other factors are associated with how countries
for boys during bad times. The female-to-male and regions respond to economic downturns.
enrollment ratios for primary, secondary, and For example, human development indicators
tertiary education rise about 2 percent dur- in Sub-Saharan Africa are among the lowest in
ing growth accelerations but fall 7 percent the world: infant and under-five mortality rates
(primary), 15 percent (secondary), and 40 are almost three times higher than the global
percent (tertiary) during decelerations. These average, life expectancy is 29 percent lower,
differences may result from household time primary school completion is 66 percent lower,
and resource allocations that favor boys over and the ratio of female to male tertiary enroll-
girls when household budgets shrink.1 The dif- ment is about half the global mean. But the
ferential impact on child schooling and child difference between the average level of social
survival is greatest in low-income countries, indicators in good and bad times is smaller for
while gender differences are smaller in mid- Sub-Saharan Africa than it is for developing
dle-income countries. Economic downturns countries as a whole (compare figures 2.1 and
also have different effects on the labor force 2.2). This finding may imply that at low levels
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 31

FIGURE 2.1 Key human development and gender indicators plummet from their overall mean during
growth decelerations, all countries

a. Life expectancy at birth b. Mortality rate


8
50
6 40
4 30

deaths, thousands
years gained/lost

20
2
10
0 0
–2 –10
–4 –20
–30
–6
–40
–8 –50
women men total infant mortality child mortality under five
growth acceleration growth deceleration (per 1,000 live births) (per 1,000)
growth acceleration growth deceleration

c. Primary completion rate d. Gender equality, ratio of girls to boys


30 45
35
percentage point change

percentage point change

20
25
10 15
5
0
–5
–10 –15
–25
–20
–35
–30 –45
girls boys total primary secondary tertiary
enrollment enrollment enrollment
growth acceleration growth deceleration
growth acceleration growth deceleration

Source: World Bank staff calculations based on data from World Development Indicators. See annex table A2.1 for levels and Arbache, Go, and Korman
(2010) for more details.
Note: Differences of sample averages during growth accelerations and decelerations from overall sample means.

TABLE 2.1 Correlation coefficients between growth acceleration and deceleration and human
development indicators
  Growth acceleration Growth deceleration

Indicator Coefficient Significance level Coefficient Significance level

Life expectancy at birth, women (years) 0.13 ** –0.22 **


Life expectancy at birth, men (years) 0.12 ** –0.25 **
Life expectancy at birth, total (years) 0.13 ** –0.23 **
Infant mortality rate (per 1,000 live births) –0.17 ** 0.21 **
Child mortality under five rate (per 1,000) –0.15 ** 0.22 **
Primary completion rate, girls (% of relevant age group) 0.16 ** –0.26 **
Primary completion rate, boys (% of relevant age group) 0.13 ** –0.23 **
Primary completion rate, total (% of relevant age group) 0.16 ** –0.26 **
Ratio of girls to boys, primary enrollment 0.17 ** –0.22 **
Ratio of girls to boys, secondary enrollment 0.1 ** –0.19 **
Ratio of women to men, tertiary enrollment 0.06 –0.18 **
Source: World Bank staff calculations based on data from World Development Indicators.
Note: Tests for differences in the means of these variables among growth accelerations, decelerations, and all country-year observations show
that they are almost all statistically significant at the 1 percent level (**).
32 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

BOX 2.2 Aggregate economic shocks and gender differences: A review of the evidence

The labor market for nonagricultural wage work by analyzing the impact of crises (see the figure below),
women (often used as a proxy for women’s access to it is important to take into account the response of
decent work) tends to behave very differently from women who are not in the labor force. In some crises,
the market for nonagricultural wage work by men. for example, in Indonesia in 1997, women entered
While the unemployment rate rarely differs between the labor force to maintain household consumption
men and women, a much smaller proportion of work- (called the added-worker effect). In the Republic of
ing-age women are in the labor force (whereas most Korea during the 1997 crisis, some women left the
men of working age are either working or unem- labor force (the discouraged-worker effect).
ployed, women may be working at home). Thus in

Possible transmission channels of economic crisis

First-round impacts Second-round impacts

IMPACT 1
Loss of employment
Drop in aggregate  for women in export-
demand/exports oriented industries IMPACT 3
Vulnerable households’


IMPACT 2 Drop in household coping strategies can


Fall in microfinance  income, increased  push women into work,
Tightened institutions’ (MFI) lending affecting human capital
credit markets
 risk of poverty
resources affects women of women and


producers (MFI borrowers children


are typically women)

Drop in
remittances

Food price
shocks

of income, the ability to improve social indica- tractions and the lack of social safety nets; and
tors is particularly limited—and therefore so is declines in aid during crises that also affect
the likely deterioration. high-income countries.

Explaining the pattern of Poor countries suffer from frequent


past crises economic contractions and high growth
volatility
Several factors contributed to negative human
development outcomes during past economic One reason for the low levels of human devel-
downturns, including the high frequency of opment in low-income countries is that they
downturns in low-income countries; the poor experience numerous crises. Of all country-
policy environment in many countries during year observations for low-income, Interna-
past crises, particularly in low-income coun- tional Development Association (IDA)-eligible,
tries; shrinking social spending during con- and Sub-Saharan African countries, nearly
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 33

BOX 2.2 (continued)

The table below summarizes country studies of the impact of crises on women’s labor market participation and
health and education outcomes.

Previous crises: Available evidence by country


Income level Labor market
at time of crisis Country effects for women Schooling Health
Low-income Côte d’Ivoire Decline in student enrollment Deteriorating child health
country for both boys and girls. for both boys and girls.
Pooled survey data Girls’ infant mortality more
from several low- sensitive than boys’ infant
income countries mortality to fluctuations
in GDP.

Lower-middle- Indonesia Added-worker effect in Decline in student enrollment Higher neonatal mortality,
income 1997–98. for young boys and girls and but overall not much
country older girls. effect on child health.
Lower body mass index
for adults.
Peru Added-worker effect in Increase in student Higher infant mortality
Lima. enrollment for both boys rate.
and girls.
Philippines Added-worker effect. Drop in high school
enrollment for both boys
and girls.
Drop in elementary school
enrollment, more for girls
than boys.
Increase in child labor, more
for boys than girls.

Upper-middle- Argentina Added-worker effect in


income urban areas during 1990s.
country
Brazil Both added- and
discouraged-worker effect
in São Paulo during 1980s.
Costa Rica Decline in student enrollment
for both boys and girls in
rural areas; higher for girls
than boys in urban areas.
Mexico Added-worker effect Increase in student Higher child mortality
during 1980s. enrollment in 1995–96, during crises.
stronger for boys than girls.
Russian Federation Deterioration in weight
for height for both boys
and girls.

High-income Korea, Rep. Discouraged-worker effect


country in Seoul in 1997–98.
United States No effect. Increase in student Improved child health
enrollment during Great outcomes.
Depression.
Source: Sabarwal, Sinha, and Buvinic 2009.
34 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.2 Key human development and gender indicators also fall below their overall means during
growth decelerations in Sub-Saharan countries, if less so

a. Life expectancy at birth Sub-Saharan Africa b. Mortality rate, Sub-Saharan Africa


2.5 20
2.0 25
1.5
10

deaths, thousands
1.0
0.5 5
years

0 0
–0.5 –5
–1.0
–10
–1.5
–2.0 –15

–2.5 –20
women men total infant mortality child mortality under five
(per 1,000 live births) (per 1,000)
growth acceleration growth deceleration
growth acceleration growth deceleration

d. Gender equality, ratio of girls to boys,


c. Primary completion rate, Sub-Saharan Africa Sub-Saharan Africa
15 30

10 20
percentage point change
percentage point change

5 10

0 0

–5 –10

–10 –20

–15 –30
girls boys total primary secondary tertiary
enrollment enrollment enrollment
growth acceleration growth deceleration
growth acceleration growth deceleration

Source: World Bank staff calculations based on data from World Development Indicators. See annex table A2.1.2 for levels and Arbache, Go, and Korman
(2010) for more details .
Note: Differences of sample averages during growth accelerations and decelerations from the overall sample means.

a quarter are decelerations; for the middle- Differences in the frequency of economic
income countries in East Asia, Europe and contraction explain a significant share of the
Central Asia, and Latin America, less than 10 differences in the average growth rate of dif-
percent are decelerations (table 2.2). Similarly, ferent groups of countries. Growth in GDP per
the share of accelerations is 37–39 percent capita during 1980–2008 was 0.6 percent a
for poorer countries but 43–53 percent for year in low-income countries and more than 2
middle-income countries.2 In addition, over- percent a year in middle-income countries. The
all growth volatility is greater in low-income slower growth in low-income countries stems
countries and in Sub-Saharan Africa than in from the greater frequency of decelerations,3
middle-income countries (see table 2.2). The not from a marked difference in growth rates
regional pattern suggests growth spillovers during booms and busts. For example, during
at the geographic level, which may be associ- periods of acceleration, low-income countries’
ated with economic geography, regional trade per capita GDP rose 3.75 percent, slightly less
arrangements, natural disasters, regional than the 4.5 percent growth rate for middle-
migration, and regional conflicts. income countries. During decelerations, low-
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 35

TABLE 2.2 Frequency of growth acceleration and deceleration, growth rates, and GDP per capita, 1980–2008
  GDP Growth acceleration Growth deceleration

GDP per capita GDP per capita GDP per capita


growth rate Standard deviation Frequency growth rate Frequency growth rate
Region, income (%) of growth (country years) (%) (country years) (%)

World 1.89 6.03 0.47 4.27 0.11 –3.81


Region
East Asia and Pacific 3.09 4.45 0.46 5.01 0.09 –2.75
Europe and Central Asia 2.20 6.65 0.53 4.79 0.08 –7.19
Latin America and the Caribbean 1.63 4.65 0.53 3.72 0.07 –2.78
Middle East and North Africa 1.41 5.51 0.43 2.89 0.06 –3.44
South Asia 3.72 2.87 0.36 4.69 — —
Sub-Saharan Africa 1.02 7.28 0.39 4.19 0.22 –3.17

Country Income category            


Developing countries 1.67 6.37 0.46 4.33 0.14 –3.87
IDA countries 0.99 6.28 0.39 3.82 0.21 –3.47
Low-income countries 0.63 6.74 0.37 3.75 0.23 –3.50
Lower-middle-income countries 1.98 5.89 0.47 4.52 0.13 –4.99
Upper-middle-income countries 2.34 6.43 0.55 4.54 0.08 –2.76
High-income, non-OECD countries 3.02 7.41 0.42 5.90 0.02 –4.62
High-income OECD countries 2.19 2.59 0.54 3.31 0.03 –2.32
Source: World Bank staff calculations based on data from World Development Indicators.
Note: — = not available. OECD = Organisation for Economic Co-operation and Development.

income countries’ per capita GDP fell 3.5 all times (figure 2.3). Causality likely moves
percent, somewhat less than in middle-income in both directions: a deterioration in institu-
countries. Thus “defensive” policies that pre- tions impairs growth, which leads to further
vent collapses should have substantial impacts institutional weaknesses, and so on. In many
on average growth by avoiding multiple col- cases both institutions and growth are affected
lapses and their negative outcomes.4 The find- by domestic violence or foreign wars. In Sub-
ing that the elasticity of poverty to growth is Saharan Africa, for example, the frequency of
lower in high-poverty countries (see chapter 1) major and minor conflicts is about 23 percent
suggests that low-income countries, where pov- during growth deceleration and 13 percent dur-
erty rates are high, need a long period of sus- ing growth acceleration. In oil- and mineral-
tained growth to reduce poverty and improve dependent economies, defects in institutional
other human development indicators. quality may be masked by the revenues gen-
erated by favorable commodity prices. These
defects become clear when prices turn down
Contractions tend to occur in severely
and revenues dry up. 5
unfavorable economic and policy
Macroeconomic variables such as invest-
environments
ment, savings, exports, imports, external
Contractions have a grave impact on human finance, and inflation deteriorate more during
development because they are marked by an downturns than they improve during upturns
overall deterioration in government effective- (see figure 2.3). During decelerations, both sav-
ness. Similar to human development indicators, ings and investment, particularly fixed private
indicators of institutional quality (political investment, decline relative to average levels
stability, voice and accountability, regulatory (as a share of GDP) by much more than they
framework, rule of law, and government effec- rise during accelerations. The increase in for-
tiveness) in developing countries perform eign direct investment as a share of GDP during
asymmetrically over the growth cycle. In other accelerations is twice as large as the drop dur-
words, the deterioration during bad times is ing decelerations (relative to the sample mean).
much greater than the improvement during The very high values for inflation during growth
good times relative to the sample averages for decelerations reflect the incidence of hyperinfla-
36 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.3 During growth decelerations, economic and institutional indicators diverge far from the
overall means

a. GDP growth and inflation b. Savings and investment


6 250 6
200
percentage point change

percentage point change


4 150 4

2 100 2
50
0 0 0
–50
–2 –100 –2

–4 –150 –4
–200
–6 –250 –6
GDP per capita CPI inflation (%) Gross Gross fixed Gross
growth rate (right axis) investment investment domestic
(left axis) (% GDP) private sector savings
(% GDP) (% GDP)
growth acceleration growth deceleration
growth acceleration growth deceleration

c. Trade d. External capital flows


10 25
percentage point change
percentage point change

15
5
05
0
–05

–5
–15

–10 –25
exports (% GDP) imports (% GDP) foreign direct private capital flows,
investment, total (% GDP)
growth acceleration growth deceleration net inflows (% GDP)
growth acceleration growth deceleration

f. Frequency of conflicts and official development,


e. Institutions (–2.5 to 2.5) assistance, Sub-Saharan Africa
0 0.08 1.5
0.06
percentage point change

percentage point change

percentage point change

–0.2 1.0
0.04
–0.4 0.02 0.5

–0.6 0 0
–0.02
–0.8 0.5
–0.04
–1 1.0
–0.06
–1.2 –0.08 –1.5
frequency ODA (% GDP)
or y

law

ne nt
bi cal

ta d

ew or
Re lity
un an

ive e
ss
k
m lat

of conflicts (right axis)


sta liti

ct m
y

of
bi
lit
co ce

fe rn
fra u
Po

le
g
ac Voi

ef ve

(left axis)
Ru

Go

growth acceleration growth deceleration


growth acceleration growth deceleration

Source: World Bank staff calculations based on data from World Development Indicators. See annex table A2.3 for levels and Arbache, Go, and Korman
(2010) for more details.
Note: Differences of sample averages during growth accelerations and decelerations from the overall sample means.
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 37

TABLE 2.3 Correlation coefficients between economic cycles and economic and institutional indicators
  Growth acceleration Growth deceleration

Indicator  Coefficient Significance level Coefficient Significance level

Final consumption (% GDP) –0.1 ** 0.13 **


Government consumption (% GDP) –0.07 ** 0.04 **
Gross capital formation (% GDP) 0.09 ** –0.17 **
Gross domestic savings (% GDP) 0.1 ** –0.13 **
Gross private fixed capital formation (% GDP) 0.19 ** –0.19 **
Imports (% GDP) 0.06 ** –0.09 **
Exports (% GDP) 0.09 ** –0.11 **
Trade (% GDP) 0.08 ** –0.11 **
Net foreign direct investment (% GDP) 0.03 –0.04 *
Private capital flows, total (% GDP) 0.04 * –0.04 *
Inflation (%) –0.06 ** 0.13 **
Institutions
Political stability –0.07 ** –0.12 **
Voice and accountability –0.07 ** –0.1 **
Regulatory framework –0.06 * –0.19 **
Rule of law –0.1 ** –0.18 **
Government effectiveness –0.05 * –0.21 **
Source: World Bank staff calculations. Indicators on institutions are from the World Bank Institute’s Worldwide Governance Indicators database,
which relies on 33 sources, including surveys of enterprises and citizens, and expert polls, gathered from 30 organizations around the world.
* Significant at the 5 percent level; ** significant at the 1 percent level.

tion during several growth collapses in Africa and private domestic spending on health and
before the 1990s and incidents of high inflation public education for over 108 developing
in the early 1990s (such as in Angola, Armenia, countries for 1995–2007.7
Azerbaijan, Belarus, Brazil, Democratic Repub- The analysis points to three key results.
lic of Congo, Peru, and Ukraine).6 First, social spending growth rates tended to
Aid to poor countries is procyclical, suggest- be volatile. Second, per capita social spending
ing that donors respond to an emerging policy levels nonetheless showed a steady upward
failure by giving less aid. Macroeconomic and trend. Third, social spending in poor countries
institutional variables are more closely corre- was subject to more pressures than in richer
lated with the incidence of deceleration than countries during contractions in GDP. The
of acceleration (table 2.3)—further evidence last result confirms that it is the low-income
of the asymmetric behavior of these indicators countries that are more likely to need help in
over the economic cycle and of the extremely protecting social expenditures during crises.
poor economic environment characterizing The volatility of public and private health
downturns in developing countries. spending is evident from the unweighted aver-
age of growth rates of social spending by coun-
try over time, a calculation that gives equal
Social spending under pressure
weight to changes for each country and does
Drops in spending on social services like edu- not allow the larger countries to dominate the
cation and health care are an important reason pattern. In particular, changes in public health
for the sharp deterioration in human develop- spending are more volatile than GDP growth
ment indicators during crises. Cutbacks in trends over time (figure 2.4). Historically, a
social spending are more worrisome during drop in GDP growth of 2 percent or more has
crises because that is when people need these had a greater than proportional effect on the
services most. Cutbacks during crises are also growth of public health spending. Growth
harmful because such disruptions have long- of private health spending (insurance and
lasting effects. This section draws on evidence out-of-pocket payments) responds in a simi-
of the impacts of GDP downturns on public lar way, although the pattern is more vola-
38 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.4 Health spending growth rate is more volatile than its per capita level or GDP growth

a. Government health expenditure b. Private health expenditure and GDP growth


and GDP growth in developing countries
14 200 9 140
12 180 8 120
160 7
10 140 100
6

% change
120
% change

8 5 80

PPP $

PPP $
100
6 4 60
80
3
4 60 40
40 2
2 1 20
20
0 0 0 0
1996 1998 2000 2002 2004 2006 1996 1998 2000 2002 2004 2006

Average of GDP growth, unweighted (left axis) Average of GDP growth, unweighted (left axis)
Average of government health expenditure per capita Average of private health expenditure per capita by country,
by country, unweighted (right axis) unweighted (right axis)
Average of growth rates of government health Average of growth rates of private expenditure by
expenditure by country, unweighted (left axis) country, unweighted (left axis)

Source: Lewis and Verhoeven 2010.

FIGURE 2.5 Public education spending is less closely tied to GDP ing level (which is also an unweighted average
growth than is health spending across countries) suggests that the volatility
of spending growth rates was affecting coun-
25 250 tries with lower levels of per capita spending
more than countries with higher spending lev-
20 200
els. Indeed, the data confirm that the negative
15 150
impacts of crises on health spending are much
% change

stronger in the lowest-income countries, where


PPP $

10 100 growth in health spending is more likely to fall


in response to a decline in GDP.
5 50 Education spending in developing countries
appears to be less closely tied to GDP growth
0 0
1996 1998 2000 2002 2004 2006 than is health spending (figure 2.5). In abso-
lute terms, the public sector spends more, on a
Average of GDP growth, unweighted (left axis)
per capita basis, on education than on health.
Average of government education expenditure per capita by country,
unweighted (right axis) There is also a modest rising trend. The data on
Average of growth rates of government education expenditure by country, education spending are weak, however, so any
unweighted (left axis)
conclusions concerning cross-country trends
Source: Lewis and Verhoeven 2010.
are subject to considerable uncertainty.8

tile than for government spending, especially


Donor funding under pressure
before 2005. Does aid to developing countries rise during
Despite the fluctuations in growth rates of crises? The sharp deterioration in human devel-
GDP and health spending, the trends in abso- opment indicators and the decline in social
lute per capita health spending continue to rise spending during growth decelerations highlight
over time. In general, private health spending a potentially important role for donors. Aid’s
rises more slowly than public health spending. contribution to welfare in developing coun-
The steady rise of the mean per capita spend- tries could be bolstered by increasing aid in
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 39

FIGURE 2.6 Aid to education and health does not appear to be closely related to GDP growth, 1998–2007

a. GDP and ODA in health trends b. GDP and ODA in education trends
500 7 1,200 7
GDP growth rate (right axis) GDP growth rate (right axis)
450
6 1,000 6
400
per capita ODA, US$

per capita ODA, US$


350 5 5
800

growth rate, %

growth rate, %
300 4 4
250 ODA health 600
spending (left axis) 3 3
200 ODA education
400
150 2 spending 2
(left axis)
100 200
1 1
50
0 0 0 0
1998 2000 2002 2004 2006 1998 2000 2002 2004 2006

Source: Lewis and Verhoeven 2010.

bad times to compensate for shortfalls in gov- funding could deepen the economic deteriora-
ernment resources. The evidence on whether tion in developing countries.
aid plays such a countercyclical role is mixed. Two deviations from these aggregate trends
Overall, aid to individual countries appears to are instructive—and encouraging. During the
be procyclical, increasing when growth rises Southeast Asian crisis of 1997, donors (most
and falling when growth slows. And growth notably the U.K. Department for Interna-
in aid to health and education sectors does not tional Development) supported core social
appear to be closely related to GDP growth in programs in Indonesia, slowing the declines
developing countries (figure 2.6). However, in education and health spending and permit-
after 2003, aid to education shows a small ting social services to continue.10 During the
response to growth, with donor financing ris- current crisis, Mexico sought loans from the
ing and falling as national education resources World Bank to compensate for budget reduc-
fall and rise. This countercyclical financing tions and to expand temporary safety nets.
suggests that donor spending is modestly com- Latvia, Lithuania, Poland, and Romania all
pensating for GDP growth shifts. received policy loans or technical assistance
Evidence on aid funding during financial from the World Bank to support reforms and
crises in donor countries is limited. A recent continued financing of safety nets and educa-
study that tracked donor allocations dur- tion and health programs. These aid responses
ing and after banking crises (1998–2007) in provided necessary finance for income support
developed countries suggests that donor fund- and social service programs—both critical for
ing is tied to economic prosperity in those bridging financial gaps during a downturn.
countries.9 Aid flows decline 20–25 percent
during banking crises in member countries of
Safety nets were uncommon in
the Organisation for Economic Co-operation
developing countries before previous
and Development (OECD) and take signifi-
crises
cant time to recover. Not all donor programs
are equally affected. But some combination of Few countries facing previous macroeconomic
the fiscal costs of crises, debt overhang after and financial crises in Asia (1997–99), Europe
the crisis ends, and perhaps erosion in public (Russian Federation 1998, Turkey 2001), and
support reduces aid flows from affected donor Latin America (1980s, 1994–95, 1999, 2001–
countries. To the extent that aid recipients and 02) had strong safety nets in place before the
donors are simultaneously affected by crises crisis. Countries had to scale up programs,
(as in the recent crisis), cutbacks in donor regardless of the fit between the original target
40 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

MAP 2.1 Around 9 million young children die before their fifth birthday

Child mortality under five rate:


Deaths per 1,000 (2008) Green
(De
≥100
35–99
20–34
10–19
<10 Canada
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: World Development Indicators.

population and the population affected by the in response to the 2002 crisis and extended it
crisis, or quickly start new ones. Mexico scaled to 2 million participants within a year. Inci-
up retraining and employment programs and dence and coverage were good, with about 80
targeted food distribution in response to the percent of the benefits concentrated among the
“tequila” crisis, even though these programs poorest 40 percent of the population. Argen-
probably did not target the most affected tina benefited from extensive experience with
populations. Safety net programs set up after an earlier, smaller workfare program. The
a crisis starts often suffer from poor initial Republic of Korea was able to quickly intro-
implementation, as with Indonesia’s Labor- duce a public works program in response to
Intensive Public Works (JPS Padat Karya) the Asian financial crisis, reaching more than
program, or take too long to scale up, as with 400,000 people within six months.12
Colombia’s Families in Action (Familias en Beyond emergency responses to natural
Acción) program.11 disasters and humanitarian crises, safety nets
Despite the difficulties, however, countries have been uncommon in low-income coun-
have managed to start effective safety net pro- tries, partly because they were viewed as tak-
grams in response to crises. Argentina estab- ing away from more productive expenditures.
lished a new workfare (Jefes de Hogar) program But support for social safety nets in the poorest
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 41

IBRD 37743
APRIL 2010

nland
en)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Portugal Spain Turkey Armenia baijan Turkmenistan of Korea
Monaco Greece Tajikistan Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire Cameroon
African Rep. Somalia Palau
Togo Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Kenya
São Tomé and Príncipe Gabon Congo Rwanda Singapore Nauru
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

countries has risen as their importance in pro- But if safety nets are not in place when shocks
tecting the poor and vulnerable during crises strike, governments might respond with price
has become evident. Low-income and fragile subsidies or other suboptimal policies, which
countries are devoting a larger share of lend- can leave an unwanted legacy of fiscal bur-
ing to public works programs and increasing den, economic distortions, lower growth, and
cash transfers and in-kind safety nets, with a greater poverty.
renewed focus on school feeding programs.
A key lesson in previous crises is the
importance of well-functioning safety nets in
What is happening in the current
responding to a crisis and promoting growth
crisis—and what is different?
and development afterward. When already There is some hope that human development
in place, safety nets can be scaled up to meet indicators have not deteriorated as much dur-
increased needs and then scaled back as the ing the current crisis as in previous crises.
crisis subsides. They can provide temporary Because the current economic crisis did not
protection for households by cushioning unem- reach most developing countries until 2009, it
ployment, contractions in public services, and is too early to arrive at a definitive conclusion
falling demand for formal and informal work. on its impact. However, rapid surveys and dis-
42 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.7 Despite intense fiscal pressures, Mexico’s federal funding for health and education is set to
rise in 2009–10

a. Health b. Education
400 12 600 17

300 11 16

Mex $, thousands
400
Mex $, thousands

percent
percent
200 10 15

200
100 9 14

0 8 0 13
2005 2006 2007 2008 2009 2010 2005 2006 2007 2008 2009 2010

total health spending (left axis) total education spending (left axis)
health spending as a % of federal education spending as a % of federal
spending (right axis) spending (right axis)

Source: Mexican government statistics.


Note: Data are estimated for 2005–07, approved for 2008 and 2009, and projected for 2010.

cussions with governments have yielded pre- employment and transfers to vulnerable pop-
liminary evidence showing that social spending ulations. Chile’s Social and Economic Stabili-
may have held up better during this crisis than zation Fund provided a countercyclical boost
in previous ones and that there has been more in spending, blunting the effect of the exter-
reliance on social safety nets. Moreover, policy nal shock.13 In Mexico, the severe contraction
regimes in developing countries had improved imposed intense fiscal pressure, but educat-
considerably before the crisis, so governments ion and health funding are nevertheless set to
might have had greater success in protecting rise 10 percent in 2009–10 (although spend-
their populations from the worst effects of the ing on education is expected to fall sharply
growth downturn. as a share of total government expenditures;
figure 2.7). El Salvador is not cutting educa-
tion funding despite the severe recession, but
Social spending held up in some regions
health spending is expected to fall from 3.4
Impacts on social sector budgets for 2008–10 percent to 3.0 percent of GDP, largely because
varied by country circumstances, specifically of reductions in the Social Security Institute’s
according to how the global downturn affected health expenditures.14
the economy and public revenues and whether
countries prepared for a possible contraction. Eastern Europe and Central Asia. Because
Eastern Europe was the hardest hit of emerg-
Latin America and the Caribbean. In Latin ing market regions, countries there were the
America social spending has remained strong, first to cut all areas of public spending. Nei-
partly because of the relatively modest size ther education nor health has necessarily been
and scope of the downturn in much of the spared. However, aggregate trends in social
region and partly because of efforts to protect spending alone are not an accurate indication
social spending. Some of the larger econo- of the impact of the crisis on outcomes. Some
mies (such as Brazil and Chile) instituted countries have directed spending reductions
social measures aimed at financing temporary to sectors with overcapacity, thus improving
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 43

BOX 2.3 Crises as opportunities for reform

Crises can present opportunities to achieve reforms and prioritized the fi nancing of effective health care
in social sector spending that will improve efficiency procedures by adjusting the list of ineligible health
and welfare over the long term. The current crisis services. The crisis made all these needed reforms
sharply reduced GDP in many countries in East- possible, and policy research informed strategic
ern Europe—by more than 15 percent in Latvia— investment decisions, largely avoiding across-the-
making it impossible to sustain social spending at board reductions or random cuts in social programs.
precrisis rates. Across-the-board cuts in education Romania responded to declining school enroll-
spending would have greatly impaired access to edu- ments and tighter budgets by substantially reducing
cation, with dismal implications for the quality of education personnel (teaching and nonteaching posi-
the workforce and long-term productivity. Instead, tions) in 2009 and by curtailing supplements to base
Latvia and Romania directed spending cuts at areas salaries. Some 18,000 teachers (6 percent) were laid
of overcapacity, through reforms that previously had off following adjustments in teaching norms and sub-
been blocked by political opposition. stantial cuts in the funds allocated to each county.
Latvia is using the stringencies imposed by the cri- The Ministry of Education, Research, and Innova-
sis to right-size its teaching force. By shifting teacher tion cut 15,000 additional public positions, consoli-
financing and management to local governments dated schools, and reduced the number of scholar-
and providing them with per capita student trans- ships for higher education. The staffi ng reductions
fers, the central government is tackling overcapac- will allow much needed adjustments to class size and
ity. This reform translates into an average 34 percent better alignment of teachers, students, and budgets.
reduction in the number of teachers and a 45 percent The ministry also has reduced the number of fee-
reduction in teacher salaries. In the health sector, the paying students, which shrinks the overall resource
government has embraced sources of efficiency gain envelope for higher education at the same time that
through restructuring. Drawing on diagnostic work budgets are being cut.
with the World Bank, the government has eliminated
excess hospital beds, invigorated outpatient care, Source: Lewis and Verhoeven 2010.

long-term efficiency and limiting the welfare large increases in minimum wages and social
impact of expenditure cuts (box 2.3). standards throughout 2009–10. Public wages
By contrast, some countries had increased were adjusted accordingly at the end of 2009,
planned spending heading into the crisis, but the subsequent wage hikes have not taken
necessitating painful reductions as govern- place because of challenges to the law in the
ment resources dwindled. In Moldova, edu- Constitutional Court. More recently, the Cabi-
cation sector employees make up about 60 net of Ministers approved salary top-ups for
percent of public employees. During two elec- secondary, vocational, and university teach-
tion campaigns in 2009, the outgoing govern- ers (equivalent to 20 percent of base salary).
ment raised teachers’ salaries 25–30 percent to With no budget yet in place for 2010, budget
align entry-level salaries with average national operations are being executed on the basis
earnings but made no commensurate increases of an operational budget that limits current
in class size or shifts in teaching loads. Other monthly spending to one-twelfth of the 2009
measures were contemplated that would fur- appropriations.
ther raise real wages if implemented. The new One indicator of the impact of the crisis
government’s challenge will be to implement on health expenditures is that pharmaceuti-
corresponding increases in class size or shifts cal spending (a good proxy for health sec-
in teaching loads. Ukraine adopted a social tor spending) has declined sharply in Eastern
standards law in November 2009 that calls for Europe. World demand continued to rise from
44 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.8 Average pharmaceutical expenditures fall in Eastern Europe, especially in the Baltics, before
beginning to rise again

1.30
1.24
average pharmaceutical expenditure

1.20
index (based on Q1 2008)

1.10 1.10
1.08

1.00 0.99
0.95
0.92
0.91 0.91
0.90 0.90
0.88

0.80

0.70
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
2007 2008 2009

Europe Estonia Latvia Lithuania Global

Source: Laing and Buysse 2010.

the first quarter of 2007 through the last quar- had to undertake fiscal tightening.16 The
ter of 2009 (with the first quarter of 2008 con- effects of the crisis have been relatively mod-
sidered the last quarter before the worldwide est in East Asia, although qualitative evidence
financial crisis), but expenditures in Eastern in six countries suggests that informal work
Europe declined in the first quarter of 2009 has surged and that migrants have returned
before beginning to rise again. The decline home temporarily, lowering overall income
was most dramatic in the Baltics, with Latvia and reducing households’ ability to pay for
cutting back pharmaceutical expenditures by social services. Households have responded
more than 25 percent between the fourth quar- by transferring children from private to pub-
ter of 2008 and the end of 2009 (figure 2.8). lic schools and reducing food consumption,
Information on social sector spending in although parents contend that they have tried
other regions is extremely limited, but scat- to shield children’s nutrition.17
tered information provides some examples. For Spending to combat HIV/AIDS (human im-
example, 16 of 19 country programs initiated munodeficiency virus/acquired immune defi-
and monitored by the International Monetary ciency syndrome) is a special case. Big increas-
Fund and implemented with the World Bank es in funding have made HIV/AIDS one of
in 2008–09 budgeted higher social spending the most important items on the development
for 2009; 9 of those countries were in Sub- agenda. Funding for HIV/AIDs programs dur-
Saharan Africa (Burundi, Republic of Congo, ing the current crisis has been largely sustained.
Côte d’Ivoire, Liberia, Malawi, Mali, Niger, The uptick in donor spending in 2008–09,
Togo, and Zambia.15 Several African coun- when the economic crisis was accelerating in
tries with poverty reduction strategies have donor countries, is encouraging. The Global
protected funding for social sectors. Some Fund to Fight AIDS, Tuberculosis, and Ma-
countries with adequate fiscal space (Kenya laria (the Global Fund) disburses quickly once
and Nigeria) have protected capital expendi- allocations are decided, but recipient country
ture, mainly for infrastructure. But there are spending has been slow. So the issue is sluggish
also examples of forced contractions in social disbursement and a new concern for efficiency
spending. Countries with precrisis fiscal and of resource use (see annex 2.2 for a detailed
debt problems (such as Ethiopia and Ghana) discussion). Almost 40 percent of the Global
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 45

FIGURE 2.9 Undisbursed HIV/AIDS grants from the Global Fund to Fight AIDS, Tuberculosis, and Malaria,
Rounds 1–7

60

50 0.49
% of total active grants

0.41
not yet disbursed

40 0.37 0.39 0.38


0.35 0.37

30 0.29
0.24
20

10

0
East Asia Europe and Latin Middle East South East Southern West Central World
and Pacific Central America and Asia Africa Africa Africa
Asia and the North Africa Sub-Saharan Africa
Carribbean

Source: Lewis 2009.

Fund resources remain undisbursed, a pos- place before the crisis that could be scaled up
sible source of additional resources if there is (table 2.5). Less funding to Africa and South
a shortfall or delay in funding flows. Almost Asia reflects the fact that existing safety nets
half the allocations to Sub-Saharan Africa are were smaller and less able to absorb funds.
undisbursed (figure 2.9). The $900 million al- Thus, where capacity was in place, lend-
located in late 2009 under Round 9 is also un- ing could be quickly leveraged. Some coun-
likely to have been disbursed yet.18 tries have been reluctant to introduce safety
net programs because of the costs involved,
The recent buildup of social safety nets although reducing across-the-board subsidies

Safety nets have been a crucial part of the


response to the crises in the hardest-hit coun- FIGURE 2.10 Food-related safety net programs are more
tries. Many countries that responded most common in Africa than elsewhere
effectively already had safety nets, which gov-
ernments were able to quickly modify and 0.7
expand. Evidence on the distribution of safety
net programs shows that programs vary con- 0.6
siderably across regions. For example, food- 0.5
percentage of countries

based programs are more common in Africa


than in other regions (figure 2.10).19 0.4
Another sign of the importance of safety
0.3
nets in responding to the crisis is the dramatic
increase in World Bank lending for safety nets 0.2
after the crisis struck—topping $3 billion in
0.1
29 countries in fiscal 2009. Elevated activity
is expected to continue in 2010–11, particu- 0
larly in low-income countries and fragile and cash transfer food for work food rations/ school feeding
stamps
postconflict settings (table 2.4).20 The regional
distribution of lending activities reflects the Africa rest of the world

dominance of Latin America, which had the Source: Wodon and Zaman 2010.
greatest number of effective safety nets in Note: Based on a March 2008 survey of 120 World Bank country teams.
46 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

TABLE 2.4 World Bank lending for safety nets before and since the planning for a pilot conditional cash trans-
food, fuel, and financial crises, 2006–11 fer program (Pantawid Pamilyang Pilipino
US$ billions
Program, or 4Ps) in 2007. It was launched
International Bank International in February 2008 for 6,000 households. As
for Reconstruction Development
and Development Association the crisis unfolded, the government acceler-
Period Loans Loans Grants Total ated and augmented the program, rolling it
out to 376,000 households by March 2009.
2006–08 (precrisis) 0.57 0.62 0.03 1.23
In mid-2009, the government announced
2009–11 (postcrisis) 4.48 1.38 0.03 5.89
plans to expand the program to as many as
Source: Data from the World Bank Business Warehouse project database for projects
classified as social safety nets (54). 1 million households by the end of 2009.
Note: Data do not include interventions funded under the Global Force Crisis Response • Before the crisis, the government of Brazil
Program.
had established a highly successful condi-
tional cash transfer program, Bolsa Famil-
TABLE 2.5 World Bank portfolio allocations to social safety nets, ia, to protect poor families. When the crisis
by region, 2009–10 hit, the government expanded the program
Amount Number
to more than 12 million families, using a
Region (US$ millions) of projects new methodology of poverty maps and
Latin America and the Caribbean 2,917 21 an income volatility index, and raised the
Europe and Central Asia 926 21 benefit level 10 percent to compensate for
East Asia and Pacific 618 9 higher food prices. The program was ex-
Sub-Saharan Africa 574 23 panded in regions where poverty reduction
South Asia 373 9
has been slow—in urban municipalities and
Middle East and North Africa 19 8
in the mid-south region—reaching 1.3 mil-
Source: Data from the World Bank Business Warehouse project database for projects
classified as social safety nets.
lion families in those areas in 2009. Anoth-
Note: Data do not include interventions funded under the Global Force Crisis Response er 600,000 families within poverty belts or
Program.
in specific vulnerable groups are expected
to join the program in 2010.
while augmenting targeted safety nets can help • In response to the food, fuel, and financial
reduce poverty without a significant drain on crises, Chile announced in April 2009 the
revenues (box 2.4). strengthening of multiple safety net pro-
Several countries expanded existing or grams. Family allowances of about $45 were
planned safety net programs in response to distributed to 1.4 million families, including
the crisis. all families in the Chile Solidario program
(around 300,000), families in the Fam-
• The Republic of Yemen, hard hit by the glob- ily Subsidy program, and families whose
al food crisis (drought has forced imports of monthly income was $555 or less. In all,
more than three-quarters of its food), ex- some 5.6 million people in the bottom 40
panded safety net programs with support percent of the income distribution will ben-
from the World Bank and the European efit, at a cost of $62 million.
Union. The cash-for-work program was • The government of Ethiopia established the
extended to an additional 22,000–26,000 Productive Safety Net Program in 2005 to
households in communities most affected by pay for participation in labor-intensive pub-
higher food prices, the share of cash transfers lic works and provide direct support to el-
to the poorest beneficiaries was increased, derly or incapacitated household members.
and 40,000–50,000 more households were The program has been expanded since, pro-
added to the cash transfer program. viding immediate assistance to 1.5 million
• The food and fuel price shocks in 2008, households when the food and fuel crises
the global economic crisis, and a recent struck, and providing additional transfers
typhoon have sharply increased poverty in to 4.4 million people as the crisis deepened.
the Philippines. The government had begun Evaluations find a positive impact on use of
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 47

BOX 2.4 Using safety nets to lower the cost of reducing poverty

Some countries have hesitated to establish safety gram. The response was easy, fast, and affordable
nets because of the cost. Safety net expenditures in because of the earlier investment.
developing countries average 1–2 percent of GDP. In 2005 Indonesia cut its fuel subsidies by $10
Expenditures on programs that are to scale and billion, using a quarter of the released funds for a
that have been evaluated as delivering significant pos- targeted cash transfer that more than compensated
itive impacts, such as Mexico’s Oportunidades and poor recipients for their losses. Another quarter of
Brazil’s Bolsa Familia, average 0.4 percent of GDP. the savings went to basic health and education pro-
Ethiopia’s largest safety net program, the Productive grams for the poor.
Safety Net, costs about 1.7 percent of GDP. In 2008 the Philippines found itself short of effec-
The introduction of a well-targeted safety net can tive policy instruments to protect the poor against
provide the political space to reduce or eliminate escalating rice prices. A key part of its multipronged
expensive and poorly targeted general price subsidies, response package, which cost some 1.3 percent of
freeing up resources to fund the targeted programs. GDP, was a program of loosely targeted and distor-
The potential for such reallocations is considerable tive rice subsidies. Realizing that this approach is
because many countries have large and costly price expensive and regressive, the government is working
subsidies. More than a third of countries recently on better safety net options—unifying administra-
surveyed by the International Monetary Fund raised tion under a new umbrella program, scaling up a
subsidies an average of 1 percent of GDP in response proxy means test for targeting households, reform-
to higher food and fuel prices. Several examples illus- ing and expanding the school feeding program, and
trate successful country experiences in switching accelerating rollout and scaling up of a conditional
from universal subsidies to targeted safety nets. cash transfer program.
In the late 1990s Mexico progressively moved
funding from price and in-kind food subsidy pro-
grams to the Oportunidades conditional cash trans- Source: Data for 87 countries for which data on safety net
fer program, probably the most positively evaluated expenditure were available from World Bank public expendi-
ture reviews, safety net assessments, social protection strategy
safety net program in a developing country. Fifteen notes, and other studies. Data coverage is low for Sub-Saharan
years later, as food prices rose dramatically, the gov- Africa, where government spending on safety nets may be low,
ernment was able to protect the poor by issuing a but where donor funding may compensate considerably. See
one-time top-up benefit to those already in the pro- also IMF 2005.

health services and caloric availability and and administratively feasible or desirable, con-
reductions in negative coping behaviors, sidering the negative incentives they might cre-
such as child labor and withdrawal from ate. Thus a part of policy reforms in developing
school. countries should be understanding what kind
of safety net program best serves various social
Safety nets are important not only in cush- assistance activities, what the implementation
ioning the effects of the crisis but also as part challenges are, and how to develop programs
of a broader poverty reduction strategy inter- for maximum effectiveness.21
acting with social insurance; health, education,
and financial services; the provision of utilities
Informal safety nets and remittances
and roads; and other policies for reducing
poverty and managing risk. Many challenges Households manage risk through informal
remain, however. Safety net programs in low- safety nets (such as crop diversification), infor-
income countries are often slight and frag- mal savings and credit associations, burial
mented and cover only a small percentage of societies, labor exchange arrangements, migra-
poor and vulnerable populations. There are tion, and emigration. Informal safety nets are
real concerns over whether they are affordable generally more effective against idiosyncratic
48 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

MAP 2.2 An infant in a developing country is ten times more likely to die
than a newborn in a developed country

Infant mortality rate:


Deaths per 1,000 live births (2008) Green
(De
≥100
70–99
40–69
10–39
<10 Canada
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: World Development Indicators.

shocks that affect only one or a few households to developing countries; they reached $338
than against systemic shocks that affect whole billion in 2008 before the full impact of the
communities. Thus it is not clear whether such financial crisis was felt. 23
risk-mitigation strategies were any more effec- The global nature of the current crisis has
tive during the recent crisis than they had been likely reduced the support that remittances can
previously. provide. Remittance flows to developing coun-
Remittances have played an important tries are estimated at $317 billion for 2009,
countercyclical role in crises that affected indi- a 6.1 percent decline from 2008. Analysis of
vidual developing countries.22 Because remit- the first nine months of 2009 shows that the
tances are unaffected by idiosyncratic shocks financial crisis has affected remittance flows
or even local or national systemic shocks, they unevenly. Remittances to Latin America and
are an important part of the household safety the Caribbean have suffered large declines
net for many poor households. More than tri- (down 13 percent in Mexico, for example),
pling since earlier in the past decade, remit- mainly because of the early effects of the crisis
tances constitute important monetary flows in the United States and Spain. Similarly, remit-
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 49

IBRD 37733
APRIL 2010

nland
en)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Portugal Spain Turkey Armenia baijan Turkmenistan of Korea
Monaco Greece Tajikistan Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire African Rep. Somalia Palau
Togo Cameroon Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Congo Kenya Singapore Nauru
São Tomé and Príncipe Gabon Rwanda
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

tances to the Middle East and North Africa increasing in some cases (up 24 percent in Paki-
have declined more than expected, plunging stan, 16 percent in Bangladesh, and 13 percent
20 percent in the Arab Republic of Egypt and in Nepal). In East Asia and the Pacific, flows
Morocco. The situation is even more serious in were also stronger than expected (up 4 percent
Europe and Central Asia, where many coun- in the Philippines).
tries are among the top recipients of migrant A study by the World Food Programme
remittances as a percentage of GDP. Tajikistan, found that families that rely on remittances
where remittances make up 50 percent of GDP, from abroad were among groups most affected
experienced a decline of more than 30 percent by the current financial crisis.24 In Armenia,
in the first half of 2009. Many other countries where remittances make up 20 percent of
in the region have experienced similar declines. GDP and are the main source of income for
By contrast, in Sub-Saharan Africa the decline 25 percent of households, the impact was felt
has been less steep and in some countries, such immediately, with remittances slumping 30
as Uganda, flows have increased. In South percent in the first quarter of 2009. In Nica-
Asia, remittances have remained strong, even ragua, a country highly dependent on remit-
50 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

tances and vulnerable to economic downturns of the current crisis—a first for the region. By
in the U.S. economy, food consumption pat- contrast, in Europe and Central Asia middle-
terns are changing and families are spending income countries that were unable to halt large
less on health and education. increases in private sector credit growth were
the hardest hit by the current crisis. They had
higher growth rates before the crisis but also
This crisis is not about domestic
larger declines after the crisis, and on balance
policy failure
they experienced lower average growth rates
Improvements in developing countries’ poli- than countries with more modest increases in
cies since the 1990s may blunt the impact of private sector credit growth.
the crisis on human development. Crises in Comparing the economic performance
low-income countries have often been driven of countries according to the quality of their
by poor governance, civil conflicts, or severely policies and institutions shows the importance
distorted macroeconomic policies. (Internal of policy reform. Although there is no perfect
shocks accounted for 89 percent of output vol- measure of the quality of the policy and insti-
atility in low-income countries from the early tutional environment in developing countries,
1960s to mid-1990s.) The failure of domestic the World Bank’s Country Policy and Institu-
institutions has been an important reason for tional Assessment (CPIA) provides a consis-
the severity of past crises on human develop- tent framework for assessing country perfor-
ment, macroeconomic variables, and the qual- mance (on a scale from 1, worst, to 6, best).25
ity of institutions. Countries with better policies or initial fiscal
But some indirect evidence suggests that positions have generally done better in the cur-
this situation may be changing and that the rent crisis (see chapter 3). And before the crisis
impacts of the current crisis on human devel- (2001–07), developing countries with 2008
opment could be less severe than in previous CPIA scores of 3.2 or better grew faster than
crises. Since the 1990s output volatility in low- countries below this cutoff (figure 2.11). Per
income countries has lessened, and the influ- capita GDP growth averaged 3.9 percent for
ence of external shocks has intensified (box countries with good policies and 1.9 percent
2.5). To the extent that lower volatility and a for fragile states with poorer policies. Coun-
reduced importance for internal shocks indi- tries with better policies also had lower infla-
cate improved policies, governments should tion, at 5.2 percent a year, compared with 6.6
be better placed to protect their people from percent for countries with poorer policies. The
the most severe impacts of the crisis. pattern is the same for countries in Sub-Saha-
Assessing the quality of policies and insti- ran Africa.26 Before the current crisis, coun-
tutions over time is difficult, but external evi- tries with better policies tended to have better
dence does indicate an improvement in many outcomes for MDG indicators such as under-
developing countries since the 1990s. Inflation five mortality, gender equality in primary and
rates have declined substantially, fewer coun- secondary education, primary school comple-
tries have unsustainable debt positions, more tion, and access to an improved water source.
countries have access to private capital markets Several empirical studies also showed that bet-
and have attracted substantial foreign direct ter policies and institutions improve the mar-
investment, financial intermediation has risen ginal contribution of growth to progress on
as a share of output, trade barriers have come human development indicators.27
down, black market exchange rate premiums
have shrunk, and civil conflict has subsided The impact of the current crisis is
in many countries. The pace of policy reform still worrisome
has varied. In Latin America weak currencies,
banking sectors, and poor fiscal management The crisis has generated predictions of rising
tended to amplify the impact of past crises, mortality rates and closed schools as govern-
whereas improvements in the policy and insti- ments reduce services in response to falling
tutional framework have cushioned the impact output and public revenues. These fears are
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 51

BOX 2.5 Are external shocks becoming more important than internal shocks for
developing countries?

Historically, developing countries have endured External shocks have become more important since
greater macroeconomic volatility than have indus- the 1990s
trial economies. A simple look at the data shows that
40
output volatility (measured as the standard deviation
of real GDP growth) has been two to three times 35
greater in developing countries than in industrial

% long-run output volatility


30
countries in the last 20 years.
Because developing countries are highly dependent 25
on primary commodities and foreign capital and have 20
greater exposure to natural disasters, policy makers
often blame external shocks, such as terms-of-trade 15
fluctuations, natural disasters, and aid volatility, for 10
countries’ uneven macroeconomic performance.
However, research shows that external shocks 5
account for only a small fraction of the variance 0
in real per capita GDP in low- and middle-income Latin America East Asia other Sub-Saharan
and the and middle-income Africa
countries. Among low-income countries, external Caribbean Pacific countries
shocks, including terms-of-trade fluctuations, global earlier years later years
economic growth, international fi nancial conditions,
natural disasters, and aid volatility, explain no more
than 11 percent of output volatility. The 89 percent agement, and control of corruption have improved
residual is probably related to internal conditions, since the early 1990s. Many middle-income coun-
such as the volatility of macroeconomic management. tries have also strengthened their fi scal position by
Among middle-income countries, external shocks reducing deficits and accumulating reserves; tamed
account for about 20 percent of output volatility. inflation through independent central banks; and
Since the 1990s, however, many developing coun- promoted local bond markets after the Asian and
tries have undergone structural transformations that Russian fi nancial crises of 1997–98.
may have calmed internal volatility and increased
the importance of external factors. Research shows
that external shocks have become more important
for developing countries in several regions during Source: Raddatz 2007, 2008a, 2008b.
the past two decades (see the figure at the right). In Note: The data for Latin America and the Caribbean, East
African countries this shift has resulted not from Asia and Pacifi c, and other middle-income countries corre-
spond to 1974–1985 (earlier) and 1986–2004 (later). The bars
an increase in the volatility of external shocks, or
for Africa correspond to 1963–89 (earlier) and 1990–2003
in countries’ vulnerability to them, but rather from (later). External shocks explaining long-run output volatil-
the taming of internal sources of volatility. In these ity include the state of the world business cycle, international
countries—among the most volatile—standard indi- fi nancial conditions, terms of trade, natural disasters, and aid
cators of democratic accountability, economic man- flows.

grounded in the experience of past crises, development point to serious problems. An


when, as noted earlier and in box 2.6, pov- important reason is the size of the shock—it
erty, hunger, health outcomes, and access to is the largest global downturn since the Great
education deteriorated sharply. Despite policy Depression. Thus while developing countries’
improvements and efforts to sustain social efforts have been important in mitigating the
spending and ramp up safety nets, preliminary impact of the crisis, the crisis nevertheless has
indications of the impact of the crisis on human been a severe setback to poverty reduction.
52 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.11 Economic performance and MDG outcomes are better with good policy

a. Economic performance of countries b. Selected MDGs by quality of policy,


by quality of policy, 2001–2007 averages of 2000–06
7 6.6 120 100

6 100 90
5.2
5
80 80
percentage

3.9

per 1,000
4

percent
60 70
3
1.9 40 60
2

1 20 50

0 0 40
GDP per capita growth inflation child mortality ratio of girls to primary access to
under five boys in primary completion improved water
CPIA < 3.2 CPIA ≥ 3.2 (per 1,000, and secondary rate, total (of source (of
left axis) education relevant age population
group) with access)

CPIA < 3.2 CPIA ≥ 3.2

Source: World Bank staff calculations.


Note: CPIA is the World Bank’s Country Policy and Institutional Assessment framework for assessing country performance; ratings range from 1 (worst) to 6
(best). Countries with a CPIA score of 3.2 or better have better policies than countries that score under 3.2.

The recent deterioration in human devel- number there would have been without the
opment indicators began with the food and economic crisis.
fuel price shocks of 2007. In some countries The problems were compounded by the
food prices almost doubled with no adjust- global economic crisis. A poverty monitoring
ment in earnings.28 In Mozambique incomes study of 13 countries suggests that in countries
were almost halved and food consump- like the Central African Republic and Ghana,
tion fell by a fifth; children’s weight for age parents were forced to take their children out
and body mass index were reduced with no of school and that in other countries they
change in height for age, indicating that the scrambled to finance their children’s continu-
price rise has seriously compromised nutri- ing attendance.30 In Serbia, Roma children
tion. The effects spilled over into the efficacy dropped out of school because of a lack of
of HIV/AIDS treatment, with lower-income clean clothes and soap. Poor households in
households showing slower improvements Cambodia and the Philippines reported cutting
than households with higher incomes and overall consumption in response to income
better access to adequate nutrition, which shocks to protect children’s school attendance.
reinforces the beneficial effects of antiret- Although little information exists on the dif-
roviral therapies. Recent analysis finds that ferential impact of the crisis on women and
the 2008 global food price spike may have men, recent surveys of East Asia do not show
increased global undernourishment by some that women have been disproportionately
6.8 percent, or 63 million people, relative to affected (box 2.7).
2007.29 Moreover, the analysis shows that Recent surveys in Armenia, Montenegro,
the sharp slowdown in global growth in 2009 and Turkey give a sense of how declines in
might have contributed to 41 million more income induced by the crisis are reducing
undernourished people compared with the household consumption.
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 53

BOX 2.6 Human development suffered severely during crises in developing


countries

Household studies from past crises suggest that the both rural and less educated women are at higher risk
impact on human development can be serious. In the of losing their infants.
lowest-income countries, poverty rises, people eat For middle-income countries the picture is less
lower-quality food, school enrollments fall, health consistent. In Latin America school attendance has
care use drops, and infant mortality rises. Even increased during crises, possibly because children are
modest reductions in food consumption for children not needed for economic activity, but infant mortal-
between birth and age two can have lasting deleteri- ity appears to have risen. In Indonesia the crisis of the
ous effects on cognitive and physical development. In late 1990s had little measurable impact on schooling
South Africa and Zimbabwe, the nutritional depriva- or health, possibly because the country was better off
tion of young children led to lower height for age and and perhaps because education and health services
shorter stature in adulthood. were better protected. But the impacts of recession
Analysis of the effects of downturns on infant mor- are far more severe for child health than for educa-
tality in Sub-Saharan Africa shows that a 1-percent tion, even in middle-income countries.
reduction in per capita GDP is associated with a rise
in the infant mortality rate of 0.34–0.48 per 1,000
live births, or 34–39 percent of the average annual Source: Wodon and Zaman 2010; Ferreira and Schady 2009;
change in the infant mortality rate. Infant girls are Dinkelman 2008; Alderman, Hoddinott, and Kinsey 2006;
more likely than boys to die during downturns, and and Gottret and others 2009.

Food consumption in Armenian households of poor households did, and the same percent-
has fallen 41 percent, and health care spend- age of poor households canceled health insur-
ing is down 47 percent. Some 50–60 percent ance. In education the wealthiest households
of households in the four lowest income quin- cut back the most—20 percent compared with
tiles have cut back on health care services and 11 percent for the lowest-income households.
drug purchases. Household reductions in food In Turkey, the poorest households have ex-
consumption are inversely related to income, perienced the largest reductions in wages and
with 20 percent of the wealthiest households self-employment income. Some 91 percent of
cutting back (noteworthy in itself) and more the poorest 20 percent of households lost in-
than 55 percent of the poorest 20 percent do- come, but even the wealthiest 20 percent ex-
ing so. Even bigger cuts are seen in spending perienced some income loss. Safety nets cover
on entertainment and expensive foods. There only 20 percent of the poorest households,
is some evidence that these cutbacks have requiring the rest to sell assets, draw down
helped protect education spending. savings, and find other informal sources of
In Montenegro unemployment figures sug- support. Among the poorest households, 75
gest that cutbacks affect almost a quarter of percent have reduced children’s food consump-
households. Safety nets cover only 18 percent tion, 29 percent have curtailed health care use,
of the poorest 20 percent of households, and and 14 percent have cut back on education
informal private transfers are disappearing as spending. Even middle-class households have
remittances shrink and informal safety nets un- trimmed spending, especially in education.
ravel. Private investments in education, health The data now available on the impact of the
insurance, and preventive health care have crisis on human development are still much
fallen, reducing resilience to further shocks too limited to draw any conclusions on the
(figure 2.12). Overall, 9 percent of households overall impact. But there is certainly evidence
reduced preventive care visits, but 25 percent of suffering as a result of the severe global
54 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

BOX 2.7 Gender differences in impacts of the crisis: Evidence from East Asia

Although the effects of the crisis have clear gender rapid qualitative assessments (including focus group
dimensions, it is not clear that women in East Asia discussions), ex ante simulations using precrisis
have been disproportionately affected. Gender- household survey data, analysis of labor force sur-
specifi c impacts would be expected because of the vey data as available, and triangulation across data
gender division of labor in the labor market and in sources.
the home, gender disparities in access to produc- Data indicate that unemployment in East Asia
tive resources, and gender dimensions of household has barely changed during the crisis, for men or
resource allocation. But precise impacts are unclear women, but that women’s participation has tended
because they depend on multiple factors including to rise. In some countries unemployment has fallen
the size of the shock, the economic structure of the more for women than for men, while increases in
country, the nature of government responses, and the labor force participation have been more marked for
speed of economic recovery. Identifying the gender women than for men (see the fi gures below), par-
impacts of the crisis is thus an empirical issue. ticularly in poorer countries, where female labor
Empirical analysis is complicated by a lack of has shifted from unpaid work to self-employment.
data. High-frequency data on the social impacts of Both quantitative and qualitative data indicate lon-
the crisis is generally not available, and the lack is ger working hours as men and women take on addi-
particularly intense for gender-disaggregated data. tional jobs to compensate for falling earnings from
Thus multifaceted approaches are needed, such as primary jobs.

Labor force participation by gender in selected East Asian countries, 2007–09

Cambodia Philippines
90 85
men
85 80 men
80 75
women
75
70
percent

percent

70
65
65
60
60
55 55
women
50 50
45 45
2007 2008 2009 2007 2008 2009

Indonesia Thailand
90 85
85 men
men 80
80 75
75
70
percent

percent

70 women
65
65
60
60
55 55
women
50 50
45 45
2007 2008 2009 2007 2008 2009

Source: World Bank staff calculations.


GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 55

BOX 2.7 (continued)

The impact of labor market shocks is driven by that women’s total work burden (paid plus unpaid
several factors, of which gender is only one. Both domestic work) has increased over the past year. In
quantitative and qualitative findings suggest that urban Thailand, women explain that their time on
simple interpretations of labor market data may be unpaid domestic work has declined a little but not
misleading. As an example, well-publicized data enough to offset rising labor market hours. In rural
show layoffs from enterprises producing garments Cambodia and the Philippines, research teams noted
and other products for export to shrinking markets, that an increased dependence on common property
sectors where female employment tends to dominate. resources, including fi rewood, has increased women’s
Less well-documented is the contraction in hours and time on domestic chores.
earnings in sectors serving domestic markets, where The welfare impacts of the crisis, by gender,
purchasing power is closely linked to the health of also appear to be nuanced. Microsimulations of the
the export sector. These sectors may be dominated poverty impacts of the crisis in Cambodia suggest
by men. Women laid off from formal sector work that male-headed households were more affected in
may be better off than men facing highly restricted urban areas, while female-headed households were
earnings in informal sector jobs. Quantitative and more affected in rural areas (see the fi gure below).
qualitative evidence from Cambodia suggests that For urban male-headed households, this finding
more male workers in the construction sector have likely reflects the impacts of the crisis on male jobs
been affected by the crisis than female workers in the in construction and tourism. The effects for rural
garment sector. Moreover, male construction work- female-headed households appear to reflect the loss
ers are more likely to be poor and have fewer eco- of remittance income in addition to more direct cri-
nomic fallbacks than female garment workers. sis impacts on household earnings. Findings from
There is no consistent cross-country pattern in rapid assessments in rural Cambodia indicate that
differences in hours of paid work by gender. In some female-headed households commonly cut back con-
countries, such as Cambodia, both men and women sumption sharply and increased their indebtedness to
have greatly increased their hours of paid work. In cope with loss of income as remittances from urban
other countries, such as Indonesia, women have areas fell. Male migrant workers—often migrant
overtaken men in hours of paid work in the past two spouses—reported being unable to return home as
years. And in other countries, such as the Philippines, often as before because of increased transportation
men and women appear to work the same number of costs and reduced earnings, meaning less male labor
paid hours. However, focus group discussions suggest on the farm during peak periods.

Impacts of the global financial crisis on male- and female-headed households in Cambodia

Poverty headcount
8
7.1
7
percentage point changes

6
due to crisis impact

5 4.6
4.3
4.0
4 3.6 3.7
3
2
1 0.5 0.4
0
Phnom Penh other urban rural Cambodia
male head of household female head of household

Source: Bruni and others forthcoming.


56 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.12 Spending cutbacks in crisis-affected households are jeopardizing future welfare in
Armenia, Montenegro, and Turkey

a. Armenia
less use of entertainment 64 70
fewer meetings with friends 59
cheaper instead of more expensive food items 58 60
reduced or stopped visits to healthcare centers 47
reduced or stopped buying medicines 41 50
reduced amount of food consumed 41

percent
stopped buying some nonfood items 40 40 stopped visiting
increased use of public transport or walking 33 health centers
30
work odd jobs 13 stopped buying
increased own food production 12 medicine
20
migration for work 10
decreased
bought second-hand items 9 food
10
other mitigation measures 3 consumption
withdrew or postponed admission to school 2
0
0 10 20 30 40 50 60 70 1 2 3 4 5
(poorest) (richest)
percent income quintile

Source: Armenia Integrated Living Conditions Survey 2009. See Ersado, forthcoming.

b. Montenegro
delayed purchase of durables 49 30
replaced expensive items with cheaper ones 44
less use of communication services 40
25
% households affected by crisis

stopped buying some nonfood items 34


made less use of information services 33 reduced
decreased total food consumption 31 20 training
restricted vacations 30
met friends less 23 reduced
15 preventive
changed means of transportation 17
care
replaced private medical care with public care 16
reduced playing sports/exercising 15 10 canceled
increased agricultural production for own use 13 insurance
took a job although not working before 12
started to buy second-hand products (more) 12 5
reduced visits to doctor for preventive care 9
left or postponed training 8
0
0 10 20 30 40 50 60 1 2 3 4 5
(poorest) (richest)
percent
asset quintile
Source: Montenegro Crisis Monitoring Survey 2009. See Hirshleifer and Azam, forthcoming.

downturn. Even if the deterioration in human Early evidence in 41 middle-income countries


development indicators has not been as severe indicates that the impact on the labor market
as in previous crises (as speculated above), the has been severe, especially in wealthier middle-
human suffering will be considerable. income countries of the Europe and Central
Although many people in middle-income Asia region. Although the number of jobs and
countries are above the threshold of the poverty their growth have been negatively affected,
MDG, they are also the hardest hit by adjust- the impact has been mostly on the quality and
ments in wage earnings and employment.31 earnings of employment (figure 2.13).
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 57

FIGURE 2.12 (continued)

c. Turkey
80
substituted into cheaper food items 73
substituted into cheaper nonfood items 65
decreased the amount of food consumption 53
60
met with friends less 49
stopped buying non-food products 48

percent
changed transportation type 31 40
32
less use of information services 30
reduced visits to the doctor for prevention control 26
20
reduced the use of health services 21 14
left courses of language, computer, others 10
5
withdrew or postponed the 6
admission to school 0
0 20 40 60 80 poorest middle richest
20% 20% 20%
% households that adopt this
asset quintile
coping mechanism (out of
respondent households) decreased food consumption
reduced use of health services
reduced education expenditures

Source: TEPAV, UNICEF, and World Bank 2009; Turkey Welfare Monitoring Survey.

• Three-quarters of the labor market adjust- growth. But fixed exchange rates worsen the
ment stems from slower growth in take-home labor market impact. On average, countries
pay, only one-quarter from less job creation. with fixed currency regimes witnessed a de-
• Earnings in most middle-income coun- cline in employment of 1.7 percentage points,
tries are falling mainly because people are compared with only 0.4 in countries with
working fewer hours. Hourly wages have floating rates. The slowdown in the wage-bill
changed little except in Europe and Central growth was also less severe for the countries
Asia, where they have declined. with moderate levels of development.
• The crisis severely affected labor markets, The nature of recent labor market adjust-
with few countries spared. It caused a sharp ments in these countries suggests that effective
slowdown in wage-bill growth, which fell policy packages should also focus on support-
by an average of 8 percentage points. The ing earnings and household income, not just
exceptions were Argentina, China, and the generating employment. Responses taken in
former Yugoslav Republic of Macedonia, developed European countries—such as par-
where wage-bill growth accelerated. tial unemployment insurance, expanded cash
• Employment has shifted away from indus- transfers to poor workers, and temporary
trial employment into services, where jobs wage subsidies—may be priority interventions
tend to be of lower productivity and offer in those countries where hours and earnings
lower wages. adjustments dominated.
• For a given decline in GDP growth, the
labor market impact was more severe in
upper-middle-income countries and in Conclusions
countries with fixed exchange rates. Because crises have very negative effects on
human development indicators, good poli-
The large impact in Europe and Central cies and institutions are essential in develop-
Asia resulted mainly from sharp drops in GDP ing countries to avert downturns in the first
58 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2.13 The crisis sharply reduced wage earnings in middle-income countries

a. Wage-bill growth b. Employment growth


Latvia Bosnia and Herzegovina
Latvia
Ukraine
Lithuania
Lithuania Bulgaria
Dominican Republic
Russian Federation
South Africa
Sri Lanka Russian Federation
Jamaica
Serbia
Mauritius
Mexico Ukraine
Egypt, Arab Rep.
Armenia
Serbia
Turkey Moldova
Romania Poland
Mexico
Poland Kazakhstan
Ecuador (urban) Peru
Chile
Venezuela, R.B. de Belarus
Bulgaria Colombia (urban)
Indonesia
South Africa China (urban)
Colombia (urban) Brazil (urban)
Venezuela, R.B. de
Belarus
Georgia
Moldova Romania
Argentina (Urban)
Indonesia
Armenia
Thailand Ecuador (urban)
Kyrgyz Republic
Peru
Thailand
Brazil (Urban) Trinidad and Tobago
Macedonia, FYR
Philippines
Malaysia
Kazakhstan Philippines
Chile Albania
Tajikistan
China (urban) Montenegro
Argentina (urban) Sri Lanka
Morocco
Macedonia, FYR Turkey
–35 –25 –15 –5 5 15 –20 –15 –10 –5 0 5
percentage point change percentage point change

Source: Khanna, Newhouse, and Paci, forthcoming.

place, dampen their negative effects when they There are some reasons for hope that the
do occur, and reduce the potential for reversal current crisis may be different for low-income
of reforms. Policy failures, particularly in low- countries. A great deal of social spending has
income countries affected by corruption and been protected so far. Policies and institutions
violent conflict, have been a major reason for had improved before the crisis. And external
the sharp deterioration in human development shocks, not domestic policy failures, were the
indicators in past crises. main causes of the current crisis. Nonetheless,
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 59

sharp that a long period of strong growth is


the impacts on progress toward the MDGs are
needed to undo the damage inflicted on devel-
already worrisome.
opment outcomes. The next chapter examines
While recovery of the global economy
the growth outlook and macroeconomic chal-
appears to be stronger than expected, small
lenges, including the fiscal tensions created by
reductions in growth could still have last-
temporary stimulus measures and protection of
ing negative consequences for poverty and
social spending.
human development. The contraction was so
60 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

Annex 2.1 Human and economic indicators during growth cycles


This annex presents more detailed information The conclusion that these indicators tend to
on the asymmetric impact of growth decel- deteriorate more in bad times than they improve
erations on human development indicators, in good times does not stem from composition
macroeconomic variables, and institutional effects. It is important to examine these effects
quality in developing countries. Tables 2A.– because the averages for each period (accelera-
–2A.3 show the average level of each indicator tions, decelerations, and other) do not reflect
during growth accelerations, growth decelera- the same number of observations or equal par-
tions, other periods, and across all times. Tests ticipation by different income groups—there
for differences in the means of these variables are more accelerations than decelerations, and
between growth accelerations, decelerations, low-income countries have greater represen-
and all country-year observations show that tation in the sample means during bad times
they are all statistically significant at the 1 per- because of the higher frequency of decelera-
cent level. tions in these countries (see main text). Because

TABLE 2A.1 Differences between sample averages: Human development and gender indicators
Otherwise
Growth Growth (not in acceleration Sample
Variable acceleration deceleration or deceleration) period
Life expectancy at birth, women (years) 72.1 63.4 69.4 70.0
Life expectancy at birth, men (years) 66.6 58.1 64.2 64.7
Life expectancy at birth, total (years) 69.2 60.7 66.7 67.3
Infant mortality rate (per 1000 live births) 27.7 59.7 39.9 35.9
Child mortality under-five rate (per 1,000) 42.4 96.3 59.3 54.3
Primary completion rate, girls (% of relevant age group) 83.2 49.8 76.3 78.6
Primary completion rate, boys (% of relevant age group) 84.5 59.6 80.2 81.4
Primary completion rate, total (% of relevant age group) 84.4 54.8 78.1 80.2
Ratio of girls to boys, primary enrollment 95.6 86.1 92.5 93.6
Ratio of girls to boys, secondary enrollment 96.9 80.7 94.8 95.3
Ratio of women to men, tertiary enrollment 107.3 65.1 106.2 105.4

Source: World Bank staff calculations based on data from World Development Indicators.
Note: Tests for differences in the means of these variables between growth accelerations, decelerations, and all country-year observations show that they
are all statistically significant at the 1-percent level.

TABLE 2A.2 Differences between sample averages: Sub-Saharan Africa


Otherwise
Growth Growth (not in acceleration Sample
Variable acceleration deceleration or deceleration) period
Life expectancy at birth, girls (years) 55.2 52.3 53.4 54.0
Life expectancy at birth, boys (years) 52.2 48.9 50.1 50.8
Life expectancy at birth, total (years) 53.7 50.5 51.7 52.3
Infant mortality rate (per 1.000 live births) 80.7 106.6 97.3 91.9
Child mortality under-five rate (per 1,000) 133.5 161.3 154.3 146.2
Primary completion rate, girls (% of relevant age group) 55.1 33.8 42.1 47.4
Primary completion rate, boys (% of relevant age group) 59.8 48.4 50.9 55.0
Primary completion rate, total (% of relevant age group) 57.4 41.0 46.5 51.1
Ratio of girls to boys, primary enrollment 89.7 77.9 82.4 85.0
Ratio of girls to boys, secondary enrollment 82.3 63.6 76.1 77.7
Ratio of women to men, tertiary enrollment 60.2 32.5 64.4 58.7

Source: World Bank staff calculations based on data from World Development Indicators.
Note: Tests for differences in the means of these variables between growth accelerations, decelerations, and all country-year observations show that they
are all statistically significant at the 1-percent level.
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 61

TABLE 2A.3 Differences between sample averages: Economic and institutional indicators
Otherwise
Growth Growth (not in acceleration Sample
Variable acceleration deceleration or deceleration) period
Final consumption (% GDP) 81.45 88.78 83.74 83.30
Government consumption (% GDP) 15.41 16.68 16.61 16.10
Gross capital formation (% GDP) 23.76 18.57 23.35 23.10
Gross domestic savings (% GDP) 18.58 11.23 16.26 16.70
Gross fixed capital formation private sector (% GDP) 16.35 10.43 13.75 14.40
Imports (% GDP) 45.80 37.45 43.85 44.10
Exports (% GDP) 40.43 30.05 36.52 37.60
Trade (% GDP) 86.23 67.50 80.37 81.70
Foreign direct investment. net inflows (% GDP) 4.48 2.07 3.56 4.00
Private capital flows, total (% GDP) 2.99 1.40 2.03 2.40
CPI inflation (%) 14.88 251.32 37.90 43.90
Institutions (–2.5 to 2.5)
Political stability –0.16 –0.65 0.03 –0.10
Voice and accountability –0.07 –0.47 0.09 –0.02
Regulatory framework –0.03 –0.82 0.15 0.01
Rule of law –0.14 –0.90 0.12 –0.07
Government effectiveness –0.04 –0.96 0.14 0.00
Frequency of conflicts (Sub-Saharan Africa) 0.13 0.23
Aid to poor countries (Sub-Saharan Africa)
ODA (% GDP) 13.80 12.10
ODA per capita (US$) 69.50 41.80
Source: World Bank staff calculations. Data for Sub-Saharan Africa from Arbache, Go, and Page (2008). Indicators on institutions are from the
World Bank Institute’s Worldwide Governance Indicators database, which relies on 33 sources, including surveys of enterprises and citizens,
and expert polls, gathered from 30 organizations around the world; they each range from –2.5 (worst) to 2.5 (best).
Note: Tests for differences in the means of these variables between growth accelerations, decelerations, and all country-year observations
show that they are all statistically significant at the 1 percent level. ODA = official development assistance.

human development indicators are generally tions with their own sample means when not in
lower in low-income than in middle-income growth decelerations (the column “otherwise”
countries, the greater frequency of low-income in the three tables), decelerations still have an
country observations drops the averages for asymmetric effect.32 Furthermore, the averages
decelerations, which could account for the for periods not in acceleration or deceleration
asymmetric relationship. However, even after (normal times) are close to the averages for the
controlling for the sample composition effects entire sample (the last column in each table),
by comparing the sample means of countries providing evidence that the economic cycles
undergoing growth decelerations and accelera- are being correctly identified.
62 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

Annex 2.2 The special case of HIV/AIDS spending


Large increases in funding have made HIV/ billion in 2008, and then declined to $2.6 bil-
AIDS (human immunodeficiency virus/acquired lion in 2009. In the last funding cycle (Round
immune deficiency syndrome) one of the most 9), demand from countries also fell.34 The U.S.
important items on the development agenda. In PEPFAR program increased its contributions
less than a decade the international community from $4.5 billion in 2007 to $6.2 billion in
has mobilized talent and financing to address 2008 and has subsequently increased its annual
HIV/AIDS with new institutions and long-term budgets. The 2010 fiscal year allocation is just
financial commitments to countries suffering shy of $7 billion, suggesting that U.S. support
from an established and growing epidemic. is continuing.35
This attention and financing have produced Fueled largely by increased donor resources,
data that outstrip that available for health care public health spending in the high-prevalence
generally, allowing a more thorough examina- countries of eastern and southern Africa has
tion of trends. The 2008–09 recession is the risen rapidly in absolute and per capita terms
first global crisis to affect international support (see map 1.2). As a share of GDP, the increases
for HIV/AIDS spending, and the responses are have gone disproportionately to people with
instructive. HIV/AIDS.36 Government spending in coun-
Roughly 33 million people have HIV/AIDS, tries that formerly had high HIV/AIDS preva-
but only a third of those are on antiretroviral lence, like Brazil and Thailand, has financed
therapy that will extend their life. There is no both prevention and treatment. Other coun-
cure for AIDS. Discontinuities in treatment tries, such as Ghana, have legally binding
create resistance to the basic (“first line”) anti- commitments ensuring treatment for people
retroviral treatment, which can lead to broader with AIDS.
drug resistance. The alternative “second line” Of 77 countries recently surveyed, most
treatment is 10–20 times more expensive. indicated that they had adequate funding
Thus antiretroviral therapy is central to meet- from governments, donors, and other sources
ing the MDG 6A to combat HIV/AIDS. Equally to finance their current HIV/AIDS programs,
important to treating those who have con- but they raised concerns about the future.37
tracted HIV/AIDS is strengthening preven- Prevention was identified as the likely vic-
tion—the only way to stem the pandemic. tim if funding fell. A further concern was the
increased cost of imported drugs and supplies
resulting from currency devaluations in some
Likely short-term effects of the crisis
countries.38 The Clinton Foundation recently
Funding for HIV/AIDS has risen sharply over obtained price concessions from manufactur-
the past decade. During 2001–05, aid commit- ers that could compensate for the exchange
ments for HIV/AIDS programs rose almost 30 rate penalty.
percent ($4.75 billion), fueled by the establish- The impact of the current downturn is not
ment of the Global Fund and by philanthropic entirely clear, but the uptick in donor spending
efforts by the Clinton Foundation, the Bill & in 2008 and 2009, when the economic crisis
Melinda Gates Foundation, and others. New was accelerating in donor countries, is encour-
sources of funding have come onstream since aging. The Global Fund disburses quickly once
2005 with the U.S. President’s Emergency Plan allocations are decided, but recipient country
for AIDS Relief (PEPFAR) and UNITAID, spending has been slow. Almost 40 percent of
which disburses much of its resources through the Global Fund resources remain undisbursed,
the Global Fund. a possible source of additional resources if there
In 2008 public and private entities allocated is a shortfall or delay in funding flows. Almost
$15.8 billion for global HIV/AIDS programs, half the allocations to Sub-Saharan Africa are
$6.7 billion of it from bilateral and European undisbursed (see figure 2.9). The $900 mil-
Union contributions.33 Pledges to the Global lion allocated in late 2009 under Round 9 is
Fund rose from $2.5 billion in 2007 to $3.0 unlikely to have been disbursed yet.39
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 63

Although countries may appear to have And what of prevention?


“adequate” funding for HIV/AIDS, the situ-
ation is more nuanced: Some donor funds Most international resources are earmarked
cannot be applied flexibly, leaving countries for treatment. But the only way to stem the
with important gaps even when they appear need for treatment and save lives is to expand
to be highly funded in aggregate terms. This is prevention initiatives.
where the unearmarked flexibility of the Inter- An in-depth evaluation of the U.S. PEPFAR
national Development Association becomes program concluded that it reduced deaths by 5
critical. percent but had no effect on prevention.40 The
The highest-prevalence regions of Africa recent multimillion dollar evaluation of the
receive the bulk of external funding (figure Global Fund noted the organization’s neglect
2A.1), but financing per current AIDS patient of prevention.41 A more modest assessment of
paints a different picture (figure 2A.2). Although the programs of the World Bank, Global Fund,
there is a general correlation between the num- and PEPFAR also concluded that prevention
ber of patients and funding across countries, was the weak link.42 The challenge is that for
financing available for each patient still lags in every HIV/AIDS patient placed on treatment,
the highest-prevalence regions of Africa. two or three newly infected people will need
Greater efficiency is imperative because the treatment for life.43
agenda has broadened and the pace of infec- Countries that have prioritized preven-
tion has not slowed. Targeting high-risk groups tion—Brazil, Rwanda, and Thailand—have
and improving management and efficiency in seen prevalence decline or remain low, despite
delivery can raise quality and efficiency. The spiraling levels in the early 1990s. Prevalence
Bahamas plan greater use of generic drugs, bet- rates in these countries contrast with those in
ter patient adherence to treatment protocols, Botswana and Swaziland, which have strug-
and a sharper focus on the cost effectiveness of gled to initiate effective prevention programs
purchases and service delivery. While not cost- as prevalence reached epidemic proportions.
less, such improvements will boost effective- The long-term trends reflect lack of attention to
ness and reduce waste, which are equivalent to prevention 5–10 years ago. But current preven-
reducing costs. They also raise the quality of tion efforts remain inadequate, and the crisis
services including health care services. could further curtail such efforts if constrained

FIGURE 2A.1 Projected Global Fund to Fight AIDS, Tuberculosis, and Malaria and U.S. PEPFAR
HIV/AIDS grants as of April 2009

2,500

2,000
US$, millions

1,500

1,000

500

0
East Asia Europe and Latin America Middle East South East Southern West Central
and Pacific Central Asia and the and North Asia Africa Africa Africa
Carribbean Africa Sub-Saharan Africa

Round 8 approved grants Rounds 1–7 active grants not yet disbursed Project fiscal 2009

Source: Lewis 2009.


64 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

FIGURE 2A.2 Projected Global Fund to Fight AIDS, Tuberculosis, and Malaria and U.S. PEPFAR
HIV/AIDS grants per AIDS patient as of April 2009

12

10
US$, millions

0
East Asia Europe and Latin Middle East South East Southern West Central World
and Pacific Central Asia America and North Asia Africa Africa Africa
and the Africa Sub-Saharan Africa
Carribbean

Source: Lewis 2009.

budgets force cutbacks in prevention. It is a schools during covariate shocks. However,


dynamic problem; new infections occur daily, some of these more adverse effects may be oc-
and so a continuous, uninterrupted response is curring in conflict or disastrous situations with
required. It may take 7–10 years for a person to institutional breakdowns so that microstudies
are not available.
become symptomatic, but even people without
2. For the entire sample of developing countries,
evident symptoms can pass on the virus and 47 percent of the 4,415 country-year obser-
infect others. Actions now will reduce the rate vations are classified as growth accelerations
at which people with the virus can pass it on, while 11 percent are classified as growth de-
underscoring the importance of antiretroviral celerations. The remaining 42 percent of ob-
therapy as a prevention measure. servations are for years in which countries
The Bill & Melinda Gates Foundation and experienced neither growth acceleration nor
others are financing extensive efforts in pre- deceleration episodes.
vention technologies, and considerable ongo- 3. To some extent this pattern may be endoge-
ing research is exploring how to discourage nous, because average income per capita tends
risky behaviors. But equal attention must go to rise in countries with more frequent growth
to actually promoting behavior change and accelerations and fall in countries with more
rolling out promising approaches where pre- frequent collapses.
4. Arbache and Page 2007.
vention lags. Because programs for prevention
5. Arbache and Page 2010.
are dwarfed by those for treatment, the bal-
6. Arbache, Go, and Page 2008. The inflation
ance deserves some recalibration to spare those
figure would have been higher had Zimbabwe
not yet infected. While neither simple nor easy, been included; it was excluded from the analy-
a push to expand prevention is warranted if sis because of missing data for other variables.
there is to be progress on Goal 6A: halting the 7. The analysis is taken from Lewis and Verhoev-
spread of HIV/AIDS by 2015. en (2010) and relies on data from the Interna-
tional Monetary Fund (IMF), the World Bank,
United National Educational, Scientific, and
Notes Cultural Organization (UNESCO, education
1. Arbache, Go, and Korman 2010. Although spending), and the World Health Organization
the aggregate figures show girls’ education is (WHO) National Health Accounts (health
affected by growth cycles, there is still a lack spending).
of microstudies that show girls are dispro- 8. The absence of a consistent time series in
portionately more likely to be pulled out of education spending data required the integra-
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 65

tion of data from UNESCO, the IMF, and the 34. The Global Fund has initiated a new $2.6 bil-
World Bank. This is in contrast to the consis- lion funding window for the next two years,
tent and much higher quality data from WHO which it estimates is insufficient. These re-
National Health Accounts. quirements are not addressed here because the
9. Dang, Knack, and Rogers 2009. focus is on financing HIV/AIDS prevention
10. Gottret and others 2009. and treatment.
11. Grosh and others 2008. 35. Kaiser Family Foundation (2009) and www.
12. Blomquist and others 2002. KKF.org provide updates of spending.
13. Ferreira and Schady 2009. 36. Case and Paxson 2009.
14. World Bank forthcoming. 37. A survey of UNAIDS and WHO country of-
15. IMF 2009. fices by the World Bank, UNAIDS, and WHO
16. High-Level Seminar on Africa Fiscal Policy for (2009) asked about possible issues as the crisis
Growth in Light of the Global Crisis, Maputo, evolved and the likely impact on HIV/AIDS
December 2009, sponsored by the World Bank programs over the next 6–12 months.
and various governments. 38. UNAIDS 2009.
17. Turk 2009. 39. Global Fund (www.theglobalfund.org/programs/
18. Global Fund (www.theglobalfund.org/pro- search/?lang=en&round=9).
grams/search/?lang=en&round=9). 40. Bendavid and Bhattacharya 2009.
19. Wodon and Zaman 2010. 41. Sherry, Mookherji, and Ryan 2009.
20. These data do not include safety net and nutri- 42. Ooman, Bernstein, and Rosenzweig 2007.
tion interventions under the World Bank’s 43. Revenga and others 2006.
Global Food Crisis Response Program, which
has funded an estimated $380 million for
safety net interventions in 21 countries, includ- References
ing grant funding for small, targeted projects in
17 low-income IDA-eligible countries (totaling Alderman, H., J. Hoddinott, and B. Kinsey. 2006.
$95 million). “Long-Term Consequences of Early Childhood
21. See Grosh and others (2008) for detailed Malnutrition.” Oxford Economic Papers 58
discussion of the design and implementation (3): 450–74.
challenges of effective safety nets. Arbache, J., D. Go, and V. Korman. 2010. “Does
22. World Bank 2006. Growth Volatility Matter for Development Out-
23. Ratha, Mohapatra, and Silwayl 2009. comes? An Empirical Investigation Using Global
24. World Food Programme 2009. The study ana- Data.” Background paper for Global Monitoring
lyzes 126 countries and focuses on five case Report 2010. World Bank, Washington, DC.
studies. Arbache, J., D. Go, and J. Page, eds. 2008. Africa
25. Countries are rated according to performance at a Turning Point? Growth, Aid, and External
in 16 areas grouped into four clusters: eco- Shocks. Washington DC: World Bank.
nomic management, structural policies, poli- Arbache, J., and J. Page. 2007. “More Growth or
cies for social inclusion and equity, and public Fewer Collapses? A New Look at Long-Run
sector management and institutions. Growth in Sub-Saharan Africa.” Policy Research
26. Arbache, Go, and Page 2008. Working Paper 4384. World Bank, Washington,
27. See, for examples, Wagstaff and Claeson DC.
(2004); Rajkumar and Swaroop (2008); and ———. 2010. “How Fragile Is Africa’s Recent
Filmer and Pritchettt (1999). Growth?” Journal of African Economies 19 (1):
28. de Walque and others 2010. 1–25.
29. Tiwari and Zaman 2010. Bendavid, E., and J. Bhattacharya. 2009. “The
30. Turk and Mason 2009. President’s Emergency Plan for AIDS Relief in
31. Khanna, Newhouse, and Paci, forthcoming. Africa: An Evaluation of Outcomes.” Annals of
32. See Arbache, Go, and Korman (2010) for Internal Medicine 150 (10): 688–95.
more discussion. Blomquist, J., J. P. Cordoba, M. Verhoeven, C.
33. The figure includes international and domestic Bouillon, and P. Moser. 2002. “Social Safety
philanthropic contributions, World Bank fi- Nets in Response to Crisis: Lessons and Guide-
nancing, government expenditures, and house- lines from Asia and Latin America.” In Towards
hold spending, but it excludes other multilat- Asia’s Sustainable Development: The Role of
eral and private sector funding. Social Protection, pp. 297–332. Paris: OECD.
66 L E S S O N S F R O M PA S T C R I S E S GLOBAL MONITORING REPORT 2010

Bruni, A. Mason, Pabon, and C. Turk. Forthcom- from Crisis Response Survey.” ECSHD Work-
ing. “Gender Dimensions of the Crisis in East ing Paper. World Bank, Europe and Central Asia
Asia and the Pacific.” World Bank, Washington, Region, Human Development Unit, Washing-
DC. ton, DC.
Case, A., and C. Paxson. 2006. “Stature and Status: IMF (International Monetary Fund). 2005. “Petro-
Height, Ability, and Labor Market Outcomes.” leum Product Subsidies: Costly, Inequitable, and
NBER Working Paper 12466. National Bureau Rising.” IMF Staff Position Note SPN/10/05.
of Economic Research, Cambridge, MA. Washington, DC.
Dang, H., S. Knack, and H. Rogers. 2009. “Inter- ———. 2009. “Fiscal Policy in Sub-Saharan Africa
national Aid and Financial Crises in Donor in Response to the Impact of the Global Crisis.”
Countries.” Policy Research Working Paper IMF Staff Position Note SPN/09/10. Washing-
5162. World Bank, Washington, DC. ton, DC.
De Walque, D., H. Kazianga, M. Over, and J. Vail- Kaiser Family Foundation. 2009. Updates of vari-
lant. 2010. “The Impact of the Food Crisis on ous information, including “The U.S. President’s
HIV/AIDS Treatment and Health Outcomes Emergency Plan for AIDS Relief (PEPFAR),”
in Mozambique.” Paper presented at World Fact sheet on U.S. Global Health Policy. Wash-
Bank Development Economics Development ington, DC (November) (www.kff.org).
Research Group Workshop, January 13, Wash- Khanna, G., D. Newhouse, and P. Paci. Forthcom-
ington, DC. ing. “Fewer Jobs or Smaller Paychecks? Labor
Dinkelman, T. 2008. “The Long-Term Effects of Market Impacts of the Financial Crisis in Mid-
Being Born in a Drought: Evidence from the dle Income Countries.” World Bank, Washing-
Cape Area Panel Study 2002–06.” Paper pre- ton, DC.
sented at the Northeastern Universities Devel- Laing, R., and I. Buysse. 2010. “Impact of the
opment Consortium Conference, November Global Crisis on the Global Pharmaceutical Sec-
8–9, Boston University, Boston, MA. tor.” Paper presented (January 7) at University
Ersado, L. Forthcoming. “The Global Economic of Utrecht, the Netherlands.
Crisis in Armenia: Policy Responses and House- Lewis, M.. 2009. “Likely Impacts of the Crisis on
hold Coping Strategies.” World Bank Working HIV/AIDS Programs.” Paper presented to HIV/
Paper. Washington, DC. AIDS Economics Reference Group Meeting,
Ferreira, F., and N. Schady. 2009. “Aggregate April 21, Washington, DC.
Economic Shocks, Child Schooling, and Child Lewis, M., and M. Verhoeven. 2010. “Financial
Health.” World Bank Research Observer 24 (2): Crises and Social Spending: The Impact of the
141–81. 2007–2009 Crisis.” Background paper for
Filmer, D., and L. Pritchett. 1999. “The Impact Global Monitoring Report 2010. World Bank,
of Public Spending on Health: Does Money Washington, DC.
Matter?” Social Science and Medicine 49 (10): Ooman, N., M. Bernstein, and S. Rosenzweig. 2007.
1309–23. Following the Funding for HIV/AIDS. Washing-
Gottret, P., V. Gupta, S. Sparkes, A. Tandon, V. ton, DC: Center for Global Development.
Moran, and P. Berman. 2009. “Protecting Pro- Pritchett, L. 2000. “Understanding Patterns of
poor Health Services during Financial Crises: Economic Growth: Searching for Hills among
Lessons from Experience.” In Innovations in Plateaus, Mountains, and Plains.” World Bank
Health System Finance in Developing and Tran- Economic Review 14: 221–50.
sitional Economies: Advances in Health Eco- Raddatz, C. 2007. “Are External Shocks Responsi-
nomics and Health Services Research, vol. 21, ble for the Instability of Output in Low-Income
ed. Dov Chernichovsky and Kara Hanson. Bin- Countries?” Journal of Development Econom-
gley, U.K.: Emerald Group Publishing. ics 84 (1): 155–87.
Grosh, M., C. del Ninno, C. Tesliuc, and A. ———. 2008a. “External Shocks and Macro-
Ouerghi. 2008. For Protection and Promotion: economic Volatility in Latin America.” World
The Design and Implementation of Effective Bank, Washington, DC.
Safety Nets. Washington, DC: World Bank. ———. 2008b. “Have External Shocks Become
Hausmann, R., L. Pritchett, and D. Rodrik. 2005. More Important for Output Fluctuations in
“Growth Accelerations.” Journal of Economic African Countries?” In Africa at a Turning
Growth 10 (4): 303–29. Point? Growth, Aid, and External Shocks, ed.
Hirshleifer, S., and M. Azam. Forthcoming. “Impact J. Arbache, D. Go, and J. Page. Washington,
of the Economic Crisis in Montenegro: Evidence DC: World Bank.
GLOBAL MONITORING REPORT 2010 L E S S O N S F R O M PA S T C R I S E S 67

Rajkumar, A., and V. Swaroop. 2008. “Public of Findings from Eight Countries.” Internal dis-
Spending and Outcomes: Does Governance cussion paper. World Bank Social Development
Matter?” Journal of Development Economics Department, Washington, DC.
86 (1): 96–111. Turk, C., and A. Mason. 2009. “Impacts of the Eco-
Ratha, D., S. Mohapatra, and A. Silwal. 2009. nomic Crisis in East Asia: Findings from Quali-
“Migration and Remittance Trends 2009: A tative Monitoring in Five Countries.” Paper
Better-than-Expected Outcome So Far, but Sig- presented at the 3rd China–ASEAN Forum on
nificant Risks Ahead.” Migration and Develop- Social Development and Poverty Reduction and
ment Brief 11. World Bank, Washington, DC. 4th ASEAN +3 High Level Seminar on Povery
Revenga, A., M. Over, E. Masaki, W. Peerapata- Reduction, September 28–30, Hanoi. (www.
napokin, J. Gold, V. T. Sathien, and S. Thanpra- adb.org/Documents/Events/2009/Poverty-
sertsuk. 2006. “The Economics of Effective Social-Development/papers.asp).
AIDS Treatment Evaluating Policy Options for UNAIDS (Joint United Nations Programme
Thailand.” World Bank, Washington, DC. on HIV/AIDS). 2009. “25th Meeting of the
Sabarwal, S., N. Sinha, and M. Buvinic. 2009. UNAIDS Programme Coordinating Board.”
“Aggregate Economic Shocks and Gender Dif- Geneva (December 8–10).
ferences: A Review of the Evidence.” World Wagstaff, A., and M. Claeson. 2004. Rising to the
Bank, Washington, DC. Challenges: The Millennium Development Goals
Sherry, J., S. Mookherji, and L. Ryan. 2009. The for Health. Washington, DC: World Bank.
Five-Year Evaluation of the Global Fund to Wodon, Q., and H. Zaman. 2010. “Higher Food
Fight AIDS, Tuberculosis and Malaria. Calver- Prices in Sub-Saharan Africa: Poverty Impact
ton, MD: ICF Macro International. and Policy Responses.” World Bank Research
TEPAV (Economic Policy Research Foundation), Observer 25 (1): 157–76.
UNICEF (United Nations Children’s Fund), and World Bank. 2006. Global Economic Prospects:
World Bank. 2009. “The Economic Crisis and Economic Implications of Remittances and
the Welfare of Families: Results from the Five Migration. Washington, DC: World Bank.
Urban Centers in Turkey.” Ankara. ———. Forthcoming. “El Salvador Public Expen-
Tiwari, S., and H. Zaman. 2010. “The Impact of diture Review.” World Bank, Washington, DC.
Economic Shocks on Global Undernourish- ———. Various years. World Development Indica-
ment.” Policy Research Working Paper 5215. tors. Washington, DC: World Bank.
World Bank, Washington, DC. World Food Programme. 2009. “Households Go
Turk, C. 2009. “Rapid, Qualitative Assessments of Hungry As Financial Crisis Bites.” Rome (June
the Impacts of the Economic Crisis: Overview 11).
3
Growth Outlook and
Macroeconomic Challenges
in Emerging Economies and
Developing Countries

T
he recovery of the global economy countries off their track of solid growth?
has been more robust than expected. The question is especially important for low-
Driven by strong internal demand in income countries because poverty is so much
many emerging economies and the recovery more pressing there than in countries with
of global trade, GDP growth in emerging and higher incomes. History does not suggest that
developing countries is projected to accelerate low-income countries can uniformly escape
to 6.3 percent in 2010, from 2.4 percent in global shocks without absorbing long-lasting
2009. Supporting the economic recovery are damage to both growth and welfare. In past
expansionary macroeconomic and, especially, crises, it has often taken several years for low-
fiscal policies. Fiscal deficits in emerging and income countries to bring growth rates back
developing countries, up by almost 3 percent into positive territory. Even so, the turnaround
of GDP in 2009, are projected to remain high in low-income countries this time is projected
in 2010. More than in previous crises, many to be faster than in previous crises, thanks to
countries sustained spending plans and raised countercyclical fiscal policies and better mac-
social spending to mitigate the effects of the roeconomic fundamentals in place at the be-
downturn on the poorest people, although the ginning of the crisis. Commodity exporters
differences among countries are wide. While are helped by the fairly quick recovery of com-
financial market conditions for emerging and modity prices. And financial systems in low-
developing countries are improving and capital income countries have been less affected by
flows are returning, international bank financ- turmoil than those in advanced economies.
ing and foreign direct investment are projected The recovery is still vulnerable, however,
to remain weak in 2010. and the rapid expansion of fiscal deficits and
Although the short- and medium-term the greater reliance on domestic sources of fi-
growth prospects for most emerging and de- nancing in many countries may not be sustain-
veloping countries are positive, the question able. External debt ratios in low-income coun-
arises: to what extent does the current shock tries, deteriorating in the short run, should be
have longer-run implications that could knock watched.

GLOBAL MONITORING REPORT 2010 69


70 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

Optimal exit policies from policy support tion and investment, private demand growth
should depend on country circumstances. in emerging Europe is expected to remain
sluggish, and several countries remain depen-
• Countries where private demand is still dent on exceptional policy stimulus. Com-
weak should continue supporting activity if modity exporters are benefiting from firmer
policy space is available. global demand for raw materials and higher
• Some countries, however, are facing financ- commodity prices. Even so, the recovery re-
ing constraints—they cannot delay adjust- mains vulnerable, most notably in advanced
ment. Donors should assist them by follow- countries and the economies of Eastern Eu-
ing up on commitments to increase aid. rope, where high unemployment, moderate
• All countries should adopt credible medium- income growth, and weaker household bal-
term fiscal adjustment plans to bolster con- ances are dampening consumption growth,
fidence in macroeconomic policies and un- posing risks for the global outlook. In addi-
dertake policy reforms to secure long-term tion, in the medium–term, growth rates in
growth. some groups of countries, especially low-
income countries, are not expected to reach
the high levels recorded before 2008.
The economic recovery Because the recovery is in an early stage and
Global economic activity is recovering from the unemployment rates are still elevated, global
deepest recession since the Second World War, inflation has remained low, although some
albeit at a moderate pace. According to the economies, especially in Asia, are showing the
International Monetary Fund’s (IMF) World first signs of price pressures. Inflation risks are
Economic Outlook, growth of global output rising in Latin America as well, where output
will increase to 4.2 percent in 2010, from a gaps in some countries are closing rapidly.
decline of 0.6 percent in 2009 (table 3.1). The
recovery, supported by improving financial
Commodity prices are recovering
conditions and rising world trade (figure 3.1),
is led by emerging economies in Asia, where Following the sharp drop in commodity prices
growth rates now exceed precrisis levels. The in late 2008, prices for most commodities re-
prospects for developing countries, including bounded sharply in 2009 and are continuing
the poorest, are improving as well, although their upward trend in 2010 as the global re-
growth rates have not yet recovered to the lev- covery gains momentum (figure 3.2). The in-
els seen in the years before the crisis.1 creases are helping to mitigate the impact of
The underlying factors driving the expan- the crisis on commodity exporters. Food prices
sion differ from country to country. While are the exception, because good harvests in
economies in Asia and Latin America are Sub-Saharan Africa and elsewhere have given
bolstered by a recovery of private consump- an opportunity to rebuild stocks. But food

TABLE 3.1 Global output


percent change
Projection

Region 2007 2008 2009 2010 2011–13

World output 5.2 3.0 –0.6 4.2 4.4


Advanced economies 2.8 0.5 –3.2 2.3 2.4
Emerging and developing economies 8.3 6.1 2.4 6.3 6.6
Central and Eastern Europe 5.5 3.0 –3.7 2.8 3.8
Commonwealth of Independent States 8.6 5.5 –6.6 4.0 4.1
Developing Asia 10.6 7.9 6.6 8.7 8.6
Middle East and North Africa 5.6 5.1 2.4 4.5 4.8
Sub-Saharan Africa 6.9 5.5 2.1 4.7 5.7
Western Hemisphere 5.8 4.3 –1.8 4.0 4.2
Source: World Economic Outlook.
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 71

FIGURE 3.1 Short-term indicators of production and trade are recovering

130 a. Exports and imports (value) 115 b. Industrial production

120 110
emerging markets
105
110
index (June 2008 = 100)

index (June 2008 = 100)


G3 countries
100
100
95
90
G3 countries 90
80
85
emerging markets
70 80
60 75
50 70
Jan. May Sept. Jan. May Sept. Jan. May Sept. Jan. May Sept. Jan. Jan. May Sept. Jan. May Sept. Jan. May Sept. Jan. May Sept. Jan.
2006 2007 2008 2009 2010 2006 2007 2008 2009 2010

Source: IMF International Financial Statistics; Bloomberg; Haver Analytics; central banks.
Note: Data are weighted by PPP-GDP, 2006.

FIGURE 3.2 Commodity price indexes rebounded strongly in 2009

600

500
index (2001 Q1 = 100)

400 copper

300 fuel
all commodities
200 food
non-fuel
100 cereals

0
Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1a Q3a
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2011

Source: IMF.
Note: Indexes are in U.S. dollars.
a. Projected

and commodity prices, relatively high by his- ably since the onset of the crisis. Bond spreads
torical standards, are projected to remain so, have declined, stock markets in both emerg-
given the prospects for further medium-term ing and developing countries have recovered
demand growth and continuing supply con- sharply, and exchange rate volatility has come
straints in many sectors. down considerably (figures 3.3–3.5). Some
borrowers—sovereigns and prime corpora-
tions in particular—quickly regained market
Financial conditions are improving,
access following a brief interruption at the
but financial flows remain below
end of 2008. Financial market access for sub-
precrisis levels
investment-grade borrowers in emerging and
Financial market conditions for emerging and developing countries has also improved. But
developing countries have improved consider- access to international bank financing remains
72 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

FIGURE 3.3 Bond spreads have declined in emerging markets and developing countries

100 1,000
90 900
80 EMBI Global issues of 800
(right axis) international
70 bonds (left axis) 700

spreads, basis points


US$, billions

60 600

50 500
40 400

30 300
20 200
10 100
0 0
Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q1
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Source: Dealogic; Bloomberg.


Note: Bond issues and spreads as of end-March 2010..

FIGURE 3.4 Share prices have recovered sharply

800

700

600 developing countries


index (January 2000 = 100)

500

400

300
emerging markets
200

100

0
Jan. Sept. May Jan. Sept. May. Jan. Sept. May Jan. Sept. May. Jan. Sept. May. Feb.
2000 2000 2001 2002 2002 2003 2004 2004 2005 2006 2006 2007 2008 2008 2009 2010

Source: IMF International Financial Statistics.


Note: Prices are in the local currency.

limited as banks in advanced economies con- such as liquidity support, deposit insurance,
tinue deleveraging. bank interventions, and recapitalizations.
Financial policies, such as improved finan- Banking sectors in many emerging economies
cial sector regulation and crisis measures, have have also benefited from higher financial mar-
contributed to avoidance of widespread bank- ket resilience, including less volatility in ex-
ing crises in emerging and developing coun- change and interest rates,2 and therefore have
tries. The public response to the financial crisis avoided negative dynamics from balance sheet
has been broad, covering several instruments, effects.
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 73

FIGURE 3.5 Exchange rates have been less volatile: Daily spot exchange rates

180

160
index (Jan. 1, 2008 = 100)

140

120

100

80

60

40
Jan. Mar. May July Sept. Nov. Jan. Mar. May July Sept. Nov. Jan. Mar.
2008 2009 2010
Mexico Brazil Indonesia Thailand
South Africa Kenya Poland

Source: Bloomberg.
Note: Exchange rates are in national currency per U.S. dollars.

FIGURE 3.6 The cost of external debt financing has come down

12 900
a. Yields (left axis) b. Spreads (right axis)
11
10 700
9

basis points
percent

8 500
7
6 300
5
4 100
composite high grade high yield composite high grade high yield

Jun. 30, 2007 Sept. 1, 2008 Apr. 12, 2010

Source: Bloomberg.

Even so, concerns about systemic risks to has also affected bank loan portfolios in many
the solvency of banks and corporations linger. countries, as evidenced by the rising shares of
The cost of external debt financing remains el- nonperforming loans (figure 3.7).
evated in some emerging and developing coun- Despite the general improvement in market
tries, where spreads on high-yield external cor- conditions, financial flows to emerging and
porate bonds are still substantially above those developing countries have not recovered to
before the collapse of Lehman Brothers in Sep- those seen in the years preceding the financial
tember 2008 (figure 3.6). In addition, some crisis (table 3.2). In emerging economies, net
countries in Eastern Europe and the Common- inflows of foreign financial resources (capital
wealth of Independent States (CIS) continue to flows and transfers) are not expected to ex-
face uncertainties as a result of high external ceed 8.2 percent of GDP this year, down from
debt refinancing needs and private sector for- an average of about 12 percent in 2007–08,
eign currency debt. The fallout from the crisis mainly because of the sharp drop in bank
74 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

FIGURE 3.7 The share of nonperforming loans to total loans has recovery of workers’ remittances (table 3.3).
been rising Although remittances to countries in Latin
America, North Africa, and the Middle East
12
were weaker than expected in 2009, they ap-
10 pear to have reached a bottom toward the end
of the year. At the same time, remittance flows
8
to South and East Asia, largely originating in
percent

6 the Gulf countries, surprised on the upside


4
in 2009, with particularly strong increases
in Bangladesh and Pakistan. Overall, remit-
2 tances to emerging and developing countries
0 are projected to increase by 2 percent in 2010,
2003–07 2008 2009 following a 6 percent decline in 2009.
emerging markets developing countries Current account imbalances in emerging
advanced countries and developing countries have been shifting in
recent years, mainly as the result of the sharp
Source: IMF 2009b. swings in world trade and terms of trade since
late 2008 (figure 3.9). Fuel-exporting coun-
TABLE 3.2 Net financial flows tries have been most hit by fluctuations in
percent of GDP the external accounts, a reflection of the high
Flows 2007 2008 2009 2010 volatility of oil prices and insufficient export
diversification. Nonfuel primary product ex-
Emerging market economies 12.6 11.4 8.7 8.2
Private capital flows, net 8.0 7.0 3.2 3.1
porters face strong fluctuations as well, but
of which: private direct investment 5.4 5.1 3.3 3.3 less so than fuel-exporting countries. Despite
Private portfolio flows 0.8 –0.5 –0.3 0.1 these fluctuations, there have been reductions
Private current transfers 4.1 3.7 3.8 3.6 in external imbalances in the past two years
Official capital flows and transfers, net 0.4 0.7 1.7 1.6 within the group of emerging and developing
Memorandum item:
Reserve assets –3.9 –1.6 –2.5 –1.9 countries. The number of emerging economies
with high balance of payments deficits and the
Developing countries 14.0 17.7 13.9 13.9 number of high surplus emerging economies
Private capital flows, net 6.6 7.7 5.2 5.3 and developing countries declined in 2009
of which private direct investment 6.6 6.2 4.8 4.7
(figure 3.10).
Private portfolio flows –0.7 –0.6 –0.4 –0.2
Private current transfers 5.6 5.8 5.2 5.1 Even with the large differences in external
Official capital flows and transfers, net 1.8 4.2 3.6 3.5 conditions among emerging and developing
Memorandum item: countries, there has been a remarkable simi-
Reserve assets –4.0 –2.3 –1.6 –1.0 larity in international reserve developments
Source: World Economic Outlook. across groups of countries and regions. Helped
Note: Equally weighted.
by the recovery in international trade and cap-
financing (figure 3.8), especially in Asia and ital flows, and the allocation of IMF special
drawing rights, almost all countries rebuilt
Latin America, and foreign direct investment.
international reserves (as measured by reserve
Developing countries are facing weak foreign
coverage in months of imports) in 2009, after
direct investment activity as well, because a decline in 2008 (figure 3.11). At the end of
overcapacity in extractive industries remains 2009, 80 percent of emerging markets and 75
considerable despite rising global demand for percent of developing countries had reserves
commodities. Overall, net financial flows are that could be considered adequate (equivalent
projected to decline to 13.9 percent of GDP in to three months of imports of goods and ser-
2010, from 15.9 percent in 2007–08. vices). For emerging economies, reserves as a
The drop in foreign direct investment in share of short-term debt also increased, and at
developing countries is partly offset by the the end of 2009 about 70 percent of emerg-
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 75

FIGURE 3.8 Bank financing to emerging markets dropped sharply in 2009

20

15
% change from previous quarter

10

–5

–10

–15

–20
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
2006 2007 2008 2009

Asia emerging Europe Latin America Middle East/Africa

Source: Bank for International Settlements (BIS).


Note: Adjusted for exchange rate changes. Changes are calculated as flow adjusted for exchange rate changes as a share of the stock in the previous quarter.

TABLE 3.3 Inflows of international remittances


US$billions
Annual average Annual average
1992–2002 2003–07 2008 2009a 2010b 2011c

Emerging market economies 58.9 177.2 283.3 266.0 271.7 279.1


Developing countries 8.4 26.1 47.1 46.4 47.5 48.8
Fragile states 2.2 5.1 9.7 8.6 8.8 9.1
Source: World Bank remittances data.
a. Remittances include workers’ remittances, compensation of employees, and migrant transfers.
b. Estimate.
c. Forecast base case scenario.

ing economies had reserves that exceeded the often not an option in previous crises. Further,
stock of short-term debt.3 stronger balance sheets and continued access
to financing, especially for prime borrowers,
helped private corporations in emerging and
Thanks to good policies, the recovery is
developing countries to deal better with ad-
stronger than in past crises
verse conditions than they had in the past. Lo-
Overall, emerging and developing countries cal bond markets have also benefited some of
weathered this global crisis better than past these countries, with larger enterprises in Asia
ones. Their financial markets and exchange and Latin America able to rely on local mar-
rates have not shown the sharp fluctuations kets for their refinancing needs.
of past crises, and the rebound in economic
activity is stronger than expected. Healthier
Nonetheless, the crisis has depressed
fiscal accounts, reduced debt, better debt ma-
disposable incomes in many countries
turity structures, low inflation, and higher
international reserves gave many countries The crisis and the pace of recovery have deeply
room for countercyclical policies that were affected disposable incomes in many coun-
76 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

FIGURE 3.9 Changes in terms of trade have swung sharply since have differed. In general, most emerging and
2008 developing countries first focused on address-
ing weakening confidence and containing the
25 2009
impact of the financial market crisis on the
20
15 2008 real economy. In a second stage, these policies
2006
10 have been followed by comprehensive efforts
5 2010 2006 2008 2006 2008 to support domestic demand and growth in
% change

2010
2007 2009 2010
0
2007
the medium term—mainly through expansion-
2007
–5 ary macroeconomic policies. In most countries
–10 these policies are still in place, and the start
–15
of the third stage—exiting from extraordinary
–20
–25
2009 policy support—has been gradual thus far.
lowest-quintile highest-quintile all countries
countries countries
Monetary policy provided support in
Source: IMF staff calculations. most countries
Note: Quintile groups are based on the average of terms-of-trade changes in 2008–09 and 2009–10.
Aided by moderate inflation trends and less
volatile exchange rates, central banks in most
tries, where a contraction in real activity was emerging and developing countries reduced
sometimes reinforced by a deterioration in the policy interest rates in 2009. About 70 percent
terms of trade (figure 3.12). In 2008–10, about of emerging economies and close to 60 percent
a third of emerging and developing countries of developing countries followed a path toward
were experiencing declines in disposable in- lower rates last year. In some countries, higher
comes, with potentially serious adverse effects policy interest rates were initially needed to
on poverty. Central and Eastern Europe has preserve market confidence. These increases
been particularly hard hit, with nine countries were more modest than in previous crises,
facing cumulative income declines, in total av- however, and in many cases were quickly re-
eraging more than 8 percent. versed. In most countries, lower interest rates
were associated with depreciations of nominal
effective exchange rates. As a result, monetary
Macroeconomic policy trends conditions in most emerging and developing
Reflecting cross-country differences in initial countries—as measured by a simple summary
conditions and the international transmission indicator incorporating nominal interest rates
of the crisis, macroeconomic policy responses and nominal effective exchange rates4—ap-

FIGURE 3.10 External imbalances have come down in emerging and developing countries

a. Current account deficit > –5% of GDP b. Current account surplus > –5% of GDP
70 16
60 14
number of countries

number of countries

50 12
10
40
8
30
6
20
4
10 2
0 0
emerging markets developing countries emerging markets developing countries

2008 2009 2010 2008 2009 2010

Source: World Economic Outlook.


GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 77

FIGURE 3.11 Almost all countries rebuilt their international reserves in 2009

7 2.5

external short-term debt


2.0

in terms of the stock of


months of imports

4 1.5

3 1.0
2
0.5
1
0 0
2006 2007 2008 2009 2006 2007 2008 2009
emerging markets developing countries

Source: World Economic Outlook.


Note: The median ratio is shown. “Stock of the external short-term debt” = outstanding (on remaining maturity basis) plus amortization scheduled on
medium- and long-term debt.

pear to have become more accommodating in FIGURE 3.12 Deteriorating terms of trade sometimes reinforce
2009 (figure 3.13).5 contraction in economic activity
The financial crisis resulted in a sharp de-
30
cline in money growth in emerging and devel- 8
25
oping countries (figure 3.14). The decline was 20
largest in countries that had seen the strongest 15
10
% change

growth in the precrisis years. But as a result 14


10 28
17 5
of the even stronger decline in nominal GDP 5
growth rates, measures of excess liquidity, 0
such as the nominal money gap, increased. –5 9 2
4 11 4
This suggests that despite the fall in money –10 9
–15
growth, additional liquidity remained avail- Africa Middle East Asia Western Central CIS
able to support corporations and households and North Hemisphere and
Africa Eastern
during the crisis period. Europe
positive terms of trade effects negative terms of trade effects
Expansionary fiscal policies support
the recovery Source: IMF Staff calculations.
Note: The figure shows, by region, average real per capita GDP growth rates adjusted for the per
capita value of net terms-of-trade changes. The numbers above and below the bars show the
Measured by the median general govern- number of countries.
ment balance, fiscal deficits in emerging and
developing countries expanded by almost 3
percent of GDP in 2009 (figure 3.15) and are corporate tax revenue as the contribution of
projected to increase further in 2010 in more key sectors in the economy (such as natural
than one-third of the countries, despite some resources and other export sectors) declined.
decline in the median balance. Some coun- Moreover, tax administrations may be facing
tries, especially emerging economies, have put bigger enforcement challenges during the cri-
stimulus plans in place. But in most countries sis and its aftermath as tax planning becomes
the widening deficit is the result of weaken- more aggressive. Many countries are more
ing revenue, including the disproportionate exposed to such challenges because of their
impact of the crisis on trade—and thus on weak administrative capacity, large informal
revenues from import tariffs—and on con- sectors, and the constrained cash positions of
sumption taxes. Some countries have also lost taxpayers.
78 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

FIGURE 3.13 Monetary policy conditions became more accommodating in 2009

a. Emerging markets with monetary b. Other developing countries with


policy loosening monetary policy loosening
90 80
80 70
70 60
60
50
percent

percent
50
40
40
30
30
20 20
10 10
0 0
MCI discount rate exchange rate MCI discount rate exchange rate
2008 2009 2008 2009

Source: IMF International Financial Statistics.


Note: Monetary policy loosening is based on Monetary Conditions Index (MCI) calculations.

FIGURE 3.14 Average year-on-year growth in money and the money gap in emerging markets

30
25

20
growth rate, %

M2
15

10
5 money gap

0
–5
Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3
2003 2004 2005 2006 2007 2008 2009

Source: IMF International Financial Statistics; Haver Analytics.


Note: The money gap is the difference between year-on-year growth rates of the M2 money supply and nominal GDP. The sample includes emerging-
market countries that have data on both for the whole sample period shown.

Despite falling revenues, emerging and de- exporting countries that faced sharp terms-of-
veloping countries as a group have allowed trade deteriorations after the collapse of oil
automatic stabilizers to work and have main- prices in the second half of 2008. Expendi-
tained previous spending plans during the tures were less affected in other emerging and
financial crisis. To some extent they have in- developing countries, especially nonfuel com-
creased social spending related to the crisis, modity exporters. Thanks to higher oil prices,
supporting domestic demand and sustain- many fuel-exporting countries will be able to
ing the recovery. But the overall numbers on reverse these policies in 2010.
spending conceal wide differences in policy
stances and conditions. About half of emerg-
But many countries are not on a
ing and developing countries cut spending in
sustainable fiscal path
2009 in reaction to the crisis, a pattern likely
to be repeated to some extent in 2010 (figure Widening government deficits pose financ-
3.16). The steepest spending cuts were in fuel- ing challenges for many countries, especially
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 79

FIGURE 3.15 Fiscal deficits expanded in 2009 those with limited access to capital markets.
Emerging markets rapidly regained access to
0.0 sovereign debt markets following the collapse
–0.5 of Lehman Brothers in September 2008, but
–1.0 developing countries with limited or no mar-
–1.5 ket access are more constrained in their op-
–2.0 tions. A country-by-country analysis of bud-
% of GDP

–2.5 get financing shows that most countries in this


–3.0
group were able to finance rising deficits with
–3.5
increased domestic and foreign financing. On
–4.0
average, budget financing needs of develop-
ing countries increased by about 3 percentage
–4.5
points of GDP in 2009, about half from do-
–5.0
emerging markets developing mestic sources (mainly domestic bank loans
countries and the drawing down of government deposits
2008 2009 2010a in the banking system) and the rest from for-
eign sources (mainly aid). In some countries,
Source: World Economic Outlook. however, governments could not mobilize sig-
a. Projected.
nificant additional foreign resources despite

FIGURE 3.16 Growth in real primary spending, 2010 projections

a. Fuel exporters b. Nonfuel primary products exporters c. Other countries


Madagascar
Equatorial Guinea Guinea-Bissau
Niger
Burkina Faso Malaysia
Chad
Hungary
Suriname
Azerbaijan Colombia
Zambia Lithuania
Angola Mexico
Mali Turkey
Russian Federation Romania
Guyana
India
Libya
Chile Thailand
Algeria Rwanda
Sierra Leone
Moldova
Venezuela, R.B. de unweighted average unweighted average
Latvia
unweighted average Peru
Poland
Burundi Brazil
Ecuador
South Africa
Mozambique Indonesia
Yemen, Rep.
Bangladesh
Mauritania
Nigeria China
Uzbekistan Ghana
Kazakhstan Pakistan
Guinea-Bissau Kyrgyz Republic
Sudan Argentina
Malawi
Uganda
Gabon Congo, Dem. Rep. of Ethiopia
–30 –20 –10 0 10 20 30 40 50 60 –30 –20 –10 0 10 20 30 40 –20 –15 –10 –5 0 5 10 15 20 25 30

% change % change % change

Source: World Economic Outlook.


80 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

MAP 3.1 How the crisis undermined GDP growth in 2009

GDP real growth rate:


Green
% change (2009) (De

< –6
–6 ≤ < –3
–3 ≤ <0
0≤ <3 Canada

3≤ <7
7≤
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: World Economic Outlook.

pressing needs. As budget deficits remain el- by higher private saving or reduced invest-
evated in 2010, many governments will con- ment. That may occur if consumers and inves-
tinue to borrow heavily in domestic markets. tors adapt their behavior to take into account
This is not sustainable. While fiscal stimulus higher future tax liabilities.
in many developing countries has supported
the recovery, there are risks of crowding out
The macroeconomic policy mix
through higher interest rates. Recent Interna-
tional Monetary Fund (IMF) research shows Most emerging and developing countries as
significant effects of fiscal deficits on interest a group appear to have supported economic
rates, which could dampen private investment activity in 2009 with a combination of expan-
and force governments to spend more on debt sionary fiscal and monetary policies (figure
service payments and less on social programs.6 3.17 and box 3.1). In some countries, expan-
These effects will be stronger when initial defi- sionary fiscal policies were combined with less
cits or debts are high. Expansionary fiscal poli- accommodating monetary conditions. Such a
cies may also become counterproductive if the policy mix is not necessarily incoherent. In fact,
positive demand effects are more than offset it may be useful in countries facing large capital
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 81

IBRD 37736
APRIL 2010

nland
en)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Portugal Spain Turkey Armenia baijan Turkmenistan of Korea
Monaco Greece Tajikistan Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire African Rep. Somalia Palau
Togo Cameroon Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Congo Kenya Singapore Nauru
São Tomé and Príncipe Gabon Rwanda
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

outflows and pressures on the exchange rate. best approach to exit from policy stimulus.7
In such a situation, rising interest rates may be The appropriate timing and nature of exit
appropriate to avoid excessive exchange rate policies depend on individual country circum-
volatility and ensure that sufficient external fi- stances. In many countries, where private de-
nancing remains available for the economy. On mand components are still cyclically weak and
average, growth in countries with this policy sufficient policy space is available, monetary
mix in 2009 was not weaker than in countries and fiscal policy should remain geared toward
that had expansionary policies on both the supporting activity. Governments in these
monetary and fiscal front. countries should lay out a credible exit strategy
to maintain confidence in the authorities’ com-
mitment to macroeconomic stability. Mon-
Adapting monetary and etary and fiscal support should be gradually
fiscal policies to changing removed when private demand is sufficiently
circumstances strong to sustain growth. In addition, to sup-
As the recovery in emerging and developing port fiscal consolidation, reforms to strengthen
countries takes hold, questions arise about the fiscal institutions could be initiated now.
82 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

FIGURE 3.17 Most countries responded with expansionary fiscal strong downward pressures on the exchange
and monetary policy rate (flexible regime) or reserves (fixed regime)
emerge.
a. Emerging markets b. Developing countries Several countries, mainly emerging econo-
mies, face high or rising rates of inflation. The
economic dynamics underlying these phenom-
ena differ from country to country. In several
countries, the first signs of rising inflationary
expectations are becoming visible in the con-
text of exceptional macroeconomic policy
support. In some other countries, inflation
rates remained stubbornly high in 2009, not-
withstanding depressed demand conditions,
monetary and fiscal loosening usually reflecting a lack of confidence in mon-
monetary and fiscal tightening etary policy. In both categories of countries, a
monetary loosening and fiscal tightening move toward a more restrictive monetary and
fiscal loosening and monetary tightening fiscal policy stance would be warranted.
Maintaining confidence in macroeconomic
Source: IMF International Financial Statistics. stability remains a priority for all countries.
Note: Fiscal conditions are defined based on the change in government balance as a percent of GDP
in 2008–09. Monetary conditions are based on the change in the MCI from 2008Q4 to 2009Q3. Credible medium-term fiscal adjustment plans
are important to manage expectations by re-
ducing the risks of crowding out and unsus-
A number of countries, however, are not tainable debt dynamics. To maintain the abil-
in a position to delay adjustment and should ity of fiscal policy to respond to future crises,
act in 2010 to reduce fiscal deficits. In some a preferable strategy would aim to reduce
cases, this reduction should be accompanied debt ratios to their precrisis levels in the me-
by a gradual tightening of the monetary policy dium term. In addition to phasing out tem-
stance. Three broad groups of countries can be
porary stimulus measures, this approach will
distinguished.
require some emerging economies to make
Most developing countries financed widen-
improvements in their structural primary bal-
ing budget deficits in 2009 by increasing reli-
ance.8 To enhance confidence that future fis-
ance on domestic sources of financing. This
cal adjustment will not lead to an appreciable
financing policy, while appropriate when the
increase in the tax burden, the medium-term
global economy was facing the risks of a fur-
ther sharp downturn, cannot be continued for adjustment plans could emphasize the follow-
very long, especially in countries with weak ing elements.
external payments positions, low reserves, or
rapidly rising debt. Without higher aid inflows, • Phasing out temporary stimulus measures
financing constraints and the need to maintain while strengthening well-targeted social
fiscal sustainability will compel many countries safety nets. A large number of emerging
to move to more prudent policies in 2010. and developing countries are supporting
Some developing economies with fiscal domestic activity with ad hoc measures,
sustainability problems may still be able to such as increased spending on public works
finance deficits in the current environment. or reductions in tax rates. Medium-term
But they could rapidly face external financ- fiscal consolidation plans should envisage
ing pressures if the perception takes hold that public investment at levels consistent with
fiscal discipline is not a priority. If there are fiscal sustainability and available financing,
no signs of rising inflation, accommodating and phase out tax reductions presented as
monetary policies can be maintained for some temporary stimulus measures. At the same
time to support domestic demand, unless time, temporary social programs should be
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 83

BOX 3.1 Quality of macroeconomic policies in low-income countries

As in previous years, IMF staff conducted surveys tion of the rapid response to the global crisis in many
among mission chiefs to gauge their assessments of countries and governance in the public sector. At the
the quality of macroeconomic policies in low-income same time, governance in monetary and financial
countries. In 2009 low-income countries made good institutions showed deterioration, while little change
progress in the quality of monetary policies, a reflec- was recorded in other areas.

a. Fiscal policy b. Composition of public spending


90 90
% shares of those countries

% shares of those countries


80 80
70 70
in each category

in each category
60 60
50 50
40 40
30 30
20 20
10 10
0 0
unsatisfactory adequate good unsatisfactory adequate good

c. Fiscal transparency d. Monetary policy


90 90
% shares of those countries

% shares of those countries

80 80
70 70
in each category

in each category

60 60
50 50
40 40
30 30
20 20
10 10
0 0
unsatisfactory adequate good unsatisfactory adequate good

e. Consistency of macroeconomic policy f. Governance in monetary and financial institutions


90 90
% shares of those countries

% shares of those countries

80 80
70 70
in each category

in each category

60 60
50 50
40 40
30 30
20 20
10 10
0 0
unsatisfactory adequate good unsatisfactory adequate good

g. Governance in the public sector h. Access to foreign exchange


90 90
% shares of those countries

% shares of those countries

80 80
70 70
in each category

in each category

60 60
50 50
40 40
30 30
20 20
10 10
0 0
unsatisfactory adequate good unsatisfactory adequate good

2003 2007 2008 2009


84 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

replaced with cost-effective, well-targeted, has helped to stabilize the macroeconomic ef-
permanent social safety nets. fects of the crisis (box 3.3).
• Structural cuts in nonpriority spending.
Governments should continue their efforts
to reduce nonpriority spending, through
Strengthening international
further improvements in public financial
policy cooperation
management and by eliminating expendi- The global crisis of confidence that eventually
ture categories such as badly targeted fuel caused the collapse in world trade in 2008 re-
and food subsidies. Although the decline quired a global response: a simultaneous fiscal
in food and fuel prices since mid-2008 has and monetary policy stimulus in countries with
allowed some countries to reduce spend- sufficient room to maneuver for such policies.
ing on inefficient general subsidies, further The prospects of sustaining the current eco-
progress could be made in replacing general nomic recovery will be enhanced if advanced,
subsidies with programs better targeted to emerging, and developing countries continue
the poor. to cooperate in the implementation of exit
• Improving revenue performance. Many strategies and policies aimed at increasing
emerging and developing countries have growth. The agreement among the Group of
room for further improvements in tax sys- Twenty (G-20) leaders at Pittsburgh to create
tems and revenue administration, including a new process for mutual policy assessments
measures aimed at widening the tax base is an important step in the right direction, but
to include informal sectors, further shifts policy cooperation cannot be limited to those
away from trade taxes to domestic taxes, countries. Enhanced policy cooperation will
and addressing governance problems. Well- be necessary in the following areas.
functioning revenue administrations, in Avoiding protectionism. Restrictions on
combination with tax systems that mini- international trade and services, government
mize distortions, lay the basis for better rev- subsidies for domestic industries, distortions
enue performance and create a more stable to foreign direct investment, informal pres-
investment climate (box 3.2). sures on banks to give preference to domestic
borrowers—all constitute serious threats to
Countries with strong fiscal policies in the the economic recovery. Political pressures to
period leading to the global crisis have been maintain financial support to domestic indus-
better able to deal with its effects than coun- tries indefinitely and to take more far-reaching
tries with weak policies. They regained faster protective measures could rise if unemploy-
access to international financing on more fa- ment remains relatively high in the coming
vorable terms—and were better able to offset years, in line with current expectations. Gov-
the effects of falling world demand with coun- ernments should eschew such protectionist
tercyclical fiscal policies in 2009 and 2010. policies and make strong efforts to reinvigorate
This experience argues in favor of a counter- the Doha Round. An ambitious Doha Round
cyclical, medium-term fiscal rule that aims to would constitute a major step toward a higher
generate savings during good years and create growth path for the world economy: a recent
room for countercyclical policies during cri- study puts potential annual GDP gains from
sis periods. Although almost 60 emerging and multilateral trade liberalization at $300 billion
developing countries have had some type of to $700 billion.10
fiscal rule since the 1990s,9 helping to main- Increasing aid levels and aid effectiveness.
tain fiscal discipline, only some of these are de- Insufficient progress has been made in enhanc-
signed to smooth out the effects of fluctuations ing aid effectiveness, and aid still falls well
in commodity export prices and other external short of the 2005 Gleneagles commitments, in
shocks. Chile and Nigeria are countries where particular for Africa. In addition, many donor
a countercyclical fiscal rule on the basis of pru- countries have reduced their aid budgets, while
dent projections of commodity export prices others face pressures to reduce aid in light of
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 85

BOX 3.2 Mobilizing additional revenue in developing countries: Key issues for tax
policy and revenue administration

The international financial crisis and its consequences reasons. Addressing these issues is key to improving
for economic activity have put additional pressure on the equity of tax systems.
an already fragile revenue situation in many develop- Some tax sources remain underexploited in many
ing countries. Although the revenue situation in most countries—excise taxes on alcohol and tobacco, and
countries is expected to improve as the effects of the environmental taxes (fuel and car taxes, for example)
crisis dissipate and temporary stimulus measures are are important examples.
phased out, some policy changes made in response The institutional framework for policy making,
to the crisis, and during the precrisis period that saw including coordination at the country level between
substantial increases in food and fuel prices, may have the various government entities with responsibility
longer-term effects on revenues. An example is the for tax policy, is deficient in many developing coun-
proposed change to the value added tax (VAT) direc- tries and often leads to fragmentation of policy deci-
tive of the West African Economic and Monetary sions with negative consequences for revenues— for
Union to allow for broader exemptions and a second example, between trade and tax policy, industrial and
(lower) rate on selected items. In addition, taxpayer tax policy, and central and local taxation. Countries
compliance may have declined in many countries, should better integrate tax policy making into macro-
posing challenges for revenue administration. economic management, and strengthen the coordina-
The policy tools that developing countries can tion mechanisms across government entities.
mobilize to deal with the potential revenue loss stem- Taxes on real property have historically yielded
ming from the crisis and other ongoing challenges very little revenue for a number of reasons, including
may be more limited today than in the 1980s and the lack of a proper framework for sharing taxation
1990s, and those that are left may involve stronger powers between central and regional levels of govern-
political commitments. The vast majority of coun- ment. This revenue source remains underexploited in
tries have already implemented broad-based con- many developing countries.
sumption taxes (typically VATs) at rates that are not
particularly low in general and with bases that are Revenue administration
generally narrow. Moreover, corporate tax rates have Tax agencies should also develop a strategy to enhance
fallen dramatically since the early 1990s (by about revenue administration. The primary objective of the
one-quarter on average), and countries have intensi- strategy should be to contain the rise in noncompli-
fied the use of tax incentives, further narrowing the ance often observed during periods of crisis. If left
tax base. unchecked, rising noncompliance could lead to sub-
Country experiences in addressing these chal- stantial forgone revenue and provide unfair competi-
lenges differ, sometimes significantly, but common tive advantages to noncompliant businesses.
areas for reform exist. To achieve this objective, the following four sets
of measures could be considered.
Tax policy
Tax bases can be broadened, especially for VAT and • Assistance to taxpayers could be expanded by
profit taxes. This is not an issue of improving tax adjusting advance payments, accelerating tax
administration to better handle the informal sector refunds, and making greater use of payment
(which is a separate and ongoing challenge); it pri- extensions.
marily means rationalizing the use of income tax • Communication with the taxpayer population
incentives (such as tax holidays) and reducing signifi- could be improved. An effective communication
cantly the reliance on VAT exemptions as a (costly program for taxpayers and other key stakeholders
and largely ineffective) social policy tool. in the tax system should aim at clearly conveying
The taxation of individuals is, in many countries, to stakeholders and the public the various elements
limited to the taxation of wages of the public sector of the tax agency’s compliance strategy.
and large enterprises. The taxation of unincorpo- • Legislative reforms that facilitate revenue admin-
rated small and midsize enterprises remains largely istration could be enacted. Needed reforms vary
elusive—both for technical and for political economy from country to country; they could include mea-

continued
86 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

BOX 3.2 (continued)

sures such as the strengthening of transfer-pricing face the prospect of declining budget allocations in
rules or the introduction of default assessments an economic downturn as governments seek to cre-
and indirect audit methods. ate fiscal room for high-priority social expenditures.
However, it should be recognized that the task of tax
Enforcement could be strengthened, including administration becomes more demanding during dif-
the reassessment of controls over the largest taxpay- ficult economic times. In this situation, substantial
ers, the intensification of arrears collection, securing cuts in tax agencies’ budgets are likely to reduce the
tax withholding, giving greater attention to loss- effectiveness of tax collection and further aggravate
reporting businesses, and enhancing the scrutiny of a decline in revenue.
cross-border transactions and offshore evasion. Tax agencies should align their near-term com-
Early warning for shifts in taxpayer compliance is pliance strategies and medium-term modernization
crucial for prevention. The sooner a tax agency can plans. Sustaining revenue collection over the medium
identify an increase in noncompliance, the faster it term will require tax agencies to address their most
can respond. Few tax agencies, however, have the fundamental weaknesses (such as poor organiza-
capacity to estimate the precise level of the overall tional and staffing arrangements, weak taxpayer
tax gap. In this situation, tax agencies should identify services and enforcement programs, and outdated
and track compliance indicators that can be more information systems). By their nature, such problems
easily measured, such as increases in late filing of tax can be addressed only over the medium term, but in
returns and growth in tax arrears. developing a compliance strategy for the economic
Government support for tax administration is crisis, tax agencies should not neglect their medium-
critical. Like all government agencies, tax agencies term goals.

tighter domestic fiscal constraints. These pres- has not been in the spotlight.11 Moreover, an
sures must be resisted. A substantial increase in extensive theoretical literature explores the
aid, at least in line with existing international possibility of low-income countries falling into
commitments, is essential to allow developing prolonged periods of underdevelopment, com-
countries, especially those in vulnerable debt monly known as poverty traps.12 Finally, cri-
situations and with limited alternative sources ses can result in sharp declines in investment in
of finance, to generate resources for higher education and health, declines that potentially
growth, improve social protection for the most can have long-lasting effects.13
vulnerable, and enhance food security.
Past growth
The medium- and long-term This section thus puts the current crisis in
economic effects of the crisis in historical perspective and examines the pros-
low-income countries pects for growth in the medium to long run.
Over the past few decades, a low-income coun- Although the uncertainties are enormous, and
try’s growth rate in one decade has generally the light that recent history can shed is limited,
been a poor predictor of its growth rate in the some preliminary and conditional answers are
next decade, while many policies and country possible.
characteristics are more stable. An emerging and Transmission mechanisms from the global
vibrant empirical literature points to growth crisis seem to vary considerably across coun-
nonlinearities—accelerations (periods of high tries. While advanced economies have primar-
growth) and growth decelerations (periods of ily suffered a financial and banking crisis, most
abrupt and severe growth slowdowns)—as an developing countries primarily were hit by an
important development fact that until recently external demand effect, although some, nota-
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 87

BOX 3.3 A fiscal rule for commodity exporters: The cases of Chile and Nigeria

Several commodity exporters have in recent years zation Fund, creating a comfortable buffer to offset
adopted medium-term frameworks for fi scal policy the sharp revenue declines in 2009.
aimed at reducing the impact of commodity price fluc- Nigeria introduced an oil-price-based fiscal rule in
tuations on the domestic economy. These frameworks 2004 as a framework for the annual budget process,
allowed countries to build up sizable reserves during which was subsequently formalized in the 2007 Fis-
the commodity price boom of 2007–08, helped to cal Responsibility Act. In the annual Medium-Term
stabilize expenditures, and created additional space Fiscal Strategy presented to parliament, expenditures
for countercyclical policies in 2009. Chile and Nige- are set on the basis of relatively prudent projections
ria illustrate the benefits of such a fiscal rule. for oil prices and production. If actual oil revenues
Since the beginning of the decade, fiscal poli- exceed the budgeted levels, the surpluses are trans-
cies in Chile have been based on a structural fi scal ferred to accounts held by the federal, state, and
surplus rule aimed at mitigating the effects of fluc- local governments at the central bank according to a
tuations in prices for copper and molybdenum, the preset intergovernmental sharing formula. Balances
country’s main commodity exports. Each year, the accumulated in the accounts can be used as a source
authorities make a calculation of structural revenue, of budget fi nancing at the various levels of govern-
consistent with potential GDP and long-term projec- ment if the actual oil price falls below the reference
tions of copper and molybdenum prices. The annual price for three consecutive months.
spending budget is set on the basis of total structural The fiscal rule is supported by a limit on the fed-
tax and nontax (mainly mining) revenue minus a eral government’s fiscal deficit of 3 percent of GDP,
structural surplus. Fiscal surpluses are used to feed enshrined in the Fiscal Responsibility Act. The fiscal
two sovereign wealth funds established under the rule helped Nigeria stabilize expenditures and accu-
2006 Fiscal Responsibility Law: the Pension Reserve mulate sizable reserves during the oil price boom of
Fund to cover the government’s long-term pension 2007–08. Although the political backing for the new
liabilities; and the Economic and Social Stabilization approach does not seem to be as strong as in Chile,
Fund, established to smooth fi scal expenditure and and lower levels of government are not bound by the
fi nance regular or extraordinary public debt amor- Fiscal Responsibility Act, the fi scal rule has served
tization. The consistent implementation of the fiscal Nigeria well thus far. Notwithstanding extraordi-
rule, which has received broad public support, and nary distributions from the central bank accounts
the sovereign wealth funds have served Chile well in in response to political pressures during the oil price
recent years. Rising copper prices since the middle boom, Nigeria accumulated suffi cient resources to
of the decade have allowed Chile to accumulate sub- avoid a contraction of public spending in 2009, reduc-
stantial reserves in the Economic and Social Stabili- ing the effects of the global economic downturn.

a. Chile b. Nigeria
100 50 crude oil,
copper, 40 price change
80
price change 30
60
growth rate, %

20
40 10
0
20 –10
0 –20
real public primary real public primary
–30
–20 spending growth spending growth
–40
–40 –50
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Source: World Economic Outlook.
88 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

MAP 3.2 Across the world, 884 million people lack access to safe water—
84 percent of them in rural areas

Access to water:
Share of population with access to improved water source, % (2006) Green
(D
<50
50–69
70–89
90–99
100 Canada
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: World Economic Indicators.

bly fuel exporters, were also hit by a terms-of- countries.14 The analysis consists of four ex-
trade and, perhaps to a lesser extent, a capital ercises, each tackling the importance of exter-
flows effect. From a methodological point of nal shocks from a slightly different angle. The
view, this difference is quite important because first is a simple event study that illustrates the
these types of external shocks are more famil- growth paths of past crises and compares these
iar to low-income countries than the financial to the current crisis. The second and third ex-
shock is to advanced countries, therefore per- ercises focus on the medium-run effects of
mitting a more credible historical analysis of the crisis. Specifically, an impulse response
the effects in low-income countries. analysis (a time-series analysis) is employed to
The historical analysis that follows fo- estimate the effects over time, complemented
cuses on external demand, terms of trade, and with five-year growth panel regressions. The
capital flows as the three main transmission last exercise is concerned with the longer-run
mechanisms of the crisis affecting low-income implications of the crisis using recently devel-
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 89

IBRD 37737
APRIL 2010

nland
Den)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Armenia
Portugal Spain Monaco
Turkey baijan Turkmenistan
Tajikistan
of Korea
Greece Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire Cameroon
African Rep. Somalia Palau
Togo Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Kenya
São Tomé and Príncipe Gabon Congo Rwanda Singapore Nauru
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

oped methods to capture possible sharp and mediately after the crisis year, it took about
very persistent drops in growth rates. three years for a turnaround to take place in
low-income countries in previous global crises.
The good news is that low-income countries
Global shocks
have tended to recover fully in the sense that
In past global crises, growth declined sharply they have reached or surpassed their precrisis
leading toward the crisis year, but low-income growth rate after about five years.
countries experienced the worst of the crises The current crisis is distinguished by more
about a year after the global low point was synchronization between low-income countries
reached (figure 3.18). In addition, recovery and global cyclical growth movement. Also,
seemed to be faster in the world economy IMF forecasts imply a more rapid V-shape re-
than in low-income countries. More precisely, covery path out of the recession than in previ-
while recovery in the world began almost im- ous crises.
90 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

FIGURE 3.18 After previous crises, low-income countries recovered on average moving around historical averages
more slowly than the world economy (figure 3.19).15

8.0 year of shock


Persistence of output loss over time
6.0 using time series or impulse response
analysis
growth rate, %

4.0

2.0 An impulse response function analysis, as in


Cerra and Saxena (2008), examines whether
0.0 terms of trade and external demand have his-
–2.0 torically been associated with severe output
losses and whether such output losses have
–4.0
–5 –4 –3 –2 –1 0 1 2 3 4 5
been permanent in low-income countries.16 Fig-
ure 3.20 presents impulse responses of output
world (3 crises) world, 2009 world, 2009 projected
losses, measured as the percentage change from
low-income countries (3 crises) low-income countries, 2009
low-income countries, 2009 projected a linear growth trend to a terms-of-trade shock
and an external demand shock, respectively.17
Source: IMF staff calculations. The solid orange line is the mean of output loss,
Note: The figure plots the average per capita GDP growth in the world and in low-income countries and the dashed lines reflect one standard devia-
five years before and five years after the global crises (centered at zero on the horizontal axis) of 1975,
1982, and 1991, and the current crisis. Also shown in dashed lines are IMF projections until 2013. tion from the mean. A key assumption is that
countries will eventually return to the growth
rate existing before the shock. This assumption
FIGURE 3.19 Growth of terms of trade and external demand in
is quite reasonable because most of the low-
low-income countries in past and current crises
income countries considered in these exercises
tend to revert to their preexisting growth trend
6 year of shock in the five years following the shock.
4 The main message is that the impact on
output is negative and highly persistent under
growth rate, %

2
both types of shock, but especially under exter-
0 nal demand shocks. Output losses continue to
rise without a sign of a reversal even 10 years
–2
after an external demand shock, mounting to
–4 a cumulative loss of over 6 percent of GDP.
This result may stem from interactions of ex-
–6
–5 –4 –3 –2 –1 0 1 2 3 4 5 ternal demand shocks with private and public
average terms of trade external demand (3 crises)
investment decisions or policy responses. The
current terms of trade, 2009 external demand, 2009 output loss path eventually becomes flat as
terms of trade, 2009 projected external demand, 2009 projected growth reaches its precrisis rate. But after a
decade, lower growth and a substantial loss of
Source: IMF staff calculations. output is likely to have detrimental effects on
Note: The figure plots the terms of trade and external demand growth in low-income countries five
years before and five years after the global crises (centered at zero on the horizontal axis) of 1975, tax revenues, income, and certainly welfare.
1982, and 1991, and the current crisis. Also shown in dashed lines are IMF projections until 2013. The impulse response analysis is replicated
for Sub-Saharan Africa (figure 3.21). One no-
table difference is that terms-of-trade shocks
Unlike previous crises in which terms-of- seem to have had a larger and more persistent
trade growth suffered a sharp downturn rela- effect than external demand shocks in the rest
tive to external demand growth, the current of low-income countries. Many Sub-Saharan
crisis is characterized by a sharp decline in countries are commodity exporters, particu-
export demand, with terms-of-trade growth larly fuel exporters, and are thus more prone to
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 91

FIGURE 3.20 Output losses are highly persistent, especially under external demand shocks

a. Terms of trade, low-income countries b. External demand, low-income countries


0
0
% deviation from baseline trend

% deviation from baseline trend


–2
–1

–2 –4

–3 –6

–4
–8
0 2 4 6 8 10 0 2 4 6 8 10
years after the shock years after the shock

Source: IMF staff calculations.


Note: Impulse response of output loss in low-income countries to terms-of-trade and external demand shocks. Dashed lines are 1 standard deviation from
the mean output loss.

FIGURE 3.21 In Sub-Saharan Africa terms-of-trade shocks have larger and more persistent effects

a. Terms of trade, Sub-Saharan Africa b. External demand, Sub-Saharan Africa


0 2
% deviation from baseline trend

% deviation from baseline trend

–2 0

–4
–2

–6
–4

–8 –6
0 2 4 6 8 10 0 2 4 6 8 10

years after the shock years after the shock

Source: IMF staff calculations.


Note: Impulse response of output loss in Sub-Saharan Africa countries to terms-of-trade and external demand shocks. Dashed lines are 1 standard deviation
from the mean output loss.

terms-of-trade shocks. This issue is explained sults are based on panel regressions that com-
further below regarding growth downbreak. bine time-series and cross-country informa-
tion,19 and the sample is restricted to nonfuel
exporters. Fuel exporters are excluded from
Regression analysis
the baseline sample because these countries’
A third exercise employs five-year panel growth experience has been heavily influenced
growth regressions as an alternative approach by external demand for fuel. In the baseline
to investigating the impact of terms-of-trade, specification, per capita growth is regressed on
external demand, and foreign direct invest- lagged per capita GDP growth, and the three
ment (FDI) shocks on medium-term per capita shock variables (growth in terms of trade and
GDP growth.18 In particular, the estimation re- external demand and the lag of the difference
92 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

TABLE 3.4 Growth regression results


Entire time period Before 1989 After 1989

(1) (2) (3) (4) (5) (6) (7) (8) (9)


Low-income Other Low-income Other Low-income Other
countries, countries, countries, countries, countries, countries,
Variables All non-fuel non-fuel All non-fuel non-fuel All non-fuel non-fuel

Lagged growth –0.209*** –0.167** –0.237** –0.577*** –0.487*** –0.662*** –0.292*** –0.287*** –0.261***
(0.066) (0.077) (0.095) (0.092) (0.096) (0.110) (0.063) (0.080) (0.083)
Growth in terms of trade 0.123*** 0.115* 0.111** 0.031 0.030 0.023 0.156** 0.131* 0.182***
(0.047) (0.064) (0.053) (0.028) (0.046) (0.028) (0.063) (0.077) (0.066)
Growth in external demand 2.603*** 1.960*** 3.419*** 1.332** 0.617 2.599** 1.727*** 1.665* 1.769**
(0.606) (0.736) (0.786) (0.609) (0.599) (1.135) (0.666) (0.938) (0.706)
Lagged change in (FDI / GDP) 0.631*** 0.221 1.010*** 0.599 –0.404 1.773*** 0.783*** 0.517* 0.953***
(0.187) (0.222) (0.270) (0.633) (0.732) (0.528) (0.243) (0.305) (0.319)

Observations 529 281 248 181 92 89 348 189 159


Number of countries 88 48 40 86 47 39 88 48 40
Source: IMF staff calculations
Robust standard errors in parentheses
*** p < 0.01; ** p < 0.05; * p < 0.1.

in FDI-to-GDP ratio) are all measured in five- ficient becomes positive and significant. That
year averages.20 Columns 1–3 in table 3.4 may not be surprising given that FDI in low-
present results for “All” nonadvanced nonfuel income countries has been plentiful only in the
countries, nonfuel low-income countries, and past decade or so.21 The broad message of this
nonfuel non-low-income countries. The com- exercise is that regression results seem to rein-
parison between low-income countries and force the impulse response findings showing
non-low-income countries is intended to pro- economically significant effects of the shocks
vide some insights into the differential effects in the medium run.
of these shocks to the two income groups.
For low-income countries, terms-of-trade
Growth downbreaks
growth and external demand growth obtain
positive and significant coefficient estimates, The analysis shows that external demand
indicating a positive impact on medium-term shocks, such as those faced by low-income
growth (column 2 of table 3.4). While the countries in 2009, cause growth to slow down
coefficient estimate on FDI for low-income not just immediately but for several years. An
countries using the entire time period in the even greater concern, though, is the risk that
sample is insignificant, it is highly significant the global crisis may cause an essentially per-
for “All” and non-low-income countries manent decline in growth in many low-income
along with the coefficient estimates for terms countries—that is, a growth “downbreak.”
of trade and external demand (columns 1 and Many low-income countries have enjoyed
3, respectively). Columns (4–9) present results relatively strong growth over the past 10–15
from splitting the sample in the periods be- years, when a favorable external environ-
fore and after 1989 (the median year in the ment prevailed. The concern is that, with the
sample). Coincidentally, “after 1989” is the global shock, this could change. Underlying
period when growth increased dramatically in this concern is the observation that, whereas
most low-income countries. Note that most of output paths in the advanced countries tend to
the effect of terms-of-trade and external de- be reasonably steady, in developing countries
mand growth for low-income countries has they are often characterized by “mountains,
been driven by variation in the period after cliffs, and plains.”22 This exercise employs the
1989 (columns 5 and 8). Even more notable methodology by Berg, Ostry, and Zettelmeyer
is that in the post-1989 sample the FDI coef- (2008) to obtain growth downbreaks (sus-
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 93

FIGURE 3.22 In low-income countries, growth downbreaks are more associated with terms-of-trade
shocks, giving hope for smoother recovery

a. Growth downbreaks vs. b. Growth downbreaks vs.


persistent terms-of-trade shock persistent external demand shock
10 10
# GDP growth downbreaks

# GDP growth downbreaks


5 5

0 0
–5 –4 –3 –2 –1 0 1 2 3 4 5 –5 –4 –3 –2 –1 0 1 2 3 4 5
year of shock year of shock

Source: Berg and others forthcoming.


Note: The left panel plots the number of GDP growth downbreaks in a large sample of low-income countries during the periods leading up to and following
a large persistent terms-of-trade shock (year 0 on the horizontal axis). A large persistent terms-of-trade shock is defined as the worst 10 percent of the distri-
bution of all terms-of-trade shocks, measured as the difference of the average three-year terms-of-trade growth before and after a year of shock. The right
panel is the same, except that the shock is to external demand, measured as partner-country real growth weighted by export shares.

tained declines in the rate of growth) in low- Notes


income countries and to explore whether
1. In this chapter, the group of developing coun-
terms-of-trade and external demand shocks
tries includes mainly low-income countries and
are correlated with such “cliffs.” some middle-income countries that are not
One pattern emerging from figure 3.22 is considered emerging economies. High-income
that persistent negative terms-of-trade shocks oil-exporting countries are excluded from this
have often coincided with growth downbreaks category.
in the past. However, persistent negative part- 2. IMF 2009g, box 1.2; IMF 2009d.
ner-country demand shocks have shown no 3. The adequacy of reserves depends on many
association with growth downbreaks. This factors, including the volatility of exports and
phenomenon may be related to the fact that imports, fluctuations in the terms of trade, the
terms-of-trade changes are usually strongest level and maturity structure of external debt,
and the vulnerability to sudden shifts in inter-
in commodity sectors, and that these sectors
national capital movements. While reserve ad-
often find it more difficult to adjust to the new equacy should be assessed country by country,
environment than do, for instance, industrial a level equivalent to three months of imports
sectors. The supply factors that produce the is often used as a rule of thumb, especially for
commodities in question cannot easily switch low-income countries. For a discussion on op-
to other uses, such as satisfying domestic de- timal reserve determination, with a focus on
mand or finding other export markets. The re- low-income countries, see Drummond and
sulting decline in foreign income could squeeze Dhasmana (2008).
imports and activity persistently, thus impeding 4. The evolution of monetary policy stance is ap-
productive activities throughout the economy. proximated by the Monetary Conditions In-
dex (MCI), a summary indicator of the impact
This remarkable observation suggests that
of policy rates and exchange rates on domestic
if indeed the current crisis has affected primar- demand. The MCI combines nominal short-
ily low-income countries through external de- term interest rates and the nominal effective
mand and not through terms of trade, there exchange rate (with a one-third weight for the
may be more reason for hope for a smoother latter) in a single index. The change in the in-
recovery.23 dicator is calculated up to 2009Q3, except for
94 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES GLOBAL MONITORING REPORT 2010

Vietnam and Rwanda, which have data only 16. Daniel Leigh very helpfully provided his Stata
until 2009Q1. The MCI is a useful indicator code and invaluable input. For methodological
of direction in the monetary policy stance: it is details, see Cerra and Saxena (2008) and IMF
simple to calculate and based on data readily 2009g, ch. 4.
available. However, it also suffers from vari- 17. The shock dummy variable for both terms of
ous caveats (including, for example, the use of trade and external demand was constructed as
common weights across diverse countries), so follows: A restricted sample was constructed in
detailed country results need to be interpreted which values below and above the 1st and 99th
with some caution. percentiles were excluded to mitigate the effects
5. In many countries that reduced rates in 2009, from extreme values. The crisis periods belong
inflation came down faster than nominal rates, to the left tail of the moving-average growth
propping up real interest rates. Temporary (based in two periods) distribution, where the
factors, such as commodity price movements, left tail is based on one standard deviation of
may have contributed to the fall in inflation, the restricted sample defined above. Results are
however, mitigating the impact of higher real qualitatively similar to two alternative shock
rates on spending and investment decisions. definitions considered.
6. See IMF 2009e and Baldacci and Kumar, 18. A similar estimation methodology was fol-
forthcoming. An increase in fiscal deficits of 1 lowed in Drummond and Ramirez (2009).
percent of GDP is found to increase 10-year 19. Using a statistical estimation method called
nominal bond yields by about 20 basis points generalized method of moments (GMM).
in the medium term, and an increase in the 20. An alternative growth regression specification
debt-to-GDP ratio of 1 percent increases rates would be the Barro-Solow type regression.
by approximately 5 basis points. Although the This alternative was not considered, because
econometric analysis is based on a sample of it suffers from the well-documented endogene-
advanced and emerging economies, it is plau- ity and omitted variable problems, which the
sible that low-income countries show similar specification used here is less subject to.
relations between deficits, debt, and interest 21. The robustness of these results to alterna-
rates. tive specifications and subsamples has been
7. For a detailed discussion of exit strategies see checked.
IMF 2010a. 22. Pritchett 2000.
8. IMF 2010b. 23. The definitions of “persistent” and “large” can
9. IMF 2009a. be found in the note to figure 3.22. It turns out
10. This represents approximately 0.5–1.2 per- that large negative external demand shocks
cent of 2008 world GDP. See Adler and others such as those experienced by many countries
(2009). in 2009 are not unprecedented for many low-
11. Hausmann, Pritchett, and Rodrik 2005; and income countries. In the sample used for figure
Berg, Ostry, and Zettelmeyer 2008. 3.22, there were 68 instances in which coun-
12. See the literature review in Azariadis and Sta- tries faced external demand shocks larger than
churski (2007) and more specifically the debt trap they faced in 2009 (assuming IMF projections
model in Kehoe and Levine (1993).
for the out-years).
13. Benhabib and Spiegel 1994; Krueger and Lin-
dahl 2001.
14. Data are from IMF. External demand is partner-
country real GDP growth, 2000 = 100, weighted
References
by trade exports to all partner countries (APR Adler, M., C. Brunel, J. C. Hufbauer, and J. J.
2009 Global Economic Environment). Terms Schott. 2009. “What’s on the Table? The Doha
of trade are for goods (World Economic Out- Round as of August 2009.” Working Paper
look [WEO] latest update). Capital flows are 09-6. Peterson Institute for International Eco-
proxied by direct investment in reporting econ- nomics. Washington, DC.
omy in billions of U.S. dollars. Azariadis, C., and J. Stachurski. 2007. “Poverty
15. Data on foreign direct investment were not Traps.” In Handbook of Economic Growth,
available to produce a similar plot. This ob- Vol. 1, Ch. 5, edited by P. Aghion and S. N.
servation is also shown in more formal growth Durlauf. Amsterdam: Elsevier Science.
regression analysis in Berg and others (work Baldacci, E., and M. Kumar. Forthcoming. “Defi-
in progress). cits, Debt, and Interest Rates.” IMF working
paper. Washington, DC.
GLOBAL MONITORING REPORT 2010 GROWTH OUTLOOK AND MACROECONOMIC CHALLENGES 95

Benhabib, J., and M. Spiegel. 1994. “The Role IMF (International Monetary Fund). 2009a. “Fis-
of Human Capital in Economic Development: cal Rules: Anchoring Expectations for Sustain-
Evidence from Cross-Country Data.” Journal of able Public Finances.” Washington, DC (www.
Monetary Economics 34: 143–73. imf.org/external/np/pp/eng/2009/121609.pdf).
Berg, A., J. D. Ostry, and J. Zettelmeyer. 2008. ———. 2009b. “Global Financial Stability Re-
“What Makes Growth Sustained.” Working port.” Washington, DC (October).
Paper 08/59. International Monetary Fund, ———. 2009c. “Regional Economic Outlook-
Sub-Sahara Africa.” Washington, DC (Fall).
Washington, DC.
———. 2009d. “Review of Recent Crisis Pro-
Berg, A., C. A. M Pattillo, H. Weisfeld, C. Papa-
grams.” Washington, DC (September 14).
georgiou, N. Spatafora, and S. P. Tokarick. ———. 2009e. “The State of Public Finances Cross
Forthcoming. “The Short-Run Effects of the Country.” IMF Staff Position Note, SPN/09/25.
Crisis in Low-Income Countries.” International Washington, DC (November).
Monetary Fund, Washington, DC. ———. 2009f. World Economic Outlook. Wash-
Cerra, V., and S. C. Saxena. 2008. “Growth Dy- ington, DC (Spring).
namics: The Myth of Economic Recovery.” ———. 2009g. World Economic Outlook. Wash-
American Economic Review 98: 439–57. ington, DC (October).
Drummond P., and A. Dhasmana. 2008. “Foreign ———. 2010a. “Exiting from Crisis Intervention
Reserve Adequacy in Sub-Saharan Africa.” Policies and Fiscal Consolidation in the Post-
Working Paper WP/08/150. International Mon- Crisis World.” Washington, DC (February).
etary Fund, Washington, DC. ———. 2010b “Strategies for Fiscal Consolida-
Drummond P., and G. Ramirez. 2009. “Spillovers tion in the Post-Crisis World.” Washington, DC
from the Rest of the World into Sub-Saharan (February).
Kehoe, T., and D. Levine. 1993. “Debt-Constrained
African Countries.” Working Paper WP/09/155.
Asset Markets.” Review of Economic Studies
International Monetary Fund, Washington,
60: 865–88.
DC. Krueger, A. B., and M. Lindahl. 2001. “Education
Erickson and others 2004<<Please complete or de- for Growth: Why and for Whom?” Journal of
lete cite>> Economic Literature 39 (4): 1101–36.
Hausmann, R., L. Prichett, and D. Rodrik. 2005. Pritchett, L. 2000. “Understanding Patterns of
“Growth Accelerations.” Journal of Economic Economic Growth: Searching for Hills among
Growth 10: 303–29. Plateaus, Mountains, and Plains.” World Bank
Economic Review 14: 221–25.
4
Outlook for the Millennium
Development Goals

H
ow will the global economic crisis numbers— additional children dying or
alter precrisis trends in the Millen- uneducated, additional people left without
nium Development Goals (MDGs)? clean water—could be large because of the
With only five years left until the target date size of the population underlying each rate.
of 2015, it is obvious that several of the Countries can achieve better development
MDGs will not be attained, globally or by a outcomes through improved policies, most
majority of countries. Many of the goals are notably shifts in expenditures, increases in
too high for low-income countries, given their domestic revenue, and better service deliv-
low starting points. Many countries, includ- ery. Stronger policies are unlikely to com-
ing low-income ones, have seen substantial pensate fully for the deterioration in human
gains in recent years, however, and entered development indicators that result from
the current crisis in a stronger position than slower growth, however. In the current con-
in past crises (chapters 1 and 2). Important text, better development outcomes will thus
questions are whether the gains will be pre- depend on the speed at which the global eco-
served, and what happens if the fragile recov- nomic recovery supports increases in devel-
ery slips into a prolonged stagnation. oping countries’ export revenues and exter-
The crisis is likely to have a lasting impact nal finance.
on human development indicators that will This chapter looks at these issues in two
not overcome even a robust economic recov- ways. It first presents alternative scenarios
ery. Although growth in emerging and devel- for progress on some key human develop-
oping countries is currently accelerating, ment–related MDGs based solely on dif-
should growth slow or deteriorate, progress ferent forecasts of GDP growth, with the
toward the MDGs will suffer even more. A results aggregated by regions. This relatively
decline in growth would have a signifi cant limited approach provides a general sense
impact on poverty and undernourishment. of the impact of the crisis and the potential
The impact of a growth slowdown on some envelope for the MDGs looking ahead over
of the other MDG indicators analyzed is the next five to ten years. The second part of
more muted, although the cost in absolute the chapter then takes into account a broader

GLOBAL MONITORING REPORT 2010 97


98 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

set of determinants of progress in the MDGs, economic growth and delivery of services
including fiscal policy (public expenditures to the poor—the very two factors likely to
and their composition plus revenue efforts), be most affected by global economic crisis.
export revenues, terms of trade, aid flows, That is why the lessons of history regarding
remittances, and foreign borrowing. This the effects of growth decelerations on various
richer analysis allows a much more robust human development indicators are examined
view of how the external economic environ- in chapter 2. Although not the only driver,
ment and developing-country policies will growth will likewise be a key factor in pro-
affect progress toward the MDGs. The scope jecting the postcrisis trends for the MDGs.
of the analysis, however, and the variables The other key factor, effective service deliv-
involved, make it extremely difficult to pro- ery, is difficult to assess even in the best of
vide comprehensive forecasts of human devel- circumstances.2
opment indicators for developing countries. The current crisis has resulted in a deterio-
Instead, this section illuminates the channels ration in human development indicators that
that influence MDG outcomes through the will have important future effects even with a
lens of two types of low-income developing- robust economic recovery. If growth were to
country structure based on natural endow- stagnate or slow, the impact on human wel-
ments—those that are resource poor and fare in developing countries would be severe.
those that are resource rich.1 Projecting the aggregate outlook for the
MDGs is fraught with difficulties (box 4.1).3
Nevertheless, it is essential to assess where
Forward analysis of the MDGs things stand in the aftermath of the crisis,
The original analytical framework under- as developing countries enter a new and less
pinning the assessment of policies as devel- favorable external environment.
oped by the World Bank and the Interna- The alternative scenarios of progress
tional Monetary Fund (IMF) in the first toward the MDGs presented here are based
Global Monitoring Report in 2004 remains on a simplified reduced-form analysis link-
very valid today for organizing this policy ing economic growth—the key variable of
assessment (figure 4.1). The two key pillars the crisis and the recovery scenarios—to the
for achieving the development outcomes are MDG indicators.4 The simulations are based

FIGURE 4.1 Framework linking policies and actions with development outcomes

POLICIES AND ACTIONS KEY INTERMEDIATE OUTCOMES


Agenda anchored in IN DEVELOPING COUNTRIES

developing countries
• Create a good enabling
climate for economic activity
• Empower and invest in
poor people stronger and sustainable
economic growth
developed countries Achieving the
• Enhance support to
MDGs
developing country reform
efforts and global public
goods
improved delivery of services
to poor people
development agencies
• Better align support to
delivery of development
results

Source: World Bank 2004a.


GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 99

BOX 4.1 Uncertainty and risk in projecting attainment of the MDGs

There are many uncertainties and risks in projecting vate expenditures. Hence, fiscal adjustment and thus
development outcomes. One is the strength and tim- the implications of slower growth for the MDGs will
ing of the economic recovery. Another is the com- vary from country to country depending on circum-
plexity of the relationships between the MDGs and stances and conditions entering the crisis.
their determinants, which are still poorly understood. Several studies point to other problems in account-
Among the MDGs the impact of economic perfor- ing for all of the influences on human development
mance on poverty is better established, although the indicators. An increase in public expenditures does
elasticity of poverty to growth can vary with country not necessarily improve education and health out-
circumstances and initial conditions. Furthermore, comes; nor does economic growth alone. Links
human development outcomes are influenced by between public expenditures and social sector out-
a wide range of factors, including the evolution of comes are weak. Supply-side factors associated
household incomes and public resources, as well as with effective service delivery are preconditions for
the consequences of supply and demand for policies, improving basic service provision—school facilities,
institutional actions, and microlevel services. Given books, health clinics, vaccination programs, qualified
the complexity and differing assumptions about the teachers and health staff, and the like. Client demand
recovery, assessments of human development out- for services and various other factors at the local
comes can be wide ranging. level—household incomes, distance and opportunity
Another important uncertainty in forecasting costs, voice and participation of clients, educational
progress toward the MDGs is fiscal adjustment— attainment of mothers, corruption, and cultural and
public expenditures and their composition are key religious norms—also matter and may vary by com-
determinants of human development indicators in munity. The empirical regularity of these potential
low-income countries. A deterioration in the mac- determinants can become difficult to establish at the
roeconomic environment may reduce government country, regional, and global levels.
income, thus endangering public expenditures essen-
tial for progress toward the MDGs. However, aid,
external borrowing, and international reserves may Source: Dinh, Adugna, and Myers 2002; Adams and Bevan
provide the fiscal space needed to protect social 2000; Filmer, Hammer, and Pritchett 2000, 2002; Devarajan
spending, while remittances may help to support pri- and Reinikka 2004; World Bank 2004.

on GDP growth because it is a major deter- surveys in more than 100 countries and
minant of progress toward the MDGs, and it assumes that the underlying income or expen-
is the only determinant that is projected for a diture distribution is relatively stable during
large group of countries and that is anchored changes in economic growth. 5 The poverty
by the short-, medium-, and long-term eco- analysis brings 31 new household surveys
nomic outlook in the International Monetary to the 2010 Global Monitoring Report and
Fund’s (IMF) World Economic Outlook new projections of per capita income growth
and the long-term growth projections that in the aftermath of the crisis. The analysis
underpin the World Bank’s Global Economic also considers four other MDGs—primary
Prospects. Because of the many uncertainties education completion, infant mortality,
described in box 4.1, these projections relat- gender equality in education, and access to
ing progress in the MDGs to alternative sce- clean water—for which aggregate quanti-
narios for GDP growth are necessarily subject tative analysis is currently feasible (future
to large margins of error and should be taken reports will expand the analysis to other
as illustrative. MDGs). The relationship between GDP
The estimated relationship between growth and each indicator is estimated for
poverty and growth is based on household each country.
100 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

MAP 4.1 In 2007, 72 million children worldwide were denied access to education

Primary completion rate:


% of relevant age group (2004-09) Green
(De
<50
50–69
70–84
85–94
≥95 Canada
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: World Development Indicators.

The results show that growth generally Three global scenarios


was significantly related to progress in the for progress on human
human development indicators. However, development–related MDGs
confi rming all the caveats mentioned above,
the estimations using growth alone accounted Three global scenarios for GDP growth
for only 30–40 percent of past variations of address the risks of the current global eco-
the MDG indicators across countries and nomic crisis: a postcrisis trend; a high-
time. These coefficients were then used to growth or precrisis trend; and a low-growth
forecast each MDG indicator for each coun- scenario.
try, based on alternative scenarios for GDP The postcrisis trend assumes a relatively
growth. Although it is certainly possible to rapid economic recovery in 2010, with
include other determinants of the MDG indi- strong growth continuing into the future, as
cators in the estimation, it is not practical to described in chapter 2.6 This is essentially the
forecast these other indicators on a country- base case forecast for growth in developing
by-country basis (box 4.2). countries after the crisis.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 101

IBRD 37741
APRIL 2010

nland
en)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Armenia
Portugal Spain Monaco
Turkey baijan Turkmenistan
Tajikistan
of Korea
Greece Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire Cameroon
African Rep. Somalia Palau
Togo Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Kenya
São Tomé and Príncipe Gabon Congo Rwanda Singapore Nauru
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

The precrisis trend gives the forecast past responses to severe external shocks in
path for the MDGs if developing countries developing countries.
had continued their impressive growth per-
formance during 2000–07, the period just
The impact on the very poor
before the global economic crisis. The impact
of the crisis on the MDGs can thus be mea- Recent economic shocks have taken a toll on
sured by comparing the postcrisis trend with the poor. The crisis left an estimated 50 mil-
this one. lion more people in extreme poverty in 2009,
The low-growth scenario assumes that and some 64 million more will fall into that
the recovery projected for the postcrisis trend category by the end of 2010 relative to a pre-
will not take place in the medium run. The crisis trend.7 New estimates suggest that the
scenario assumes little or no growth for large global spike in food prices in 2008 may
about five years, when it begins to slowly have led the incidence of undernourishment
recover. This scenario follows the pattern of to rise by around 63 million people, while
102 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

BOX 4.2 Estimating the impact of growth on human development indicators

The relationship between GDP growth and the jumps from 30–40 percent using growth as the sole
MDGs was estimated taking into account a policy explanatory variable to 80 percent when the index of
index reflecting the country’s level of policy and policy and initial conditions are taken into account.
institutions plus a set of initial conditions (for exam- These estimations help to refi ne the understanding
ple, adult female literacy rate, urbanization, ethnic of the relationship between growth and the human
fractionalization, level of income, and location by development indicators. However, it did not prove
geographical region).a Several policy indexes had a possible to use the policy index or the initial condi-
signifi cant relationship with the MDG indicators. tions in the alternative scenarios for future progress
Among these, the World Bank’s Country Policy and in the MDG indicators, given the difficulties involved
Institutional Assessment (CPIA) rating was selected in forecasting these variables in many countries.
because it is a broader measure than one based solely
on governance indexes. It covers economic manage- a. The use of the policy index is similar to the empirical works
ment, structural policies, policies for social inclu- of Wagstaff and Claeson (2004), Rajkumar and Swaroop
sion and equity, and public sector management and (2002), and Filmer and Pritchett (1999). The CPIA is avail-
institutions. The explanatory power of the equations able in the World Development Indicators.

the crisis itself may have led to an additional reductions in the poverty rate, to 15 percent
41.3 million undernourished people, or 4.4 in 2015, well below the MDG target of 20.4
percent more undernourished people in 2009 percent (table 4.1). Nevertheless, the crisis has
than would have been the case without the imposed a lasting cost on poverty reduction.
economic crisis.8 Had the crisis not interrupted the rapid eco-
A rapid economic recovery (the postcrisis nomic progress made by developing countries
trend) would improve the situation for many through 2007 (the precrisis trend), the pov-
of the extremely poor and lead to substantial erty rate at $1.25 a day would have fallen to

TABLE 4.1 Poverty in developing countries, alternative scenarios, 1990 –2020


Region and scenario 1990 2005 2015 2020

Global level
Percentage of the population living on less than $1.25 a day
Postcrisis 41.7 25.2 15.0 12.8
Precrisis 41.7 25.2 14.1 11.7
Low-growth 41.7 25.2 18.5 16.3

Number of people living on less than $1.25 a day (millions)


Postcrisis 1,817 1,371 918 826
Precrisis 1,817 1,371 865 755
Low-growth 1,817 1,371 1132 1053

Sub-Saharan Africa
Percentage of the population living on less than $1.25 a day
Postcrisis 57.6 50.9 38.0 32.8
Precrisis 57.6 50.9 35.9 29.9
Low-growth 57.6 50.9 43.8 39.9

Number of people living on less than $1.25 a day (millions)


Postcrisis 296 387 366 352
Precrisis 296 387 346 321
Low-growth 296 387 421 428
Source: World Bank staff calculations using the PovcalNet database.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 103

about 14 percent by 2015, implying that an are projected to miss the poverty reduction
additional 53 million people would have been MDG at poverty lines of both $1.25 and $2
lifted out of extreme poverty. Things could a day. However, the poverty rates in these
be worse than the postcrisis trend, however. countries are very low to start with (about 4
If the economic outlook deteriorates to the percent at $1.25 a day and about 9 percent
low-growth scenario, the poverty rate could at $2 a day in 2005), so a higher poverty line
fall only to 18.5 percent, with an additional of $4 to $5 a day is more meaningful for this
214 million people living in absolute poverty group of countries.
by 2015 (relative to the postcrisis trend). Overall, the projection for the $2 a day
On current or postcrisis growth trends, poverty threshold is less promising. In the
poverty in Sub-Saharan Africa is projected postcrisis trend, 2 billion people, or one-third
to drop to 38 percent by 2015—more than of the population of developing countries—
9 percentage points short of its target. Before more than half of the 1990s level—remain in
the crisis the region had been on a path to poverty at $2 a day.
reach a poverty rate of 35.9 percent, which
would have lifted another 20 million people
Impact on selected human development
out of poverty by 2015. If growth stagnates
indicators
into the low-growth scenario, the trend gap
could more than double, implying an addi- The crisis will have serious and lasting costs
tional 55 million people remaining in extreme and gaps for other human development
poverty by 2015. indicators as well (figure 4.2 and table 4.2).
The long-term nature of the cumulative According to the projections for 2015, as a
effects becomes clearer when global projec- result of the crisis:
tions are extended 10 years forward. The
postcrisis trend suggests that by 2020, 826 • The number of infants dying would increase
million people (12.8 percent) in developing by 55,000. Without the crisis, 260,000
countries will be living on less than $1.25 a additional children under the age of five
day, implying that 71 million more people could have been prevented from dying in
will be living in absolute poverty in 2020 as 2015. The cumulative total from 2009 to
a result of the crisis. The low-growth sce- 2015 could reach 265,000 and 1.2 million,
nario would result in a rise of 227 million respectively. The consequences for infant
living in absolute poverty compared with the mortality in Africa are grave, with some
postcrisis trend. The corresponding increases 30,000–50,000 additional infant deaths
in poverty for Sub-Saharan Africa in 2020 in 2009, virtually all of them girls.9 The
are 31 million more people in poverty for tragedy is not just these added deaths—
the postcrisis trend and 76 million more for more than 3 million infants die in Africa
the low-growth scenario. The five additional every year, a number that could be reduced
years would leave Sub-Saharan Africa still through better policies and interventions.
short of halving poverty, the MDG target for • Some 350,000 more students will fail to
2015. complete primary school.
Poverty rates vary considerably among the • Some 100 million more people will lose
other regions (annex tables 4A.1 and 4A.2). access to safe drinking water.
Even in the low-growth scenario, the East
Asia and Pacific region more than meets its The impact on gender equality in educa-
poverty target, in large part because of Chi- tion and on access to safe water is muted
na’s success in reducing poverty. South Asia, in these scenarios (although even small
on the strength of India’s achievement, meets changes in these indicators can translate into
the poverty target in the postcrisis trend but large numbers of people affected) because
not in the low-growth scenario. Middle- these indicators are influenced by forces
income countries in Europe and Central Asia that change only slowly. For example, the
104 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 4.2 The long-run effect of slower growth on selected MDGs is worrisome

a. MDG 2: Primary completion rate a. MDG 2: Primary completion rate by 2015


96 100 100
94
96
92
percent

percent
90 92 91.8 91.5
90.4
88
88
86
84 86
2007 2009 2011 2013 2015 2017 2019 target precrisis postcrisis low-growth
low-growth postcrisis precrisis

b. MDG 3: Gender parity in primary b. MDG 3: Gender parity in primary


and secondary education and secondary education by 2015
98.0 100
100
99
97.0 98
percent

percent
97 96.5
96.0 96
96 95.6
95.0 95
94
94.0 93
2007 2009 2011 2013 2015 2017 2019 target precrisis postcrisis low-growth
low-growth postcrisis precrisis

c. MDG 4: Child mortality under five c. MDG 4: Child mortality under five by 2015
75 80
68.1 68.6 69.5
73
deaths per 1,000

deaths per 1,000

60
71
40 33.7
69
67 20

65 0
2007 2009 2011 2013 2015 2017 2019 target precrisis postcrisis low-growth
low-growth postcrisis precrisis

d. MDG 7.c: Access to safe drinking water d. MDG 7.c: Access to safe drinking water by 2015
17 16
14
15
12 11
13 9.6 10.1
percent

percent

11 8
9
4
7
5 0
2007 2009 2011 2013 2015 2017 2019 target precrisis postcrisis low-growth
low-growth postcrisis precrisis

Source: World Bank staff calculations.


GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 105

TABLE 4.2 Trends for other MDG human development indicators by region and alternative economic
scenarios
2015
MDG and region Target 1991 2007 Postcrisis Precrisis Low-growth

MDG 2: Primary completion rate (%)            


East Asia and Pacific 100 101 98 100 100 99.3
Europe and Central Asia 100 93 98 99.9 100 99.9
Latin America and the Caribbean 100 84 100 97.9 100 97.7
Middle East and North Africa 100 78 90 94.9 95.6 93.6
South Asia 100 62 80 82.4 91.7 81.9
Sub-Saharan Africa 100 51 60 67.3 67.6 66.7
All developing countries 100 78 85 91.5 91.8 90.4

MDG 3: Ratio of girls to boys in primary


and secondary education (%) Target 1991 2007

East Asia and Pacific 100 89 99 100 100 100


Europe and Central Asia 100 100 102 99.4 100 97.8
Latin America and the Caribbean 100 98 103 100 100 100
Middle East and North Africa 100 78 96 95.6 98.2 94.7
South Asia 100 70 89 92.7 94.4 92.1
Sub-Saharan Africa 100 79 86 89.7 89.9 89.1
All developing countries 100 83 95 96.0 96.5 95.6

MDG 4: Child mortality under five


(per 1,000) Target 1990 2007      
East Asia and Pacific 19 56 27 24.6 18.6 24.9
Europe and Central Asia 17 50 23 18.8 15.4 21.7
Latin America and the Caribbean 18 55 26 23.7 19.7 25.4
Middle East and North Africa 26 78 38 36.7 29.2 37.3
South Asia 42 125 78 76.0 62.7 76.6
Sub-Saharan Africa 61 183 146 139.5 138.7 141.0
All developing countries 34 101 74 68.6 68.1 69.5

MDG 7.c: Access to improved water source


(% population w/access) Target 1990 2006

East Asia and Pacific 16 32 13 3.3 0.6 4.1


Europe and Central Asia 5 10 5 0 0 1.8
Latin America and the Caribbean 8 16 9 5.4 4.5 7.1
Middle East and North Africa 6 11 12 8.3 7.4 10.0
South Asia 13 27 13 9.3 5.1 10.2
Sub-Saharan Africa 26 51 42 39.1 38.8 39.8
All developing countries 12 24 14 10.1 9.6 11
Source: World Bank staff estimates.

participation by girls in school reflects in In general, the impact of the low-growth


part the educational level of the mother, and scenario on development outcomes will be
access to safe water is affected by the degree cumulative and long term (figure 4.3).
of urbanization. The impact of slower growth
on the MDGs increases, however, as the time • If the baseline scenario (the postcrisis trend)
horizon is extended further into the future holds up, human development indicators
(for example, fewer girls being educated now will continue to improve albeit less rapidly
means that eventually women of childbearing owing to the extended impact of the crisis.
age will have less education). By 2015 the differences between the gains
106 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 4.3 The long-run effect of slower growth is especially worrisome in Sub-Saharan Africa

a. MDG 2: Primary completion rate, Sub-Saharan Africa a. MDG 2: Primary completion rate by 2015,
70 Sub-Saharan Africa
100 100
69
90
68
80

percent
percent

67 67.6
70 67.3 66.7
66
60
65
50
64
40
63 target precrisis postcrisis low-growth
2008 2010 2012 2014 2016 2018 2020
low-growth postcrisis precrisis

b. MDG 3: Gender parity in primary and secondary b. MDG 3: Gender parity in primary and secondary
education, Sub-Saharan Africa education by 2015, Sub-Saharan Africa
91.5 100 100
91
90.5 94

percent
percent

90
89.9 89.7 89.1
89.5
88
89
88.5
88 82
target precrisis postcrisis low-growth
87.5
2008 2010 2012 2014 2016 2018 2020
low-growth postcrisis precrisis

c. MDG 4: Child mortality under five,


Sub-Saharan Africa c. MDG 4: Child mortality under five by 2015,
Sub-Saharan Africa
147
160
138.7 139.5 141
145
deaths per 1,000

120
deaths per 1,000

143
80
141 61.1

139 40

137 0
target precrisis postcrisis low-growth
135
2008 2010 2012 2014 2016 2018 2020
low-growth postcrisis precrisis

d. MDG 7.c: Access to safe drinking water, d. MDG 7.c: Access to safe drinking water by 2015,
Sub-Saharan Africa Sub-Saharan Africa
41.5 45
41 38.8 39.1 39.8
40.5
30 25.7
percent

40
percent

39.5
39 15
38.5
38
0
37.5 target precrisis postcrisis low-growth
37
2008 2010 2012 2014 2016 2018 2020
low-growth postcrisis precrisis

Source: World Bank staff estimates.


Note: The precrisis period is 2000–07.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 107

projected in the postcrisis trend and those each of the two representative economies are
for the precrisis trend will become discern- given in box 4.3.
ible, especially for human development These simulations use the World Bank’s
outcomes such as primary school comple- Maquette for MDG Simulations (MAMS),
tion and infant mortality. a model that analyzes the implications of
• Like the compounding effects of interest strategic choices for economic outcomes,
rates, these gaps will intensify from 2015 to including changes in human development
2020. A look at the long-term impact reveals indicators (see box 4A.1 in the annex for
that the projected slide in human develop- more discussion).11 MAMS’ main contribu-
ment outcomes will become damaging and tion is its integration of government services
irreversible unless action is taken now. and their impact on the economy, including
• The world needs to avoid economic stag- on the MDGs and the labor market, within a
nation. If the growth trend in developing standard recursive dynamic computable gen-
countries becomes sluggish for a long time, eral equilibrium framework. Several MAMS
as in the low-growth scenario, development features are useful for assessing the impact of
outcomes will deteriorate or stall, as hap- alternative scenarios on MDGs. The model
pened in many low-income countries in incorporates a formal representation of the
Sub-Saharan Africa during the 1970s and production of government services (educa-
1980s. tion, health, and infrastructure) that takes
into account demand, supply, and efficiency.
Spending strategies under less It allows for complementarity or synergy
favorable circumstances effects across the MDGs—for example, bet-
ter access to clean water may improve health,
What can developing countries do if the exter- which may boost school attendance, labor
nal economic environment remains unfavor- productivity, and economic growth. It shows
able, and what impacts might their policy the economywide repercussions of scaling
and spending strategies have on development up (or down) human development services,
outcomes? The three global growth scenarios including the impact on economic growth,
provided a broad picture of the likely impact relative prices, the exchange rate, and the
of the crisis on poverty. But these scenarios allocation of resources between government
cannot be used to explore the scope for miti- and nongovernment sectors. And it makes
gating the effects of external shocks on pov- possible the consideration of sequencing and
erty through appropriate policy adjustments. time-related trade-offs through a recursive
For this purpose, the broad country cover- treatment of dynamics that tracks indicators
age achieved in the scenarios given above is over time.
set aside in favor of a richer analysis of the
impact of policies.
The low-income, resource-poor country
To begin, low-income countries are divided
into two groups—those that are resource rich The analysis for the low-income, resource-
and those that are resource poor.10 A repre- poor archetype (LIRP) considers four cases
sentative economy of each type is then con- (the reference year for the analysis is 2009,
structed based on the average indicators for and the simulation period is 2010–20):
all of the low-income countries in that group
(tables 4A.1 and 4A.2 in the annex sum- • The base case is relatively optimistic. It
marize the social and economic indicators assumes that GDP growth recovers by
that characterize each country archetype; 2011 to the growth rate in 2008. The
for the most part they correspond to the lat- annual growth rate in 2012–20 is slightly
est median statistics from the World Bank’s higher than in 2011 (see figure 4.4 for
World Development Indicators database. The GDP growth under different LIRP cases).
assumptions and data used in constructing Growth in foreign aid is slower after 2010
108 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

BOX 4.3 Assumptions for the archetype countries

The low-income, resource-rich (LIRR) archetype has archetype and 61.8 percent for the LIRR—the
a natural resource that it exports. The government median values for the countries in each group. The
receives 70 percent of the income, and foreign inves- poorest statistics for the LIRR result partly from the
tors get the rest. In 2009 government income from the “natural resource curse” associated with past con-
natural resource was 8.4 percent of GDP. All output fl icts and corruption; see, for example, Collier and
of the natural resource commodity is exported and Goderis (2007). Median GDP per capita is $598 for
accounts for 56 percent of the value of total exports. the LIRP and $482 for the LIRR. Similarly, both the
Government borrowing is 2.6 percent of GDP, and LIRP and the LIRR are assumed to have the median
foreign debt is 49 percent of GDP. The country value of their group for share of the population with
receives no debt relief during the simulation period. access to clean water (MDG 7); the under-five mor-
The low-income, resource-poor (LIRP) archetype is tality rate (MDG 4); and selected education indica-
more dependent than the LIRR on foreign aid, which tors, including the gross completion rate for primary
equals about 6.5 percent of GDP, and its foreign debt school (MDG 2) and gross enrollment rates at all
is higher, at 65 percent of GDP. Like the LIRR, it three levels (primary, secondary, and tertiary). The
receives no debt relief during the analysis period. analysis looks especially at the evolution of MDGs
The poverty headcount rate at $1.25 a day (the 1, 2, 4, and 7.
indicator for MDG 1) is 49.6 percent for the LIRP

Government and nongovernment payments and foreign debt of archetype


countries, 2009
percent of GDP
Low-income, Low-income,
Payment resource-poor countries resource-rich countries
Income from natural resource n.a. 8.4
Foreign aid 6.5 1.2
Taxes 20.2 16.0
Private borrowing 0.5 0.4
Foreign borrowing 4.0 2.6
Foreign debt 65.0 48.8
Foreign direct investment 1.9 1.7
Remittances 1.3 1.2
Source: Go and others, forthcoming.
Note: n.a. = not applicable.

than in the previous decade, reflecting in productivity growth. World prices, FDI,
a decline in GDP growth in donor coun- and foreign aid all grow at slower rates
tries.12 Remittance growth and foreign than in the base case (25 percent of base
direct investment (FDI) fall relative to the case rates). The growth slowdown for for-
previous decade, also reflecting a decline in eign aid and other government receipts
GDP growth in the countries from which leads to reduced development spending
the payments flow. By 2015 world prices (defined as spending on education, health,
have recovered to 2008 precrisis levels. water and sanitation, and infrastructure),
• The low-aid case represents an extreme, as the government fails to reduce spending
negative case with a weak recovery in GDP in other areas. Remittances are assumed to
growth (to just 40 percent of real GDP grow at the same annual rate as in the base
growth in the base case), driven by a deteri- case because these payments are based on
orating external environment and a decline personal connections, and there is little
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 109

reason to expect them to respond nega- FIGURE 4.4 Annual GDP growth for LIRP under four cases
tively to slower growth in the developing
countries. Low-income, resource-poor archetype
• In the low-aid internal 1 case the govern- 8.0

ment makes internal adjustments to offset 7.0

GDP annual growth, %


the effects on the MDGs of a weak recov- 6.0

ery in GDP and reduced growth in foreign 5.0

inflows. The government reduces growth in 4.0

nondevelopment spending (to 90 percent of 3.0

such spending in the base case), increases 2.0


receipts from domestic taxes (by half a 1.0
0.0
percentage point of GDP over the base 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
case), and uses the resulting fiscal space to base case low-aid case
expand development spending. low-aid internal 2 case low-aid internal 1 case
• In the low-aid internal 2 case, the govern-
ment further improves policies and service Source: Go and others, forthcoming.
delivery relative to the low-aid internal 1
case, resulting in a moderately higher GDP
growth (55 percent of the base-case rate).13 the increase in taxes (predominantly indi-
rect taxes in low-income countries) reduces
Slow growth in the low-income, resource- expenditures or incomes of the poor.
poor country results in a severe deteriora- Under the low-aid internal 2 case, a more
tion in human development indicators. All substantial improvement in MDG indicators
four of the MDGs covered by the analysis can be accomplished by combining improved
(poverty, primary school gross completion fiscal policies with policies that improve over-
rate, under-five mortality rate, and share of all productivity. Progress toward the MDGs
population with access to safe water) decline improves relative to the low-aid internal 1
in the low-aid simulation relative to the base and the low-aid cases, although not enough
case (figure 4.5). By 2020 the poverty rate is to catch up with the base case.
more than 20 percentage points higher, the Thus policy adjustments to support devel-
under-five mortality rate 15 points higher, opment spending and improve overall eco-
and the share of the population with access nomic productivity are critical to limiting the
to safe water 4 percentage points lower in the impact on human development indicators of
low-aid case than in the base case. The gross an externally induced decline in GDP growth
primary school completion rate improves in (for example, the current crisis). However, to
all scenarios, as students enrolled in lower the extent that policies cannot maintain trend
grades (reflecting recent strong expansion growth in the face of an external shock, then
in primary enrollment) proceed through the a deterioration in human development indi-
primary level. Because of a natural decline in cators is inevitable. This fact highlights the
the intake of out-of-cohort students, progress importance of a global response to the crisis
tends to level off.14 that focuses on ensuring strong flows of aid,
With better expenditure management limiting the deterioration in developing coun-
and internal effort in the low-aid internal 1 tries’ access to external fi nance, and main-
case, including a government shift in expen- taining open export markets to permit trade
ditures to protect development spending and expansion at more attractive world prices.
increased domestic tax collection, all the
MDGs (except poverty reduction) do better
The low-income, resource-rich country
than under the low-aid case. The poverty
rate in the low-aid internal 1 case is margin- The pattern of results for the low-income,
ally higher than in the low-aid case, because resource-rich archetype (LIRR) is similar to
110 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

MAP 4.2 Tuberculosis kills around 1.3 million people a year, or 3,500 a day

Tuberculosis rate:
Incidence of tuberculosis per 100,000 people (2008) Green
(De
≥500
250–499
100–249
50–99
<50 Canada
no data

United States

Bermuda
(UK)

The Bahamas
Cayman Is. (UK)
Cuba Turks and Caicos Islands (UK)
Mexico
Haiti
C
Belize Jamaica
Guatemala Honduras
El Salvador Nicaragua
Costa Rica Panama
R.B. de Guyana
Venezuela Suriname
Colombia French Guiana (Fr)

Ecuador
Kiribati

Peru Brazil
Samoa Cook Is. (NZ) French Polynesia (Fr)

American
Samoa (US) Bolivia
Fiji Tonga
Dominican Puerto British Virgin Paraguay
Republic Rico (US) Islands (UK)
Anguilla (UK)
U.S. Virgin Antigua and Barbuda
Islands (US)
St. Kitts Guadeloupe (Fr)
Chile Uruguay
and Nevis Argentina
This map was produced by the Netherlands Dominica
Map Design Unit of The World Bank. Antilles (Neth) Montserrat (UK) Martinique (Fr)
The boundaries, colors, denominations St. Lucia
Aruba St. Vincent and
and any other information shown on Barbados
(Neth) The Grenadines
this map do not imply, on the part of
The World Bank Group, any judgment Grenada
on the legal status of any territory, or Trinidad
any endorsement or acceptance of and Tobago
such boundaries. R.B. de Venezuela

Source: World Development Indicators.

that of the LIRP, including GDP growth rates the MDGs but is not sufficient to bring the
in figure 4.4. Under the optimistic base case, country up to the path of the base case (figure
which, unlike the other scenarios, includes 4.6). A resource-rich country has the ability
a strong recovery in the world price of the to draw down reserves accumulated from its
natural resource export, all MDG indica- resource exports or to increase government
tors continue to improve. Internal adjustment foreign borrowing, in both cases creating a
(that is, the government reduces growth in capital inflow, captured by the government.
nondevelopment spending, increases domes- This option, incorporated into the low-aid
tic taxes, and uses the resulting fiscal space internal 2 simulation, can move progress on
to expand spending on education, health, the MDGs closer to the base path. 15 But at
water and sanitation, and infrastructure) in the level reported, the LIRR country cannot
the context of stagnant export prices for the make up for the impact of the financial crisis
natural resource improves progress toward on MDGs through internal adjustment alone.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 111

IBRD 37732
APRIL 2010

nland
en)

Faeroe Norway
Iceland Islands Finland
(Den) Sweden
The Netherlands Russian Federation
Estonia
Isle of Man (UK)
Denmark Russian Latvia
Fed. Lithuania
United
Ireland Kingdom Germany Poland Belarus
Channel Islands (UK) Belgium
Ukraine
Luxembourg Moldova Kazakhstan
Liechtenstein Mongolia
France Italy Romania
Switzerland
Andorra Bulgaria Georgia Uzbekistan Kyrgyz
Azer- Rep. D.P.R.
Portugal Spain Turkey Armenia baijan Turkmenistan of Korea
Monaco Greece Tajikistan Japan
Cyprus Syrian Rep. of
Gibraltar (UK) Arab Islamic Rep. Afghanistan China
Tunisia Malta Lebanon Korea
Rep. Iraq of Iran
Morocco Israel
Jordan Kuwait
West Bank and Gaza Bahrain Pakistan Bhutan
Nepal
Algeria Libya Arab Rep. Qatar
Former
Spanish of Egypt Saudi
Sahara Arabia United Arab Bangladesh
Emirates India Myanmar Lao
Mauritania Oman P.D.R.
Cape Verde Mali N. Mariana Islands (US)
Niger Rep. of Thailand Vietnam
Senegal Chad Eritrea Yemen Guam (US)
The Gambia Burkina Sudan
Djibouti Cambodia Philippines Federated States of Micronesia Marshall Islands
Guinea-Bissau Guinea Faso
Benin Sri
Sierra Leone Côte Ghana Nigeria Central Ethiopia
Lanka Brunei
Liberia D’Ivoire African Rep. Somalia Palau
Togo Cameroon Malaysia
Maldives
Equatorial Guinea Uganda Kiribati
Congo Kenya Singapore Nauru
São Tomé and Príncipe Gabon Rwanda
Dem. Rep. of Burundi Seychelles
Papua Solomon
Congo Indonesia New Guinea Islands
Tanzania Comoros
Tuvalu
Timor-Leste
Angola Malawi
Zambia Mayotte Fiji
(Fr) Vanuatu
Zimbabwe Mauritius
Namibia Madagascar
Botswana Mozambique Réunion (Fr) New
Poland Caledonia
Australia (Fr)
Swaziland
Ukraine
Germany

Czech Rep.
Slovak Rep. South Lesotho
Africa
Austria
Hungary
Slovenia Romania
Croatia New
Zealand
Bosnia and
San Herz. Serbia
Bulgaria

Marino
Montenegro Kosovo
FYR
Macedonia
Vatican Italy Albania
City Greece

Summary and conclusions are resource poor. While understanding the


prospects for progress toward the MDGs is
This chapter presented forecasts of MDG of crucial importance as the world looks for-
outcomes at the global level and for Sub- ward to 2015 and beyond, it should be recog-
Saharan Africa based solely on alternative nized that such analysis inevitably is fraught
assumptions for growth in developing coun- with difficulties given data gaps and still-
tries. It also explored the scope for policy limited knowledge about the processes that
improvements to mitigate the impact of determine these outcomes.
slower growth on progress toward the MDGs The projections given here indicate that
through simulations using two archetypical the economic crisis will lead to a deteriora-
low-income countries, one representing those tion across all MDGs, extending beyond
that are resource rich and the other those that 2015. In all the growth scenarios, the world
112 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

FIGURE 4.5 Simulated MDG outcomes for the LIRP archetype under alternative cases

a. MDG 1: Extreme poverty and hunger b. MDG 2: Primary completion rate


70 100
90
60 80

percent
percent

70
50
60
50
40
40
30 30
2009 2011 2013 2015 2017 2019 2009 2011 2013 2015 2017 2019

base case low-aid case base case low-aid case


low-aid internal 1 case low-aid internal 1 case
low-aid internal 2 case low-aid internal 2 case

c. MDG 4: Child mortality under five d. MDG 7c: Access to safe drinking water
145 68

67
140
deaths per 1,000

66
135
percent

65
130
64
125 63

120 62
2009 2011 2013 2015 2017 2019 2009 2011 2013 2015 2017 2019

base case low-aid case base case low-aid case


low-aid internal 1 case low-aid internal 1 case
low-aid internal 2 case low-aid internal 2 case

Source: World Bank staff calculations using the Maquette for MDG Simulations (MAMS). See Go and others, forthcoming.

will meet the MDG of halving its headcount rates and will have the most difficulty achiev-
poverty rate using a poverty line of $1.25 a ing its regional poverty reduction targets.
day. However, the poverty rate in 2015 is con- The projected impact of alternative sce-
siderably higher in the low-growth scenario narios for growth on the other MDGs ana-
(18.5 percent) than in the postcrisis trend lyzed here—completion of primary school,
(15 percent), which assumes a rapid recovery under-five mortality rate, gender equality in
from the crisis. The rough magnitude of the education, and access to safe water—is more
projected effects on hunger is similar. Under- limited, although small changes in these per-
lying these figures are considerable regional centages may involve large numbers of peo-
variations. Sub-Saharan Africa poses the ple. This muted effect reflects the presence
greatest challenge—it has the highest poverty of significant lags, perhaps most obviously
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 113

FIGURE 4.6 Simulated MDG outcomes for the LIRR archetype under alternative cases

a. MDG 1: Extreme poverty and hunger b. MDG 2: Primary completion rate


80 90
75 80
70
70

percent
percent

65
60
60
50
55
40
50
45 30
2009 2011 2013 2015 2017 2019 2009 2011 2013 2015 2017 2019

base case low-aid case base case low-aid case


low-aid internal 1 case low-aid internal 1 case
low-aid internal 2 case low-aid internal 2 case

c. MDG 4: Child mortality under five d. MDG 7.c: Access to safe drinking water
140 66

135 64
deaths per 1,000

130 62
percent

125 60

120 58

115 56

110 54
2009 2011 2013 2015 2017 2019 2009 2011 2013 2015 2017 2019

base case low-aid case base case low-aid case


low-aid internal 1 case low-aid internal 1 case
low-aid internal 2 case low-aid internal 2 case

Source: World Bank staff calculations using the Maquette for MDG Simulations (MAMS). See Go and others, forthcoming.

in education. The negative effects of slower development and tax increases) lead to some
growth will make themselves more strongly improvement in the MDGs compared with
felt in the long run, however. a scenario with no improvement in policies.
Country-level simulations for the two However, the improvement from internal
low-income archetypes indicate that, if the efforts alone falls far short of that required
global economic environment and domestic to achieve the base-case levels of the MDG
GDP growth recover rapidly, continued prog- indicators. Thus, while policy matters, bet-
ress will take place across the MDGs that ter development outcomes hinge critically on
are covered here (poverty, primary comple- a rapid global recovery that improves export
tion, under-five mortality, and access to safe conditions, terms of trade, and capital flows
water). If the global recovery is weak, internal for low-income countries. Chapter 5 turns to
efforts (including spending switches toward this subject.
114 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

Annex: Forecast, Tools, and Data

BOX 4A.1 MAMS: A tool for country-level analysis of development strategies

MAMS (Maquette for MDG Simulations) is an econ- education, students successfully complete their grade,
omywide simulation model developed at the World repeat it, or drop out of their cycle. Student perfor-
Bank to analyze development strategies. The model mance depends on educational quality (quantity of
integrates a dynamic recursive computable general services per student), household welfare, public infra-
equilibrium model with an additional module that structure, wage incentives, and health status.
links specific MDG or poverty-related interventions A MAMS country database is a synthesis of infor-
to progress on poverty and other MDGs. This link mation from a variety of sources, structured to meet
is made possible by a disaggregation of government the requirements of the model. The model parameters
activities into functions related to MDG services are defi ned using this data. The main components of
(education, health, and water and sanitation) and the database are a social accounting matrix and other
infrastructure as well as a residual for other govern- data that reflect the functioning of the economy, with
ment activity. The government fi nances its activities some emphasis on human development and infrastruc-
from domestic taxes, domestic borrowing, and for- ture. More specifically, the information is primarily
eign aid (borrowing and grants). The private sector related to stock data (for labor and other production
disaggregation varies between applications; where factors, students, and population) and elasticities
private provision of MDG services is important, such (related to substitutability in production, consump-
services are included, complementing the contribution tion, and trade as well as to responses in MDG indica-
of government services to MDG progress. The factors tors to various determinants). For the simulations, it is
of production in the model typically include three also necessary to provide assumptions about the evo-
types of labor, each of which is linked to an educa- lution of policies and other factors that are exogenous
tion cycle: those with incomplete secondary education to the model.
(unskilled), those with completed secondary educa- The government policies that may be considered
tion but incomplete tertiary (semi-skilled), and those include spending—its level and allocation across differ-
with completed tertiary (skilled). The labor force ent areas, including education, health, and infrastruc-
variable depends on the functioning of the education ture—and fi nancing—policies for taxation, domestic
system in the model. The other factors of production and foreign borrowing, and foreign aid. Economic per-
include public capital stocks by government activity formance is measured by the evolution of:
and a private capital stock. Growth in the stock of
government infrastructure capital contributes to over- • poverty and other MDG targets
all growth by adding to the productivity of other pro- • macro-indicators, including GDP (split into pri-
duction activities. vate and government consumption and investment,
MAMS covers MDGs in the areas of poverty, edu- exports, and imports); the composition of the gov-
cation, health, and water and sanitation. For poverty, ernment budget, the balance of payments, and the
a log-normal distribution is assumed; other applica- savings-investment balance; total factor productiv-
tions have used microsimulations. For other MDGs, ity; and domestic and foreign debt stocks
a set of functions links the level of each indicator to • sectoral structure of production, employment,
a set of determinants. The determinants include the incomes, and trade
delivery of relevant services and other indicators, also • the labor market, including unemployment and the
allowing for the recognition that achievements in one educational composition of the labor force
MDG can have an impact on other MDGs. Other
than education, service delivery for other MDGs is Note: For more information about MAMS, see www.world
expressed relative to the size of the population. In bank.org/mams.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 115

TABLE 4A.1 Alternate scenarios for poverty reduction, based on a poverty line of $1.25 a day, by region
Scenario Region or country 1990 2005 2015 2020 1990 2005 2015 2020

Postcrisis Percentage of the population living on Number of people living on


less than $1.25 a day less than $1.25 a day (millions)

East Asia and Pacific 54.7 16.8 5.9 4.0 873 317 120 83
China 60.2 15.9 5.1 4.0 683 208 70 56
Europe and Central Asia 2.0 3.7 1.7 1.2 9 16 7 5
Latin America and the
Caribbean 11.3 8.2 5.0 4.3 50 45 30 27
Middle East and North Africa 4.3 3.6 1.8 1.5 10 11 6 6
South Asia 51.7 40.3 22.8 19.4 579 595 388 352
India 51.3 41.6 23.6 20.3 435 456 295 268
Sub-Saharan Africa 57.6 50.9 38.0 32.8 296 387 366 352
Total 41.7 25.2 15.0 12.8 1,817 1,371 918 826

Precrisis Percentage of the population living on Number of people living on less than
less than $1.25 (2005 PPP) a day $1.25 (2005 PPP) a day (millions)

East Asia and Pacific 54.7 16.8 5.5 3.5 873 317 111 73
China 60.2 15.9 5.0 3.9 683 208 69 55
Europe and Central Asia 2.0 3.7 1.5 1.1 9 16 7 5
Latin America and the
Caribbean 11.3 8.2 4.6 3.9 50 45 28 25
Middle East and North Africa 4.3 3.6 1.7 1.4 10 11 6 6
South Asia 51.7 40.3 21.5 17.9 579 595 367 326
India 51.3 41.6 22.7 19.6 435 456 283 259
Sub-Saharan Africa 57.6 50.9 35.9 29.9 296 387 346 321
Total 41.7 25.2 14.1 11.7 1,817 1,371 865 755

Low-growth Percentage of the population living on Number of people living on less than
less than $1.25 (2005 PPP) a day $1.25 (2005 PPP) a day (millions)
East Asia and Pacific 54.7 16.8 7.8 5.8 873 317 159 122
China 60.2 15.9 6.0 4.7 683 208 82 67
Europe and Central Asia 2.0 3.7 2.5 2.2 9 16 11 10
Latin America and the
Caribbean 11.3 8.2 6.5 5.7 50 45 39 36
Middle East and North Africa 4.3 3.6 3.3 2.7 10 11 12 11
South Asia 51.7 40.3 28.6 24.6 579 595 489 447
India 51.3 41.6 29.4 25.2 435 456 367 333
Sub-Saharan Africa 57.6 50.9 43.8 39.9 296 387 421 428
Total 41.7 25.2 18.5 16.3 1,817 1,371 1,132 1,053
Source: World Bank staff calculations, using PovcalNet.
116 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

TABLE 4A.2 Alternate scenarios for poverty reduction, based on a poverty line of $2.00 a day, by region
Scenario Region or country 1990 2005 2015 2020 1990 2005 2015 2020

Postcrisis Percentage of the population living on Number of people living on


less than $2.00 a day less than $2.00 a day (millions)

East Asia and Pacific 79.8 38.7 19.4 14.3 1,274 730 394 299
China 84.6 36.3 16.0 12.0 961 473 220 168
Europe and Central Asia 6.9 8.9 5.0 4.1 32 39 22 18
Latin America and the
Caribbean 19.7 16.6 11.1 9.7 86 91 67 62
Middle East and North Africa 19.7 16.9 8.3 6.6 44 52 30 26
South Asia 82.7 73.9 57.0 51.0 926 1,091 973 926
India 82.6 75.6 58.3 51.9 702 828 728 686
Sub-Saharan Africa 76.2 73.0 59.6 55.4 391 555 574 595
Total 63.2 47.0 33.7 29.8 2,754 2,557 2,060 1,926

Precrisis Percentage of the population living on Number of people living on less than
less than $2.00 (2005 PPP) a day $2.00 (2005 PPP) a day (millions)

East Asia and Pacific 79.8 38.7 18.6 13.4 1,274 730 379 280
China 84.6 36.3 15.7 11.8 961 473 216 166
Europe and Central Asia 6.9 8.9 4.5 3.7 32 39 20 16
Latin America and the
Caribbean 19.7 16.6 10.3 8.8 86 91 62 56
Middle East and North Africa 19.7 16.9 8.0 6.1 44 52 29 24
South Asia 82.7 73.9 55.5 49.0 926 1,091 946 890
India 82.6 75.6 57.2 50.9 702 828 715 674
Sub-Saharan Africa 76.2 73.0 57.6 52.4 391 555 555 563
Total 63.2 47.0 32.6 28.4 2,754 2,557 1,991 1,830

Low-growth Percentage of the population living on Number of people living on less than
less than $2.00 (2005 PPP) a day $2.00 (2005 PPP) a day (millions)
East Asia and Pacific 79.8 38.7 22.2 18.1 1,274 730 451 379
China 84.6 36.3 16.9 13.6 961 473 233 191
Europe and Central Asia 6.9 8.9 7.1 6.2 32 39 31 27
Latin America and the
Caribbean 19.7 16.6 14.5 12.9 86 91 88 82
Middle East and North Africa 19.7 16.9 14.1 11.4 44 52 52 45
South Asia 82.7 73.9 63.9 57.8 926 1,091 1,089 1,049
India 82.6 75.6 64.6 57.9 702 828 808 766
Sub-Saharan Africa 76.2 73.0 65.1 62.5 391 555 627 671
Total 63.2 47.0 38.2 34.9 2,754 2,557 2,338 2,254
Source: World Bank staff calculations, using PovcalNet.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 117

TABLE 4A.3 Detailed data for archetypes


Median values by archetype, selected variables
Low-income, resource-poor Low-income, resource-rich
Variable (LIRP) (LIRR)

Poverty headcount ratio at $1.25 a day (% of population) 49.6 61.8


Poverty headcount ratio at $2.00 a day (% of population) 76.7 80.5
Elasticity of poverty to income –1.01 –1.01
Poverty headcount ratio at national poverty line (% of population) 44.2 45.4
Primary school completion rate, total (% gross) 55.8 59.9
Primary school enrollment (% gross) 95.9 95.2
Secondary school enrollment (% gross) 31.6 35.5
Tertiary school enrollment (% gross) 3.2 4.7
Under-five mortality rate (per 1,000) 115.2 141.6
Maternal mortality ratio, modeled estimate (per 100,000 live births) 720.0 825.0
Maternal mortality ratio, national estimate (per 100,000 live births) 478.0 613.0
Improved water source (% of population with access) 65.0 60.0
Improved sanitation facilities (% of population with access) 30.0 31.5
Foreign direct investment, net inflows (% of GDP) 2.7 6.5
Foreign direct investment, net outflows (% of GDP) 0.0 0.0
Foreign direct investment inflow outflows (% of GDP) 2.7 6.5
Net current transfers, remittances (% of GDP) 8.9 4.5
Official current transfers, receipts, foreign aid (% of GDP) 2.5 1.7
External debt stocks (% GNI) 29.9 49.5
External debt stocks private (% GNI) 0.0 0.0
External debt stocks public (% GNI) 29.9 49.5
External debt stocks public, median (% GDP) 28.0 48.6
Gross fixed capital formation (% of GDP) 20.8 18.3
Gross fixed capital formation, private (% of GDP) 10.7 11.3
Labor force participation rate (% of total population ages 15–64) 74.3 71.4
Resource exports (% of GDP) 0.4 19.0
Resource exports (% of merchandise exports) 3.4 67.9
Mining value added (% of GDP) 0.7 3.3
Interest payment on private external debt (% of GDP) 0.0 0.0
Interest payment on public external debt (% of GDP) 0.3 0.5
Source: World Bank 2009b.
118 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS GLOBAL MONITORING REPORT 2010

Notes tendency for the indicator to level off would


be weaker, especially for the base case.
1. Resource intensity is an important factor in 15. In the model this is done by increasing foreign
the performance of low-income countries and borrowing, which reduces the net asset posi-
has been used to classify developing countries tion of the country relative to the rest of the
in several studies; see Collier and O’Connell world and is equivalent to drawing down for-
(2006); IMF (2006); Ndulu and others (2007); eign exchange reserves or liquidating foreign
and Arbache, Go, and Page (2008). investment financed by the natural resource in
2. World Bank 2004b. the past. Here, the annual growth rates in for-
3. Even short-term assessments are necessar- eign borrowing are assumed to be twice the
ily projections because of the infrequency annual growth rates in the base. As a result,
of the underlying data. Household surveys the foreign debt stock in foreign currency is
of incomes and expenditures are generally 30 percent higher in 2020 for the low-aid
undertaken only every five or more years in internal 2 case than for the other scenarios.
many developing countries. Relative to GDP, the foreign debt stock is
4. The estimation uses a logistic function, simi- around 10 percentage points higher in 2020
lar to Clemens, Kenny, and Moss (2007) but for the low-aid internal 2 case than for the
with per capita income as a key determinant low-aid internal 1 case (which has a slightly
instead of a time trend. Income rather than slower rate of GDP growth and similar evolu-
social spending is used as the independent tion for the exchange rate). References
variable because of data and other difficulties
with fiscal adjustment and public expendi-
tures. The logistic curve was used for the pro- References
jections because it has a smoother transition
across income levels, although the elasticity Adams, C. S., and D. Bevan. 2000. “The Cash
form (double-log regressions) by income level Budget as Restraint: The Experience of Zam-
or region yielded similar results. bia.” In Investment and Risk in Africa, ed. P.
5. World Bank PovcalNet database. Collier and C. Patillo. New York: St. Martin’s
6. See also IMF (2010); World Bank (2010b). Press.
7. World Bank 2003, p. 41. These calculations Arbache, J., D. Go, and J. Page. 2008. “Is Africa’s
update estimates found in Ravallion (2009) Economy at a Turning Point?” In Africa at
and World Bank (2009a). a Turning Point? ed. D. Go and J. Page, pp.
8. Tiwari and Zaman 2010; World Bank 2010a. 13–85. Washington, DC: World Bank.
9. Friedman and Schady 2009. Bourguignon, François, Carolina Diaz-Bonilla,
10. The low-income countries are disaggregated and Hans Lofgren. 2008. “Aid, Service Deliv-
into resource rich and resource poor using ery and the Millennium Development Goals
data on exports of fuel ore and minerals as a in an Economywide Framework.” In The
share of merchandise exports. See table 4A.3 Impact of Macroeconomic Policies on Pov-
in the annex for more details. erty and Income Distribution: Macro-Micro
11. See Bourguignon, Diaz-Bonilla, and Lofgren Evaluation Techniques and Tools, ed. Fran-
(2008) and Lofgren and Diaz-Bonilla (2010) çois Bourguignon, Maurizio Bussolo, and Luiz
as well as www.worldbank.org/mams. A. Pereira da Silva, pp. 283–315. Washington,
12. The growth rate is set at 3 percent, the DC: World Bank.
assumed annual GDP growth in developed Clemens, M. A., C. J. Kenny, and T. J. Moss.
countries from 2010 onward. 2007. “The Trouble with the MDGs: Con-
13. This is 15 percent higher than the annual fronting Expectations of Aid and Development
GDP growth in the pessimistic scenario with Success.” World Development 35 (5): 735–51.
internal adjustment (low-aid internal 1 case). Collier, P., and S. O’Connell. 2006. “Opportu-
14. The primary gross completion rate (MDG nities and Choices.” Explaining African Eco-
2) is defi ned as the total number of primary nomic Growth, ch. 2 of synthesis vol. Nairobi:
school graduates (regardless of age) as a share African Economic Research Consortium.
of the total population of the theoretical Collier, P., and B. Goderis. 2007. “Commodity
graduation age. If MDG 2 were measured by Prices, Growth, and the Natural Resource
the net completion rate (the number of grad- Curse: Reconciling a Conundrum.” Oxford
uates of the theoretical right age as a share University, Centre for the Study of African
of the total population of the same age), the Economies, Oxford, U.K.
GLOBAL MONITORING REPORT 2010 OUTLOOK FOR THE MILLENNIUM DEVELOPMENT GOALS 119

Devarajan S., and R. Reinikka. 2004. “Making Ndulu, B. J., L. Chakroborti, L. Lijane, V.
Services Work for the Poor.” Journal of Afri- Ramachandran, and J. Wolgin. 2007. Chal-
can Economies 13 (Supp. 1): i142–66. lenges of Africa Growth: Opportunities, Con-
Dinh, H., A. Adugna, and B. Myers. 2002. “The straints, and Strategic Directions. Washington,
Impact of Cash Budgets on Poverty Reduc- DC: World Bank.
tion in Zambia: A Case Study of the Confl ict Rajkumar, A., and V. Swaroop. 2008. “Public
between Well-Intentioned Macroeconomic Spending and Outcomes: Does Governance
Policy and Service Delivery to the Poor.” Policy Matter?” Journal of Development Economics
Research Working Paper 2914. World Bank, 86 (1): 96–111.
Washington, DC. Ravallion, M. 2009. “The Crisis and the World’s
Filmer, D., J. Hammer, and L. Pritchett. 2000. Poorest.” Development Outreach, World Bank,
“Weak Links in the Chain: A Diagnosis of Washington, DC (December).
Health Policy in Poor Countries.” World Bank Tiwari, S., and H. Zaman 2010. “The Impact of
Research Observer 15 (2): 188–224. Economic Shocks on Global Undernourish-
———. 2002. “Weak Links in the Chain II: A Pre- ment” Policy Research Working Paper 5215.
scription for Health Policy in Poor Countries.” World Bank, Washington, DC.
World Bank Research Observer 17 (1): 47–66. Wagstaff, A., and M. Claeson. 2004. Rising to
Filmer, D., and L. Pritchett. 1999. “The Impact the Challenges: The Millennium Development
of Public Spending on Health: Does Money Goals for Health. Washington, DC: World
Matter?” Social Science and Medicine 49 (10): Bank.
1309–23. World Bank. 2004a. Global Monitoring Report
Friedman, J., and N. Schady. 2009. “How Many 2004: Policies and Actions for Achieving the
More Infants Are Likely to Die in Africa as a Millennium Development Goals and Related
Result of the Global Financial Crisis?” Policy Outcomes. Washington, DC: World Bank.
Research Working Paper 5023. World Bank, ———. 2004b. World Development Report
Washington, DC. 2004: Making Services Work for Poor People.
Go, D., H. Lofgren, S. Robinson, and K. Thier- Washington, D.C.: World Bank.
felder. Forthcoming. “The Impact of the ———. 2009a. “Protecting Progress: The Chal-
Global Economic Crisis on MDGs in Arche- lenge Facing Low-Income Countries in the
typical Developing Countries.” World Bank, Global Recession.” Background paper prepared
Washington, DC. for the G-20 Leaders’ Meeting, Pittsburgh, PA,
IMF (International Monetary Fund). 2006. September 24–25.
Regional Economic Outlook: Sub-Saharan ———. 2009b. World Development Indicators.
Africa. Washington, DC. Washington, DC: World Bank.
———. 2010. World Economic Outlook. Wash- ———2010a. “Food Price Watch.” Washington,
ington, DC (January). DC (February).
Lofgren, H., and C. Diaz-Bonilla. 2010. “MAMS: ———. 2010b. Global Economic Prospects 2010.
An Economywide Model for Development Washington, DC: World Bank.
Strategy Analysis.” World Bank, Washington,
DC.
5
The International Community and
Development: Trade, Aid, and the
International Financial Institutions

T
he global economic crisis severely progress is required to strengthen aid effec-
reduced developing-country external tiveness and improve aid allocation. Reach-
resources by drastically curtailing their ing agreement on the Doha Round would
export revenues and their access to private support an open trading environment and
capital flows. As elaborated in previous chap- generate substantial increases in market
ters, the