Sie sind auf Seite 1von 258

Wor l d Eco no m i c a nd F i na nci a l S ur v e y s

World Economic Outlook


April 2017

Gaining Momentum?

I N T E R N A T I O N A L M O N E T A R Y F U N D
2017 International Monetary Fund

Cover and Design: Luisa Menjivar and Jorge Salazar


Composition: AGS, An RR Donnelley Company

Cataloging-in-Publication Data

Joint Bank-Fund Library

Names: International Monetary Fund.


Title: World economic outlook (International Monetary Fund)
Other titles: WEO | Occasional paper (International Monetary Fund) | World economic and
financial surveys.
Description: Washington, DC : International Monetary Fund, 1980- | Semiannual | Some
issues also have thematic titles. | Began with issue for May 1980. | 1981-1984: Occasional
paper / International Monetary Fund, 0251-6365 | 1986-: World economic and financial
surveys, 0256-6877.
Identifiers: ISSN 0256-6877 (print) | ISSN 1564-5215 (online)
Subjects: LCSH: Economic developmentPeriodicals. | International economic relations
Periodicals. | Debts, ExternalPeriodicals. | Balance of paymentsPeriodicals. |
International financePeriodicals. | Economic forecastingPeriodicals.
Classification: LCC HC10.W79

HC10.80

ISBN 978-1-47556-465-5 (paper)


978-1-47559-214-6 (Web PDF)
978-1-47559-211-5 (ePub)
978-1-47559-212-2 (Mobi)

The World Economic Outlook (WEO) is a survey by the IMF staff published twice a
year, in the spring and fall. The WEO is prepared by the IMF staff and has benefited
from comments and suggestions by Executive Directors following their discussion of the
report on April 4, 2017. The views expressed in this publication are those of the IMF
staff and do not necessarily represent the views of the IMFs Executive Directors or their
national authorities.

Recommended citation: International Monetary Fund. 2017. World Economic Outlook:


Gaining Momentum? Washington, April.

Publication orders may be placed online, by fax, or through the mail:


International Monetary Fund, Publication Services
P.O. Box 92780, Washington, DC 20090, U.S.A.
Tel.: (202) 623-7430 Fax: (202) 623-7201
E-mail: publications@imf.org
www.imfbookstore.org
www.elibrary.imf.org
CONTENTS

Assumptions and Conventions x

Further Information and Data xi

Preface xii

Foreword xiii

Executive Summary xv

Chapter 1. Global Prospects and Policies 1


Recent Developments and Prospects 1
The Forecast 13
Risks 22
Policy Priorities 29
Scenario Box 1. Permanent U.S. Fiscal Expansions 37
Box 1.1. Conflict, Growth, and Migration 40
Box 1.2. Tackling Measurement Challenges of Irish Economic Activity 43
Special Feature: Commodity Market Developments and Forecasts, with a Focus on
the Role of Technology and Unconventional Sources in the Global Oil Market 52
References 63

Chapter 2. Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment 65
Introduction 65
Emerging Market and Developing Economy Growth Performance over Time 67
How Important Are External Conditions? 69
How Do External Conditions Influence the Occurrence of Growth Episodes? 76
The Role of Policies and Structural Attributes in Mediating the Impact of External Conditions 82
Taking Stock: What Does the Current Environment Imply for Growth Prospects in
Emerging Market and Developing Economies? 87
Conclusion 88
Box 2.1. Within-Country Trends in Income per Capita: The Cases of Brazil, Russia, India, China,
and South Africa 89
Box 2.2. Growing with Flows: Evidence from Industry-Level Data 91
Box 2.3. The Evolution of Emerging Market and Developing Economies Trade Integration
with Chinas Final Demand 94
Box 2.4. Shifts in the Global Allocation of Capital: Implications for Emerging Market and
Developing Economies 97
Annex 2.1. Data 100
Annex 2.2. Channels through Which Emerging Market and Developing Economies Have
Narrowed Income Differentials with Advanced Economies 102
Annex 2.3. Estimation of the Impact of External Conditions on Emerging Market and
Developing Economy Growth 104
Annex 2.4. Identification of Growth Episodes 106

International Monetary Fund | April 2017 iii


WORLD ECONOMIC OUTLOOK: GAINING MOMENTUM?

Annex 2.5. Estimation of the Influence of External Conditions on the Likelihood of


Experiencing Persistent Accelerations and Reversals 109
Annex 2.6. Analysis of Domestic Attributes in Mediating the Impact of External Conditions 113
References 118

Chapter 3. Understanding the Downward Trend in Labor Income Shares 121


Introduction 121
Trends in the Labor Share of Income: Key Facts 126
Drivers of the Labor Share of Income: Key Concepts and Mechanisms 127
Analyzing Trends in the Labor Share of Income: Empirical Analysis 133
Summary and Policy Implications 140
Box 3.1. Technological Progress and Labor Shares: A Historical Overview 142
Box 3.2. The Elasticity of Substitution between Capital and Labor: Concept and Estimation 145
Box 3.3. Routine Tasks, Automation, and Economic Dislocation around the World 148
Box 3.4. Adjustments to the Labor Share of Income 152
Annex 3.1. Wages and Deflators 155
Annex 3.2. A Theoretical Model of Relative Cost of Capital, Offshoring, and Labor Shares
of Income in Advanced Economies and Emerging Market and Developing Economies 156
Annex 3.3. Country Coverage and Data 159
Annex 3.4. Methodology 161
Annex 3.5. Robustness and Additional Tables 162
References 170

Statistical Appendix 173


Assumptions 173
Whats New 174
Data and Conventions 174
Country Notes 175
Classification of Countries 175
General Features and Composition of Groups in the World Economic Outlook Classification 176
Table A. Classification by World Economic Outlook Groups and Their Shares in Aggregate GDP,
Exports of Goods and Services, and Population, 2016 177
Table B. Advanced Economies by Subgroup 178
Table C. European Union 178
Table D. Emerging Market and Developing Economies by Region and Main Source of
Export Earnings 179
Table E. Emerging Market and Developing Economies by Region, Net External Position, and
Status as Heavily Indebted Poor Countries and Low-Income Developing Countries 180
Table F. Economies with Exceptional Reporting Periods 182
Table G. Key Data Documentation 183
Box A1. Economic Policy Assumptions Underlying the Projections for Selected Economies 193
List of Tables
Output (Tables A1A4) 198
Inflation (Tables A5A7) 205
Financial Policies (Table A8) 210
Foreign Trade (Table A9) 211
Current Account Transactions (Tables A10A12) 213
Balance of Payments and External Financing (Table A13) 220

iv International Monetary Fund | April 2017


CONTENTS

Flow of Funds (Table A14) 224


Medium-Term Baseline Scenario (Table A15) 227

World Economic Outlook, Selected Topics 229

IMF Executive Board Discussion of the Outlook, April 2017 237

Tables
Table 1.1. Overview of the World Economic Outlook Projections 2
Scenario Table 1. The Impact of Fiscal Measures on the Deficit 38
Table 1.2.1. Ireland: Balance of Payments and International Investment Position 44
Annex Table 1.1.1. European Economies: Real GDP, Consumer Prices, Current
Account Balance, and Unemployment 46
Annex Table 1.1.2. Asian and Pacific Economies: Real GDP, Consumer Prices,
Current Account Balance, and Unemployment 47
Annex Table 1.1.3. Western Hemisphere Economies: Real GDP, Consumer Prices,
Current Account Balance, and Unemployment 48
Annex Table 1.1.4. Commonwealth of Independent States Economies: Real GDP,
Consumer Prices, Current Account Balance, and Unemployment 49
Annex Table 1.1.5. Middle East, North African Economies, Afghanistan, and Pakistan:
Real GDP, Consumer Prices, Current Account Balance, and Unemployment 50
Annex Table 1.1.6. Sub-Saharan African Economies: Real GDP, Consumer Prices,
Current Account Balance, and Unemployment 51
Table 1.SF.1. Unconventional Oil Production, 2016 57
Table 2.2.1. Industry Growth with Low versus High Levels of Capital Inflows 92
Table 2.2.2. Capital Inflows and Industry Growth 93
Annex Table 2.1.1. Data Sources 100
Annex Table 2.1.2. Sample of Emerging Market and Developing Economies Included in the Analyses 101
Annex Table 2.1.3. Pairwise Correlation between External Conditions Variables 102
Annex Table 2.3.1. Estimation Results from Linear Panel Growth Regression 105
Annex Table 2.3.2. Estimation Results from Linear Panel Growth Regression: Robustness Exercises 105
Annex Table 2.4.1. Persistent Acceleration Episodes 107
Annex Table 2.4.2. Reversal Episodes 109
Annex Table 2.5.1. Logistic Estimates of the Effects of External Conditions Variables on the
Odds Ratio of Persistent Accelerations 110
Annex Table 2.5.2. Logistic Estimates of the Effects of External Conditions Variables on the
Odds Ratio of Reversals 110
Annex Table 2.6.1. Logistic Estimates of the Effects of Policy Variables on the Odds Ratio of
Persistent Accelerations 115
Annex Table 2.6.2. Logistic Estimates of the Effects of Policy Variables on the Odds Ratio of Reversals 116
Annex Table 3.3.1. Country Coverage 160
Annex Table 3.3.2. Data Sources 160
Annex Table 3.5.1. Baseline Aggregate Results 163
Annex Table 3.5.2. Stacked Aggregate Results 164
Annex Table 3.5.3.A. Aggregate Results, Robustness (User Cost) 165
Annex Table 3.5.3.B. Aggregate Results, Robustness (Alternative Measure of Offshoring) 166
Annex Table 3.5.4. Aggregate Results, Robustness (Other Robustness Checks) 166
Annex Table 3.5.5. Aggregate Results, Robustness (Measurement Issues) 167
Annex Table 3.5.6. Baseline Sectoral Results 167

International Monetary Fund | April 2017 v


WORLD ECONOMIC OUTLOOK: GAINING MOMENTUM?

Annex Table 3.5.7. Aggregate Results by Skill Level 168


Annex Table 3.5.8. Sectoral Results by Skill Level 168
Annex Table 3.5.9. Sectoral Results by Skill Level, Controlling for Skill Composition 169
Annex Table 3.5.10. Sectoral Results by Skill Level, Controlling for Policy and Institution Variables 169
Online Tables
Table B1. Advanced Economies: Unemployment, Employment, and Real GDP per Capita
Table B2. Emerging Market and Developing Economies: Real GDP
Table B3. Advanced Economies: Hourly Earnings, Productivity, and Unit Labor Costs in Manufacturing
Table B4. Emerging Market and Developing Economies: Consumer Prices
Table B5. Summary of Fiscal and Financial Indicators
Table B6. Advanced Economies: General and Central Government Net Lending/Borrowing
and General Government Net Lending/Borrowing Excluding Social Security Schemes
Table B7. Advanced Economies: General Government Structural Balances
Table B8. Emerging Market and Developing Economies: General Government Net
Lending/Borrowing and Overall Fiscal Balance
Table B9. Emerging Market and Developing Economies: General Government Net Lending/Borrowing
Table B10. Selected Advanced Economies: Exchange Rates
Table B11. Emerging Market and Developing Economies: Broad Money Aggregates
Table B12. Advanced Economies: Export Volumes, Import Volumes, and Terms of Trade in
Goods and Services
Table B13. Emerging Market and Developing Economies by Region: Total Trade in Goods
Table B14. Emerging Market and Developing Economies by Source of Export Earnings:
Total Trade in Goods
Table B15. Summary of Current Account Transactions
Table B16. Emerging Market and Developing Economies: Summary of External Debt and Debt Service
Table B17. Emerging Market and Developing Economies by Region: External Debt by Maturity
Table B18. Emerging Market and Developing Economies by Analytical Criteria: External
Debt by Maturity
Table B19. Emerging Market and Developing Economies: Ratio of External Debt to GDP
Table B20. Emerging Market and Developing Economies: Debt-Service Ratios
Table B21. Emerging Market and Developing Economies, Medium-Term Baseline Scenario:
Selected Economic Indicators
Figures
Figure 1.1. Global Activity Indicators 4
Figure 1.2. Recent Trends in Global Production 5
Figure 1.3. Global Trade and Fixed Investment Growth 5
Figure 1.4. Commodity and Oil Markets 6
Figure 1.5. Global Inflation 7
Figure 1.6. Advanced Economies: Monetary and Financial Market Conditions 8
Figure 1.7. Real Effective Exchange Rate Changes, August 2016March 2017 8
Figure 1.8. Emerging Market Economies: Interest Rates 9
Figure 1.9. Emerging Market Economies: Equity Markets and Credit 10
Figure 1.10. Emerging Market Economies: Capital Flows 10
Figure 1.11. Revisions to 2016 Growth and Output Gaps in 2015 11
Figure 1.12. GDP Growth, 19992021 12
Figure 1.13. Emerging Markets: Terms-of-Trade Windfall Gains and Losses 13
Figure 1.14. Total Factor Productivity 13
Figure 1.15. Fiscal Indicators 14

vi International Monetary Fund | April 2017


CONTENTS

Figure 1.16. Global Current Account Balances 20


Figure 1.17. Net International Investment Position 21
Figure 1.18. Growth for Creditors and Debtors 22
Figure 1.19. Risks to the Global Outlook 28
Figure 1.20. Recession and Deflation Risks 29
Scenario Figure 1. Fiscal Stimulus in the United States 37
Figure 1.1.1. Conflict-Related Fatalities and Number of Countries Affected by Conflict 40
Figure 1.1.2. Global GDP Shares of Conflict-Affected Countries and Impact
of Conflict on Growth 41
Figure 1.1.3. Impact of Conflict Onset 42
Figure 1.2.1. Irish National Accounts 43
Figure 1. SF.1. Commodity Market Developments 52
Figure 1. SF.2. Unconventional Oil, Proven Reserves, and Production, 2016 56
Figure 1. SF.3. Evolution of Research and Development Expenditure in Select
Integrated Oil and Service Companies 58
Figure 1. SF.4. Historical Evolution of Global Capital and Operational Expenditures 59
Figure 1. SF.5. Growth in Unconventional World Oil Production and Real Oil Prices 60
Figure 1. SF.6. Global Oil Supply Cost Curve and Breakeven Prices 60
Figure 1. SF.7. North American Shale Oil Wells at Different West Texas Intermediate
Oil Prices and Cost Deflation Scenarios 61
Figure 1. SF.8. Unconventional Oil Production Outlook Vintages 61
Figure 1. SF.9. Unconventional Oil Outlook 62
Figure 2.1. Contribution to Global Output and Consumption Growth 67
Figure 2.2. Emerging Market and Developing Economies, Relative Income in
Purchasing-Power-Parity Terms 68
Figure 2.3. Distribution of Income per Capita in EMDEs in the 1970s and the 2010s 68
Figure 2.4. Change in Real Income per Capita in EMDEs Relative to the United States over Decades 69
Figure 2.5. Elasticity of Medium-Term GDP per Capita Growth in EMDEs with Respect to
External Conditions 71
Figure 2.6. Average Contribution to GDP per Capita Growth 73
Figure 2.7. Relative Average Contribution to GDP per Capita Growth among External
Conditions Variables 73
Figure 2.8. Average Contribution of Terms of Trade to GDP per Capita Growth, by
Groups of Economies 74
Figure 2.9. Variance of GDP per Capita Growth Accounted for by Each External Conditions Variable 75
Figure 2.10. Contribution of Other Common Factors to GDP per Capita Growth and
Selected Global Variables 75
Figure 2.11. Growth Episodes in EMDEs, 19702015 77
Figure 2.12. Cumulative Growth during Episodes, 19702015 77
Figure 2.13. Normalized GDP per Capita during Growth Episodes and Their Aftermath, 19702015 78
Figure 2.14. Cumulative Growth Rate of Real Income per Capita during Episodes versus
Average Growth Rate of Real Income per Capita during 19702015 79
Figure 2.15. Event Analysis: Persistent Accelerations and Reversals, 19702015 79
Figure 2.16. Event Analysis: Persistent and Nonpersistent Accelerations, 19702015 80
Figure 2.17. Change in the Probability of Occurrence of Growth Episodes, 19702015 81
Figure 2.18. Domestic Attributes across Persistent Accelerations and Reversals, 19702015 83
Figure 2.19. Change in Marginal Effect of External Conditions When Domestic Attributes Improve 85
Figure 2.20. Actual and Projected External Conditions for Emerging Market and
Developing Economies 87
Figure 2.1.1. Decomposition of Selected Emerging Market Economies by Province 89

International Monetary Fund | April 2017 vii


WORLD ECONOMIC OUTLOOK: GAINING MOMENTUM?

Figure 2.2.1. Capital Inflows and Industry Growth, 19982010 92


Figure 2.3.1. Value Added in Chinas Final Demand 94
Figure 2.3.2. Relative Changes in Country Exposures to Chinas Final Demand 95
Figure 2.3.3. Sector Composition of Value Added in Chinas Final Demand 95
Figure 2.3.4. Sector Composition of Commodity-Exporting Economies Foreign Value Added 96
Figure 2.4.1. EMDEs: Current Account Balance by Group and Net Capital Inflows by Type 97
Figure 2.4.2. Distribution of EMDEs Average Current Account Balances, 200016 98
Figure 2.4.3. Correlation between Capital Flows and per Capita Real GDP Growth 98
Annex Figure 2.1.1. Correlation between Country-Specific External Conditions Variables and Global
Variables over Time 102
Annex Figure 2.2.1. Changes in Levels of Selected Variables Relative to the United States 103
Annex Figure 2.4.1. Persistent Acceleration Episodes by Region 108
Annex Figure 2.4.2. Reversal Episodes by Region 108
Annex Figure 2.5.1. Change in the Odds Ratio of Occurrence of Growth Episodes, 19702015 111
Annex Figure 2.5.2. Change in the Odds Ratio of Occurrence of Growth Episodes by Subsamples,
19702015 112
Annex Figure 2.5.3. Change in the Probability of Occurrence of Growth Episodes (Marginal Effect)
Using Seven-Year Durations, 19702015 112
Annex Figure 2.5.4. Change in the Probability of Occurrence of Persistent Accelerations
(Marginal Effect) by Type of Acceleration, 19702015 113
Annex Figure 2.6.1. Change in the Probability of Occurrence of Growth Episodes (Marginal Effect),
19702015 117
Annex Figure 2.6.2. Reversals: Change in the Marginal Effect of External Financial Conditions
When Selected Domestic Attributes Improve 117
Figure 3.1. Evolution of the Labor Share of Income 122
Figure 3.2. Labor Shares and Income Inequality 122
Figure 3.3. Distribution of Estimated Trends in Labor Shares, 19912014 123
Figure 3.4. Estimated Trends in Labor Shares by Country and Sector 127
Figure 3.5. Labor Share Evolutions and Labor Force Composition by Skill Level 128
Figure 3.6. Trends in Potential Drivers of Labor Shares 129
Figure 3.7. Change in the Relative Price of Investment and Capital Intensity 130
Figure 3.8. Changes in Global Value Chain Participation and Capital Intensity 132
Figure 3.9. Evolution of the Adjusted Labor Share of Income 133
Figure 3.10. Shift-Share Analysis 134
Figure 3.11. Aggregate Results 136
Figure 3.12. Heterogeneity across Sectors and Countries 137
Figure 3.13. Sectoral Results, Advanced Economies 138
Figure 3.14. Contributions to Aggregate Labor Share Change by Skill, 19952009 139
Figure 3.1.1. Labor Share and Inequality in the United Kingdom 142
Figure 3.2.1. Change in Labor Share versus Change in Relative Price of Investment, 19922014 146
Figure 3.2.2. Distribution of Initial Routine Exposure, 199095 146
Figure 3.2.3. Estimated Elasticity of Substitution by Two-Digit Industry 147
Figure 3.2.4. Elasticity of Substitution versus Routine Exposure by Sector, 19922014 147
Figure 3.3.1. Initial Routine Exposure across Industries, 19952000 149
Figure 3.3.2. Routine Exposure across Country Groups and over Time, 19902015 150
Figure 3.3.3. Initial Routine Exposure and Subsequent Change in Routine Exposure, 19902015 150
Figure 3.3.4. Structural Transformation and Routine Exposure, 19902015 151
Figure 3.4.1. Adjustments to the Labor Share of Income in the United States, 19482016 153
Figure 3.4.2. Adjustments to the Labor Share of Income in Large Advanced Economies, 19802014 153

viii International Monetary Fund | April 2017


CONTENTS

Figure 3.4.3. Long Changes in Unadjusted and Adjusted Labor Shares, 19912014 154
Figure 3.4.4. Long Changes in Self-Employment and Depreciation, 19912014 154
Annex Figure 3.1.1. Decomposition of the Labor Share of Income, 19912014 155
Annex Figure 3.1.2. Product Wages, Consumption Wages, and Productivity in Manufacturing 156
Annex Figure 3.2.1. Impact of the Costs of Capital and Offshoring on the Set of Tasks Offshored
from a High-Wage Country to a Low-Wage Country 158
Annex Figure 3.4.1. Estimated Trends in Labor Shares across the World 161
Annex Figure 3.4.2. Heterogeneity in the Evolution of Key Drivers of the Labor Share 162

International Monetary Fund | April 2017 ix


WORLD ECONOMIC OUTLOOK: TENSIONS FROM THE TWO-SPEED RECOVERY

ASSUMPTIONS AND CONVENTIONS

A number of assumptions have been adopted for the projections presented in the World Economic Outlook (WEO).
It has been assumed that real effective exchange rates remained constant at their average levels during February 1 to
March 1, 2017, except for those for the currencies participating in the European exchange rate mechanism II (ERM
II), which are assumed to have remained constant in nominal terms relative to the euro; that established policies of
national authorities will be maintained (for specific assumptions about fiscal and monetary policies for selected econo-
mies, see Box A1 in the Statistical Appendix); that the average price of oil will be $55.23 a barrel in 2017 and $55.06 a
barrel in 2018 and will remain unchanged in real terms over the medium term; that the six-month London interbank
offered rate (LIBOR) on U.S. dollar deposits will average 1.7 percent in 2017 and 2.8 percent in 2018; that the three-
month euro deposit rate will average 0.3 percent in 2017 and 0.2 percent in 2018; and that the six-month Japanese
yen deposit rate will yield on average 0.0 percent in 2017 and 2018. These are, of course, working hypotheses rather
than forecasts, and the uncertainties surrounding them add to the margin of error that would in any event be involved
in the projections. The estimates and projections are based on statistical information available through April 3, 2017.
The following conventions are used throughout the WEO:
. . . to indicate that data are not available or not applicable;
between years or months (for example, 201617 or JanuaryJune) to indicate the years or months
covered, including the beginning and ending years or months; and
/ between years or months (for example, 2016/17) to indicate a fiscal or financial year.
Billion means a thousand million; trillion means a thousand billion.
Basis points refers to hundredths of 1 percentage point (for example, 25 basis points are equivalent to of 1
percentage point).
Data refer to calendar years, except in the case of a few countries that use fiscal years. Please refer to Table F in
the Statistical Appendix, which lists the economies with exceptional reporting periods for national accounts and
government finance data for each country.
For some countries, the figures for 2016 and earlier are based on estimates rather than actual outturns. Please
refer to Table G in the Statistical Appendix, which lists the latest actual outturns for the indicators in the national
accounts, prices, government finance, and balance of payments indicators for each country.
On October 1, 2016, the Chinese renminbi joined the U.S. dollar, euro, yen, and British pound in the IMFs
SDR basket.
Nauru is the latest country added to the WEO database, expanding it to a total of 192 countries.
Belarus redenominated its currency by replacing 10,000 old Belarusian rubles with 1 new Belarusian ruble.
Local currency data for Belarus are expressed in the new currency starting with the April 2017 WEO database.
In the tables and figures, the following conventions apply:
If no source is listed on tables and figures, data are drawn from the WEO database.
When countries are not listed alphabetically, they are ordered on the basis of economic size.
Minor discrepancies between sums of constituent figures and totals shown reflect rounding.
As used in this report, the terms country and economy do not in all cases refer to a territorial entity that is
a state as understood by international law and practice. As used here, the term also covers some territorial entities
that are not states but for which statistical data are maintained on a separate and independent basis.
Composite data are provided for various groups of countries organized according to economic characteristics or
region. Unless noted otherwise, country group composites represent calculations based on 90 percent or more of
the weighted group data.
The boundaries, colors, denominations, and any other information shown on the maps do not imply, on the
part of the International Monetary Fund, any judgment on the legal status of any territory or any endorsement or
acceptance of such boundaries.

x International Monetary Fund | April 2017


WORLD ECONOMIC OUTLOOK: TENSIONS FROM THE TWO-SPEED RECOVERY

FURTHER INFORMATION AND DATA

This version of the World Economic Outlook (WEO) is available in full through the IMF eLibrary (www.elibrary.
imf.org) and the IMF website (www.imf.org). Accompanying the publication on the IMF website is a larger compila-
tion of data from the WEO database than is included in the report itself, including files containing the series most
frequently requested by readers. These files may be downloaded for use in a variety of software packages.
The data appearing in the WEO are compiled by the IMF staff at the time of the WEO exercises. The histori-
cal data and projections are based on the information gathered by the IMF country desk officers in the context
of their missions to IMF member countries and through their ongoing analysis of the evolving situation in each
country. Historical data are updated on a continual basis as more information becomes available, and structural
breaks in data are often adjusted to produce smooth series with the use of splicing and other techniques. IMF
staff estimates continue to serve as proxies for historical series when complete information is unavailable. As a
result, WEO data can differ from those in other sources with official data, including the IMFs International
Financial Statistics.
The WEO data and metadata provided are as is and as available, and every effort is made to ensure their
timeliness, accuracy, and completeness, but it cannot be guaranteed. When errors are discovered, there is a con-
certed effort to correct them as appropriate and feasible. Corrections and revisions made after publication are
incorporated into the electronic editions available from the IMF eLibrary (www.elibrary.imf.org) and on the IMF
website (www.imf.org). All substantive changes are listed in detail in the online tables of contents.
For details on the terms and conditions for usage of the WEO database, please refer to the IMF Copyright and
Usage website (www.imf.org/external/terms.htm).
Inquiries about the content of the WEO and the WEO database should be sent by mail, fax, or online forum
(telephone inquiries cannot be accepted):
World Economic Studies Division
Research Department
International Monetary Fund
700 19th Street, N.W.
Washington, D.C. 20431, U.S.A.
Fax: (202) 623-6343
Online Forum: www.imf.org/weoforum

International Monetary Fund | April 2017 xi


PREFACE

The analysis and projections contained in the World Economic Outlook are integral elements of the IMFs
surveillance of economic developments and policies in its member countries, of developments in international
financial markets, and of the global economic system. The survey of prospects and policies is the product of a
comprehensive interdepartmental review of world economic developments, which draws primarily on information
the IMF staff gathers through its consultations with member countries. These consultations are carried out in par-
ticular by the IMFs area departmentsnamely, the African Department, Asia and Pacific Department, European
Department, Middle East and Central Asia Department, and Western Hemisphere Departmenttogether with
the Strategy, Policy, and Review Department; the Monetary and Capital Markets Department; and the Fiscal
Affairs Department.
The analysis in this report was coordinated in the Research Department under the general direction of Maurice
Obstfeld, Economic Counsellor and Director of Research. The project was directed by Gian Maria Milesi-Ferretti,
Deputy Director, Research Department, and Oya Celasun, Division Chief, Research Department.
The primary contributors to this report were Rabah Arezki, Claudia Berg, Christian Bogmans, Mai Chi Dao,
Mitali Das, Bertrand Gruss, Zska Kczn, Toh Kuan, Weicheng Lian, Akito Matsumoto, Malhar Nabar, Marcos
Poplawski-Ribeiro, and Petia Topalova.
Other contributors include Gavin Asdorian, Felicia Belostecinic, Patrick Blagrave, Emine Boz, Benjamin Carton,
Luis Cubeddu, Jihad Dagher, Silvia Domit, Romain Duval, Angela Espiritu, Rachel Yuting Fan, Emily Forrest,
Clara Galeazzi, Mitko Grigorov, Meron Haile, Mahnaz Hemmati, Benjamin Hilgenstock, Ava Yeabin Hong, Benja-
min Hunt, Deniz Igan, Hao Jiang, Alimata Kini-Kabor, Lama Kiyasseh, Douglas Laxton, Jungjin Lee, Olivia Ma,
Trevor Charles Meadows, Joannes Mongardini, Mico Mrkaic, Natalija Novta, Emory Oakes, Evgenia Pugacheva,
Michael Stanger, Susanna Mursula, Kadir Tanyeri, Nicholas Tong, Jilun Xing, Yuan Zeng, and Qiaoqiao Zhang.
Joseph Procopio from the Communications Department led the editorial team for the report, with production
and editorial support from Michael Harrup, Christine Ebrahimzadeh, and Linda Kean and editorial assistance
from Lucy Scott Morales, Sherrie Brown, and Vector Talent Resources.
The analysis has benefited from comments and suggestions by staff members from other IMF departments, as
well as by Executive Directors following their discussion of the report on April 4,2017. However, both projections
and policy considerations are those of the IMF staff and should not be attributed to Executive Directors or to their
national authorities.

xii International Monetary Fund | April 2017


FOREWORD

C
onsistently good economic news since of this report examines aspects of their convergence
summer 2016 is starting to add up to a toward higher incomes.
brightening global outlook. The economic Richer countries have continued to grow as well,
upswing that we have expected for some but with less impressive income gains over the past
time seems to be materializing: indeed, the World Eco- 10 years when compared with previous decades, and
nomic Outlook (WEO) raises its projection for 2017 certainly when compared with the more successful
global growth to 3.5 percent, up from our recently emerging market and developing economies. It is not
forecast 3.4 percent. Our 2018 forecast holds steady, surprising, therefore, that attitudes about interna-
with the world economys growth set to rise to 3.6 tional trades effects on jobs and wages, as measured
percent next year. The expected growth improvements by leading surveys, tend to be more positive in poorer
in 2017 and 2018 are broadly based, although growth economies.
remains tepid in many advanced economies, and com- These findings may have less to do with the overall
modity exporters continue to struggle. growth disparity between rich and poor countries
At the same time, however, the upgrade to our than with the failure of growth gains in rich econo-
2017 forecast is modest, and longer-term poten- mies to substantially reach those in the lower parts of
tial growth rates remain subdued across the globe the income distribution in recent decades. Inequality
compared with past decades, especially in advanced remains substantial within poorer countries, but with
economies. Moreover, while there is a chance growth more room for catch-up and higher growth under the
will exceed expectations in the near term, significant right policies, they have been able to lift substantially
downside risks continue to cloud the medium-term the incomes of even their poorest citizens. Interna-
outlook, and indeed may have intensified since our tional trade has been a key element in those success
last forecast. The gathering recovery remains vulner- stories.
able to a range of downside risks, which Chapter 1 of Global trends in inequality are related to trade, but
this WEO describes. owe much, and in many countries more, to technol-
One salient threat is a turn toward protectionism, ogy changesinsofar as one can conceptually separate
leading to trade warfare. Mainly in advanced econo- technological advance (which facilitates trade) from
mies, several factorslower growth since the 201011 trade itself (which spreads technological know-how).
recovery from the global financial crisis, even slower Chapter 3 of this report explores how the forces of
growth of median incomes, and structural labor- technology and trade have tended to lower labors
market disruptionshave generated political support share of national income in many countries. A fall
for zero-sum policy approaches that could undermine in labors GDP share could, in theory, be a benign
international trading relationships, along with multi- response to economic developments that raise workers
lateral cooperation more generally. real incomesfor example, fast productivity growth
An approach to international economic policy that benefits capital even more than labor. However,
based on collaboration among countries took root where a fall in labors share coincides with stagnant
after World War II and has evolved in scope and geo- median incomes and a worsening income distribu-
graphic breadth. This evolution has not always been tion, as has been the case in a number of advanced
smooth, as a history of financial and currency crises in economies, political pressures to roll back economic
recent decades attests, but the global economys cop- integration with trading partners can follow.
ing mechanisms have until now proved resilient. One Capitulating to those pressures would result in
result has been a notable surge in growth in a number a self-inflicted wound, leading to higher prices for
of emerging market and developing economies, some consumers and businesses, lower productivity, and
of which have reached high-income status. Chapter 2 therefore, lower overall real income for households.

International Monetary Fund | April 2017 xiii


WORLD ECONOMIC OUTLOOK: GAINING MOMENTUM?

Governments should instead follow trade policies ity where helpful, more effective investment in educa-
consistent with maximum productivity, supplement- tion, and changes to housing and credit markets that
ing those with other policies that better distribute facilitate worker mobility.
the gains from foreign trade internally, improve Many of these policies not only ease economic
the skills and adaptability of their workforces, and adjustment, but they also raise potential output over
smooth the process of adjustment for those adversely the longer term. They are key components of the set
affected by the need for economic reallocation. of monetary, fiscal, structural, and financial sector
Unfortunately, governments often find it harder to policies that will strengthen and secure the recovery
make such domestic improvements than to restrict over time.
trade. But they need to be aware that the gains that The global economy seems to be gaining momen-
such an approach may yield for some at home come tumwe could be at a turning point. But even as
at the expense of others in the domestic economy things look up, the postWorld War II system of
in addition to foreign trade partners. Even the international economic relations is under severe strain
sectoral gains from curbing cross-border economic despite the aggregate benefits it has deliveredand
integration disappear, and losses worsen, when trade precisely because growth and the resulting economic
partners retaliate in kind. adjustments have too often entailed unequal rewards
Policymakers instead must do the hard work of and costs within countries. Policy must address these
investing in their economies, especially in people, disparities head-on to ensure the stability of an open,
to create greater resilience to a host of potential and collaborative trading system that benefits all.
ongoing structural changesincluding the changing
modalities of globalization. Useful reforms can focus Maurice Obstfeld
on active labor market policies, greater tax progressiv- Economic Counsellor

xiv International Monetary Fund | April 2017


EXECUTIVE SUMMARY

With buoyant financial markets and a long-awaited advanced economies, the pickup is primarily driven by
cyclical recovery in manufacturing and trade under way, higher projected growth in the United States, where
world growth is projected to rise from 3.1 percent in activity was held back in 2016 by inventory adjust-
2016 to 3.5 percent in 2017and 3.6 percent in 2018, ment and weak investment.
slightly above the October 2016 World Economic Although changes to the global growth forecast for
Outlook (WEO) forecast. But binding structural 2017 and 2018 since the October 2016 WEO are
impediments continue to hold back a stronger recovery, small, there have been meaningful changes to fore-
and the balance of risks remains tilted to the downside, casts for country groups and individual countries. In
especially over the medium term. With persistent struc- line with stronger-than-expected momentum in the
tural problemssuch as low productivity growth and second half of 2016, the forecast envisages a stronger
high income inequalitypressures for inward-looking rebound in advanced economies. And while growth
policies are increasing in advanced economies. These is still expected to pick up notably for the emerging
threaten global economic integration and the cooperative market and developing economies group, weaker-
global economic order that has served the world economy, than-expected activity in some large countries has led
especially emerging market and developing economies, to small downward revisions to the groups growth
well. Against this backdrop, economic policies have an prospects for 2017.
important role to play in staving off downside risks For advanced economies, projected growth has
and securing the recovery. On the domestic front, poli- been revised upward in the United States, reflecting
cies should aim to support demand and repair balance the assumed fiscal policy easing and an uptick in
sheets where necessary and feasible; boost productivity, confidence, especially after the November elec-
labor supply, and investment through structural reforms tions, which, if it persists, will reinforce the cyclical
and supply-friendly fiscal measures; upgrade the public momentum. The outlook has also improved for
infrastructure; and support those displaced by structural Europe and Japan based on a cyclical recovery in
transformations such as technological change and global- global manufacturing and trade that started in the
ization. At the same time, credible strategies are needed second half of 2016.
in many countries to place public debt on a sustain- The downward revisions to growth forecasts for
able path. Adjusting to lower commodity revenues and emerging market and developing economies result
addressing financial vulnerabilities remain key challenges from a weaker outlook in several large economies,
for many emerging market and developing economies. especially in Latin America and the Middle East,
A renewed multilateral effort is also needed to tackle reflecting continued adjustment to the decline in
common challenges in an integrated global economy. their terms of trade in recent years, oil produc-
The world economy gained speed in the fourth tion cuts, and idiosyncratic factors. The 2017
quarter of 2016 and the momentum is expected to and 2018 growth forecasts have been marked
persist. Global growth is projected to increase from an up for China, reflecting stronger-than-expected
estimated 3.1 percent in 2016 to 3.5 percent in 2017 policy support, as well as for Russia, where activ-
and 3.6 percent in 2018. ity appears to have bottomed out and higher oil
Activity is projected to pick up markedly in prices bolster the recovery.
emerging market and developing economies because
conditions in commodity exporters experiencing mac- Since the U.S. election, expectations of looser
roeconomic strains are gradually expected to improve, fiscal policy in the United States have contributed to
supported by the partial recovery in commodity a stronger dollar and higher U.S. Treasury interest
prices, while growth is projected to remain strong rates, pushing up yields elsewhere as well. Market
in China and many other commodity importers. In sentiment has generally been strong, with notable

International Monetary Fund | April 2017 xv


WORLD ECONOMIC OUTLOOK: GAINING MOMENTUM?

gains in equity markets in both advanced and emerg- Noneconomic factors, including geopolitical ten-
ing market economies. Stronger activity and expecta- sions, domestic political discord, risks from weak
tions of more robust global demand going forward, governance and corruption, extreme weather events,
coupled with agreed restrictions on oil supply, have and terrorism and security concerns
helped commodity prices recover from their troughs
of early 2016. These risks are interconnected and can be mutually
Headline inflation has been picking up in advanced reinforcing. For example, an inward turn in policies
economies due to higher commodity prices, but core could be associated with increased geopolitical tensions
inflation dynamics remain subdued and heterogeneous as well as with rising global risk aversion; noneconomic
(consistent with diversity in output gaps). Core infla- shocks can weigh directly on economic activity as
tion has improved little where it had been the weakest well as harm confidence and market sentiment; and a
(for instance, in Japan and parts of the euro area). faster-than-anticipated tightening of global financial
Headline inflation has also picked up in many emerg- conditions or a shift toward protectionism in advanced
ing market and developing economies due to higher economies could exacerbate capital outflow pressures
commodity prices, but in a number of cases it has in China.
receded as pass-through from the sharp currency depre- Policy choices will therefore be crucial in shaping
ciations in 2015 and early 2016 continues to fade. the outlook and reducing risks. Priorities for mac-
Risks remain skewed to the downside, however, roeconomic demand management are increasingly
especially over the medium term, with pervasive uncer- differentiated, given the diversity in cyclical positions.
tainty surrounding policies. Buoyant market senti- In economies with slack and persistently weak core
ment implies that there is now more tangible upside inflation, cyclical demand support remains necessary,
potential for the near term, but in light of the sources including to stave off pernicious hysteresis effects. In
of uncertainties discussed below, a sharp increase in economies where output is close to or above potential,
risk aversion is possible. Risks to medium-term growth fiscal policy should aim at strengthening safety nets
appear more clearly negative, also because policy sup- and increasing potential output. At the same time,
port in the United States and China will have to be credible strategies are needed in many countries to
unwound or reversed down the road to avoid unsus- place public debt on a sustainable path.
tainable fiscal dynamics. More generally, downside risks Following a lackluster recovery from the global
stem from several potential factors: financial crisis, and in the aftermath of the sharp
An inward shift in policies, including toward protec- adjustment of global commodity prices, many econo-
tionism, with lower global growth caused by reduced mies are seeking to enhance growth potential, inclu-
trade and cross-border investment flows siveness, and resilience. Actions to bolster potential
A faster-than-expected pace of interest rate hikes in output are indeed urgent given persistent headwinds
the United States, which could trigger a more rapid from population aging in advanced economies, the
tightening in global financial conditions and a sharp ongoing adjustment to lower terms of trade and the
dollar appreciation, with adverse repercussions for need to address financial vulnerabilities in emerging
vulnerable economies market and developing economies, as well as slug-
An aggressive rollback of financial regulation, which gish total factor productivity growth in both groups.
could spur excessive risk taking and increase the Chapter 2 documents that trade openness, exchange
likelihood of future financial crises rate flexibility, and strong institutions help emerging
Financial tightening in emerging market economies, market and developing economies enhance the growth
made more likely by mounting vulnerabilities in impulse from external conditions. Facing ever-present
Chinas financial system associated with fast credit risks of global financial volatility, emerging market
growth and continued balance sheet weaknesses in economies vulnerable to an adverse turn in external
other emerging market economies financial conditions would benefit from adopting
Adverse feedback loops among weak demand, low stronger risk management practices and containing bal-
inflation, weak balance sheets, and anemic produc- ance sheet mismatches.
tivity growth in some advanced economies operating Preserving the global economic expansion will also
with high levels of excess capacity require policymakers to avoid protectionist measures

xvi International Monetary Fund | April 2017


Executive Summary

and to do more to ensure that gains from growth are the need to make growth more inclusive. Possible policy
shared more widely. Chapter 3 documents that wages levers include more progressive taxation; investments in
have not kept up with productivity in many econo- skills, lifelong learning, and high-quality education; and
mies over much of the past three decades, leading to a other efforts to enhance the occupational and geographi-
decline in labors share of national income. The chapters cal mobility of workers to ease and hasten labor market
analysis suggests that technological change and trade adjustments to structural transformations.
integrationboth of which are drivers of medium- Many of the challenges that the global economy
and long-term growthhave likely contributed to the confronts call for individual country actions to be
decline. The chapters findings suggest that technologi- supported by multilateral cooperation. Key areas for
cal change has been the dominant driver of the labor collective action include preserving an open trading
share in advanced economies whereas trade integration system, safeguarding global financial stability, achieving
(and the attendant increase in the capital intensity of equitable tax systems, continuing to support low-
production) has been the dominant driver in the case of income countries as they pursue their development
emerging market economies. These findings highlight goals, and mitigating and adapting to climate change.

International Monetary Fund | April 2017 xvii


1
CHAPTER

GLOBAL PROSPECTS AND POLICIES

Global economic activity is picking up with a long- securing the recovery, as stressed in previous WEOs. On
awaited cyclical recovery in investment, manufacturing, the domestic front, policies should support demand and
and trade. World growth is expected to rise from 3.1per- balance sheet repair where necessary and feasible; boost
cent in2016 to 3.5percent in2017 and 3.6percent productivity through structural reforms, well-targeted
in2018, slightly above the October2016 World Eco- infrastructure spending, and other supply-friendly fiscal
nomic Outlook (WEO) forecast. Stronger activity and policy measures; and support those displaced by structural
expectations of more robust global demand, coupled with transformations, such as technological change and global-
agreed restrictions on oil supply, have helped commodity ization. Credible strategies are needed in many countries
prices recover from their troughs in early2016. Higher to place public debt on a sustainable path. Adjusting
commodity prices have provided some relief to commod- to lower commodity revenues and addressing financial
ity exporters and helped lift global headline inflation vulnerabilities remain key challenges for many emerg-
and reduce deflationary pressures. Financial markets are ing market and developing economies. The world also
buoyant and expect continued policy support in China needs a renewed multilateral effort to tackle a number
and fiscal expansion and deregulation in the United of common challenges in an integrated global economy.
States. If confidence and market sentiment remain strong,
short-term growth could indeed surprise on the upside.
But these positive developments should not distract from Recent Developments and Prospects
binding structural impediments to a stronger recovery and
a balance of risks that remains tilted to the downside, World Economy Gaining Momentum
especially over the medium term. Structural problemssuch Economic activity gained some momentum in the
as low productivity growth and high income inequal- second half of2016, especially in advanced econ-
ityare likely to persist. Inward-looking policies threaten omies. Growth picked up in the United States as
global economic integration and the cooperative global firms grew more confident about future demand, and
economic order, which have served the world economy, inventories started contributing positively to growth
especially emerging market and developing economies, well. (after five quarters of drag). Growth also remained
A faster-than-expected pace of interest rate hikes in the solid in the United Kingdom, where spending proved
United States could tighten financial conditions elsewhere, resilient in the aftermath of the June2016 referen-
with potential further U.S.dollar appreciation straining dum in favor of leaving the European Union (Brexit).
emerging market economies with exchange rate pegs to the Activity surprised on the upside in Japan thanks to
dollar or with material balance sheet mismatches. More strong net exports, as well as in euro area countries,
generally, a reversal in market sentiment and confidence such as Germany and Spain, as a result of strong
could tighten financial conditions and exacerbate existing domestic demand.
vulnerabilities in a number of emerging market economies, Economic performance across emerging market
including Chinawhich faces the daunting challenge and developing economies has remained mixed.
of reducing its reliance on credit growth. A dilution Whereas Chinas growth remained strong, reflecting
of financial regulation may lead to stronger near-term continued policy support, activity has slowed in India
growth but may imperil global financial stability and because of the impact of the currency exchange ini-
raise the risk of costly financial crises down the road. In tiative, as well as in Brazil, which has been mired in
addition, the threat of deepening geopolitical tensions a deep recession. Activity remained weak in fuel and
persists, especially in the Middle East and North Africa. nonfuel commodity exporters more generally, while
Against this backdrop, economic policies have an geopolitical factors held back growth in parts of the
important role to play in staving off downside risks and Middle East and Turkey.

International Monetary Fund | April 2017 1


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Table 1.1. Overview of the World Economic Outlook Projections


(Percent change, unless noted otherwise)
Difference from January Difference from October
Projections 2017 WEO Update1 2016 WEO1
2016 2017 2018 2017 2018 2017 2018
World Output 3.1 3.5 3.6 0.1 0.0 0.1 0.0
Advanced Economies 1.7 2.0 2.0 0.1 0.0 0.2 0.2
United States 1.6 2.3 2.5 0.0 0.0 0.1 0.4
Euro Area 1.7 1.7 1.6 0.1 0.0 0.2 0.0
Germany 1.8 1.6 1.5 0.1 0.0 0.2 0.1
France 1.2 1.4 1.6 0.1 0.0 0.1 0.0
Italy 0.9 0.8 0.8 0.1 0.0 0.1 0.3
Spain 3.2 2.6 2.1 0.3 0.0 0.4 0.2
Japan2 1.0 1.2 0.6 0.4 0.1 0.6 0.1
United Kingdom 1.8 2.0 1.5 0.5 0.1 0.9 0.2
Canada 1.4 1.9 2.0 0.0 0.0 0.0 0.1
Other Advanced Economies3 2.2 2.3 2.4 0.1 0.0 0.0 0.0
Emerging Market and Developing Economies 4.1 4.5 4.8 0.0 0.0 0.1 0.0
Commonwealth of Independent States 0.3 1.7 2.1 0.2 0.3 0.3 0.4
Russia 0.2 1.4 1.4 0.3 0.2 0.3 0.2
Excluding Russia 1.8 2.5 3.5 0.0 0.2 0.2 0.6
Emerging and Developing Asia 6.4 6.4 6.4 0.0 0.1 0.1 0.1
China 6.7 6.6 6.2 0.1 0.2 0.4 0.2
India4 6.8 7.2 7.7 0.0 0.0 0.4 0.0
ASEAN-55 4.9 5.0 5.2 0.1 0.0 0.1 0.0
Emerging and Developing Europe 3.0 3.0 3.3 0.1 0.1 0.1 0.1
Latin America and the Caribbean 1.0 1.1 2.0 0.1 0.1 0.5 0.2
Brazil 3.6 0.2 1.7 0.0 0.2 0.3 0.2
Mexico 2.3 1.7 2.0 0.0 0.0 0.6 0.6
Middle East, North Africa, Afghanistan, and Pakistan 3.9 2.6 3.4 0.5 0.1 0.8 0.2
Saudi Arabia 1.4 0.4 1.3 0.0 1.0 1.6 1.3
Sub-Saharan Africa 1.4 2.6 3.5 0.2 0.2 0.3 0.1
Nigeria 1.5 0.8 1.9 0.0 0.4 0.2 0.3
South Africa 0.3 0.8 1.6 0.0 0.0 0.0 0.0
Memorandum
European Union 2.0 2.0 1.8 0.2 0.0 0.3 0.0
Low-Income Developing Countries 3.6 4.7 5.3 0.0 0.1 0.2 0.1
Middle East and North Africa 3.8 2.3 3.2 0.6 0.1 0.9 0.2
World Growth Based on Market Exchange Rates 2.4 2.9 3.0 0.1 0.0 0.1 0.1
World Trade Volume (goods and services) 2.2 3.8 3.9 0.0 0.2 0.0 0.3
Imports
Advanced Economies 2.4 4.0 4.0 0.2 0.2 0.1 0.2
Emerging Market and Developing Economies 1.9 4.5 4.3 0.3 0.4 0.4 0.2
Exports
Advanced Economies 2.1 3.5 3.2 0.1 0.2 0.0 0.8
Emerging Market and Developing Economies 2.5 3.6 4.3 0.1 0.3 0.0 0.1
Commodity Prices (U.S. dollars)
Oil6 15.7 28.9 0.3 9.0 3.9 11.0 5.1
Nonfuel (average based on world commodity export
weights) 1.9 8.5 1.3 6.4 0.4 7.6 0.6
Consumer Prices
Advanced Economies 0.8 2.0 1.9 0.3 0.0 0.3 0.0
Emerging Market and Developing Economies7 4.4 4.7 4.4 0.2 0.0 0.3 0.2
London Interbank Offered Rate (percent)
On U.S. Dollar Deposits (six month) 1.1 1.7 2.8 0.0 0.0 0.4 0.7
On Euro Deposits (three month) 0.3 0.3 0.2 0.0 0.0 0.1 0.2
On Japanese Yen Deposits (six month) 0.0 0.0 0.0 0.0 0.0 0.1 0.1
Note: Real effective exchange rates are assumed to remain constant at the levels prevailing during February 1March 1, 2017. Economies are listed on the
basis of economic size. The aggregated quarterly data are seasonally adjusted.
1Difference based on rounded figures for the current, January 2017 World Economic Outlook Update, and October 2016 World Economic Outlook forecasts.
2Japans historical national accounts figures reflect a comprehensive revision by the national authorities, released in December 2016. The main revisions are the

switch from the System of National Accounts 1993 to the System of National Accounts 2008 and the updating of the benchmark year from 2005 to 2011.
3Excludes the G7 (Canada, France, Germany, Italy, Japan, United Kingdom, United States) and euro area countries.
4For India, data and forecasts are presented on a fiscal year basis and GDP from 2011 onward is based on GDP at market prices with FY2011/12 as a base year.

2 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Table 1.1 (continued)

Year-over-Year Q4-over-Q48

Projections Projections
2015 2016 2017 2018 2015 2016 2017 2018
World Output 3.4 3.1 3.5 3.6 3.2 3.2 3.5 3.6
Advanced Economies 2.1 1.7 2.0 2.0 1.8 2.0 2.0 2.0
United States 2.6 1.6 2.3 2.5 1.9 2.0 2.3 2.5
Euro Area 2.0 1.7 1.7 1.6 2.0 1.7 1.7 1.5
Germany 1.5 1.8 1.6 1.5 1.3 1.8 1.7 1.5
France 1.3 1.2 1.4 1.6 1.2 1.2 1.9 1.4
Italy 0.8 0.9 0.8 0.8 1.0 1.0 0.8 0.8
Spain 3.2 3.2 2.6 2.1 3.5 3.0 2.3 2.1
Japan2 1.2 1.0 1.2 0.6 1.2 1.6 1.0 0.6
United Kingdom 2.2 1.8 2.0 1.5 1.7 1.9 1.7 1.5
Canada 0.9 1.4 1.9 2.0 0.4 1.9 1.7 2.0
Other Advanced Economies3 2.0 2.2 2.3 2.4 2.0 2.4 2.4 2.6
Emerging Market and Developing Economies 4.2 4.1 4.5 4.8 4.4 4.4 4.8 5.0
Commonwealth of Independent States 2.2 0.3 1.7 2.1 2.8 0.7 1.6 1.6
Russia 2.8 0.2 1.4 1.4 3.0 0.4 1.6 1.3
Excluding Russia 0.5 1.8 2.5 3.5 ... ... ... ...
Emerging and Developing Asia 6.7 6.4 6.4 6.4 6.8 6.5 6.5 6.3
China 6.9 6.7 6.6 6.2 6.8 6.8 6.4 6.1
India4 7.9 6.8 7.2 7.7 8.5 6.9 7.8 7.6
ASEAN-55 4.8 4.9 5.0 5.2 4.9 4.8 5.1 5.3
Emerging and Developing Europe 4.7 3.0 3.0 3.3 4.9 3.4 2.1 3.4
Latin America and the Caribbean 0.1 1.0 1.1 2.0 1.1 1.1 1.6 2.1
Brazil 3.8 3.6 0.2 1.7 5.8 2.5 2.0 1.7
Mexico 2.6 2.3 1.7 2.0 2.4 2.4 0.9 3.0
Middle East, North Africa, Afghanistan, and Pakistan 2.7 3.9 2.6 3.4 ... ... ... ...
Saudi Arabia 4.1 1.4 0.4 1.3 4.3 1.2 0.4 2.0
Sub-Saharan Africa 3.4 1.4 2.6 3.5 ... ... ... ...
Nigeria 2.7 1.5 0.8 1.9 ... ... ... ...
South Africa 1.3 0.3 0.8 1.6 0.3 0.4 1.0 1.9
Memorandum
European Union 2.4 2.0 2.0 1.8 2.3 2.0 1.9 1.8
Low-Income Developing Countries 4.6 3.6 4.7 5.3 ... ... ... ...
Middle East and North Africa 2.6 3.8 2.3 3.2 ... ... ... ...
World Growth Based on Market Exchange Rates 2.7 2.4 2.9 3.0 2.4 2.6 2.9 2.9
World Trade Volume (goods and services) 2.7 2.2 3.8 3.9 ... ... ... ...
Imports
Advanced Economies 4.4 2.4 4.0 4.0 ... ... ... ...
Emerging Market and Developing Economies 0.8 1.9 4.5 4.3 ... ... ... ...
Exports
Advanced Economies 3.7 2.1 3.5 3.2 ... ... ... ...
Emerging Market and Developing Economies 1.4 2.5 3.6 4.3 ... ... ... ...
Commodity Prices (U.S. dollars)
Oil6 47.2 15.7 28.9 0.3 43.4 16.2 13.5 2.0
Nonfuel (average based on world commodity export
weights) 17.4 1.9 8.5 1.3 19.1 9.8 3.9 1.0
Consumer Prices
Advanced Economies 0.3 0.8 2.0 1.9 0.4 1.2 1.9 2.0
Emerging Market and Developing Economies7 4.7 4.4 4.7 4.4 4.7 4.0 4.1 3.9
London Interbank Offered Rate (percent)
On U.S. Dollar Deposits (six month) 0.5 1.1 1.7 2.8 ... ... ... ...
On Euro Deposits (three month) 0.0 0.3 0.3 0.2 ... ... ... ...
On Japanese Yen Deposits (six month) 0.1 0.0 0.0 0.0 ... ... ... ...
5Indonesia, Malaysia, Philippines, Thailand, Vietnam.
6Simple average of prices of U.K. Brent, Dubai Fateh, and West Texas Intermediate crude oil. The average price of oil in U.S. dollars a barrel was $42.84 in

2016; the assumed price based on futures markets is $55.23 in 2017 and $55.06 in 2018.
7Excludes Argentina and Venezuela. See country-specific notes for Argentina and Venezuela in the Country Notes section of the Statistical Appendix.
8For World Output, the quarterly estimates and projections account for approximately 90 percent of annual world output at purchasing-power-parity weights.

For Emerging Market and Developing Economies, the quarterly estimates and projections account for approximately 80 percent of annual emerging market
and developing economies output at purchasing-power-parity weights.
International Monetary Fund | April 2017 3
WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.1. Global Activity Indicators Indicators of Economic Activity

Global economic activity gained momentum in the fourth quarter of 2016.


In the second half of2016, the stronger global
Manufacturing PMIs and consumer condence increased noticeably in advanced momentum in demandinvestment in particular
economies in the last few months of 2016 and early 2017. They also recovered to a resulted in marked improvements in manufacturing
more modest extent in emerging market economies.
and trade, which were very weak in late2015 and
20 1. World Trade, Industrial Production, and Manufacturing PMI 8
early2016 (Figure1.1, panel 1).
(Three-month moving average; annualized percent change, Production of both consumer durables and capital
15 unless noted otherwise) 6 goods rebounded in the second half of2016 (Fig-
10 Manufacturing PMI (deviations from 50; right scale) 4 ure1.2). A number of factors contributed to these
Industrial production
developments: a gradual global recovery in invest-
5 2
ment, supported by infrastructure and real estate
0 0 investment in China, reduced drag from adjustment
5 2 to lower commodity prices, and the end of an inven-
World trade volumes
tory cycle in United States. Forward-looking indica-
10 4
2011 12 13 14 15 16 Feb. tors, such as purchasing managers indices, suggest
17 continued strength in manufacturing activity into
14 2. Manufacturing PMI 3. Consumer Condence 130 early2017.
(Three-month moving (Index, 2010 = 100)
12 125 Consistent with indications of firming global
average; deviations
10 from 50) Advanced economies1
120
manufacturing activity, global trade is showing some
Emerging market
8 Advanced economies1
economies2 115
signs of recovery after a long period of weakness
Emerging market
6
economies2 World (Figure1.3, panel 1). As discussed in Chapter2 of
110
4 the October2016 WEO, trade growthin partic-
105
2 ular, growth in importsis strongly correlated with
0 100
investment dynamics. This pattern is illustrated for
2 95
a cross-section of advanced economies (Figure1.3,
4 90
2012 13 14 15 16 Feb. 2012 13 14 15 16 Feb. panel 2) and emerging market economies (Figure1.3,
17 17 panel 3) for2016. Panel 3, in particular, highlights the
GDP Growth sharp contractions in trade and investment in several
(Annualized semiannual percent change) commodity exporters during2016, a pattern similar to
October 2016 WEO April 2017 WEO the one for the previous year. The gradual stabilization
4 4. Advanced Economies 5. Emerging Market and 9 of macroeconomic conditions in these economies, also
Developing Economies supported by some rebound in commodity prices,
8
3 should lead to a gradual recovery in imports and
7 investment in2017 and beyond, as discussed in more
2 6 detail in the section titled The Forecast.

5 Commodity Prices and Markets


1
4 Alongside the pickup in economic activity, com-
modity prices have also strengthened (see the Com-
0 3
2011: 13: 15: 17: 18: 2011: 13: 15: 17: 18: modity Special Feature for more details). The IMFs
H1 H1 H1 H1 H2 H1 H1 H1 H1 H2 Primary Commodities Price Index increased by 15per-
Sources: CPB Netherlands Bureau for Economic Policy Analysis; Haver Analytics; cent between August2016 and February2017that
Markit Economics; and IMF staff estimates. is, between the reference periods for the October2016
Note: CC = consumer condence; PMI = purchasing managers index.
1Australia, Canada (PMI only), Czech Republic, Denmark, euro area, Hong Kong SAR and the current WEO reports (Figure1.4). Some of
(CC only), Israel, Japan, Korea, New Zealand (PMI only), Norway (CC only), the strongest price increases were for fuels:
Singapore (PMI only), Sweden (CC only), Switzerland, Taiwan Province of China, Oil prices increased by some20percent between
United Kingdom, United States.
2Argentina (CC only), Brazil, China, Colombia (CC only), Hungary, India (PMI only), August2016 and February2017, in part due to the
Indonesia, Latvia (CC only), Malaysia (PMI only), Mexico (PMI only), Philippines (CC agreement by the Organization of the Petroleum
only), Poland, Russia, South Africa, Thailand (CC only), Turkey, Ukraine (CC only).
Exporting Countries (OPEC) and other producers
to cut oil production. Stronger activity and expec-

4 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Figure 1.2. Recent Trends in Global Production Figure 1.3. Global Trade and Fixed Investment Growth

The production of both consumer durables and capital goods recovered in late Real import growth picked up in the second half of 2016, consistent with the
2016, after several quarters of lackluster growth or contraction. rming in investment.
1. Real Merchandise Imports
8 1. Global Production of Consumer Durable Goods 12 (Seasonally adjusted quarter-over-quarter annualized
(Quarter-over-quarter annualized percent change) percent change)
6 Euro area Other 9 China
Other countries1
Japan Consumer durables World
Euro area, Japan, United
4 United States 6 States
2 3

0 0

2 3

4 6
2015:Q1 15:Q2 15:Q3 15:Q4 16:Q1 16:Q2 16:Q3 16:Q4 2015:Q1 15:Q2 15:Q3 15:Q4 16:Q1 16:Q2 16:Q3 16:Q4

9 2. Global Production of Capital Goods 2. Trade and Fixed Investment Growth in Advanced Economies
(Quarter-over-quarter annualized percent change) 6 (Percent, 2016) BEL
6 Euro area Other 5 NLD KOR
Japan Consumer durables PRT TWN FRA NZL
4 ESP
3 United States DEU SWE
3 GBR ITA
FIN
0 Import growth 2 HKG
AUT
DNK
USA CHE
1
3 AUS NOR
0
SGP
6 1
CAN
2 JPN
9
2015:Q1 15:Q2 15:Q3 15:Q4 16:Q1 16:Q2 16:Q3 16:Q4 3
4 2 0 2 4 6 8
Fixed investment growth
Source: IMF staff estimates.
Note: Euro area data are through November 2016. Other = Brazil, India, Korea,
Norway, Sweden, Switzerland, Taiwan Province of China, Turkey, United Kingdom. 3. Trade and Fixed Investment Growth in Emerging Market and
25 Developing Economies
20 PAK PHL
(Percent, 2016) MAR
POL ROU VNM
15
10 HUN TUR IND
5 ARG CHN
Import growth

tations of more robust future global demand also MEX KWT


BGD
0
contributed to strengthening oil prices since their 5 DZA PER MYS ARE
troughs in early2016. Following some weakening in 10 BRA
ZAF
CHL
IDN
recent weeks, oil prices stood at about $50 a barrel 15 KAZ THA
20 SAU COL
as of end-March, still some 12percent stronger than 25 NGA RUS
in August2016. 30
30 20 10 0 10 20 30
Natural gas prices have increasedas of Febru-
Fixed investment growth
ary2017 the average price for Europe, Japan, and
the United States was up by about19percent rela- Source: IMF staff estimates.
tive to August2016. In Europe, natural gas prices Note: Data labels in the gure use International Organization for Standardization
have risen following higher oil prices. While prices (ISO) country codes.
1Other countries = Brazil, Bulgaria, Canada, Czech Republic, Denmark, Hong Kong
in Asia and the United States initially rose because SAR, Korea, Malaysia, Mexico, Peru, Singapore, South Africa, Sweden, Switzerland,
of expectations of strong winter demand, a fairly Taiwan Province of China, Thailand, Turkey, United Kingdom.
mild winter led to subdued demand for gas-fired
power generation and helped contain gas prices.
Coal prices have rallied, with the average of Aus-
tralian and South African prices in February2017
more than20percent higher than in August2016.
That rally has followed government-led reductions

International Monetary Fund | April 2017 5


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.4. Commodity and Oil Markets in coal production in China and production and
shipment outages in Australia.
Commodity prices have strengthened as global economic activity has gained
momentum. Among nonfuel commodities, metal prices have
increased by 23.6percent and agricultural commodity
200 1. Real Commodity Price Indices prices by 4.3percent.
(Deated using U.S. consumer price index; index, 2014 = 100)
160 Metal prices have been supported by higher real
estate investment and capacity reduction efforts in
120
China and the anticipated fiscal policy easing in the
80 United States.
APSP Food
40 Metal
Among agricultural commodities, food prices rose by
4.9percent as excess supply eased, especially for grains
0 and vegetable oils. Prices have increased for most
2011 12 13 14 15 16 17 Dec.
18 items, except for a few, including rice and cocoa beans.

1,400 2. Global Capital Expenditures in Oil and Gas Industries Inflation Developments
(Annualized, billions of U.S. dollars)
1,200
Australia Middle East Africa The increase in commodity prices has contrib-
1,000 Asia Lat. Am. United States and Canada
Europe Russia uted to a recovery in global inflation since August
800
(Figure1.5). The increase in global producer price
600
inflation has been particularly marked, reflecting
400
both the greater weight of commodities in producer
200
price indices when compared with consumer price
0
2014 15 16 17: indices and their importance as intermediate inputs
Q4
in production. Notably, Chinas producer prices have
42 3. Global Oil Demand and OECD Oil Inventories 4 emerged from deflation after four years, reflecting
Oil inventories (days of consumption) higher raw material prices as well as efforts to reduce
40 Change in global oil demand (IEA, 3 excess industrial capacity and recovering real estate
million barrels per day, right scale)
38 investment.
2
36 Global consumer price inflation has also ticked
1 up as the retail prices of gasoline and other energy-
34
related products have increased. The uptick has been
32 0 especially strong for advanced economies, where
2011 12 13 14 15 16 17:
Q4 12-month consumer price inflation in February
stood slightly above 2percent (more than double the
4. World Oil Production Growth
(Change in million barrels a day produced, unless noted
average annual inflation rate of 0.8percent in2016).
6 160
otherwise) Non-OPEC APSP crude oil ($/bbl, By contrast, core inflation has increased much lessif
140
4 OPEC right scale) at alland remains well below central bank targets in
120
2 almost all advanced economies. In emerging market
100
economies, the revival in headline consumer inflation
0 80
is more recent, as the impact of higher fuel prices
60
2 has only of late started to outweigh the downward
40
pressure from the fading of earlier exchange rate
4 20
2006 07 08 09 10 11 12 13 14 15 16 Feb. depreciations.
17
Near- and longer-term inflation expectations also
Sources: IMF, Primary Commodity Price System; International Energy Agency (IEA); remain subdued. Survey-based consumer price inflation
Organisation for Economic Co-operation and Development; and IMF staff estimates. expectations for2017 have only very recently stopped
Note: In panel 2, 2017 projections are based on investment plans. APSP = average
petroleum spot price; bbl = barrel; Lat. Am. = Latin America (Argentina, Brazil, falling for advanced economies, and expected infla-
Chile, Colombia, Mexico, Peru, Uruguay); OPEC = Organization of the Petroleum tion for the next 10 years has only recently registered
Exporting Countries.
an increase after declining steadily in2015 and2016
(Figure1.5, panels 5 and 6).

6 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Financial Market Developments Figure 1.5. Global Ination


(Three-month moving average; annualized percent change, unless noted
Market sentiment has strengthened since August, otherwise)
reflecting generally positive data on the outlook as well
Higher commodity prices have pushed up global headline ination. Core ination
as expectations of a fiscal stimulus, higher infrastruc- remains subdued, especially in advanced economies.
ture investment, and deregulation in the United States. World1 Advanced Emerging market and
With stronger future demand suggesting more economies developing economies1
inflation pressure and a less gradual normalization of 28 1. Consumer Price Ination 2. Producer Price Ination 18
U.S.monetary policy, long-term nominal and real 24 (Percent change from 14
interest rates have risen substantially since August, 20 a year ago) 10
16 6
especially since the U.S.elections in November (Fig- 12
2
ure1.6). As of end-March, nominal yields on 10-year 8
4 2
U.S.Treasury bonds had increased by some 85 basis 0 6
points compared with August and 55 basis points 4 10
2007 09 11 13 15 1718: 2013 14 15 16 Feb.
compared with just before the U.S.election. Long- Q4 17
term rates increased sharply in the United Kingdom Consumer price index ination Core consumer price index ination
as well, reflecting spillovers from higher U.S.rates and
3 3. Advanced Economies 7 4. Emerging Market and
expectations of a less accommodative monetary policy Developing Economies
stance going forward, given rising inflation pressure. 2 6
The increase in core euro area long-term yields after 1 5
August was more moderateabout 40 basis points in 0 4
Germanybut Italian yields rose more sharply (about 1 3
120 basis points), reflecting elevated political and 2 2
banking sector uncertainties. The U.S.Federal Reserve 2013 14 15 16 Feb. 2013 14 15 16 Feb.
17 17
raised short-term interest rates in December2016
Advanced economies Emerging market and developing
and March2017, as expected, with markets pricing economies (right scale)
in two additional rate increases by the end of2017
1.9 5. Ination Expectation 3.7 2.0 6. 10-Year Ination 4.0
or early2018. In most other advanced economies, 1.8 for 2017 Expectation
the monetary policy stance has remained broadly 1.7 (Percent) 1.9 (Percent) 3.8
1.6 3.6
unchanged. 1.5 3.6
1.8
Equity markets in advanced economies have regis- 1.4 3.4
1.3 3.5
tered sizable gains in recent months, amid strengthening 1.2 1.7 3.2
consumer confidence and positive macroeconomic 1.1
1.0 3.4 1.6 3.0
data. As discussed in more detail in the April2017 Jan. Apr. Jul. Oct. Jan.Mar. Jan. Jul. Jan. Jul. Jan.
Global Financial Stability Report (GFSR), gains have 16 16 16 16 17 17 15 15 16 16 17
been notable for sectors that are particularly exposed to Unit labor cost (Average, 201315) Unit labor cost (2016:Q4)
potential fiscal stimulus measures as well as for financial Compensation (Average, 201315) Compensation (2016:Q4)

stocks. Higher valuations of financial stocks reflect both 5 7. Nominal Unit Labor Cost and Compensation
(Quarterly percent change, annual rate)
welcome developments, such as the favorable impact of 4
steepening yield curves and higher growth on expected 3
2
profitability, as well as factors that could heighten
1
downside risks, such as the possibility of some rollback
0
in financial regulation in the United States. 1
With widening interest differentials, the U.S.dol- 2
United States Euro area Japan Other adv. Eur. Other adv.
lar has strengthened in real effective terms by about
3.5percent between August2016 and late March2017
Sources: Consensus Economics; Haver Analytics; IMF, Primary Commodity Price
(Figure1.7, panel 1), whereas the euro and especially System; and IMF staff estimates.
the Japanese yen have weakened. Note: Other adv. Eur. = other advanced Europe (Iceland, Norway, Sweden,
Switzerland, United Kingdom); Other adv. = other advanced economies (Australia,
In emerging market economies, financial conditions Canada, New Zealand). All quarterly data are seasonally adjusted.
have been diverse. Long-term interest rates on local-cur- 1
Excludes Argentina and Venezuela.
rency bonds rose in the aftermath of the U.S.elections,

International Monetary Fund | April 2017 7


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.6. Advanced Economies: Monetary and Financial Figure 1.7. Real Effective Exchange Rate Changes,
Market Conditions August 2016March 2017
(Percent, unless noted otherwise) (Percent)

With markets expecting a less gradual normalization of U.S. monetary policy, long-
term nominal real rates have risen in the United States, pushing up longer-term The U.S. dollar, Korean won, Taiwanese dollar, and Australian dollar have
rates elsewhere as well. Equity markets in advanced economies have registered strengthened in real effective terms since August, while the euro, and especially
strong gains in recent months. the Japanese yen, have weakened. The Turkish lira and the Malaysian ringgit have
depreciated in real effective terms, while the Indian rupee and the currencies of
commodity exporting emerging market economiesin particular the Russian
4.0 1. U.S. Policy Rate 2. Policy Rate Expectations 1 4.5 rublehave gained. The Mexican peso has also strengthened in recent weeks and
Expectations1 (Percent; dashed lines 4.0 now stands little changed relative to August.
3.5
are from the October 3.5
3.0 Sep. 16, 2015
2016 WEO)
Sep. 16, 2016 3.0
2.5 Nov. 8, 2016 United States 8 1. Advanced Economies1
2.5
Apr. 3, 2017 Euro area 6
2.0 United Kingdom 2.0
1.5 4
1.5
1.0 2
1.0
0.5 0
0.5 0.0 2
0.0 0.5 4
15 16 17 18 19 Apr. 2016 17 18 19 Apr.
6 Nov. 8, 2016 relative to Aug. 2016
20 20
Latest relative to Nov. 8, 2016
8
6 3. Key Interest Rates 2 4. Credit Spreads3 1,200 10
(Basis points) USA EA JPN GBR SWE CHE KOR TWN SGP CAN NOR AUS NZL
U.S. average 30-year
5 1,000
xed-rate mortgage U.S. high grade
4 U.S. high yield 24 2. Emerging Market Economies1
Euro high grade 800
20
3 United States Euro high yield
600 16
2 12
Germany 400 8
1
Japan 4
0 200
0
1 0 4
2013 14 15 16 Mar. 2013 14 15 16 Apr. 8
17 Nov. 8, 2016 relative to Aug. 2016
17
12 Latest relative to Nov. 8, 2016
16
200 5. Equity Markets 6. Price-to-Earnings Ratios3 35 ZAF CHN IND IDN MYS PHL THA HUN POL RUS TUR BRA CHL COL MEX PER
180 (Index, 2007 = 100;
United States
160 national currency) Japan 30
Germany Source: IMF staff calculations.
140
Italy Note: EA = euro area. Data labels in the gure use International Organization for
120 25 Standardization (ISO) country codes.
100 1Latest data available are for March 31, 2017.

80 20
60
MSCI Emerging Market
40 15 especially in emerging Europe, but have since declined
Euro Stoxx S&P 500
20 TOPIX
10
(Figure1.8). Policy rate changes since August also reflect
0
2013 14 15 16 Mar. 2013 14 15 16 Apr. this diversitywith rate hikes in Mexico and Turkey
17 17
and cuts in Brazil, India, and Russiaas do changes in
EMBI (Emerging Market Bond Index) spreads.
Sources: Bank of Spain; Bloomberg, L.P.; Haver Analytics; Thomson Reuters
Datastream; and IMF staff calculations. Equity markets in emerging market and developing
Note: MSCI = Morgan Stanley Capital International; S&P = Standard & Poors; economies have strengthened since August, staging
TOPIX = Tokyo Stock Price Index.
1Expectations are based on the federal funds rate futures for the United States, the a strong recovery so far this year after weakening in
sterling overnight interbank average rate for the United Kingdom, and the euro the immediate aftermath of the U.S.election (Fig-
interbank offered forward rate for the euro area; updated April 3, 2017.
2Interest rates are 10-year government bond yields, unless noted otherwise. Data ure1.9). However, they generally remain below their
are through March 31, 2017. post-financial-crisis peaks, reached in2011.
3Data are through April 3, 2017.
A few emerging market currencies have depreciated
substantially in recent monthsmost notably the

8 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Turkish lira and, to a lesser extent, the Malaysian ring- Figure 1.8. Emerging Market Economies: Interest Rates
gitwhile the currencies of some commodity export- The evolution of nancial market conditions has been diverse across emerging
ers, especially Russia, have appreciated (Figure1.7, market economies. Long-term government bond yields in local currency rose
together with bond yields in advanced economies after the U.S. election in
panel 2). The Mexican peso, which had depreciated November, but have since retreated in most countries.
sharply in the aftermath of the U.S.election, has
Emerging Europe China
strengthened in recent weeks and now stands little Emerging Asia excluding China Latin America
changed relative to August. Preliminary data point to
13 1. Policy Rate
sharp nonresident portfolio outflows from emerging (Percent)
12
markets in the wake of the U.S.election, following a 11
few months of solid inflows, but a turnaround in more 10
recent weeks (Figure1.10, panel 1). 9
8
7
6
Key Forces Shaping the Outlook
5
The main forces shaping the outlook differ, to some 4
2012 13 14 15 16 Mar.
extent, between advanced economies and emerg- 17
ing market and developing economies. Among the 10 2. Real Policy Rates 1
(Percent)
advanced economies group, the U.S.economy is 8 August 2016
August 2016 average
projected to gather steam as a result of expansionary 6 March 2017
fiscal policy. Elsewhere, especially in Europe, the cycli- March 2017 average
4
cal recovery from the crises of200809 and201112
2
will help keep growth modestly above potential over
the next few years. Looking to the medium term, 0

however, demographic headwinds and weak trend 2


BRA CHL CHN COL IDN IND MEX MYS PER PHL POL RUS THA TUR ZAF
productivity are likely to restrain growth, as discussed
in the October2016 WEO. Among emerging market 14 3. 10-Year Government Bond Yields2
(Percent)
and developing economies, especially those that rely 12
heavily on energy or metal exports, the adjustment to 10
lower commodity prices remains a key influence on
8
the outlook, in both the short and medium term. The
6
slowdown of productivity growth in the past few years
is also a medium-term challenge for many emerging 4
market and developing economies. 2
2013 14 15 16 Mar.
17
Continued Cyclical Recovery in Advanced Economies 600 4. EMBI Sovereign Spreads2
(Basis points)
As discussed in Chapter1 of the October2016 500
WEO, the recovery from the crises of200809 400
and201112 is ongoing in many advanced economies. 300
Output is still below potential, and unemployment is
200
above2008 levels in many countries, especially in euro
100
area economies with high borrowing spreads during
the201112 sovereign debt crisis. The cyclical rebound 0
2013 14 15 16 Mar.
that normally follows deep recessions, supported by 17
accommodative monetary policy, has been slow in many Sources: Bloomberg L.P.; Haver Analytics; IMF, Balance of Payments and
countries in a context of gradual repair of impaired International Investment Position Statistics database; and IMF staff calculations.
balance sheets (through temporarily high private and Note: Emerging Asia excluding China comprises India, Indonesia, Malaysia, the
Philippines, and Thailand; emerging Europe comprises Poland, Romania, Russia,
public sector savings) and the associated weakening of and Turkey; Latin America comprises Brazil, Chile, Colombia, Mexico, and Peru.
the monetary policy transmission mechanism. The tight- EMBI = J.P. Morgan Emerging Markets Bond Index. Data labels in the gure use
International Organization for Standardization (ISO) country codes.
ening in fiscal policy in many economies between2011 1
Deated by two-year-ahead World Economic Outlook ination projections.
2
and2015 also put a brake on the postcrisis recovery. Data are through March 31, 2017.

International Monetary Fund | April 2017 9


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.9. Emerging Market Economies: Equity Markets and Figure 1.10. Emerging Market Economies: Capital Flows
Credit
Net ows into emerging market funds turned negative in the immediate aftermath
of the November 8 election in the United States, but were positive in the rst three
Equity prices are up, relative to August, in most emerging market economies. months of 2017. Capital inows into emerging market economies declined
Credit dynamics are heterogeneous across emerging market economies. somewhat in the third quarter of 2016 while capital outows picked up modestly;
both were little changed in the fourth quarter. Reserves continue to decline for the
Emerging Asia Latin America group, driven largely by continued reserve decumulation in China.
1. Equity Markets excluding China
200 (Index, 2007 = 100) Emerging Europe China
40 1. Net Flows in Emerging Market Funds Bond
180 30 (Billions of U.S. dollars)
May 22, Equity
160 20 2013 EM-VXY
140 10
120 0
10
100 Greek Irish 1st ECB
20
80 crisis LTROs
30 crisis
60 40
2010 11 12 13 14 15 16 Mar.
40 17
2012 13 14 15 16 Mar.
17 15 2. Capital Inows China Emerging Europe
Real Credit Growth1 (Percent of GDP) Saudi Arabia Emerging Asia excluding China
12
(Year-over-year percent change) Total Latin America
9
25 2. 3. 40
COL IDN 6
BRA CHN
20 MYS RUS
IND MEX 3
TUR 30
15 0
10 20 3
5 6
10 2007 08 09 10 11 12 13 14 15 16:
0 Q4
5 15 3. Capital Outows Excluding Change in Reserves
0
10 12 (Percent of GDP)
China Emerging Europe
15 10 9 Saudi Arabia Emerging Asia excluding China
2012 13 14 15 Jan. 2012 13 14 15 Jan.
6 Total Latin America
17 17
3
Credit-to-GDP Ratio1
(Percent) 0
3
85 4. 230 5. 30
MEX (right scale) 6
75 210 2007 08 09 10 11 12 13 14 15 16:
BRA COL CHN 25 Q4
IDN IND MYS
65 190
RUS TUR 20 15 4. Change in Reserves
55 170 12 (Percent of GDP) China Emerging Europe
15 Saudi Arabia Emerging Asia excluding China
45 150 9 Total Latin America
10 6
35 130
3
25 110 5
0
15 90 0 3
2006 08 10 12 14 16: 2006 08 10 12 14 16:
Q4 Q4 6
2007 08 09 10 11 12 13 14 15 16:
Q4
Sources: Bloomberg L.P.; Haver Analytics; IMF, International Financial Statistics
(IFS) database; and IMF staff calculations. Sources: Bloomberg L.P.; EPFR Global; Haver Analytics; IMF, International Financial
Note: Data labels in the gure use International Organization for Standardization Statistics (IFS) database; and IMF staff calculations.
(ISO) country codes. Note: Capital inows are net purchases of domestic assets by nonresidents. Capital
1Credit is other depository corporations claims on the private sector (from IFS),
outows are net purchases of foreign assets by domestic residents. Emerging Asia
except in the case of Brazil, for which private sector credit is from the Monetary excluding China comprises India, Indonesia, Malaysia, the Philippines, and Thailand;
Policy and Financial System Credit Operations published by Banco Central do emerging Europe comprises Poland, Romania, Russia, and Turkey; Latin America
Brasil, and China, for which credit is total social nancing after adjusting for local comprises Brazil, Chile, Colombia, Mexico, and Peru. ECB = European Central Bank;
government debt swap. EM-VXY = J.P. Morgan Emerging Market Volatility Index; LTROs = longer-term
renancing operations.

10 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Barring unforeseen developments, continued recov- Figure 1.11. Revisions to 2016 Growth and Output Gaps
ery and gradual closing of output gaps are projected in 2015
(Percent)
to keep growth modestly above potential in many
advanced economies over the next few years. The
Growth surprises for 2016 tended to be larger in countries with greater excess
pattern of growth surprises for2016 suggests that the capacity, suggesting that the cyclical recovery may be gaining momentum.
cyclical recovery may be firming up. Indeed, growth
in2016 is estimated to have exceeded expectations to 2.5
a greater extent in countries with deeper output gaps,
NZL
especially in Europe (Figure1.11). Policy actions to 2.0
y = 0.07x + 0.09, R 2 = 0.04
accelerate the cleanup of balance sheets and demand (0.07)
SGP
1.5
support would help entrench the recovery in countries

Revisions to 2016 growth


ISR IRL
operating with significant excess capacity, as discussed 1.0 NLD
CHE
in the section titled Policy Priorities. GRC DEU
ESP JPN BEL
FIN AUT
0.5
GBR
Adjustment to Terms-of-Trade Changes in Emerging
PRT FRA TWN
Market and Developing Economies 0.0
AUS KOR
ITA CZE
As discussed in a number of previous WEO
0.5 HKG
reports, the slowdown in Chinaalong with com- NOR
modity price fluctuationshas been the key driver of CAN SWE DNK
1.0
economic performance in emerging market and devel- USA
oping economies, especially in commodity exporters.1 1.5
6 5 4 3 2 1 0 1
Panel 1 of Figure1.12 shows Chinas growth rate
Output gaps
and the purchasing-power-parity GDP-weighted
aggregate growth rates for commodity exporters Source: IMF staff estimates.
and the remaining emerging market and developing Note: 2016 growth revisions are differences between current growth estimates for
2016 and projections in the April 2016 World Economic Outlook. Japans latest
economies. The growth profiles of commodity and gures reect comprehensive methodological revisions adopted in December
noncommodity exporters are quite similar until2011, 2016. The number in parentheses in the regression equation is the standard error
when a growth downturn begins for commodity of the estimated coefcient on the output gap. Data labels in the gure use
International Organization for Standardization (ISO) country codes.
exporters against the backdrop of falling non-oil
commodity prices. For emerging market and devel-
oping economies as a group, the decline in growth
between2011 and2016 was 2.2percentage points, share of the 1.6percentage point decline in growth
with about two-thirds of this decline attributable to between2011 and2016 is attributable to the drastic
weaker growth in commodity exporters (Figure1.12, slowdown in Nigeria, an oil exporter that in2016
panel 2)the rest being accounted for by slower accounted for more than20percent of purchas-
growth in China and in other emerging market ing-power-parity GDP of low-income countries and
and developing economies.2 Commodity exporters about half of the GDP of commodity exporters in this
account for most of the projected pickup in emerging country group. Panel 3 of Figure1.12 also underscores
market and developing economy growth in201719, the broad stability of growth in low-income countries
even though their projected growth recovery is rela- that are not primarily commodity exportersa group
tively modest compared with the striking decline in of countries in which Bangladesh and Vietnam have
their growth rates over the past five years. large weightsas well as the milder slowdown in
A broadly similar picture holds for low-income low-income commodity exporters excluding Nigeria
developing countries (Figure1.12, panel 3). The lions when compared with all commodity exporters.
Panel 1 of Figure1.13 illustrates the windfall
1See, for instance, Chapter4 of the April 2014 WEO, Chapter2 gains and losses in emerging market and developing
of the October 2015 WEO, and Chapter1 of the April 2016 WEO. economies arising from commodity price fluctuations
2The negative impact of the large decline in Chinese growth on
(see also the April2016 WEO and the October2016
aggregate growth in emerging market and developing economies is
attenuated by Chinas rising weight in the group, which reflects a WEO for related discussions). Commodity exporters
growth rate substantially above most of the rest of the group. suffered sizable income losses during2015 and2016.

International Monetary Fund | April 2017 11


WORLD ECONOMIC OUTLOOK: GaINING MOMENTUM?

Figure 1.12. GDP Growth, 19992021 Although commodity price forecasts suggest some
(Percent) recovery in prices during2017 and beyond, the fore-
cast gains are expected to be much more modest than
Among emerging market and developing economies, growth rates have diverged
markedly since 2011 between the commodity-exporter and -importer groups.
the losses already incurred. This suggests that, for many
Growth in exporters is projected to pick up over 201719, but to remain below the of these countries, the period ahead will be one of
average growth rate for 200010. Growth in importers is projected to remain protracted adjustmentparticularly in those econo-
buoyant.
mies in which revenues from commodities account for
18 1. Growth in Emerging Market and Developing Economies an important fraction of government revenues (see the
16 Commodity exporters
discussion in the April2017 Fiscal Monitor). The need
14 Noncommodity EMDEs excluding China for a protracted period of fiscal consolidation is one
12 China important reason the recovery in commodity exporters
10
is forecast to be subdued.
8
6
Productivity Headwinds
4
2 Medium-term growth rates in both advanced and
0 emerging market economies will be shaped largely by
2
1999 2001 03 05 07 09 11 13 15 17 19 22 the pace of total factor productivity (TFP) growth.
GDP projections in the April2017 WEO incorporate
1.5 2. Contributions to Changes in EMDE Growth a gradual recovery in TFP growth rates from recent
1.0 weak levels. Nonetheless, TFP growth is projected to
0.5 stay below the pace registered before the global finan-
0.0 cial crisis, especially in emerging market economies
0.5 (Figure1.14, panel 1).
1.0 Commodity exporters The persistent decline in TFP growth in recent
1.5 China
India years and its projected slow recovery, in part, reflect
2.0
Other noncommodity exporters the legacies of the financial crisis. New evidence
2.5
suggests that in advanced economies, notably in
3.0
201116 201622 Europe, high levels of corporate debt and nonper-
forming bank loans have constrained investment in
18 3. Growth in Low-Income Developing Countries capital goods and intangible assets, slowing the pace
16 Nigeria of capital-embodied technological change (Fig-
14 Others
Commodity exporters excluding Nigeria and Yemen ure1.14, panels 2 and 3) (Adler and others2017).
12
In a number of advanced economies, the boom-bust
10
8 cycle also appears to have increased the misallocation
6 of capital within and across sectors, dragging down
4 productivity growth.
2 Subdued TFP growth prospects also reflect unfavor-
0
able trends that started before the crisis. The broadly
2
1999 2001 03 05 07 09 11 13 15 17 19 22 synchronized slowdown in productivity growth ahead
of the global financial crisis can be traced to forces
Source: IMF staff estimates. that weakened technological innovation or diffusion,
Note: Commodity exporters includes fuel and nonfuel primary products exporters,
as indicated in Table D of the Statistical Appendix, plus Brazil and Peru. EMDE = including the waning effects of the earlier boom in the
emerging market and developing economy. adoption of information and communications technol-
ogies (Fernald2014), population aging (Feyrer2007),
decelerating global trade integration (Ahn and Duval,
forthcoming), slowing human capital accumulation,
and taxation policies (Chapter2 of the April2017
Fiscal Monitor). In emerging market economies, the
fading effects of earlier structural reforms and struc-
tural transformationwhereby resources are real-

12 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Figure 1.13. Emerging Markets: Terms-of-Trade Windfall Figure 1.14. Total Factor Productivity
Gains and Losses (Percent)
Total factor productivity slowed sharply following the 2008 09 crisis, both in
Commodity exporters are set to experience some windfall gains from higher advanced and emerging market economies. While some recovery is expected,
commodity prices in 2017 and beyond, but these gains will be modest compared productivity growth is not projected to return to its precrisis pace. A key factor
with the losses experienced in 2015
16. behind the slowdown has been weak investmentand the associated slow
pace of adoption of capital-embodied technologies. The drop in investment
was abrupt and sustained in advanced economies, but more gradual in
1. Terms-of-Trade Windfall Losses 1
10 emerging market economies.
(Percent of GDP)

0 1. Total Factor Productivity Growth, 200022


5 (PPP GDP-weighted; dashed lines indicate period average)
10 4 Global EMDEs
2.8%
nancial AEs
2015 3 crisis 1.9%
1.3%
20 2016 2
201718 (Aug. 2016 commodity prices)
201718 (Jan. 2017 commodity prices) 1
30
0
1.0% 0.3% 0.7%
40 1
SAU DZA IRN COL AUS MEX ARG
KAZ NGA RUS MYS CAN BRA IDN 2
2000 02 04 06 08 10 12 14 16 18 20 22
Gross Fixed Capital Formation, 200014
8 2. Terms-of-Trade Windfall Gains 1 (Share of stock of physical capital)
(Percent of GDP)
0.08 2. Advanced Economies 0.11 3. Emerging Market
6 2015 Economies
2016 0.10
0.07
201718 (Aug. 2016 commodity prices)
4 201718 (Jan. 2017 commodity prices)
0.06 0.09
2
0.05 0.08 Excluding China
0
0.04 0.07
2 2000 02 04 06 08 10 12 14 2000 02 04 06 08 10 12 14
USA FRA DEU ITA JPN IND PAK
EGY TUR POL ESP CHN THA KOR 4. Estimated Contribution of Capital Accumulation to Total
Factor Productivity Growth 1
1.0
AEs EMDEs
Source: IMF staff estimates. 0.8 Pre-GFC Post-GFC Pre-GFC Post-GFC
Note: Data labels in the gure use International Organization for Standardization 0.6
(ISO) country codes.
1Gains (losses) for 201718 are simple averages of annual incremental gains 0.4
(losses) for 2017 and 2018. The windfall is an estimate of the change in 0.2
disposable income arising from commodity price changes. The windfall gain in 0.0
year t for a country exporting x U.S. dollars of commodity A and importing m U.S. 0.2
dollars of commodity B in year t 1 is dened as (ptAxt-1 ptBmt-1) / Yt-1, in 0.4
which ptA and ptB are the percentage changes in the prices of A and B between
0.6
year t 1 and year t, and Y is GDP in year t 1 in U.S. dollars. See also Gruss 200307 200814 201314 200307 200814 201314
(2014). vs 200002 vs 200307 vs 200307 vs 200002 vs 200307 vs 200307

Sources: Penn World Table 9.0; and IMF staff estimates.


located from low-productivity to high-productivity Note: Weighted averages are reported for each income group. AEs = advanced
economies; EMDEs = emerging market and developing economies. GFC = global
sectors and firmsseem to have accounted for part of nancial crisis; PPP = purchasing power parity. Advanced economies comprise
the TFP slowdown. Australia, Austria, Belgium, Canada, Denmark, France, Germany, Israel, Italy, Japan,
Korea, the Netherlands, Norway, Singapore, Spain, Sweden, Switzerland, Taiwan
Province of China, the United Kingdom, and the United States. Emerging market and
The Forecast developing economies comprise Brazil, Chile, China, Colombia, Egypt, India,
Indonesia, Iran, Malaysia, Mexico, Pakistan, the Philippines, Poland, Russia, South
Policy Assumptions Africa, Thailand, Turkey, and the United Arab Emirates. In panel 1, TFP growth data
for 2015 and 2016 are estimates, and those for 201722 are forecasts based on
After providing mild support to economic activity projections in the World Economic Outlook for GDP, gross xed capital formation,
and employment.
in2016, fiscal policy at the global level is projected 1Panel 4 shows the estimated contribution of capital accumulation to the change in

to be broadly neutral in2017 and2018. The overall total factor productivity growth between stated periods. 90 percent condence
neutral stance masks substantial variation across bands are reported. See details in Adler and others (2017).

International Monetary Fund | April 2017 13


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.15. Fiscal Indicators countries and important changes relative to the
(Percent of GDP, unless noted otherwise)
October2016 WEO assumptions. Among advanced
economies, the fiscal stance (measured by the fiscal
Fiscal policy is projected to be broadly neutral at the global level in 2017 and 2018 ,
but this overall neutral stance masks considerable diversity across countries. impulse) in2017 is forecast to be expansionary in
Canada, France, and Germany; contractionary in
4 1. Change in the Structural Fiscal Balance Australia, Korea, and the United Kingdom; and
(Percentage points) broadly neutral in Japan and the United States
3 2013 2014 2015
2016 2017 2018 (Figure1.15).3 For the advanced economies as a
2 October 2016 WEO
whole, and the United States in particular, the
1
projected neutral fiscal stance in2017 represents
0 a slight easing relative to the October2016 WEO
1 assumptions. In2018, the forecast assumes a sizable
2 fiscal stimulus in the United States, reflecting the
Advanced Emerging market and
economies developing economies anticipated changes in U.S.federal government tax
policy. The U.S.fiscal deficit is assumed to widen by
4 2. Change in the Structural Fiscal Balance 2percentage points of GDP by2019, which entails
(Percentage points) a fiscal impulse of 1percent of GDP, with about
3 2013 2014 2015
2016 2017 2018 equally sized decreases in the personal and corporate
2 October 2016 WEO
income tax burdens, concentrated in2018 and2019,
1
and no change in infrastructure spending for the
0 time being.4 In emerging market and developing
1 economies as a group, fiscal adjustment is expected
2 to detract slightly from economic activity in2017
United States Japan1 France and Greece, Ireland,
Germany Italy, Portugal, Spain and2018, albeit with marked differences across
countries and regions.
2 3. Fiscal Balance On the monetary policy front, the forecast assumes
a less gradual normalization of policy interest rates
0
in advanced economies than projected in the Octo-
2
ber2016 WEO, particularly in the United Kingdom
4
and the United States. With the anticipated widening
6 World
of the U.S.fiscal deficit, monetary policy is projected
AEs
8 EMDEs to be moderately less accommodative than previously
10 expected because of stronger demand and inflation
2001 04 07 10 13 16 19 22
pressure. The U.S.policy interest rate is projected to
rise by 75 basis points in2017 and 125 basis points
180 4. Gross Public Debt
World AEs 2 in2018, reaching a long-term equilibrium rate of
160
140
Major AEs 2,3 EMDE Asia just below 3percent in2019. In other advanced
Other EMDEs Latin America and
120 the Caribbean
economies, the forecast assumes that monetary policy
100 will remain very accommodative. Short-term rates
80 are projected to remain negative in the euro area
60 through2018 and close to zero in Japan over the
40
forecast horizon. The assumed monetary policy stances
20
1950 60 70 80 90 2000 10 22 across emerging market economies vary, reflecting these
economies diverse cyclical positions.
Source: IMF staff estimates.
Note: AEs = advanced economies; EMDEs = emerging market and developing
economies.
1
Japans latest gures reect comprehensive methodological revisions adopted in 3The fiscal impulse is defined as the change in the structural fiscal
December 2016. balance as a share of potential output.
2
Data through 2000 exclude the United States. 4The projection for fiscal policy in the United States is the one
3
Canada, France, Germany, Italy, Japan, United Kingdom, United States. IMF staff sees as the most likely among a wide range of possible
scenarios.

14 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Other Assumptions ber2016 WEO. These revisions notwithstanding, the


Global financial conditions are assumed to remain broad story remains unchanged: over the near and
accommodative, though somewhat tighter than fore- medium term, most of the projected pickup in global
cast in the October2016 WEO. As discussed in the growth will stem from stronger activity in emerging
April2017 GFSR, an easing of lending conditions market and developing economies.
in major economies is expected to offset the antici- Economic activity in advanced economies as a
pated rise in interest rates, while the normalization of group is now forecast to grow by 2.0percent in2017
monetary policy in the United States and the United and2018, 0.2percentage point higher than expected
Kingdomeven if faster than previously thoughtis in October2016. The stronger outlook in advanced
expected to proceed smoothly, without triggering large economies reflects a projected cyclical recovery in
and protracted increases in financial market volatility. global manufacturing, signs of which were already vis-
With the exception of several vulnerable economies, ible at the end of2016, and an uptick in confidence,
most emerging markets are expected to face generally especially after the U.S.elections, which are expected
accommodative financial conditions, with higher pol- to fuel the cyclical momentum. As also noted in the
icy rates partially offset by a recovery in risk appetite, January2017 WEO Update, this forecast is partic-
as reflected in the recent decline in sovereign bond ularly uncertain in light of potential changes in the
spreads and the uptick in most equity markets. The policy stance of the new U.S.administration and their
forecast also incorporates a firming of commodity global spillovers.
prices. Oil prices are expected to rise to an average of Growth in the group of emerging market and
$55 a barrel in201718, compared with an average developing economies is forecast to rise to 4.5percent
of $43 a barrel in2016. Nonfuel commodity prices, and 4.8percent, respectively, in2017 and2018, from
in particular for metals, are expected to strengthen an estimated outturn of 4.1percent in2016. This
in2017 relative to their2016 averages as a result of projected upturn reflects, to an important extent, a
substantial infrastructure spending in China, expecta- stabilization or recovery in a number of commodity
tions of fiscal easing in the United States, and a general exporters, some of which underwent painful adjust-
pickup in global demand. Finally, negotiations on the ments following the drop in commodity prices, and
future economic relations between the United King- strengthening growth in India, partially offset by a
dom and the European Union are assumed to proceed gradual slowdown of the Chinese economy. Neverthe-
without raising excessive uncertainty, and the arrange- less, as emphasized in previous WEOs, the outlook for
ments are expected to eventually settle in a manner emerging market and developing economies remains
that avoids a very large increase in economic barriers. uneven and generally below these economies average
performance in200015. A variety of factors weigh
on their outlooks, including Chinas transition to a
Global Outlook for201718 more sustainable pattern of growth that is less reliant
World growth, estimated as in the October2016 on investment and commodity imports; a protracted
WEO, at 3.1percent in2016, is projected to increase adjustment to structurally lower commodity reve-
to 3.5percent in2017 and 3.6percent in2018an nues in some commodity exporters; high debt levels
upward revision of 0.1percentage point for2017 everywhere; sluggish medium-term growth prospects
relative to October. Together with the modest in advanced economies; and domestic strife, politi-
change in the forecast for the overall global growth cal discord, and geopolitical tensions in a number of
rate, projections of the strength of economic activ- countries (see Box1.1).
ity across country groups have also shifted. In line
with the stronger-than-expected pickup in growth
in advanced economies and weaker-than-expected Growth Outlook for the Medium Term
activity in some emerging market economies in the Global growth is forecast to increase marginally
latter half of2016, the forecast for201718 envisions beyond2018, reaching 3.8percent by2022. This
a rebound in activity in advanced economies that is pickup in global activity comes entirely from develop-
faster than previously expected, while growth in2017 ments in emerging market and developing economies,
is forecast to be marginally weaker in emerging mar- where growth is projected to increase to 5percent
ket and developing economies relative to the Octo- by the end of the forecast period. These economies

International Monetary Fund | April 2017 15


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

impact on global activity is further boosted by their in2018), but to increase modestly in France
rising world weight. This forecast assumes continued (1.4percent in2017 and 1.6percent in2018). The
strengthening of growth in commodity exporters, medium-term outlook for the euro area as a whole
albeit to rates much more modest than in200015 remains dim, as projected potential growth is held
(Figure1.12); an acceleration of activity in India back by weak productivity, adverse demographics,
resulting from the implementation of important struc- and, in some countries, unresolved legacy problems
tural reforms; and a successful rebalancing of Chinas of public and private debt overhang, with a high
economy to lower, but still high, trend growth rates. level of nonperforming loans.
Advanced economies more modest medium-term Growth in the United Kingdom is projected to be
growth rates reflect the structural headwinds they face 2.0percent in2017, before declining to 1.5percent
once output gaps have closed: diminished growth of in2018. The 0.9percentage point upward revision
the labor force as populations age, and persistently to the2017 forecast and the 0.2percentage point
low productivity growth, barring significant structural downward revision to the2018 forecast reflect the
reform efforts (Adler and others2017). stronger-than-expected performance of the U.K.
economy since the June Brexit vote, which points
to a more gradual materialization than previously
Growth Outlook for Individual Countries and Regions
anticipated of the negative effects of the United
Advanced Economies Kingdoms decision to leave the European Union.
The U.S.economy is projected to expand at a These effects include reduced consumer purchasing
faster pace in2017 and2018, with growth forecast power following the pounds depreciation and its
at 2.3 and 2.5percent, respectively, a cumulative gradual pass-through to prices and the impact of
increase in GDP of percentage point relative uncertainty on private investment. Though highly
to the October2016 forecast. The stronger near- uncertain, medium-term growth prospects have
term outlook reflects the momentum from the also diminished in the aftermath of the Brexit vote
second half of2016, driven by a cyclical recovery in because of the expected increase in barriers to trade
inventory accumulation, solid consumption growth, and migration, as well as a potential downsizing of
and the assumption of a looser fiscal policy stance. the financial services sector amid possible barriers to
The anticipated shift in the policy mix so far has cross-border financial activity.
buoyed financial markets and strengthened business In Japan, a comprehensive revision of the national
confidence, which could further fuel the current accounts led to an upward revision of historical
momentum. Over a longer horizon, however, the growth rates and placed the2016 growth estimate
outlook for the U.S.economy is more subdued. at 1.0percent, significantly higher than projected
Potential growth is estimated at only 1.8percent, in the October2016 WEO. The growth momen-
weighed down by an aging population and weaker tum, fueled by stronger-than-expected net exports
TFP growth. in2016, is expected to continue into2017, with
The euro area recovery is expected to proceed at a growth forecast at 1.2percent. The pace of expan-
broadly similar pace in201718 as in2016. The sion is expected to weaken thereafter, with the
modest recovery is projected to be supported by a assumed withdrawal of fiscal support and a recovery
mildly expansionary fiscal stance, accommodative of imports offsetting the impact of stronger antici-
financial conditions, a weaker euro, and beneficial pated foreign demand and Tokyo Olympicsrelated
spillovers from a likely U.S.fiscal stimulus; political private investment. Over the medium term, a
uncertainty as elections approach in several coun- shrinking labor force will weigh on Japans growth
tries, coupled with uncertainty about the European prospects, although its per capita income growth
Unions future relationship with the United King- rates are projected to remain near the levels seen
dom, is expected to weigh on activity. Output in the over the past several years.
euro area is expected to grow by 1.7percent in2017 In most other advanced economies, the pace of
and 1.6percent in2018. Growth is forecast to activity is expected to accelerate.
soften in Germany (1.6percent in2017 and 1.5per- oo In Switzerland, growth is projected to rise
cent in2018), Italy (0.8percent in2017 and2018), modestly to 1.4percent in2017 and 1.6percent
and Spain (2.6percent in2017 and 2.1percent in2018, supported by sustained external and

16 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

domestic demand and the waning effects of the vulnerabilities associated with the reliance on near-
past appreciation of the Swiss franc. term policy easing and credit-financed investment.
oo The pace of expansion of Swedens economy is Elsewhere in emerging and developing Asia, growth
expected to moderate to a still-robust 2.7percent is projected to remain robust, even if somewhat
in2017 and 2.4percent in2018. The slowdown lower than anticipated in the October2016 WEO.
from the very strong growth in201516 is partly In India, the growth forecast for2017 has been
a result of normalization of public consumption trimmed by 0.4percentage point to 7.2percent, pri-
and moderation of high investment growth, marily because of the temporary negative consump-
which outweigh some strengthening in private tion shock induced by cash shortages and payment
consumption. disruptions from the recent currency exchange
oo Growth in commodity-exporting advanced econ- initiative. Medium-term growth prospects are favor-
omies is projected to recover. In2017 it is forecast able, with growth forecast to rise to about 8percent
to rise to 1.2percent in Norway, 1.9percent in over the medium term due to the implementation
Canada, and 3.1percent in Australia. The accel- of key reforms, loosening of supply-side bottle-
eration in activity will be supported by accommo- necks, and appropriate fiscal and monetary policies.
dative monetary policies, supportive fiscal policies Economic activity is forecast to accelerate slightly
or infrastructure investment, improving sentiment in2017 in four ASEAN-5 economies (Indonesia,
following the upturn in commodity prices, and less Malaysia, Philippines,Vietnam). The fifth, Thailand,
drag from declining investment in the commodity is projected to recover from a temporary dip in tour-
sector (Australia, Norway). Canadas economy also ism and consumption in late2016. Growth in2017
stands to benefit from the stronger U.S.outlook is projected to be 5.1percent in Indonesia, 4.5per-
and the appreciation of the U.S.dollar. cent in Malaysia, 6.8percent in the Philippines,
oo Among other advanced economies in Asia, a and 6.5percent in Vietnam. In these economies,
pickup in growth for2017 is projected in Hong the near-term pickup in growth is underpinned to
Kong Special Administrative Region (to 2.4per- a significant extent by stronger domestic demand
cent), Taiwan Province of China (to 1.7percent), and, in the Philippines, by higher public spending in
and Singapore (to 2.2percent), partly because of particular.
the expected recovery in Chinas import demand. A weaker-than-previously-expected recovery is pro-
By contrast, a marginal decline in growth is fore- jected to take hold in Latin America and the Carib-
cast in Korea (to 2.7percent in2017, 0.3percent- bean, with growth forecast at 1.1percent in2017
age point less than forecast in the October2016 and 2.0percent in2018 (0.5 and 0.2percentage
WEO), reflecting weaker private consumption point lower than in the October2016 WEO).
growth due to the expiration of temporary sup- Within the region, the growth outlook differs
portive measures, ongoing political uncertainty, substantially across countries. While activity in most
and high household debt. commodity exporters is expected to be supported
by the recovery in commodity prices, domestic
fundamentals continue to play a key role in the out-
Emerging Market and Developing Economies look of some large countries. At the same time, the
Growth in China is projected at 6.6percent outlook for Mexico, one of the largest economies in
in2017, slowing to 6.2percent in2018. The the region, has weakened.
upward revision to near-term growththe2017 oo Growth in Mexico is projected to moderate to
forecast is 0.4percentage point higher than in 1.7percent in2017 and 2.0percent in2018.
the October2016 WEO and the2018 fore- The 1.2percentage point cumulative growth
cast is 0.2percentage point higherreflects the downgrade over the two years reflects subdued
stronger-than-expected momentum in2016 and prospects for investment and consumption in the
the anticipation of continued policy support in the face of tighter financial conditions and increased
form of strong credit growth and reliance on public uncertainty about future U.S.Mexico trade
investment to achieve growth targets. The medi- relations. These factors more than offset the pos-
um-term outlook, however, continues to be clouded itive impact of a stronger U.S.outlook and the
by increasing resource misallocation and growing depreciation of the currency. Continued imple-

International Monetary Fund | April 2017 17


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

mentation of structural reforms in the areas of supported by improved confidence and rising real
energy, labor markets, competition, telecommu- incomes, including from a higher minimum wage,
nications, and the financial sector is projected to but growth is projected to soften slightly to 2per-
boost growth by about percentage point over cent in2017 due to the adverse impact on indus-
the medium term. trial production of the recent trade blockade in the
oo Among commodity exporters, Brazil is expected eastern part of Ukraine.
to emerge from one of its deepest recessions, Economic prospects in emerging and developing
with growth forecast at 0.2percent in2017 and Europe are relatively favorable, with the exception of
1.7percent in2018 (0.3percentage point lower Turkey. For the group as whole, growth is projected
and 0.2percentage point higher, respectively, to remain at 3.0percent in2017 and strengthen
relative to the October2016 WEO forecast). The to 3.3percent in2018. In Turkey, after a sharp
gradual recovery will be supported by reduced slowdown in growth in the third quarter of2016,
political uncertainty, easing monetary policy, and a modest acceleration in activity is projected, with
further progress on the reform agenda. After a growth reaching 2.5percent in2017 based on
contraction last year, activity in Argentina is also stronger net exports and a moderate fiscal stimu-
set to expand by 2.2percent in2017, thanks to lus. The outlook is clouded by heightened political
stronger consumption and public investment, uncertainty, security concerns, and the rising burden
and 2.3percent in2018, reflecting the gradual of foreign-exchange-denominated debt caused by the
rebound of private investment and exports. Ven- lira depreciation. Growth in the rest of the region
ezuela remains mired in a deep economic crisis, is expected to pick up after a temporary slowdown,
with output forecast to contract by 7.4percent as rising wages in some countries support strong
in2017 and 4.1percent in2018, as monetization domestic consumption growth.
of fiscal deficits, extensive economic distortions, In sub-Saharan Africa, a modest recovery is foreseen
and severe restrictions on intermediate goods in2017. Growth is projected to rise to 2.6percent
imports fuel rapidly rising inflation. Higher in2017 and 3.5percent in2018, largely driven by
commodity prices will help strengthen growth specific factors in the largest economies, which faced
in2017 in Chile (1.7percent) and Colombia challenging macroeconomic conditions in2016.
(2.3percent). After contracting by 1.5percent in2016 because of
The near-term outlook for the Commonwealth disruptions in the oil sector coupled with foreign
of Independent States has improved, with growth exchange, power, and fuel shortages, output in
projected to rise to 1.7percent in2017 (0.3per- Nigeria is projected to grow by 0.8percent in2017
centage point higher than forecast in the Octo- as a result of a recovery in oil production, continued
ber2016 WEO). Russia is poised to exit recession, growth in agriculture, and higher public investment.
with growth reaching 1.4percent in2017 (follow- In South Africa, a modest recovery is expected, with
ing a cumulative contraction of about 3percent growth forecast at 0.8percent in2017 as commod-
in the previous two years). The pickup in activity ity prices rebound, drought conditions ease, and
reflects firming oil prices and a recovery in domestic electricity capacity expands. Angolas growth is also
demand attributable to easing financial conditions expected to turn positive in2017 (to 1.3percent),
and improved confidence. At the same time, Russias driven by an expansion in the non-oil sector because
potential growth will remain subdued at about of higher public spending and better terms of trade.
1.5percent barring reforms, slowing a convergence The outlook for the region, however, remains sub-
toward advanced economy per capita income levels. dued: output growth is expected only moderately to
Higher oil prices and the improved outlook for exceed population growth over the forecast horizon,
Russia will support activity elsewhere in the region, having fallen short in2016. Many commodity
given tight linkages through trade, investment, exporters still need to adjust fully to structurally
and remittances. Among oil exporters, growth in lower commodity revenues because commodity
Kazakhstan is now projected to reach 2.5percent pricesthe recent rebound notwithstanding
in2017, 1.9percentage points higher than forecast remain low (restraining stronger growth in Nigeria,
in October, as a result of higher oil production and Angola, and oil exporters within the Economic
stronger external demand. In Ukraine, activity is Community of Central African States). Many of the

18 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

largest non-resource-intensive countries will find it in2017 from 4.4percent last year, mostly reflecting
increasingly hard to sustain growth through higher higher commodity prices.
public capital spending, as they have done in the In the United States, consumer price inflation is
past, in the face of rising public debt and a slowing picking up relatively strongly with the recovery in
credit cycle. energy prices, from 1.3percent in2016 to a pro-
The near-term outlook for the Middle East, jected 2.7percent in2017. Core inflation, however,
North Africa, Afghanistan, and Pakistan region has remains relatively subdued and is forecast to rise
weakened, with growth forecast to be 2.6percent more gradually, reaching its medium-term objective
in2017, 0.8percentage point lower than pro- of 2percent personal consumption expenditure
jected in the October2016 WEO. The subdued inflation targeted by the Federal Reserve by2018,
pace of expansion reflects lower headline growth as economic slack diminishes and wage growth
in the regions oil exporters, driven by the Novem- strengthens.
ber2016 OPEC agreement to cut oil production, Inflation is also picking up in the euro area, to
which masks the expected pickup in non-oil about 1.7percent in2017 from 0.2 last year, partly
growth as the pace of fiscal adjustment to struc- reflecting base effects from energy and food prices.
turally lower oil revenues slows. Continued strife But core inflation remains subdued and the output
and conflict in many countries in the region also gap is still negative; as such, headline inflation will
detract from economic activity. Growth in Saudi only gradually approach the European Central Banks
Arabia, the regions largest economy, is expected objective of below but close to 2percent over the
to slow to 0.4percent in2017 because of lower next few years, reaching 1.9percent in2022. Higher
oil production and ongoing fiscal consolidation, energy prices, the recent weakening in the yen, and
before picking up to 1.3percent in2018. Growth slowly building wage-price pressures are expected to
rates in most other countries in the Cooperation lift inflation in Japan as well. However, with inflation
Council of the Arab States of the Gulf are similarly expectations rising only slowly, the increase in infla-
projected to dip in2017. By contrast, activity in tion is projected to be quite subdued, with inflation
most of the regions oil importers is expected to rates staying well below the Bank of Japans target
continue to accelerate, with growth rising from throughout the forecast horizon.
3.7percent in2016 to 4.0percent in2017 and In all remaining advanced economies, except Nor-
4.4percent in2018. In Pakistan, a broad-based way, consumer price inflation rates are expected to
recovery is expected to continue at a healthy pace, rise in2017. In the United Kingdom, the pounds
with growth forecast at 5percent in2017 and depreciation and the increase in energy prices
5.2percent in2018, supported by ramped-up are projected to push inflation up to 2.5percent
infrastructure investment. In Egypt, comprehen- in2017, before it gradually subsides to the Bank of
sive reforms are expected to deliver sizable growth Englands target of 2percent in the next few years.
dividends, lifting growth from 3.5percent in2017 Average headline inflation is expected to return
to 4.5percent in2018. to positive territory in Singapore and Switzerland
in2017.
The projected path of inflation rates among
Inflation Outlook for201718 emerging market and developing economies shows
With the uptick in commodity prices, a broad- considerable diversity. Inflation in China is expected
based increase in headline inflation rates is projected to pick up to 2.4percent in2017 and to 3percent
in both advanced and emerging market and devel- over the medium term as slack in the industrial
oping economies. In nearly all advanced economies, sector and downward pressure on goods prices
inflation rates are expected to be higher in2017 than diminish. A pickup in inflation is also forecast in
in2016. For the advanced group as a whole, infla- Mexico and Turkey in2017, reflecting mostly the
tion is forecast to be 2.0percent in2017, up from liberalization of gasoline prices in Mexico as well
0.8percent in2016, and to stabilize at about that as the significant depreciation of both countries
level over the next few years. Inflation in emerging currencies. By contrast, inflation rates in Brazil and
market and developing economies (excluding Argen- Russia are expected to continue to decline, reflect-
tina and Venezuela) is projected to rise to 4.7percent ing a combination of negative output gaps and the

International Monetary Fund | April 2017 19


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.16. Global Current Account Balances External Sector Outlook


Global trade is estimated to have grown by 2.2per-
Global current account imbalances narrowed marginally in 2016. In general,
current account balances tended to increase in debtor countries but decline in cent in2016 in volume terms, the slowest pace
creditorschanges that would help stabilize the international investment since2009, and below the 2.4 growth rate of world
positions. Imbalances are projected to remain stable in 2017 but widen again from
2018 onward. GDP at market exchange rates. The further slowdown
is attributable to developments in advanced economies,
1. Global Current Account Balance whose exports and imports slowed substantially relative
(Percent of world GDP) to2015. Weaker trade growth was related to an invest-
4 Afr. and ME Japan China
3 Eur. creditors Adv. Asia Oil exporters ment slowdown and inventory adjustment, especially
2 during the first part of the year. At the same time,
1
there are signs of recovery, as discussed earlier, which
0
should lead to a pickup in trade growth in201718, as
demand and especially capital spending recover.
1
After declining to about percent in2015, trade
2
United States Other adv. Em. Asia Discrepancy growth in emerging market and developing economies
3
Eur. debtors Lat. Am. CEE showed some signs of recovery, rising to an estimated
4
2002 04 06 08 10 12 14 16 18 20 22 2.2percent in2016. This recovery was underpinned
by stronger trade growth in China and India as well
2. IIP 2015 and Change in Current Account Balance as in Russia and the Commonwealth of Independent
1.5 (Percent of GDP)
States, where the contraction in imports moderated from
Change in current-account-to-GDP

1.0 Lat. Am. Japan


Other adv. the dramatic pace in 2015. Trade growth is projected
0.5 Eur. debtors Em. Asia to increase further in201718, as a gradual recovery
ratio, 201516

Afr. and ME Oil exporters


in investment by commodity exporters boosts import
0.0
CEE Eur. creditors growth. As a result, global trade is projected to grow at a
Adv. Asia
0.5 United States rate of close to 4percent in201718 (close to 1 percent-
China
1.0 age point above world growth at market exchange rates).
Preliminary data suggest that global current account
1.5
100 50 0 50 100 150 imbalances in2016 narrowed marginally (Figure1.16,
Net IIP, 2015 panel 1). Among creditor countries, the current
account balance in fuel exporters worsened slightly,
Source: IMF staff estimates.
Note: Adv. Asia = advanced Asia (Hong Kong SAR, Korea, Singapore, Taiwan
reflecting the further decline in oil prices, and the
Province of China); Afr. and ME = Africa and the Middle East (Democratic Republic surplus in China contracted. These developments more
of the Congo, Egypt, Ethiopia, Ghana, Jordan, Kenya, Lebanon, Morocco, South than offset the increase in the current account sur-
Africa, Sudan, Tanzania, Tunisia); CEE = central and eastern Europe (Belarus,
Bulgaria, Croatia, Czech Republic, Hungary, Poland, Romania, Slovak Republic, plus in Japan, which was driven primarily by a sharp
Turkey, Ukraine); Em. Asia = emerging Asia (India, Indonesia, Pakistan, Philippines, decline in the volume and price of energy imports.
Thailand, Vietnam); Eur. creditors = European creditors (Austria, Belgium,
Denmark, Finland, Germany, Luxembourg, Netherlands, Norway, Sweden, Among debtor countries, current account balances
Switzerland); Eur. debtors = European debtors (Cyprus, Greece, Ireland, Italy, strengthened in nonfuel-exporting Latin American
Portugal, Spain, Slovenia); IIP = international investment position; Lat. Am. = Latin
America (Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay); Other adv. =
countries, reflecting the impact of weak domestic
other advanced economies (Australia, Canada, France, Iceland, New Zealand, demand on imports, as well as in emerging Asia and
United Kingdom); Oil exporters = Algeria, Azerbaijan, Iran, Kazakhstan, Kuwait, in euro area debtor countries, also helped by further
Nigeria, Oman, Qatar, Russia, Saudi Arabia, United Arab Emirates, Venezuela.
terms-of-trade gains.
While there is, of course, no normative presumption
dissipation of the effects of past currency depreci- that current account deficits and surpluses should be
ations, supply shocks, and/or administrative price compressed, the IMFs2016 External Sector Report
increases. Inflation in2017 is expected to remain highlights how, in2015, current account imbalances
at double-digit levels in a few large economies in in some of the worlds largest economies were too large
sub-Saharan Africa (for example, Nigeria, Angola, in relation to country-specific norms consistent with
Ghana), reflecting, among other factors, the pass- underlying fundamentals and desirable policies. The
through of large depreciations. forthcoming2017 External Sector Report will update

20 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

those assessments. Current account balances in2016 Figure 1.17. Net International Investment Position
generally tended to increase in debtor countries and
Creditor and debtor positions are estimated to have widened in 2016 and are
decrease in creditor countries, thereby moving in a projected to widen further over the medium term.
stabilizing direction (Figure1.16, panel 2). The global
current account forecasts indicate broad stability of 1. Global International Investment Position
imbalances in2017 but a widening of deficits starting 40
(Percent of world GDP)
in2018, as a projected fiscal expansion would lead to Afr. and ME Japan China
30 Eur. creditors Adv. Asia Oil exporters
stronger domestic demand in the United States and a
20
higher current account deficit (Figure1.16, panel 1).
10
Despite the narrowing of flow imbalances, creditor
0
and debtor positions are estimated to have widened
10
in2016 and are projected to widen further over the
20
medium term in relation to world GDP (Figure1.17,
30 United States Other adv. Em. Asia
panel 1).5 On the debtor side, the increase is explained Eur. debtors Lat. Am. CEE
entirely by rising net external liabilities in the United 40
2005 07 09 11 13 15 17 19 21 22
States, where the current account deficit is projected
to widen over the next few years. In contrast, net 2. Net IIP, 2016, and Projected Changes, 201622
40 (Percent of GDP)
external liabilities are projected to shrink further in

Projected change in IIP, 2016 22


Eur. creditors
euro area debtor countries. Among creditor countries, 30 Adv. Asia
the increase in net external claims reflects primarily 20 Eur. debtors Japan
Em. Asia
the projected continuation of large current account
10 Lat. Am.
surpluses in European creditor countries (such as
CEE
Germany and the Netherlands) and in advanced Asian 0
Afr. and ME China
economies. Oil exporters
10 Other adv.
The assessment of net international investment United States
positions is becoming increasingly complex as these 20
100 50 0 50 100 150
positionsalongside national accounts figurescan be Net IIP, 2016
affected by financial decisions related to the corporate
structure of large multinational companies, with no clear Source: IMF staff estimates.
repercussions for external sustainability (or any tangible Note: Adv. Asia = advanced Asia (Hong Kong SAR, Korea, Singapore, Taiwan
Province of China); Afr. and ME = Africa and the Middle East (Democratic Republic
effects on employment and living standards). A case in of the Congo, Egypt, Ethiopia, Ghana, Jordan, Kenya, Lebanon, Morocco, South
point is Ireland, where the relocation of entire balance Africa, Sudan, Tanzania, Tunisia); CEE = central and eastern Europe (Belarus,
Bulgaria, Croatia, Czech Republic, Hungary, Poland, Romania, Slovak Republic,
sheets by multinational companies, and in particular Turkey, Ukraine); Em. Asia = Emerging Asia (India, Indonesia, Pakistan, Philippines,
intellectual property products, led to a very large upward Thailand, Vietnam); Eur. creditors = European creditors (Austria, Belgium,
Denmark, Finland, Germany, Luxembourg, Netherlands, Norway, Sweden,
revision in the stock of intangible capital in the country
Switzerland); Eur. debtors = European debtors (Cyprus, Greece, Ireland, Italy,
in2015 (about 300billion, larger than Irish GDP). Portugal, Spain, Slovenia); IIP = international investment position; Lat. Am. = Latin
There was a corresponding increase in Irish net external America (Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay); Other adv. =
Other advanced economies (Australia, Canada, France, Iceland, New Zealand,
liabilities, which thus exceeded200percent of GDP, United Kingdom); Oil exporters = Algeria, Azerbaijan, Iran, Kazakhstan, Kuwait,
as well as a sharp upward revision to growth. Box1.2 Nigeria, Oman, Qatar, Russia, Saudi Arabia, United Arab Emirates, Venezuela.

5Predicting the evolution of the net international investment posi-


discusses the repercussions of these financial operations
tion is particularly difficult given the importanct role of valuation
changes arising from movements in exchange rates and other asset for domestic and external accounts in Ireland and the
prices. These changes have contributed to a sharp widening in the measurement challenges they pose.
U.S. net liability position in recent years, as the appreciation of the Panel 2 of Figure1.17 shows how creditor and
U.S. dollar has reduced the dollar value of U.S. external assets, and
to corresponding improvements in countries experiencing sharp debtor positions are projected to evolve over201622
exchange rate depreciations and holding dollar assets. Valuation as a share of domestic GDP. It highlights further
changes have also been notable in the United Kingdom, where the growth in creditor positions among both European
depreciation of the pound has turned the country into a net creditor
as of 2016, by boosting the domestic-currency value of foreign-
creditor countries and advanced economies in Asia
currency assets. in the range of 2530percentage points of GDP;

International Monetary Fund | April 2017 21


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.18. Growth for Creditors and Debtors among debtor countries the largest reduction in net
(Percent) liabilities is projected for euro area debtor countries
(over 18percentage points of GDP). The projected
Among creditor countries and regions, net external demand in 201516 supported
output growth in oil exporters and Japan, whereas it detracted from growth in deterioration in the U.S.net external position is about
China and advanced Asia. Among debtors, net external demand has added to 8percentage points of GDP.
growth in Latin America and in European debtor countries, while it has deducted
Figure1.18 looks at global rebalancing from a
from growth in the United States.
different but related anglenamely, the contribution
Net external contribution to Domestic demand contribution to
to a countrys or a regions growth rate from domestic
growth growth demand and from net external demand. In the aftermath
Total of the global financial crisis, the growth rate of creditor
countries, in the aggregate, has exceeded that of debtor
10 1. Growth for Creditors
countries, reflecting to a significant extent rapid growth
8 in China. Among creditor countries and regions, the fig-
6 ure shows that during201516, the contribution of net
4 external demand to growth in China, smaller advanced
Asian economies, and European creditor countries has
2
declined. It has, however, increased in Japan and espe-
0
cially in oil exporters, where domestic demand has been
2 contracting, dragging down the demand for imports.6
4 Among debtor countries, those in Latin America display
1516

1516

1516

201014

1516
1516
201014

201014

201014

201014

a pattern similar to the one for oil exporters for the


same reasons. Among other debtor regions, net external
Eur. creditors China Japan Adv. Asia Oil exporters demand has been supporting growth in euro area debtor
countries, albeit to a lesser extent than in201014 in
light of recovery in their domestic demand.
10 2. Growth for Debtors
The shifting constellation of global macroeconomic
8
policies and associated exchange rate movements could
6 lead flow imbalances to widen again, generating a
4 further expansion of stock imbalances. In the future,
2 stronger reliance on domestic demand growth in a
0
number of creditor countries, especially those with the
policy space to support it, would help sustain world
2
growth while facilitating global rebalancing. In the
4 United States, which already has close to full employ-
1516

1516

1516

1516
201014

201014
1516

201014

201014

201014
1516

201014

ment, fiscal policy measures designed to gradually


United Eur. debtors Other adv. Latin Em. Asia CEE enhance productive capacity along with demand,
States America anchored in a medium-term fiscal consolidation plan
to bring down the rising ratio of public debt to GDP,
Source: IMF staff calculations. would result in a more sustained growth impact and
Note: Adv. Asia = advanced Asia (Hong Kong SAR, Korea, Singapore, Taiwan
Province of China); CEE = central and eastern Europe (Belarus, Bulgaria, Croatia,
help contain external imbalances.
Czech Republic, Hungary, Poland, Romania, Slovak Republic, Turkey, Ukraine); Em.
Asia = emerging Asia (India, Indonesia, Pakistan, Philippines, Thailand, Vietnam);
Eur. creditors = European creditors (Austria, Belgium, Denmark, Finland, Germany, Risks
Luxembourg, Netherlands, Norway, Sweden, Switzerland); Eur. debtors = European
debtors (Cyprus, Greece, Ireland, Italy, Portugal, Spain, Slovenia); Latin America =
Risks Remain Tilted to the Downside
Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay; other Adv. = other WEO growth forecasts represent the IMF staffs
advanced economies (Australia, Canada, France, Iceland, New Zealand, United
Kingdom); Oil exporters = Algeria, Azerbaijan, Iran, Kazakhstan, Kuwait, Nigeria, modal scenariothe growth path the staff sees as
Oman, Qatar, Russia, Saudi Arabia, United Arab Emirates, Venezuela.

6Given the very large terms-of-trade losses discussed in the first sec-

tion, current account balances have actually worsened in oil exporters,


despite the sharp import contraction (Figure1.16, panel 2).

22 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

the most likely within the distribution of possible increase in tariffs or other trade barriers would harm
outcomes. Outturns may differ from the baseline both the U.S.economy and its trading partners, espe-
forecast if key macroeconomic policies are different cially if there are retaliatory responses. In Europe, the
than assumed or if economic and noneconomic shocks coming elections offer a platform for such protectionist
materialize. The former factor is particularly salient policy tendencies to enter the mainstream.
at this time, given the high uncertainty surrounding Most economists agree that raising barriers to trade
policies going forward. would reduce aggregate output and lower well-being.
Risks to the baseline forecast remain tilted to the As shown in Scenario Box1 of the October2016
downside, more so over the medium term. But near- WEO, a country that hikes tariffs can expect to see its
term upside potential has risen in recent months. In price level rise and output fall, especially if its trading
particular, gains in business and consumer sentiment partners retaliate. The analysis also shows that a broad-
in advanced economies since last fall, as reflected in based increase in import costs caused by heightened
survey outcomes and equity prices, could underpin global trade protectionism would put a dent in global
stronger momentum in consumption and invest- output. The damage could be even higher in light of
ment in the short term. If followed through by the increasing fragmentation of production processes
supply-friendly reforms and policies, the momentum across countries (Koopman, Wang, and Wei2014;
could become entrenched and sustain the pickup Yi2003,2010). Higher import costs could do partic-
in activity for longer. Another source of short-term ular harm to the purchasing power of lower-income
upside risk is the possibility of policy easing greater groups in advanced economies, whose consumption
than assumed in the baseline in the United States and baskets tend to skew toward heavily traded goods
China. For instance, pending specifics, the baseline (Fajgelbaum and Khandelwal2016). Further to such
forecast for the United States does not incorporate immediate adverse effects on demand, a persistent,
additional public infrastructure investment. But the protection-induced reduction of trade could also
size and composition of fiscal policy easing may also harm supply-side potential. As competitive pressures
be modest and less growth friendly than assumed in to innovate weaken, and the cross-border diffusion of
the baseline, as discussed below. new technologies slows, productivity growth would
There are five primary areas of uncertainty affecting suffer over time. Similarly, curbing immigration flows
the forecast, most pointing to downside risks relative would hinder opportunities for skill specialization
to the baseline. in advanced economies, limiting a positive force for
productivity and income growth over the long term
Disruption of Global Trade, Capital Flows, and (Chapter4 of the October2016 WEO).
Migration The negative repercussions of protectionism could
As noted in Chapter3, a number of middle-skill be even larger if the disruption of international
jobs in advanced economies have been lost as a economic linkages leads to a more generalized decline
result of technological change since the early1990s. in cross-border cooperation. As coordinated solu-
And the slow recovery from the crises of200809 tions to multilateral challenges become more elusive,
and201112 in countries where the distribution of heightened perceptions of policy ineffectiveness could
income has continued to favor the highest earners magnify the output costs of negative shocks, including
has left little room for those with lower incomes to those discussed further below.
advanceor in some cases, even preservetheir living So far, signs of a potential inward-looking tilt in
standards. The resultnotably in the United States policies have not had a noticeable impact on economic
and parts of Europehas been growing disillusion- sentiment indicators in advanced economies. For
ment with globalization. There is a palpable risk that instance, despite the increased possibility of greater
legitimate equity concerns could trigger protectionist impediments to trade and migration down the road,
policy actions under the pressure of mounting skep- private sector confidence and spending in the United
ticism toward trade, immigration, and multilateral Kingdom have remained resilient in the aftermath of
engagement. In the United States, the authorities have the Brexit vote. This resilience could reflect still-high
declared their intention to reopen existing trade agree- expectations of a favorable outcome; the backdrop of
ments. If well executed, and mutually agreeable, such an improving global economy may also have helped
efforts could benefit all signatories; by contrast, an mask some of the concerns. Nonetheless, growing

International Monetary Fund | April 2017 23


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

salience of a future increase in trade costs will likely the U.S.current account deficit would widen more.
gradually dampen expectations of future real earnings The associated widening in global imbalances in such a
and weigh on investment and hiring. Such head- scenario could intensify the demand for trade protec-
winds could be magnified if the negotiations on new tion and retaliatory responses.
trade agreements are drawn out and contribute to an Fiscal sustainability would require any increase in
increase in uncertainty. A case in point is Mexico, the U.S.federal deficit to be reversed at some point.
where financial market conditions have tightened That is, a fiscal policy shift that results in sustained
noticeably because of fears of protectionist policy widening of the fiscal deficit would essentially shift
changes in the United States. demand from the future to the present, support-
ing short-term activity but imposing a drag on
The U.S.Policy Agenda U.S.growth over the medium term. To illustrate these
Several aspects of the U.S.policy agenda contribute considerations, Scenario Box1 discusses the potential
to uncertainty around the U.S.and global growth pro- consequences of an increase in U.S.federal govern-
jections, in particular the size and composition of any ment spending and tax cuts using stylized scenarios. It
fiscal policy easing, and the impact of a possible reform contrasts a scenario in which the changes yield a strong
of the corporate tax system (toward destination-based increase in U.S.potential output with one in which
cash flow taxation). the positive supply effects are more limited (but still
positive) and both U.S.and global financial conditions
The U.S.Fiscal Policy Stance tighten more rapidly. The IMF staffs baseline growth
The projections for the April2017 WEO were projections for the United States would fall between
prepared before crucial details of U.S.fiscal policy these two cases. In both hypothetical scenarios, fiscal
changesincluding the overall amount and compo- adjustment is undertaken five years into the simulation
sition of easingwere known. Uncertainty about the horizon to stabilize public debt, which requires a larger
U.S.policy actions and their effects on U.S.aggregate contraction in the primary deficit in the second sce-
demand, potential output, the government budget nario than in the first, given the more limited increase
deficit, and the value of the U.S.dollar suggests in potential output.
a wide range of upside and downside risks to the In the United States, output rises above the baseline
current baseline forecast for the United States, in path in both cases, an output gap opens up, mone-
both the near and the medium term. Global spillovers tary policy tightens, the U.S.dollar appreciates, and
are thus also uncertain and will vary across coun- the U.S.current account deficit widens given the
tries, depending on their economic linkages with the increase in U.S.permanent income. These effects
United States and their sensitivity to changes in global are generally stronger in the first case, in which the
financial conditions, as discussed in Chapter3 of the impact on potential output is more favorable. The
April2017 GFSR. increased demand for foreign saving by the United
A sustained noninflationary increase in output States raises the global interest rate in both cases,
in the United States, underpinned by a significant but more in the second case owing to the assumed
expansion of the U.S.capital stock and a lasting rise faster normalization in U.S.and global term premia.
in labor force participation, should be associated with The permanent increase in the level of U.S.public
a moderate pace of interest rate increases under the debt also adds to the upward pressure on the global
Federal Reserves price stability mandate. By contrast, interest rate. The dollar depreciates over the longer
if a large fiscal stimulus does not lead to a significant term, given the assumed permanent decline in
increase in supply potential, or if the inflation response U.S.public sector saving.
to the rise in demand is larger than expected, a steeper The impact on most other economies is initially
path for interest rates would be necessary to contain positive under the first scenario because the larger
inflation. The weaker fiscal position could lead markets increase in U.S.imports outweighs the negative
to deliver faster normalization of the term premium effect on demand of higher global interest rates. In
causing tighter overall financial conditions both in the the second scenario, the boost from U.S.imports
United States and globallywhich could put stress on to foreign output is more limited, given a smaller
many emerging market and some low-income econo- rise in U.S.demand, and is more than offset by
mies. The dollar would appreciate more sharply, and the adverse impact from the sharper tightening in

24 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

financial conditions. Once U.S.fiscal policy tightens cash flow tax (discussed in detail in Box1.1 of the
in the medium term, the positive demand spill- Fiscal Monitor). If the proposal is implemented, the
overs weaken and output falls below baseline in all full and immediate expensing of investment under the
economies in both scenarios because of permanently destination-based tax would be expected to meaning-
tighter financial conditions. fully boost U.S.business investment and output.
A replacement of the U.S.corporate income tax
A number of factors are not captured in the sim- with a destination-based cash flow tax could generate
ulations. On the upside, productivity gains in the large international spillovers through several channels.
United States could spill over to some extent on other As discussed in Box1.1 of the Fiscal Monitor, the
economies, boosting permanent incomes and demand change would generate strong incentives for profit
there as well. A more generalized rise in productiv- and production shifting into the United States. Other
ity would temper the widening of the U.S.current countries might then take measures to protect their
account deficit, the increase in global interest rates, and own tax bases or ultimately also move toward destina-
the attendant negative ramifications for other econo- tion-based taxation.
mies. On the downside, the initial appreciation of the A cash flow tax with full expensing of capital would
dollar could generate financial and real stress among be expected to raise the U.S.household saving rate
emerging market economies with de jure or de facto and put downward pressure on global interest rates.
currency pegs to the U.S.dollar and/or balance-sheet The effects of the change on U.S.competitiveness,
vulnerabilities (associated with currency mismatches) however, would likely be limited. The border adjust-
aspects not captured in the model simulations but ment inherent in destination-based taxationwhich
elaborated further below. Finally, as noted in Scenario exempts exports from revenues and does not allow
Box1, a similar growth-friendly fiscal policy imple- firms to deduct the cost of imports from their tax
mented in a deficit-neutral way would lead to an even basewould in the simplest textbook case strengthen
higher long-term level of GDP. the dollar relative to all other currencies and/or raise
All in all, the simulations point to the downside domestic prices and wages, so as to leave the trade bal-
risks associated with deficit-financed U.S.fiscal policy ance unchanged. A sharp appreciation of the U.S.dol-
easing, especially in the medium term. The scenar- lar, however, would generate deflation pressure in
ios highlight how the ultimate impact of the policy economies whose currencies are tied to the U.S.dollar
changes on the U.S.economy itself depend on whether and could impose financial stress on countries whose
the measures successfully lift U.S.potential output. private or public balance sheets contain significant
They also underscore the possible negative inter- currency mismatches. In addition, the border adjust-
national repercussions of the policy easing through ment may prove inconsistent with existing World
tighter global financial conditions. Trade Organization rules, which may lead to trade
disputes with trading partners, posing risks to the
U.S. Corporate Tax Reform open trading system.
Beyond a shift to a more expansionary fiscal policy,
potentially far-reaching tax policy changes are being Financial Deregulation
considered in the United States, including a structural As discussed in Chapter1 of the April2017 GFSR,
overhaul of the corporate income tax. The U.S.corpo- the postcrisis reform agenda has strengthened oversight
rate tax system has well-documented shortcomings and of the financial system, raised capital and liquid-
distortions. It is too complex, has a narrow base and ity buffers of individual institutions, and improved
a marginal rate that is too high, is rife with legislated cooperation among regulators.A wholesale dilution
exemptions, favors debt financing, and incentivizes or backtracking on important steps taken since the
a range of cross-border avoidance and tax planning global financial crisis in enhancing the resilience of the
mechanisms to lower U.S.tax liabilities.7 One spe- financial system would raise the probability of costly
cific proposal now under discussion is to replace the financial crises in the future. Deregulation in one
U.S.corporate income tax with a destination-based country may also lead to deregulation in others in the
highly interconnected international financial system.
7See Box6 of the 2016 IMF Article IV Staff Report on the A failure to complete the global reform agenda and
United States. allowing regulatory fragmentation across borders would

International Monetary Fund | April 2017 25


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

also hurt countries outside the central standard-setting in late2016, causing losses for leveraged bond inves-
bodies, in particular emerging market economies, tors and pushing up bond yields sharply. Segments
which rely heavily on a strong global standard to level of the repurchase arrangement market began to seize
the playing field and support financial stability at a up, leading the authorities to take actions to provide
time when threats to their domestic financial stability broad-based liquidity support in December2016. This
have risen. episode of market turmoil serves as a reminder of the
elevated risks associated with existing vulnerabilities in
Tightening of Economic and Financial Conditions in Chinas financial system, as discussed in Chapter1 of
Emerging Market Economies the April2017 GFSR.
Emerging market and developing economies have The baseline forecast assumes limited progress in
accounted for the bulk of the downward revisions to tackling the corporate debt overhang and reining
global growth in recent years and have been a source in credit, and a policy preference for maintaining
of uncertainty around the WEO forecasts. Most of relatively high GDP growth in the near term. The
the downward revisions to growth have been in China resulting persistent resource misallocation, however,
and India, especially during201113; in commodity raises the risk of a disruptive adjustment in China in
exporters following the201516 plunge in oil prices; the medium term.
and, to a lesser extent, in Middle Eastern economies External triggers, such as a shift toward protection-
suffering from conflict (see Box1.1). ism in advanced economies or domestic shocks, could
Many emerging market economies have gone lead to a broader tightening of financial conditions
through bouts of financial volatility over the past few in China, possibly exacerbated by capital outflow
years. Some large commodity exporters and other pressures, with an adverse impact on demand and
stressed economies have also weathered substantial output. As demonstrated by market jitters in the
exchange rate movements, while China has experi- second half of2015 and early2016, spillovers onto
enced a swing from net capital inflows to sizable net other economies from turbulence in China can be
outflows. Though it proved short lived for most, the large, operating mainly through commodity prices
tightening of financial conditions across emerging and global financial risk aversion (Chapter4 of the
market economies in the immediate aftermath of the October2016 WEO).
U.S.election is a reminder that many countries in this
group remain vulnerable to sudden shifts in global Vulnerabilities in Other Emerging Market and
market sentiment. Developing Economies
Compared with past episodes of capital inflow
Risks from Continued Rapid Credit Expansion in China slowdowns, emerging market economies have seen
Chinese authorities are expected to maintain fewer financial sector problems in recent years,
emphasis on protecting macroeconomic stability in despite entering the episode with highly leveraged
the run-up to the leadership transition later this year. corporate sectors and, in some cases, experiencing
Progress with demand-side rebalancing and reducing sharp losses in earnings driven by adverse shifts in
excess industrial capacity has continued, but so has their terms of trade (Chapter2 of the April2016
the reliance on stimulus measures to maintain high WEO). The improvement in emerging market
rates of growth and the Chinese economys dangerous economies ability to cope with external volatility is
dependence on rapidly expanding credit, intermediated testimony to better macroeconomic policy manage-
through an increasingly opaque and complex financial ment and in particular the beneficial role of exchange
system. Recent months have seen a return of capital rate flexibility in smoothing shocks. Credit booms are
outflows, reflecting market expectations of renminbi waning in many economies (with the key exception
depreciation against the dollar and narrowing yield of China), and corporate leverage, in most cases, has
differentials as global interest rates increased. Though peaked and continues to decline from a high level.
Chinese equity markets have remained tranquil, in But underlying fragilities remain, and in some cases,
stark contrast to the turmoil of August2015 and Janu- corporate sector buffers could be wearing thin after
ary2016, bond markets have seen bouts of turbulence. a period of macroeconomic strains and financial vol-
Efforts by the Peoples Bank of China to tighten short- atility. More generally, reduced profitability, still-el-
term liquidity pushed up repurchase arrangement rates evated corporate debt, limited policy space, and, in

26 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

some cases, weak bank balance sheets suggest that A downshift in inflation expectations, higher
some emerging market economies remain potentially expected real interest rates, debt service difficulties,
exposed to tighter global financial conditions, capital and negative feedback to demand
flow reversals, and the adverse balance sheet implica- Weak investment and slower adoption of capital-
tions of sharp currency depreciations (Chapter1 of embodied technological change, lower productivity
the April2017 GFSR). Such strains could material- growth, and weaker expected profitability, reinforc-
ize, for example, if the projected fiscal policy easing ing the sluggishness in investment
in the United States proves to be more inflationary A prolonged period of high unemployment leading
than expected, requiring a faster pace of monetary some job seekers to drop out of the workforce or
policy tightening and triggering a faster normalization become unemployable as a result of skill erosion
of U.S.term premia (a possibility discussed above),
or if there is a marked shift toward protectionist With a slightly firmer outlook for demand in
policy actions in advanced economies. As elaborated advanced economies, fears of such debilitating cycles
in Chapter2, a weakening growth impulse from a have receded somewhat. Steepening yield curves have
less supportive external environment could lead to also alleviated some of the concerns about the prof-
persistent and durable shifts in growth outcomes for itability of banks and other financial intermediaries
emerging market and developing economies, raising and their ability to support the recovery. Nevertheless,
financial vulnerabilities as well. in parts of Europe, the cyclical recovery in output,
In the baseline forecast, recoveries in a relatively employment, and inflation remains incomplete under
small number of stressed economiesmost of which a large burden of nonperforming loans, and banking
are commodity exportersaccount for an important system profitability is challenged by structural features,
portion of the global growth pickup in201718. such as high costs and overbanking (Chapter1 of the
The pace of these recoveries could fall short of the April2017 GFSR). In the absence of a more concerted
baseline projections if domestic reforms to tackle effort to clean up balance sheets, consolidate and raise
structural problems are delayed, harming confidence. the cost effectiveness of banking systems, maintain
Likewise, in many commodity-exporting low-income demand, and enact productivity-enhancing reforms,
economies where fiscal buffers are exhausted, further these economies will continue to confront weak infla-
delays in policy adjustments could lead to disor- tion dynamics and investment and remain susceptible
derly conditions and weaker growth than currently to the danger of self-reinforcing adverse feedback
projected. A reversal of foreign direct investment loops. As growth and core inflation prospects in core
and other capital flows from China could also put euro area economies strengthen, there is also a risk that
significant strain on a number of low-income econo- euro area monetary policy tightens, weighing on the
mies that rely increasingly on such financing for key recovery in countries with high unemployment and
infrastructure projects. large output gaps. A sluggish recovery in incomes can,
Even in emerging market economies where growth in turn, fuel pressures for an inward turn in policies
has remained resilient in recent years, in some cases and the adoption of protectionist measures, further
because of favorable terms-of-trade shifts, investor harming demand both at home and abroad.
sentiment could falter and growth could disappoint
if policymakers do not implement needed structural Noneconomic Factors
reforms, tackle debt overhangs, and undertake neces- Geopolitical tensions as well as domestic strife and
sary fiscal adjustments. idiosyncratic political problems have been on the rise
in recent years, burdening the outlook for various
Weak Demand and Balance Sheet Problems in Parts regions. Most notable are the civil wars and domestic
of Europe conflicts in parts of the Middle East and Africa, the
One common theme running through several recent tragic plight of refugees and migrants in neighboring
WEO reports has been weak demand in a number of countries and in Europe, and acts of terror world-
advanced economies and its possibly pernicious and wide. For many of the severely affected countries, the
long-lasting effects on inflation and supply potential. baseline scenario assumes a gradual easing of tensions.
These effects could, in principle, work through three However, these episodes may turn out to be more
channels: protracted, holding back recovery in these countries.

International Monetary Fund | April 2017 27


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.19. Risks to the Global Outlook Weak governance and large-scale corruption can also
A fan chart analysis suggests that risks to the global growth outlook remain undermine confidence and popular support, taking a
skewed to the downside. heavy toll on domestic activity.
6 1. Prospects for World GDP Growth
1 Other noneconomic factors weighing on growth
(Percent change) include the persistent effects of a drought in eastern
5
and southern Africa and the spread of the Zika virus.
4 If these factors intensify, the hardship in directly
WEO baseline
3
affected countries, especially smaller developing econo-
90 percent condence interval mies, would deepen (IMF2016). Increased geopolitical
2 70 percent condence interval
50 percent condence interval tensions and terrorism could also take a toll on global
1 90 percent condence interval from October 2016 WEO market sentiment and broader economic confidence.
90 percent condence interval from April 2016 WEO
0
2014 15 16 17 18
1.5 2. Balance of Risks Associated with Selected Risk Factors 2 Fan Chart
(Coefcient of skewness expressed in units of the underlying
1.0 variables) A fan chart analysisbased on equity and commod-
ity market data as well as the dispersion of inflation
0.5
and term spread projections of private sector forecast-
0.0 erscorroborates the assessment that risks remain
skewed to the downside for2017 and2018. The analy-
0.5 Current year (April 2016 WEO)
Current year (April 2017 WEO) sis suggests a narrower dispersion of outcomes around
1.0 Balance of risks for Next year (April 2016 WEO) the current- and next-year baseline than a year ago,
Next year (April 2017 WEO)
1.5 consistent with the more optimistic tone in financial
Term spread S&P 500 Ination risk Oil market risks
markets and reduced uncertainty in the aftermath of
Dispersion of Forecasts and Implied Volatility 3 the Brexit vote in June2016 and the U.S.elections
80 3. 1.2 125 4. 0.5 in November. Nonetheless, the analysis continues to
GDP (right scale) Term spread
70 suggest that the balance of risks to the outlook are
VIX (left scale) (right scale)
1.0 100 tilted to the downside. As illustrated in Figure1.19,
60 Oil (left scale) 0.4
50 although the width of the 90percent confidence inter-
0.8 75
40 0.3 val has diminished for both the current- and next-year
30 0.6 50 growth forecasts, the decline is slightly greater for the
20 0.2
upper part of the interval, pointing to a somewhat
0.4 25 more pronounced downward skew of risks than in
10
October2016.
0 0.2 0 0.1
2006 08 10 12 14 Jan. 2006 08 10 12 14 Jan. The probability of a recession over a four-quarter
17 17
horizon (first quarter of2017fourth quarter of2017)
Sources: Bloomberg L.P.; Chicago Board Options Exchange (CBOE); Consensus has declined in most regions, relative to the probabil-
Economics; Haver Analytics; and IMF staff estimates. ity computed in October2016 for the third quarter
1
The fan chart shows the uncertainty around the April 2017 World Economic
Outlook (WEO) central forecast with 50, 70, and 90 percent condence intervals. As
of2016second quarter of2017 (Figure1.20). Stron-
shown, the 70 percent condence interval includes the 50 percent interval, and the ger cyclical momentum and the anticipated U.S.fiscal
90 percent condence interval includes the 50 and 70 percent intervals. See stimulus have lifted the growth outlook in advanced
Appendix 1.2 of the April 2009 WEO for details. The 90 percent intervals for the
current-year and one-year-ahead forecasts from the October 2016 WEO and April economies, while the increase in external demand and
2016 WEO are shown. the rise in commodity prices have boosted growth
2
The bars depict the coefcient of skewness expressed in units of the underlying
variables. The values for ination risks and oil price risks enter with the opposite prospects in emerging Asia and selected commod-
sign since they represent downside risks to growth. ity exporters. Deflation risksas measured by the
3
GDP measures the purchasing-power-parity-weighted average dispersion of GDP
growth forecasts for the Group of Seven economies (Canada, France, Germany, Italy, estimated probability of a decline in the price level
Japan, United Kingdom, United States), Brazil, China, India, and Mexico. VIX is the relative to one year agoremain elevated for the euro
CBOE Standard & Poors (S&P) 500 Implied Volatility Index. Term spread measures
the average dispersion of term spreads implicit in interest rate forecasts for
area and Japan because the pass-through of higher
Germany, Japan, the United Kingdom, and the United States. Oil is the CBOE crude commodity prices to headline inflation is projected
oil volatility index. Forecasts are from Consensus Economics surveys. Dashed lines to fade next year and core inflation remains weak,
represent the average values from 2000 to the present.
especially in Japan.

28 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Policy Priorities Figure 1.20. Recession and Deation Risks


(Percent)
Global economic activity is picking up speed, but
the potential for disappointments remains high, and The probability of recession has declined in most regions, except in Japan where it
momentum is unlikely to be sustained in the absence is broadly unchanged. Deation risks remain elevated in Japan and the euro area.
of efforts by policymakers to implement the right set
of policies and avoid missteps. For many economies, 70 1. Probability of Recession, 2017:Q22018:Q1
continued demand support and well-targeted structural 60 October 2016 WEO: 2016:Q42017:Q3
reforms to lift supply potential and broaden economic 50
opportunities across the skills spectrum remain key
40
goals. The precise combination of priorities differs
30
across individual economies, depending on their
cyclical conditions, structural challenges, and needs for 20
enhancing resilience. 10
An overarching challenge for policymakers will 0
be to safeguard global economic integration and the United Euro area Japan Emerging Latin Rest of the
States Asia America 5 world
cooperative global economic order, which have been
critical sources of productivity growth and resilience 30 2. Probability of Deation, 2018:Q21
over the past several decades. A body of research has October 2016 WEO: 2017:Q4
documented that economic integration together with 25

technological progress has increased the efficient use of 20


global resources, boosted incomes, and expanded access
15
to goods and services.8Hundreds ofmillions were
lifted out of poverty through this process, helping to 10
reduce global income inequality. 5
However, amid weak growth and rising inequality,
popular support for international trade and immi- 0
United Euro area Japan Emerging Latin Rest of the
gration has eroded in some advanced economies. As States Asia America 5 world
documented in Chapter3, wages have not kept up
with productivity in many economies over much of Source: IMF staff estimates.
Note: Emerging Asia comprises China, Hong Kong SAR, India, Indonesia, Korea,
the past three decades, leading to a decline in labors Malaysia, the Philippines, Singapore, Taiwan Province of China, and Thailand; Latin
share of national income. Moreover, the declines have America 5 comprises Brazil, Chile, Colombia, Mexico, and Peru; Rest of the world
comprises Argentina, Australia, Bulgaria, Canada, Czech Republic, Denmark, Israel,
been much harsher for those in lower- and middle-skill New Zealand, Norway, Russia, South Africa, Sweden, Switzerland, Turkey, the
groups, potentially contributing to worsening income United Kingdom, and Venezuela. October 2016 WEO data refer to simulations run in
September 2016.
distributions and income polarization within coun- 1
Deation is dened as a fall in the price level on a year-over-year basis in the
tries. As this process coincided with deepening global quarter indicated in the gure.
economic integration, the economic model of free
movement of goods and factors of production, which
has guided policymaking over the past several decades,
is being increasingly questioned as a politically viable only a relatively small role. The analysis does find
mechanism for delivering broad-based growth. How declining labor shares in emerging market economies
much of the deterioration in income distributions and to be closely linked to trade integration. However, this
the decline in the labor share of income can be traced reflects the fact that, with the rise in global production
to cross-border economic integration? The analysis in sharing, trade has been increasingly accompanied by
Chapter3 suggests that the bulk of the decline in the investment flows and capital deepeninga devel-
labor share in advanced economies is attributable to opment that is otherwise beneficial to capital-scarce
technological change, with trade integration playing emerging market economies. Nonetheless, amid grow-
ing recognition that the gains from growth often are
8For
not broadly shared, support for inward-looking pro-
a recent summary, see Baldwin (2016). See also Fajgelbaum
and Khandelwal (2016), Costinot and Rodrguez-Clare (2013), and tectionist measures and restrictions on the cross-border
Wacziarg and Welch (2008). movement of people is gaining ground.

International Monetary Fund | April 2017 29


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Rolling back economic integration would not some cases) remains. Monetary policy therefore
address these legitimate distributional concerns, which must continue to chart an accommodative course,
are to a significant extent the consequence of tech- relying on unconventional strategies, as needed, to
nological change, especially in advanced economies. help raise inflation expectations and lower the real
Heightened restrictions on trade and capital flows costs of borrowing for households and firms. But
would impose broad economic costs, harming consum- accommodative monetary policy alone cannot lift
ers and producers alike, with the potential to leave all demand sufficiently and can potentially generate
countries worse off if protectionism begets retaliation. undesirable side effects (as discussed in the Octo-
Instead, the challenge will be to preserve the gains ber2016 GFSR). Fiscal supportcalibrated to
from cross-border economic integration while ramping the amount of space available and oriented toward
up domestic policy efforts to ensure that those gains policies that protect the vulnerable and lift medi-
are shared more broadly. Well-targeted initiatives can um-term growth prospectsalso remains essential
help workers adversely affected by structural transfor- for generating momentum and reducing the risk
mations find jobs in expanding sectors. Short-term that a prolonged shortfall in demand erodes supply
measures include active labor market policies combined capacity or unmoors medium-term inflation expec-
with social safety nets to smooth the loss of income. tations. In cases in which postponing fiscal adjust-
In the longer term, adequate education, skill building ment is either not possible or too risky, its speed and
and retraining, and policies to facilitate reallocation, composition should be configured to minimize the
such as housing and credit access, will be needed to drag on output. And support for demand must be
attain inclusive and sustainable growth in a context of accompanied by efforts to address corporate debt
continued rapid technological progress and economic overhangs and decisively repair bank balance sheets
integration. Such efforts require public resources, so (addressing a legacy of nonperforming loans and
progressive taxes and well-targeted transfer policies will strengthening operational efficiency, as discussed
also have an increasingly important role to play (see in the October2016 GFSR and the October2016
Chapter1 of the April2017 Fiscal Monitor). Fiscal Monitor).
In those advanced economies where output is
close to or above potential, well-anchored inflation
PoliciesAdvanced Economies expectations should allow for monetary policy to be
The recent uptick in momentum notwithstand- normalized gradually. Desirable changes to the fiscal
ing, advanced economies as a group continue to face policy stance depend on country circumstances,
modest current and prospective economic growth, including public debt dynamics. Fiscal policy should
characterized by sluggish productivity dynamics, low aim at strengthening safety nets (including to help
investment, and, in some cases, persistently low core with the integration of refugees in some cases) and
inflation. These features reflect, to a large extent, increasing longer-term potential output.
the interplay between subdued demand, dimin- Structural reforms are needed across advanced econ-
ished growth expectations, and aging populations. omies to enhance productivity, investment, and labor
A cross-cutting theme for economies therefore is the supply. Specific priorities vary across countries and
need to lift potential output. At the same time, the include measures to boost labor force participation
cyclical conditions of individual economies continue to through reforms to labor taxes and social benefits,
diverge. In Germany, the United States, and a number well-targeted infrastructure investments, corpo-
of other advanced economies in Europe and Asia, out- rate income tax reform and tax incentives to boost
put is either close to or above potential. By contrast, research and development, facilitation of improve-
output remains significantly below potential in France, ments in human capital by investing in education
Italy, Portugal, Spain, and especially in Greece. These and health care, and elimination of product and labor
heterogeneous cyclical positions call for differentiated market distortions to boost private sector dynamism.9
macroeconomic policy stances.
In those advanced economies where output gaps 9As discussed in Chapter3 of the April 2016 WEO, removing
barriers to entry into product and service markets can also raise
are still negative and wage pressures and inflation
near-term activity, but labor market reforms may require supportive
expectations for the next few years are muted, the macroeconomic policies to lessen possible dampening effects on
risk of persistently low inflation (or deflation, in near-term growth and inflation when the economy is weak.

30 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

As discussed earlier, resisting a retreat from global oo A critical priority for boosting growth and
economic integration also needs to be a part of the limiting downside risks in the euro area is to
agenda for strengthening growth. accelerate banks balance sheet repair and the
resolution of nonperforming loans, including
through a combination of greater supervi-
Country-Specific Priorities sory encouragement, insolvency reform, and
In the United States, the economy regained the development of distressed debt markets.
momentum in the second half of2016, with Completion of the banking union, including
strong job creation, solid growth in disposable by introducing a common deposit insurance
income, and robust consumer spending. The econ- program with a common effective fiscal back-
omy is close to full employment, but core personal stop, also remains critical. These actions would
consumption expenditure inflation is only slowly strengthen the transmission of monetary policy
inching up toward the Federal Reserves 2percent accommodation to the real economy and facil-
target, suggesting that monetary policy can con- itate the consolidation and restructuring of the
tinue to tighten at a gradual, data-dependent pace. banking sector.
A credible deficit- and debt-reduction strategy is oo Greater centralized investment in public infra-
needed to open up space for policies to improve structure will help countries with continued
social outcomes and lift productive capacity while demand shortfalls that lack fiscal space or need
putting the debt ratio firmly on a downward path. to consolidate because of high and rising debt
The fiscal stance should remain neutral this year, burdens. Where consolidation is required, it
and fiscal consolidation could start in2018. Struc- should be undertaken in a gradual and growth-
tural and fiscal policies should seek to upgrade the friendly manner. In countries with fiscal space,
public infrastructure, boost labor force partici- such as Germany, fiscal policy should be geared
pation, and enhance human capital. Skill-based toward bolstering productive capacity as well as
immigration reform, job training, paid family demand. In turn, this would help reduce their
leave, and child care assistance are key priorities in current account surpluses, support intra-euro-
this regard. Complementing the fiscal consolida- area rebalancing, and generate positive demand
tion plan, a comprehensive reform of the business spillovers for others.
tax code geared toward simplification and fewer oo Synergies between structural reforms and demand
exemptions would encourage job creation and management policies should be exploited to
investment, ultimately enhancing fiscal sustain- the extent possible. Where demand is still weak
ability. Any changes to financial regulation should but fiscal space is lacking, budget-neutral fiscal
strive to avoid a buildup of financial stability support can enhance the effects ofpublic admin-
risks. While potential changes to the existing istration or labor market reforms. Product and
framework could lower existing regulatory burdens labor market reforms are needed to encourage
for small and community banks, there is a need to business dynamism, raise labor force participation
strengthen the regulation and supervision of non- rates, and address labor market duality. Reforms
bank financial institutions, particularly as financial to complete the single market would help boost
activity continues to shift to these less-regulated productive capacity.
entities. oo Refugee integration into the workforce should
In theeuro area, with inflation expectations still be facilitated through swift processing of asylum
below target and several economies still operating applications, language training and assistance in
significantly below capacity, the European Central job search, better recognition of migrants skills
Bank should maintain its current accommodative through credential systems, and support for
stance. Additional easing may be needed if core migrant entrepreneurship.
inflation fails to pick up. Critically, monetary policy In Japan, growth was stronger than expected
will be more effective if supported by measures to in2016. Inflation appears to be bottoming out,
clean up balance sheets, strengthen the financial helped by higher fresh food prices and fading
sector, use fiscal space where available, and accelerate downward pressure from the earlier yen apprecia-
structural reforms. Specifically, tion. Net exports were the main driver of growth

International Monetary Fund | April 2017 31


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

in2016, with fiscal policy also supportive of the PoliciesEmerging Market and Developing Economies
positive economic momentum. Despite a tighten-
Emerging market and developing economies have
ing labor market, wage demands are not stronger
operated in a complicated external environment
than in the past few years and thus are unlikely to
in recent years, characterized by generally sluggish
kindle much-needed positive wage-price dynam-
demand from advanced economies, a sharp correction
ics. The Bank of Japans monetary easing through
in commodity prices followed by a recovery since
asset purchases and negative deposit rates, and the
the first quarter of2016 (albeit to levels well below
introduction of quantitative and qualitative easing
previous peaks), and spells of relatively benign finan-
with yield curve control, have been critical to
cial conditions interspersed with recurrent spikes in
preventing another bout of deflation, but the low
market volatility.
and declining neutral real rate and low nominal
As discussed in Chapter2, some aspects of the exter-
rates constrain monetary policy effectiveness.
nal environment are likely to be less supportive going
Continued efforts to raise inflation expectations
forward than in the past, while others remain uncer-
to further lower real rates thus remain neces-
tain. Weaker potential output growth across advanced
sary, including through a further upgrade to the
economies, together with a possible increase in trade
Bank of Japans communication framework. To
barriers in some, could translate into generally subdued
attain a durable increase in inflation and growth,
demand growth for emerging market and developing
a comprehensive policy approach that enhances
economies. An additional element that may weigh on
monetary accommodation with a supportive
commodity exporters in particular is Chinas neces-
fiscal stance and reforms to labor market policies
sary transition to slower, more sustainable, consump-
is needed. Elements of such a package would
tion- and services-based growth. External financial
include reforms to diminish labor market duality
and increase labor force participation by women conditions facing emerging market and developing
and older workers while admitting more foreign economies are likely to remain uncertain. A gradual,
workers, lowering entry barriers in retail trade generalized tightening is expected as U.S.monetary
and services, improving the provision of capital policy normalizes, but this tightening will likely
for new ventures, and supporting stronger cor- be accompanied by a continued search for yield in
porate governance to discourage companies from emerging market investment opportunities as long as
accumulating excess cash reserves. A credible fiscal returns remain modest in a low-growth environment
consolidation over the medium termbased on a in advanced economies. A third, important element of
gradual preannounced increase in the consump- the external environmentthe terms of trademay
tion tax, social security reform, and a broadening improve for a subset of emerging market and develop-
of the tax baseremains critical. ing economies with the bottoming out of commod-
In the United Kingdom, a principal challenge will be ity prices, but the outlook for export prices remains
to successfully navigate the exit from the Euro- subdued compared with the past. By contrast, for
pean Union and negotiate the new arrangements importers, the windfall gains from lower commodity
for economic relations with the European Union prices will diminish.
and other trading partners. The adverse impact on Although this combination of factors may provide a
medium-term output would be lower if the new weaker growth impulse for emerging market and devel-
arrangements limit new economic barriers. The oping economies than had been the case for long inter-
current accommodative monetary policy stance is vals since2000, the analysis in Chapter2 points to the
appropriate because growth is expected to slow and role of domestic policies that can help these countries
domestic cost pressures to remain contained. On secure growth prospects in an increasingly complicated
the fiscal front, the envisioned path of steady but external environment. Country-specific priorities will
gradual fiscal consolidation and the moderate relax- necessarily differ, based on levels of development and
ation of the targets strike an appropriate balance individual circumstances. But, in general, a policy
between providing an anchor for medium-term orientation that protects trade integration, permits
objectives and allowing room for short-term exchange rate flexibility, and ensures that vulnerabilities
maneuvering amid elevated uncertainty about the stemming from high external imbalances and public
economic outlook. debt are contained is likely to help emerging market

32 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

and developing economies extract the most out of a capital markets by reining in shadow products and
weaker external growth impulse and help sustain con- strengthening the supervisory framework.
vergence to higher levels of income. Indias economy has grown at a strong pace in
With ever-present risks of global financial volatility, recent years owing to the implementation of critical
sharp currency movements, and capital flow reversals, structural reforms, favorable terms of trade, and
it will be important for economies with large and ris- lower external vulnerabilities. Beyond the immediate
ing nonfinancial debt, unhedged foreign liabilities, or challenge of replacing currency in circulation follow-
heavy reliance on short-term borrowing to adopt stron- ing the November2016 currency exchange initia-
ger risk management practices and contain balance tive, policy actions should focus on reducing labor
sheet mismatches. Decisive actions toward improving and product market rigidities to ease firm entry and
domestic governance, institutions, and the business exit, expand the manufacturing base, and gainfully
environment can help reduce country risk perceptions employ the abundant pool of labor. Policy actions
and thereby act as a powerful countervailing force should also consolidate the disinflation under way
against the expected tightening in global financial since the collapse in commodity prices through agri-
conditions. cultural sector reforms and infrastructure enhance-
ments to ease supply bottlenecks; boost financial
Country-Specific Priorities stability through full recognition of nonperforming
The near-term outlook for China has strengthened loans and raising public sector banks capital buffers;
in recent months, with policy support expected to and secure the public finances through continued
maintain steady growth in the run-up to the lead- reduction of poorly targeted subsidies and struc-
ership transition in late2017. The complex process tural tax reforms, including implementation of the
of rebalancing is advancing on multiple fronts, recently approved nationwide goods and services tax.
rotating activity away from industry to services In Brazil, the pace of contraction has diminished,
and reorienting demand from exports and invest- but investment and output had yet to bottom out
ment to consumption. Progress lags along one at the end of2016, while fiscal crises in some states
critical dimension, however: the continued heavy continue to deepen. Inflation has continued to
reliance on credit to support activity compounds surprise on the downside, allowing for prospects
the considerable risks that have accrued in recent of faster monetary easing. Growth is projected to
years from the rapid buildup of corporate and local recover gradually and remain moderate. Against this
government debt, funded through an increasingly backdrop, Brazils macroeconomic prospects hinge
opaque financial system. With vulnerabilities con- on the implementation of ambitious structural
tinuing to accumulate, the macro policy mix needs economic and fiscal reforms. To underpin medi-
to focus on containing the problems at their source um-term fiscal consolidation, the focus should be on
by accepting slower and more sustainable growth reforms that address unsustainable expenditure man-
outcomes; reducing the pace of credit growth dates, including in the social security system, but
closer to that of nominal GDP; raising policy rates; there is also merit in undertaking actions to achieve
and cutting off-budget public sector investment a more front-loaded reduction in the fiscal deficit.
while increasing on-budget allocations for social Reforms to boost potential growth are needed not
assistance, health expenditure, unemployment ben- only to restore and improve living standards after
efits, and restructuring funds. Together with these the deep recession, but also to facilitate the fiscal
measures, structural reform priorities to improve consolidation. Imperatives for lifting investment
efficiency include deregulating sectors dominated and productivity include addressing long-standing
by state-owned enterprises to facilitate entry; infrastructure bottlenecks, simplifying the tax code,
decisively restructuring those that are unprofitable and reducing barriers to trade.
and replenishing bank buffers, as needed, once the In South Africa, following the decline in commodity
losses are appropriately accounted for; and accel- prices and amid perceptions of weakening gov-
erating household residency reforms to facilitate ernance and rising policy uncertainty, economic
more efficient matching of labor market vacan- growth gradually softened and came to a near
cies with job seekers. An intensified focus is also standstill in2016. The projected near-term recovery
needed on containing financial risks in domestic remains insufficient to keep pace with population

International Monetary Fund | April 2017 33


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

growth. In the baseline scenario of a moderate since mid-2014 has been a major setback for commod-
resumption of growth this year, monetary policy ity-exporting low-income developing countries, where
can remain on hold unless inflation expectations policies have been slow to adjust to the large income
rise or external financing becomes challenging. loss. Three years after commodity prices fell from their
Envisioned fiscal measures appropriately strike a peak, fiscal deficits remain wide, external positions are
balance between maintaining debt sustainability and weaker, debt is rising, and depreciated currencies
safeguarding the fragile economic recovery. If growth although they help cushion the adverse terms-of-trade
prospects were to falter, additional measuressuch shockhave, in some cases, led to higher inflation and
as slower public sector wage increases and a moder- pushed up external debt. Although most commodity
ate increase in consumption taxeswould be needed exporters are set to record positive growth in2017,
to stabilize the debt ratio. With monetary and fiscal their medium-term growth prospects are subdued. By
policies constrained by the need to keep inflation contrast, low-income countries with more diversified
and the rising public debt in check, reforms in prod- export bases have recorded relatively strong growth
uct and labor markets that allow greater entry by and are expected to continue to grow at a healthy rate,
new firms and reduce impediments to job creation with the benefit of lower oil bills outweighing the drop
are urgently needed to strengthen confidence, invest- in remittances and weaker demand from commodity
ment, and growth. Such reforms would lower the exporters. Robust growth, however, has not always
cost of crucial inputs for businesses and of services translated into improved fiscal and external current
for workerssuch as in electric power generation, account positions, reflecting limited progress in adopt-
telecommunications, and transportation. ing countercyclical policies, but also public investment
In Russia, the economy is projected to continue to support activity. Many low-income developing
its nascent recovery in2017. Inflation is expected economies have been also hit by idiosyncratic shocks,
to fall further toward the central banks inflation such as conflicts and security disruptions (Afghanistan,
target over the course of2017, providing the Chad, South Sudan, Yemen, parts of Nigeria), and
conditions for the central bank to gradually resume natural disasters (Haiti, Ethiopia, Malawi). Some still
monetary policy easing, with due attention to endure the persistent growth-dampening effect of the
external risks and the need to build the credibility Ebola outbreak (Guinea, Liberia, Sierra Leone).
of the newly introduced inflation-targeting regime. With such divergent prospects, the appropriate
The reestablishment of a three-year fiscal frame- courses of action in the near term differ across low-
work will help facilitate the consolidation required income developing countries.
by lower oil revenues. However, to sustain the sig- Commodity exporters need to continue and, in some
nificant adjustment, better-targeted and more per- cases, accelerate the process of adjusting to structur-
manent reforms to the pension system, subsidies, ally lower commodity prices based on comprehensive
and tax exemptions are needed. The adoption of a and internally consistent sets of policies. Fiscal policy
revised fiscal rule would help reduce policy uncer- needs to be better calibrated to contain debt accu-
tainty and cement the fiscal adjustment. Improve- mulation while protecting outlays that are key to
ments to financial supervision and regulation as growth prospects, such as priority capital expenditures
well as a stronger resolution framework are needed and social spending. In many countries, improving
to make the financial system more resilient and domestic revenue mobilization and continued ratio-
improve credit allocation. Raising medium-term nalization of spending needs, along with concessional
growth prospects will necessitate a diversification of financing, are necessary to underpin successful
the economy, accelerated institutional reforms, and adjustment processes. Monetary tightening may also
an improved business climate. be needed in a number of countries, either to defend
pegged exchange rates or to contain inflation resulting
from the side effects of exchange rate flexibility and
PoliciesLow-Income Developing Countries depreciation. Enhanced financial sector regulation
Among low-income economies, the economic pros- and supervision will be required to manage foreign
pects of commodity-exporting countries continue to currency exposures in balance sheets.
diverge from those with more diversified export bases. Policy priorities for diversified low-income devel-
The sharp realignment of global commodity prices oping countries vary, given the diversity of coun-

34 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

try circumstances. However, an overarching goal aggregate demand, in particular investment). Rolling
for these economies should be to strike a better back temporary barriers to trade introduced since the
balance between spending for developmental and global financial crisis and further reducing trade costs
social needs and improving public debt sustainabil- would support the nascent recovery in trade, revving
ity, rebuilding fiscal positions and foreign reserves up an important engine of global productivity growth.
holdings while growth is strong to enhance resil- To that end, it is critical to preserve the multilateral
ience against potential future shocks. Stronger debt rules-based trading system and press ahead with an
management will also help those exposed to global ambitious trade agenda at the global level. Addressing
financial markets better cope with volatility in capi- tariff barriers in sectors where they remain high, such
tal inflows. as agriculture, and implementing commitments under
the Trade Facilitation Agreement, which went into
Near-term challenges notwithstanding, low-income effect in February2017, can significantly reduce trade
developing countries should not lose sight of their costs in traditional areas. Advancing trade reforms in
longer-term objectives reflected in the United Nations services and in frontier areas, such as digital trade,
Sustainable Development Goals. In that context, many and improving cooperation in investment policies
of the policies that would set these economies on a have the potential to make large contributions to
sustainable macroeconomic trajectory in the near term cross-border flows and global growth. However, as dis-
will also help achieve sustained growth and resilience cussed, further trade liberalization should go hand in
in the long term, a precondition for convergence and hand with domestic policies to support individuals and
attaining the development goals. In particular, efforts communities that may be at risk of being left behind.
to create fiscal space by enhancing domestic resource
mobilization and improving the efficiency of govern- Cooperation on International Taxation Issues
ment spending and debt management, steps to reorient As increased capital mobility across borders has
fiscal spending to protect the vulnerable and address fueled international tax competition, governments have
infrastructure gaps, and measures to improve financial found it more challenging to finance their budgets
sector resilience and deepen financial inclusion, will without imposing higher taxes on labor income or
help achieve macroeconomic stabilization, overall eco- implementing regressive consumption taxes. Policy-
nomic resilience, and durable and inclusive growth. makers can achieve equitable tax systems (that prevent
an increasing share of after-tax income from accruing
to owners of capital) in the future only if their national
Multilateral Policies efforts to tackle tax evasion and avoidance are backed
To put the pickup in global growth on a firmer up with multilateral cooperation on these fronts. If
footing and sustain improvements in global living firms continue to face pronounced incentives to shift
standards over the medium term, supporting national profits across borders for tax planning and avoidance,
policy efforts with continued multilateral cooperation popular support for trade and investment flows may
in a number of areas will be vital. Such cooperation wane further. Box1.1 of the April2017 Fiscal Monitor
is particularly needed for preserving an open, rules- discusses the implications of proposals for corporate
based multilateral trading system, maintaining global tax reform in the United States that aim to reduce the
financial stability, cracking down on tax evasion and incentives for profit shifting by U.S.firms.
limiting tax avoidance, and addressing longer-term
challenges facing the global economy. Maintaining Global Financial Stability
Efforts to strengthen the resilience of the global
Maintaining a Rules-Based, Open Multilateral financial system must continue, including by recapital-
Trade System, with Broadly Shared Gains izing institutions and cleaning up balance sheets where
As documented in Chapter2 of the October2016 necessary, ensuring effective national and international
WEO, the slower pace of new trade reforms and an banking resolution frameworks, and addressing emerg-
uptick in protectionist measures have contributed to ing risks from nonbank intermediaries. A stronger
the remarkable slowdown in global trade in recent global safety net can protect economies with robust
years (although their estimated contribution to the fundamentals that may nevertheless be vulnerable to
trade slowdown is smaller than that of the weakness in cross-border contagion and spillovers in the context of

International Monetary Fund | April 2017 35


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

elevated downside risks to the global outlook. Closer such as meeting the2015 Sustainable Development
cross-border regulatory cooperation is also required to Goals, providing financial support to vulnerable
limit the withdrawal of correspondent banking rela- economies and fragile states, mitigating and adapting
tionships that provide low-income countries access to to climate change, and preventing the spread of global
the international payments system. epidemics. Risks stemming from noneconomic factors
with cross-border repercussions, such as the ongoing
Longer-Term Challenges refugee crisis, further underscore the case for insti-
Finally, multilateral cooperation is also indispensable tuting globally funded vehicles to help the exposed
for addressing important longer-term global challenges, economies cope with the strains.

36 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Scenario Box 1. Permanent U.S.Fiscal Expansions


This box uses the IMFs G20 Model (G20MOD) Scenario Figure 1. Fiscal Stimulus in the United
to illustrate the impact of two alternative U.S.fiscal States
expansions relative to a baseline scenario with no (Percent deviation from case with no change in U.S. scal
change in U.S.fiscal policy. Both expansions use iden- policy)
tical instruments: reduced labor income taxes, reduced Highly productive scal stimulus
corporate income taxes, and increased infrastructure Less productive scal stimulus
spending. However, differences in the efficacy of the 1.5 1. U.S. Real GDP 2. U.S. Real 4
infrastructure spending and labor tax cuts, and the Effective
Exchange Rate 3
way the public debt is eventually stabilized, lead to dif-
1.0
ferent macroeconomic outcomes, as discussed below. 2

Assumptions 1
0.5
In both cases, the fiscal expansion is debt financed 0
for the first four years (201821), and monetary policy
0.0 1
in the United Statesresponds endogenously to the 2017 19 21 23 25 2017 19 21 23 25
change in demand. It is assumed that monetary policy
1.5 3. U.S. Policy Rate 4. U.S. Long-Term 1.0
in both Japan and the euro area would accommodate
Interest Rate
any positive increase in demand, but would have 0.8
1.0
no conventional policy space to respond to negative
0.6
developments. Households and firms are assumed to 0.5
learn gradually about the changes in fiscal policy and 0.4
their permanent nature. In both cases, after four years 0.0
0.2
(2022) the fiscal authority needs to adjust policy to
stabilize the debt-to-GDP ratio. 0.5 0.0
2017 19 21 23 25 2017 19 21 23 25
In the first case, the fiscal expansion is highly pro-
ductive (blue lines in Scenario Figure1)the increase 6. U.S. Government
2.5 5. U.S. Government 8
in public infrastructure spending is assumed to have Decit-to-GDP Debt-to-GDP
7
a strong positive impact on output, and the cuts in 2.0 Ratio Ratio
6
labor income taxes are assumed to be broad based. In 5
1.5
the second case, the fiscal expansion is less productive 4
(red lines in Scenario Figure1)the infrastructure 1.0 3
spending is assumed to be unproductive, and the tax 0.5
2
cuts are assumed to go mostly to wealthier households 1
with a very low marginal propensity to spend the 0.0 0
2017 19 21 23 25 2017 19 21 23 25
additional income on consumption. In the second
case, it is also assumed that financial markets deliver a 0.50 7. Advanced 8. Emerging 0.12
faster normalization in the U.S.term premium than in Economies Real Markets Real
0.25 GDP1 GDP 0.08
the case of no change in fiscal policy (25 basis points
in2018 and an additional 25 basis points in2019). 0.04
0.00
This faster normalization in the U.S.term premium is 0.00
transmitted into the term premium worldwide, consis- 0.25
0.04
tent with the empirical correlations in the IMFs2014 0.50 0.08
Spillover Report.
0.75 0.12
Once policy needs to adjust to stabilize debt, in the 2017 19 21 23 25 2017 19 21 23 25
highly productive case, the fiscal authority partially
cuts back the initial increase in infrastructure spending Source: IMF staff estimates.
1Excluding the United States.
to simply maintain the new higher level of the public
capital stock (Scenario Table1). Half of the remaining
required adjustment comes from reducing tax expen-

International Monetary Fund | April 2017 37


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Scenario Box 1. (continued)


Scenario Table 1. The Impact of Fiscal Measures on the Deficit
(Percent of no-change-in-fiscal-measures case GDP)
2017 2018 2019 2020 2021 2022
Highly Productive Fiscal Measures
Capital Income Taxes 0 0.375 0.750 0.750 0.750 0.750
General Labor Income Taxes 0 0.375 0.750 0.750 0.750 0.330
Productive Infrastructure 0 0.250 0.500 0.500 0.500 0.150
Spending
Tax, Expenditures 0 0 0 0 0 0.320
Total Change in the Deficit 0 1.000 2.000 2.000 2.000 0.200
Less Productive Fiscal Measures
Capital Income Taxes 0 0.375 0.750 0.750 0.750 0.750
Labor Income Taxes for the 0 0.375 0.750 0.750 0.750 0
Wealthy
General Labor Income Taxes 0 0 0 0 0 0.530
Unproductive Infrastructure 0 0.250 0.500 0.500 0.500 0
Spending
Total Change in the Deficit 0 1.000 2.000 2.000 2.000 0.220
Source: IMF staff assumptions for the scenario analysis.

ditures, and the other half comes from higher labor Mexico, which have strong trade links with the United
income taxes. In the less productive case, the increase States, and the assumption that monetary policy in the
in unproductive infrastructure spending is completely euro area and Japan does not tighten in the face of the
unwound and the tax cuts to the wealthy are com- increase in external demand. In the highly productive
pletely reversed. The remaining adjustment required case, the spillovers to emerging market economies are
to stabilize debt comes in the form of higher general also positive in the short term, but modest. Under
labor income taxes. In both cases, these adjustments the less productive fiscal expansion, the short-term
stabilize the public-debt-to-GDP ratio roughly 5per- spillovers become negative both for other advanced
centage points above its prestimulus level. economies and for emerging market economies for
two reasons. First, with lower U.S.demand in the less
Results productive case, the direct trade spillovers are smaller.
When the fiscal measures are highly productive, Second, the faster normalization of term premiums
U.S.GDP rises notably, peaking at 1percent above worldwide tightens financial conditions, which is
the no-policy-change case in2021. When fiscal particularly onerous for advanced economies that have
measures are less productive, U.S.GDP rises by limited or no conventional monetary policy space with
roughly half that amount by2021. With a smaller which to respond.
increase in U.S.output in the less productive case, Once U.S. fiscal policy needs to be tightened to
the deficit and debt as a share of GDP both rise by stabilize public debt, the withdrawal of stimulus tem-
more. In both cases, U.S.monetary policy tightens in porarily lowers U.S.GDP relative to its level in2021
response to higher demand and inflation, and higher in both cases. However, because capital income taxes
real U.S.interest rates lead to an appreciation of the are assumed to be permanently lower in both cases,
U.S.dollar. In the less productive case, the U.S.policy thereby raising the returns to private capital, real GDP
rate tightens by less, but the faster normalization of subsequently recovers as firms continue investing to
the term premium and thus higher long-term interest raise the private capital stock to its higher desired level.
rates leads to more upward pressure on the currency In the highly productive fiscal expansion, this effect
in the near term. With regard to spillovers to the rest is reinforced by the permanently higher level of the
of the world, in the highly productive case, other public capital stock, which raises private productivity,
advanced economies benefit the most in the short further increasing the return to private capital. With
term, with GDP roughly 0.2percent higher. This U.S.output permanently higher in the long term and
outcome reflects inclusion in this group of Canada and with no change in the relative price of U.S.tradable

38 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Scenario Box 1. (continued)


and nontradable goods, the U.S.dollar would need to investment in infrastructure) rather than simply from
depreciate to maintain external stability. the initial fiscal expansion. Simulations show that a
In the long term, the spillovers to all economies similarly growth-friendly fiscal policy implemented in
outside the United States are small, but negative, a deficit-neutral way (financed by a reduction in tax
because the permanently higher level of U.S.public expenditures and lower government consumption)
debt raises global real interest rates. The increase in would lead to a higher long-term level of GDP. In
global interest rates in turn permanently raises the the short term, GDP would be lower compared with
cost of capital, which more than offsets the increase the deficit-financed expansion, with policy rates and
in the return to private capital coming from higher long-term interest rates correspondingly lower. The
U.S.demand. dollar would appreciate by less, but there would be
It is important to note that the positive effects no subsequent need for additional tightening of fiscal
on U.S. GDP over the medium and long term arise policy, and with lower medium-term debt long-term
from the beneficial supply-side effects of some tax interest rates would be a bit lower. Both factors sup-
and expenditure changes (notably the reduction in port medium-term GDP, the first on a temporary basis
corporate income tax rates and the increase in public and the second on a permanent basis.

International Monetary Fund | April 2017 39


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 1.1. Conflict, Growth, and Migration


Conflict has been on the rise since the early 2000s. Figure 1.1.1. Conict-Related Fatalities and
The incidence of conflict, defined as the number of Number of Countries Affected by Conict
countries that have had at least 100 conflict-related
deaths per 1 million people, has risen in recent years 200 1. 19892016 16
from low levels in the early 2000s (Figure1.1.1, panel 180 Fatalities (thousands) 14
1).1 Although the total annual number of conflict- 160 Number of countries
12
140 (right scale)
related deaths is still relatively low from a historical
120 10
perspective, its increase in recent years has been quite
100 8
sharp, reflecting the very deadly conflicts in Afghan-
80 6
istan, Iraq, and Syria (Figure1.1.1, panel 2). Over 60
time, the nature of conflict has changed: there was 4
40
more interstate conflict between World War II and the 20 2
1990s, and there has been more internal civil war since 0 0
1989 95 2000 05 10 16
the 1990s (Blattman and Miguel 2010). The location
of conflict has also shifted, from sub-Saharan Africa in
the 1990s to the broader Middle East region, espe- 700 2. Conict-Related Fatalities since World War II
(Thousands)
cially since 2010.2 600 PRIO (state based)
Conflict leads not only to immeasurable humani- IISS (all conict)
500 UCDP (all conict)
tarian suffering, but also to significant economic losses UCDP (state based)
that can persist for years. Empirical research points to 400
conflict as one of the factors that can hold back eco- 300
nomic development (Rodrik 1999; Besley and Persson 200
2008). It can also ignite large refugee flows and may
100
affect the economies of countries near and far for an
extended period. 0
1946 55 65 75 85 95 2005 16
The tragic rise in conflict has also weighed on
global GDP growth in recent years, given the increas-
Sources: International Institute for Strategic Studies (IISS)
ing number of economies experiencing strife, the Armed Conict database; Peace Research Institute Oslo
severe effect of some of these episodes on economic (PRIO) Battle Deaths data set v. 3.1; UN (2016); Uppsala
activity, and the considerable size of some of the Conict Data Program (UCDP) Georeferenced Event data set
v. 5.0 and Battle-Related Deaths data set v. 5.0; and IMF
affected economies. The countries currently involved staff calculations.
in conflict accounted for 1.02.5 percent of GDP Note: In panel 1, a country is considered in conict if in any
in 2010, depending on the precise threshold used year 100 people or more are killed per 1 million population.
In panel 2, state-based conicts are those in which at least
to define the incidence of conflict (Figure1.1.2, one of the conict parties is a state. All conicts can also
panel1).3 In some countries, the difference between include one-sided violence (for example, violence against
civilians perpetrated by rebel groups) and nonstate conict
(for example, organized rebel or communal group ght).
The authors of this box are Natalija Novta and Evgenia Fatalities that are not attributed to a specic country are
Pugacheva. excluded. The UCDP all-conict estimate of fatalities
1The choice of different thresholds does not change the thrust excludes the Rwanda outlier in 1994 (501,958 dead).
of the findings. In Figure1.1.1, a country is considered in
conflict in a given year if there are more than 100 conflict-related
deaths per 1 million people in the country. In many previous
conflict studies, conflict incidence is defined as an absolute
number of conflict-related deaths; however, this approach makes
it mechanically harder for smaller countries to pass the threshold, 200205, 200609, and 201015. If calculated separately each
even if they are experiencing significant conflict (see Mueller year, the share of global GDP in conflict-affected countries
2016). mechanically declines during the period of conflict because
2Middle East, including Afghanistan, Israel, North Africa, the GDP of conflict-affected countries typically drops during
Palestine, and Pakistan. conflict (Mueller 2013; Cerra and Saxena 2008). To limit this
3Three definitions of conflict are used, based on severityif mechanical effect, in panel 1 of Figure1.1.2, the percentage of
there are at least 50, 100, or 150 conflict-related deaths per global GDP that a country represents is recorded in the first
million people in the country and for three different periods: year of the period.

40 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Box 1.1 (continued)


Figure 1.1.2. Global GDP Shares of Conict- which lowers human capital both individually and in
Affected Countries and Impact of Conict on the aggregate (see Blattman and Miguel 2010; Justino
Growth 2007 and 2009). Furthermore, conflict typically leads
to lower investment (as investors lose confidence),
3.5 1. Global GDP Shares of Conict-Affected Countries changes in household saving and consumption (Voors
at Different Levels of Conict Intensity and others 2012), and capital flight (Collier, Hoeffler,
3.0 (Percent) and Pattillo 2004).
2.5 200205
200609 During 19892016, outbreaks of conflict are
2.0 201016 estimated to have reduced output per capita by a
1.5 cumulative 18 percent over the subsequent 10 years,
on average (Figure1.1.3, panel 1).4 Restricting the
1.0
analysis to state-based conflicts and using data for a
0.5 longer period point to losses of about 5 percent after
0.0 10 years (Figure1.1.3, panel 2).5 The econometric
50 killed per 100 killed per 150 killed per
million million million
finding of a persistent loss of output holds true if the
conflict variable is defined as the share of lost lives in
260 2. Preconict GDP Forecast versus Actual
the population or with a dummy variable indicating
(Index, year before conict = 100) conflict incidence in a given year. In the first case
220 Preconict GDP path (forecast) (conflict fatalities), the cumulative loss in output after
Postconict GDP path 10 years is about 5 percent, and in the second case
180 Postconict GDP path
(forecast to 2022)
(annual conflict incidence), the cumulative loss is
140 about 7 percent (not shown in figures). These losses
100 build up as conflict evolves.6
Iraq Libya
60 Central African Ukraine
Yemen
Republic
20
2012 2012 2010 2013 2014
4The local projection method of Jorda (2005) and Teulings

and Zubanov (2014) is used to estimate the impact of conflict


Sources: UN (2016); Uppsala Conict Data Program on GDP over the subsequent 10-year horizon. The following
Georeferenced Event data set v. 5.0 and Battle-Related type of equation is estimated:
Deaths data set v. 5.0; and IMF staff calculations.
Note: In panel 1, GDP shares are based on the rst year yit+h yit1 = h1 cit + h2cit1 + h1
j=1 3 cit+hj + 1yit1
hj h
within the bin (using 2011 data for South Sudan and
rescaling all 2010 numbers). In panel 2, conict onset is the + hi + rth + hit, h = 0,,10,
rst year of conict in which the number of deaths exceeds
100 per 1 million population (after at least four consecutive in which yit is log GDP per capita (or log number of refugees,
years without passing that threshold). for migration), cit are conflict variables (the onset of conflict,
percentage of population killed, or conflict incidence), ih are
country fixed effects, rth are time fixed effects, and h is the hori-
zon. The reported findings are generally robust to the addition of
various controls.
5The longest series for conflict-related deaths, which is

preconflict GDP forecasts and actual GDP during compiled by the Peace Research Institute Oslo, starts in 1946,
conflict is dramatic (Figure1.1.2, panel 2). but covers only state-based conflicts. The Uppsala Conflict Data
Program provides data on fatalities from all types of conflict
Economic Recovery from Conflict Is Slow (including non-state-based actors, one-sided violence against
civilians, and so on) starting in 1989.
The onset of conflict can hurt GDP per capita 6The econometric estimates would be biased if low growth

growth in many ways, such as by directly reducing the caused the conflict rather than resulting from it. However, the
workforce or hampering labor productivity. The neg- results do not change much if the World Economic Outlook
ative effects of conflict can be large over the medium (WEO) GDP per capita forecast for the current year, made the
year before the conflict, is controlled for in the regressions (based
and long term if peoples health is permanently dam-
on the level of GDP per capita projections from different vin-
aged, they leave the country as refugees or economic tages of the WEO). Overall, the results are very similar to those
migrants, or they are prevented from attending school, from regressions that do not control for GDP forecasts.

International Monetary Fund | April 2017 41


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 1.1 (continued)


Figure 1.1.3. Impact of Conict Onset Emigration from Conflict-Ridden Areas Remains
(Percent; years on x-axis) High for a Long Time
Refugee populations tend to grow for many years
Estimate 95 percent condence interval after conflict begins, potentially placing a significant
burden on other economies (Figure1.1.3, panel 3).
Economic Output after Conict Onset After a conflict erupts, neighboring economies are
10 1. All Conicts 2. State-Based 10 typically the first to receive a large influx of refugees,
5 Conicts 5 but if these countries do not offer much economic
0 0 opportunity, refugees may eventually prefer to move
5 5 to advanced economies. Panel 4 of Figure1.1.3 shows
10 10 that refugee populations in advanced economies
15 15 remain on the rise 10 years after the beginning of a
20 20 conflict.
25 25
30 30
35 35
1 2 4 6 8 10 1 2 4 6 8 10

Refugee Stocks after Conict Onset


350 3. Overall 4. In Advanced 160
Economies
300 120
250 80
200
40
150
0
100
50 40
0 80
1 2 4 6 8 10 1 2 4 6 8 10

Sources: UN (2016); UNHCR (2016); Uppsala Conict Data


Program Georeferenced Event data set v. 5.0 and Battle-
Related Deaths data set v. 5.0; and IMF staff calculations.
Note: t = 0 is the year of the shock. Conict onset is the rst
year of conict in which the number of deaths exceeds 100
per 1 million population (after at least four consecutive years
without passing that threshold).

42 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Box1.2. Tackling Measurement Challenges of Irish Economic Activity


On July 12, 2016, the Central Statistics Office of Figure 1.2.1. Irish National Accounts
Ireland disseminated unprecedented revisions to some
of the countrys main macroeconomic statistics. GDP 260 1. GDP and Gross National Income
growth in real terms for 2015 was revised from a (Billions of 2014 euros)
240
preliminary figure of 7.8 percent to a record 26.3 per-
cent, growth in the gross national income (GNI) was 220 GDP
revised from 5.7 percent to 18.7 percent, and revisions Gross national income
200
to exports and imports resulted in an increase in net
exports of more than 35 billion (about 17 percent 180
of the preliminary 2015 GDP) estimate reported in
160
March 2016 in 2015 (Figure1.2.1).
The revisions conform to international standards 140
2006 07 08 09 10 11 12 13 14 15
the System of National Accounts (2008 SNA) and the
European System of Accounts (ESA 2010)and the
500 2. Trade and Domestic Demand 300
new data were disseminated according to an estab- (Billions of 2014 euros, except where noted)
lished revision cycle. Exports Imports 250
400
The unusually large revisions are explained to a Total domestic demand (index, 200
great extent by relocations of entire balance sheets 300 2010 = 100)
and their related activity to Ireland. More specifically, Net exports (index, 2010 = 100) 150
the revisions were driven mainly by: (1) a significant 200
100
increase in external contract manufacturing activity
attributable to Ireland, and (2) the relocation and use 100 50
of intellectual property products. From a statistical
0 0
perspective, the increase in contract manufacturing 2010 11 12 13 14 15
activity through redomiciliation means that all value
added derived from this type of production is now Sources: Central Statistics Ofce Ireland; and IMF staff
recorded in Ireland. This will have an impact on calculations.
production, exports, imports, and taxation. Even when
actual physical manufacturing of goods is carried out
abroad, the payment to the manufacturer is treated as The intellectual property product relocations were
importation of services, and the final output of this mostly recorded as other changes in the international
activity, once sold (exported), contributes to exports investment positionimplying a sharp downward
in an amount that includes the cost of intermediate revision to the net international investment position.
inputs (including manufacturing services), license fees, This is because the intellectual property product
other production costs, and profit margins. transfer resulted in much larger intercompany debt
The relocation of intellectual property products in foreign direct investment liabilities (Table1.2.1).1
has several direct effects on national accounts, the If these relocations had been recorded in the balance
balance of payments, and the international investment of payments, the effects on GDP would have been
position. Net exports are affected because: (1) the fees the same, but the Irish accounts would have shown
that firms located in Ireland charge foreign companies an additional very large one-off increase in imports of
to manufacture patented products result in an increase services and a correspondingly large one-off current
in services exports, and (2) firms located in Ireland account deficit, along with a one-off increase in gross
producing patented products no longer pay the fee fixed capital formation in 2015.
associated with relocated intellectual property prod- The relocation of balance sheets (dominated by
ucts, which reduces services imports. GDP and GNI intellectual property) is not new, but the scale observed
are also affected because the increase in fixed assets
implies an increase in the estimates of depreciation. 1The transfer of intellectual property capital to Ireland was

financed by loans to the relevant Irish affiliates from other enti-


ties in the group and hence resulted in a sharp increase in foreign
The author of this box is Michael Stanger. direct investment liabilities in the form of debt.

International Monetary Fund | April 2017 43


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 1.2 (continued)


Table 1.2.1 Ireland: Balance of Payments and International Investment Position
(Billions of euros)
International International
Investment Position Balance of Investment Position
Direct Investment (end of 2014) Payments (2015) Other Changes (end of 2015)
Assets 522.8 91.6 114.3 728.8
Release 2015:Q4
Liabilities 311.5 90.7 2.2 400.0
Assets 510.2 149.9 155.1 815.2
Release 2016:Q1
Liabilities 342.7 169.8 283.1 795.6
Assets 12.6 58.3 40.8 86.4
Revisions
Liabilities 31.2 79.2 285.3 395.6
Sources: Central Statistics Office of Ireland (for data on balance of payments and international investment position); Other Changes derived residually.

in 2015 is exceptionalit added about 300 billion companies and products with substantial intellectual
to Irelands capital stock and a similar amount to its property content, retained earnings are typically sizable
net external liabilities. Activity attributable to goods because they need to offset the relatively rapid depreci-
for processing (that is, contract manufacturing) also ation of intellectual property capital.
increased significantly. Together, these two factors had a As a consequence of these relocations, the use
substantial impact on Irelands macroeconomic statistics, of standard headline measuressuch as domestic
particularly given the small size of the economy. production, national income, domestic demand, and
net exportsare less applicable to economic activity
Need for Additional Measures to Understand in Ireland. For instance, the conventional measures of
Complex New Arrangements fixed capital formation and domestic demand contain
The acquisition of foreign-owned intellectual significant components related to the nondomestic
property assets adds to capital formation, and any economy. Additional measures to reflect the level of
subsequent revenue from licensing adds to Irelands activity within the domestic economy are therefore
GDP if licenses are charged; this has not happened required.
significantly to date. Moreover, the growth of capital
formation significantly increases standard measures of Strategy to Address Measurement Issues
labor productivity and alters their relationship with The Central Statistics Office of Ireland convened the
domestically generated GDP and employment. Economic Statistics Review Group to provide direction
The inclusion of contract manufacturing activity in on how best to meet user needs for a better under-
statistical accounts increases output (exports), imports, standing of Irish economic activity in the context of
GDP, and GNI, but leaves domestic employment a highly globalized economy.2 The group finalized its
mostly unchanged. GDP is a measure of production report in December 2016, and in February 2017 the
and thus includes value added that accrues to foreign Central Statistics Office published its response to the
investors. GNI, in contrast, is a measure of income, reports recommendations, including a timetable for
and Irelands GNI is significantly lower than its GDP implementation.
because GNI does not include the income paid abroad Based on the reports recommendations, GDP and
or the retained earnings of foreign direct investors in GNI will remain the key international standard indica-
Ireland. However, GNI does include retained earn- tors, and new analytical presentations and supplemen-
ings on foreign investment that is not direct (many tary statistics will be made available. Annual aggregates
corporate relocations to Ireland entail foreign invest- will be developed first, followed by quarterly series
ment that is not directthat is, individual owners where feasible and appropriate. Most recommenda-
fall short of the 10 percent threshold that classifies an
investment as direct). In those cases, corporate entities
are considered Irish, and their retained earnings are 2The Economic Statistics Review Group includes policymak-
treated as Irish income, even though retained earnings
ers, analysts, regulators, business and trade union representatives,
ultimately accrue to foreign shareholders through their academics, and members of the international statistics commu-
impact on stock prices. Furthermore, in the case of nity represented by Eurostat and the IMF.

44 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Box 1.2 (continued)


tions are to be implemented between mid-2017 and into two broadly defined, foreign and domestic,
the end of 2018, in particular the following: subsectors, as this sector accounts for most of the
A reliable indicator of the size of the economy that multinational enterprises operating in Ireland.
is relatively immune to relocations. The recom- The same detail is needed for the entire system of
mended indicator is an adjusted GNI that is an national accounts, the balance of payments, and the
extension of the standard GNI and takes into international investment position.
account the retained earnings of redomiciled firms Additional detail on cross-border economic activ-
and depreciation on foreign-owned domestic capital ities to allow for the monitoring of the domestic
assets. Corresponding adjusted presentations of the macroeconomic situation, which would provide
balance of payments and international investment increased detail on gross fixed capital formation,
position data are also proposed. domestic demand, exports, and imports. Along the
A standard set of structural macroeconomic indica- same lines, an additional breakdown of the indus-
tors that better describe economic activity by multi- trial production index is proposed.
national-company-dominated and domestic sectors. A number of initiatives to enhance the communica-
This includes a breakdown of the nonfinancial tion strategy to make it easier for users to under-
sector in the annual Institutional Sector Accounts stand major statistical releases.

International Monetary Fund | April 2017 45


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Table 1.1.1. European Economies: Real GDP, Consumer Prices, Current Account Balance, and Unemployment
(Annual percent change, unless noted otherwise)
Real GDP Consumer Prices1 Current Account Balance2 Unemployment3
Projections Projections Projections Projections
2016 2017 2018 2016 2017 2018 2016 2017 2018 2016 2017 2018
Europe 2.0 2.0 2.0 0.9 2.5 2.4 2.4 2.3 2.4 ... ... ...
Advanced Europe 1.8 1.8 1.7 0.4 1.8 1.6 2.9 2.9 2.9 8.7 8.3 8.0
Euro Area4,5 1.7 1.7 1.6 0.2 1.7 1.5 3.4 3.0 3.0 10.0 9.4 9.1
Germany 1.8 1.6 1.5 0.4 2.0 1.7 8.5 8.2 8.0 4.2 4.2 4.2
France 1.2 1.4 1.6 0.3 1.4 1.2 1.1 0.9 0.5 10.0 9.6 9.3
Italy 0.9 0.8 0.8 0.1 1.3 1.3 2.7 2.0 1.8 11.7 11.4 11.0
Spain 3.2 2.6 2.1 0.2 2.4 1.4 2.0 1.5 1.6 19.6 17.7 16.6
Netherlands 2.1 2.1 1.8 0.1 0.9 1.4 9.6 9.2 9.1 5.9 5.4 5.3
Belgium 1.2 1.6 1.5 1.8 2.0 1.7 1.0 0.9 1.0 8.0 7.8 7.6
Austria 1.5 1.4 1.3 1.0 2.1 1.8 2.4 2.4 2.2 6.1 5.9 5.9
Greece 0.0 2.2 2.7 0.0 1.3 1.4 0.6 0.3 0.0 23.8 21.9 21.0
Portugal 1.4 1.7 1.5 0.6 1.2 1.4 0.8 0.3 0.4 11.1 10.6 10.1
Ireland 5.2 3.5 3.2 0.2 0.9 1.5 4.7 4.7 4.7 7.9 6.5 6.3
Finland 1.4 1.3 1.4 0.4 1.4 1.6 1.1 1.3 1.2 8.8 8.5 8.3
Slovak Republic 3.3 3.3 3.7 0.5 1.2 1.5 0.4 0.3 0.2 9.7 7.9 7.4
Lithuania 2.3 2.8 3.1 0.7 2.8 2.0 0.9 1.6 1.5 7.9 7.4 7.2
Slovenia 2.5 2.5 2.0 0.1 1.5 2.0 6.8 5.5 5.1 7.9 7.0 6.6
Luxembourg 4.0 3.7 3.5 0.1 1.4 1.3 4.8 5.1 5.1 6.4 5.9 5.7
Latvia 2.0 3.0 3.3 0.1 2.8 2.5 1.5 1.1 1.4 9.6 9.4 9.2
Estonia 1.6 2.5 2.8 0.8 3.2 2.5 2.7 1.4 0.9 6.9 8.3 8.9
Cyprus 2.8 2.5 2.3 1.2 1.5 1.4 2.4 2.5 2.5 12.9 11.3 10.2
Malta 5.0 4.1 3.5 0.9 1.5 1.6 5.8 5.5 5.3 4.8 4.7 4.7
United Kingdom5 1.8 2.0 1.5 0.6 2.5 2.6 4.4 3.3 2.9 4.9 4.9 5.1
Switzerland 1.3 1.4 1.6 0.4 0.4 0.7 12.0 10.8 10.5 3.3 3.0 2.9
Sweden 3.3 2.7 2.4 1.1 1.4 1.6 4.7 4.6 4.2 7.0 6.7 6.7
Norway 1.0 1.2 1.9 3.6 2.6 2.5 4.6 5.7 5.7 4.8 4.5 4.2
Czech Republic 2.4 2.8 2.2 0.7 2.3 1.8 1.1 1.2 0.7 4.0 3.8 4.2
Denmark 1.1 1.5 1.7 0.3 0.6 1.1 8.1 7.5 7.2 6.2 5.8 5.8
Iceland 7.2 5.7 3.6 1.7 2.2 2.6 8.0 6.9 6.7 3.0 3.0 3.3
San Marino 1.0 1.2 1.3 0.6 0.7 0.8 ... ... ... 8.6 8.0 7.4
Emerging and Developing Europe6 3.0 3.0 3.3 3.2 5.7 5.5 1.9 2.8 2.8 ... ... ...
Turkey 2.9 2.5 3.3 7.8 10.1 9.1 3.8 4.7 4.6 10.8 11.5 11.0
Poland 2.8 3.4 3.2 0.6 2.3 2.3 0.3 1.7 1.8 6.1 5.5 5.3
Romania 4.8 4.2 3.4 1.6 1.3 3.1 2.4 2.8 2.5 6.0 5.4 5.2
Hungary 2.0 2.9 3.0 0.4 2.5 3.3 4.3 3.7 3.0 4.9 4.4 4.3
Bulgaria5 3.4 2.9 2.7 1.3 1.0 1.8 4.2 2.3 2.0 7.7 7.1 6.9
Serbia 2.8 3.0 3.5 1.1 2.6 3.0 4.0 4.0 4.0 15.9 16.0 15.6
Croatia 2.9 2.9 2.6 1.1 1.1 1.1 3.9 2.8 1.8 15.0 13.9 13.5
Note: Data for some countries are based on fiscal years. Please refer to Table F in the Statistical Appendix for a list of economies with exceptional reporting periods.
1Movements in consumer prices are shown as annual averages. Year-end to year-end changes can be found in Tables A6 and A7 in the Statistical Appendix.
2Percent of GDP.
3Percent. National definitions of unemployment may differ.
4Current account position corrected for reporting discrepancies in intra-area transactions.
5Based on Eurostats harmonized index of consumer prices except for Slovenia.
6Includes Albania, Bosnia and Herzegovina, Kosovo, FYR Macedonia, and Montenegro.

46 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Annex Table 1.1.2. Asian and Pacific Economies: Real GDP, Consumer Prices, Current Account Balance, and Unemployment
(Annual percent change, unless noted otherwise)
Real GDP Consumer Prices1 Current Account Balance2 Unemployment3
Projections Projections Projections Projections
2016 2017 2018 2016 2017 2018 2016 2017 2018 2016 2017 2018
Asia 5.3 5.5 5.4 2.3 2.9 2.9 2.5 2.1 2.0 ... ... ...
Advanced Asia 1.6 1.9 1.6 0.5 1.4 1.3 4.6 4.6 4.5 3.6 3.5 3.5
Japan 1.0 1.2 0.6 0.1 1.0 0.6 3.9 4.2 4.3 3.1 3.1 3.1
Korea 2.8 2.7 2.8 1.0 1.8 1.9 7.0 6.2 6.1 3.7 3.8 3.6
Australia 2.5 3.1 3.0 1.3 2.0 2.4 2.6 2.8 2.9 5.7 5.2 5.1
Taiwan Province of China 1.4 1.7 1.9 1.4 1.4 1.3 14.2 14.8 15.0 3.9 4.0 4.0
Singapore 2.0 2.2 2.6 0.5 1.1 1.8 19.0 20.1 19.2 2.1 2.1 2.1
Hong Kong SAR 1.9 2.4 2.5 2.6 2.6 2.7 5.1 3.0 3.1 3.3 3.2 3.2
New Zealand 4.0 3.1 2.9 0.6 1.5 2.0 2.7 2.5 3.1 5.1 5.0 4.8
Macao SAR 4.0 2.8 1.7 2.4 2.0 2.2 27.1 29.5 30.5 1.9 2.0 2.0
Emerging and Developing Asia 6.4 6.4 6.4 2.9 3.3 3.3 1.3 0.8 0.7 ... ... ...
China 6.7 6.6 6.2 2.0 2.4 2.3 1.8 1.3 1.2 4.0 4.0 4.0
India4 6.8 7.2 7.7 4.9 4.8 5.1 0.9 1.5 1.5 ... ... ...
ASEAN-5 4.9 5.0 5.2 2.4 3.6 3.7 2.2 1.6 1.1 ... ... ...
Indonesia 5.0 5.1 5.3 3.5 4.5 4.5 1.8 1.9 2.0 5.6 5.4 5.2
Thailand 3.2 3.0 3.3 0.2 1.4 1.5 11.4 9.7 7.8 0.8 0.7 0.7
Malaysia 4.2 4.5 4.7 2.1 2.7 2.9 2.0 1.8 1.8 3.5 3.4 3.2
Philippines 6.8 6.8 6.9 1.8 3.6 3.3 0.2 0.1 0.3 5.5 6.0 5.5
Vietnam 6.2 6.5 6.3 2.7 4.9 5.0 4.7 4.1 3.4 2.4 2.4 2.4
Other Emerging and Developing
Asia5 5.5 6.1 6.3 5.6 5.9 5.6 1.0 2.0 2.6 ... ... ...
Memorandum
Emerging Asia6 6.4 6.4 6.4 2.8 3.2 3.2 1.4 0.9 0.8 ... ... ...
Note: Data for some countries are based on fiscal years. Please refer to Table F in the Statistical Appendix for a list of economies with exceptional reporting periods.
1Movements in consumer prices are shown as annual averages. Year-end to year-end changes can be found in Tables A6 and A7 in the Statistical Appendix.
2Percent of GDP.
3Percent. National definitions of unemployment may differ.
4See country-specific notes for India in the Country Notes section of the Statistical Appendix.
5Other Emerging and Developing Asia comprises Bangladesh, Bhutan, Brunei Darussalam, Cambodia, Fiji, Kiribati, Lao P.D.R., Maldives, Marshall Islands, Micronesia,

Mongolia, Myanmar, Nauru, Nepal, Palau, Papua New Guinea, Samoa, Solomon Islands, Sri Lanka, Timor-Leste, Tonga, Tuvalu, and Vanuatu.
6Emerging Asia comprises the ASEAN-5 (Indonesia, Malaysia, Philippines, Thailand, Vietnam) economies, China, and India.

International Monetary Fund | April 2017 47


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Table 1.1.3. Western Hemisphere Economies: Real GDP, Consumer Prices, Current Account Balance, and Unemployment
(Annual percent change, unless noted otherwise)
Real GDP Consumer Prices1 Current Account Balance2 Unemployment3
Projections Projections Projections Projections
2016 2017 2018 2016 2017 2018 2016 2017 2018 2016 2017 2018
North America 1.7 2.2 2.4 1.4 2.8 2.4 2.6 2.7 3.2 ... ... ...
United States 1.6 2.3 2.5 1.3 2.7 2.4 2.6 2.7 3.3 4.9 4.7 4.6
Canada 1.4 1.9 2.0 1.4 2.0 2.1 3.3 2.9 2.7 7.0 6.9 6.8
Mexico 2.3 1.7 2.0 2.8 4.8 3.2 2.7 2.5 2.7 4.3 4.4 4.4
Puerto Rico4 1.8 3.0 2.5 0.2 1.5 0.5 ... ... ... 11.8 12.6 12.1
South America5 2.7 0.6 1.8 ... ... ... 1.9 1.9 2.1 ... ... ...
Brazil 3.6 0.2 1.7 8.7 4.4 4.3 1.3 1.3 1.7 11.3 12.1 11.6
Argentina 2.3 2.2 2.3 ... 25.6 18.7 2.6 2.9 3.4 8.5 7.4 7.3
Colombia 2.0 2.3 3.0 7.5 4.5 3.2 4.4 3.6 3.3 9.2 9.5 9.3
Venezuela 18.0 7.4 4.1 254.9 720.5 2,068.5 2.4 3.3 2.1 21.2 25.3 28.2
Chile 1.6 1.7 2.3 3.8 2.8 3.0 1.4 1.4 1.7 6.5 7.0 6.8
Peru 3.9 3.5 3.7 3.6 3.1 2.6 2.8 1.9 2.0 6.7 6.7 6.7
Ecuador 2.2 1.6 0.3 1.7 0.3 0.6 1.1 0.9 0.1 5.2 5.7 5.8
Bolivia 4.1 4.0 3.7 3.6 4.0 5.0 5.4 3.9 2.6 4.0 4.0 4.0
Uruguay 1.4 1.6 2.6 9.6 7.7 7.5 1.0 1.5 1.6 7.9 7.8 7.8
Paraguay 4.1 3.3 3.7 4.1 4.0 4.0 0.6 1.4 0.5 5.1 5.4 5.5
Central America6 3.8 3.9 4.1 2.1 2.8 3.5 3.0 3.1 3.2 ... ... ...
Caribbean7 3.4 3.6 4.2 2.8 4.3 4.3 3.4 3.7 3.8 ... ... ...
Memorandum
Latin America and the Caribbean8 1.0 1.1 2.0 5.6 4.2 3.7 2.1 2.1 2.3 ... ... ...
East Caribbean Currency Union9 1.9 2.4 2.3 0.2 1.7 1.6 11.7 13.8 13.8 ... ... ...
Note: Data for some countries are based on fiscal years. Please refer to Table F in the Statistical Appendix for a list of economies with exceptional reporting periods.
1Movements in consumer prices are shown as annual averages. Year-end to year-end changes can be found in Tables A6 and A7 in the Statistical Appendix.
2Percent of GDP.
3Percent. National definitions of unemployment may differ.
4Puerto Rico is a territory of the United States but its statistical data are maintained on a separate and independent basis.
5Includes Guyana and Suriname. Data for Argentinas and Venezuelas consumer prices are excluded. See country-specific notes for Argentina and Venezuela in the Country

Notes section of the Statistical Appendix.


6Central America comprises Belize, Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama.
7The Caribbean comprises Antigua and Barbuda, The Bahamas, Barbados, Dominica, Dominican Republic, Grenada, Haiti, Jamaica, St. Kitts and Nevis, St. Lucia, St. Vincent

and the Grenadines, and Trinidad and Tobago.


8Latin America and the Caribbean comprises Mexico and economies from the Caribbean, Central America, and South America. Data for Argentinas and Venezuelas con-

sumer prices are excluded. See country-specific notes for Argentina and Venezuela in the Country Notes section of the Statistical Appendix.
9Eastern Caribbean Currency Union comprises Antigua and Barbuda, Dominica, Grenada, St. Kitts and Nevis, St. Lucia, and St. Vincent and the Grenadines as well as

Anguilla and Montserrat, which are not IMF members.

48 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Annex Table 1.1.4. Commonwealth of Independent States Economies: Real GDP, Consumer Prices, Current Account
Balance, and Unemployment
(Annual percent change, unless noted otherwise)
Real GDP Consumer Prices1 Current Account Balance2 Unemployment3
Projections Projections Projections Projections
2016 2017 2018 2016 2017 2018 2016 2017 2018 2016 2017 2018
Commonwealth of Independent States4 0.3 1.7 2.1 8.3 5.7 5.3 0.2 1.6 1.8 ... ... ...
Net Energy Exporters 0.2 1.7 2.0 7.9 5.2 4.9 0.4 2.2 2.5 ... ... ...
Russia 0.2 1.4 1.4 7.0 4.5 4.2 1.7 3.3 3.5 5.5 5.5 5.5
Kazakhstan 1.1 2.5 3.4 14.6 8.0 7.2 6.1 4.0 2.8 5.0 5.0 5.0
Uzbekistan 7.8 6.0 6.0 8.0 8.6 8.8 1.4 2.1 1.6 ... ... ...
Azerbaijan 3.8 1.0 2.0 12.4 10.0 8.0 3.8 1.3 3.8 6.0 6.0 6.0
Turkmenistan 6.2 6.5 6.3 3.5 6.0 6.2 21.0 12.8 11.5 ... ... ...
Net Energy Importers 1.1 1.6 2.7 11.0 9.5 8.2 4.7 4.9 4.6 ... ... ...
Ukraine 2.3 2.0 3.2 13.9 11.5 9.5 3.6 3.6 2.9 8.8 9.0 8.7
Belarus 3.0 0.8 0.6 11.8 9.3 8.7 4.3 4.7 5.0 1.0 1.0 1.0
Georgia 2.7 3.5 4.0 2.1 5.7 2.4 12.4 12.9 12.5 ... ... ...
Armenia 0.2 2.9 2.9 1.4 2.0 3.5 2.9 3.2 2.9 18.8 18.9 18.9
Tajikistan 6.9 4.5 5.0 5.9 5.8 6.0 5.1 5.5 5.1 ... ... ...
Kyrgyz Republic 3.8 3.4 3.8 0.4 3.6 5.2 9.4 12.0 12.1 7.5 7.4 7.3
Moldova 4.0 4.5 3.7 6.4 5.5 5.9 3.4 3.8 4.0 4.2 4.3 4.2
Memorandum
Caucasus and Central Asia5 2.4 3.1 4.1 10.4 7.9 7.2 6.2 3.8 3.0 ... ... ...
Low-Income CIS Countries6 6.1 5.1 5.2 5.8 7.0 7.1 2.1 1.9 2.2 ... ... ...
Net Energy Exporters Excluding Russia 2.2 3.1 4.1 11.5 8.3 7.6 5.9 3.2 2.3 ... ... ...
Note: Data for some countries are based on fiscal years. Please refer to Table F in the Statistical Appendix for a list of economies with exceptional reporting periods.
1Movements in consumer prices are shown as annual averages. Year-end to year-end changes can be found in Table A7 in the Statistical Appendix.
2Percent of GDP.
3Percent. National definitions of unemployment may differ.
4Georgia, Turkmenistan, and Ukraine, which are not members of the Commonwealth of Independent States (CIS), are included in this group for reasons of geography and

similarity in economic structure.


5Caucasus and Central Asia comprises Armenia, Azerbaijan, Georgia, Kazakhstan, the Kyrgyz Republic, Tajikistan, Turkmenistan, and Uzbekistan.
6Low-Income CIS countries comprise Armenia, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan.

International Monetary Fund | April 2017 49


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Table 1.1.5. Middle East, North African Economies, Afghanistan, and Pakistan: Real GDP, Consumer Prices, Current
Account Balance, and Unemployment
(Annual percent change, unless noted otherwise)
Real GDP Consumer Prices1 Current Account Balance2 Unemployment3
Projections Projections Projections Projections
2016 2017 2018 2016 2017 2018 2016 2017 2018 2016 2017 2018
Middle East, North Africa, Afghanistan,
and Pakistan 3.9 2.6 3.4 5.1 7.6 7.4 3.4 1.1 0.8 ... ... ...
Oil Exporters4 4.0 1.9 2.9 4.6 5.8 6.3 2.7 0.4 0.6 ... ... ...
Saudi Arabia 1.4 0.4 1.3 3.5 3.8 5.1 3.9 1.5 2.0 5.7 ... ...
Iran 6.5 3.3 4.3 8.9 11.2 11.0 6.3 5.3 5.1 12.5 12.5 12.5
United Arab Emirates 2.7 1.5 4.4 1.8 2.8 3.7 2.4 3.5 3.9 ... ... ...
Algeria 4.2 1.4 0.6 6.4 4.8 4.3 16.4 12.3 10.2 10.5 11.7 13.2
Iraq 10.1 3.1 2.6 0.4 2.0 2.0 7.3 4.4 4.9 ... ... ...
Qatar 2.7 3.4 2.8 2.7 2.6 5.7 2.2 0.7 0.6 ... ... ...
Kuwait 2.5 0.2 3.5 3.2 4.2 3.6 2.7 8.2 7.1 2.1 2.1 2.1
Oil Importers5 3.7 4.0 4.4 6.2 11.4 9.5 4.8 4.9 4.3 ... ... ...
Egypt 4.3 3.5 4.5 10.2 22.0 16.9 5.6 5.3 3.9 12.7 12.6 11.8
Pakistan 4.7 5.0 5.2 2.9 4.3 5.0 1.1 2.9 3.0 6.0 6.0 6.1
Morocco 1.5 4.4 3.9 1.6 1.2 1.5 3.9 2.6 2.0 9.4 9.3 9.5
Sudan 3.0 3.7 3.6 17.8 23.2 16.0 5.8 4.7 4.3 20.6 19.6 18.6
Tunisia 1.0 2.5 3.1 3.7 3.9 3.8 9.0 8.6 8.1 14.0 13.0 12.0
Lebanon 1.0 2.0 2.5 0.8 2.6 2.0 16.0 15.5 14.9 ... ... ...
Jordan 2.1 2.3 2.5 0.8 2.3 2.5 9.4 8.6 7.4 ... ... ...
Memorandum
Middle East and North Africa 3.8 2.3 3.2 5.4 8.1 7.7 3.7 1.0 0.6 ... ... ...
Israel6 4.0 2.9 3.0 0.5 0.7 1.4 3.6 3.4 3.4 4.8 4.8 4.8
Maghreb7 2.6 6.2 2.0 5.7 5.6 5.4 14.1 9.0 8.3 ... ... ...
Mashreq8 3.9 3.3 4.2 8.7 19.3 14.9 7.2 7.4 6.1 ... ... ...
Note: Data for some countries are based on fiscal years. Please refer to Table F in the Statistical Appendix for a list of economies with exceptional reporting periods.
1Movements in consumer prices are shown as annual averages. Year-end to year-end changes can be found in Tables A6 and A7 in the Statistical Appendix.
2Percent of GDP.
3Percent. National definitions of unemployment may differ.
4Includes Bahrain, Libya, Oman, and Yemen.
5Includes Afghanistan, Djibouti, and Mauritania. Excludes Syria because of the uncertain political situation.
6Israel, which is not a member of the economic region, is included for reasons of geography but is not included in the regional aggregates.
7The Maghreb comprises Algeria, Libya, Mauritania, Morocco, and Tunisia.
8The Mashreq comprises Egypt, Jordan, and Lebanon. Syria is excluded because of the uncertain political situation.

50 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

Annex Table 1.1.6. Sub-Saharan African Economies: Real GDP, Consumer Prices, Current Account Balance, and Unemployment
(Annual percent change, unless noted otherwise)
Real GDP Consumer Prices1 Current Account Balance2 Unemployment3
Projections Projections Projections Projections
2016 2017 2018 2016 2017 2018 2016 2017 2018 2016 2017 2018
Sub-Saharan Africa 1.4 2.6 3.5 11.4 10.7 9.5 4.0 3.8 3.7 ... ... ...
Oil Exporters4 1.7 0.7 1.9 18.8 18.3 16.2 1.4 0.7 0.2 ... ... ...
Nigeria 1.5 0.8 1.9 15.7 17.4 17.5 0.6 1.0 1.0 12.7 ... ...
Angola 0.0 1.3 1.5 32.4 27.0 17.8 4.3 3.8 3.2 ... ... ...
Gabon 2.3 1.0 2.7 2.1 2.5 2.5 9.0 8.3 6.3 ... ... ...
Chad 6.4 0.3 2.4 1.1 0.2 1.8 8.8 4.7 6.2 ... ... ...
Republic of Congo 2.7 0.6 8.8 3.6 1.3 2.1 28.5 4.7 12.1 ... ... ...
Middle-Income Countries5 1.9 2.5 3.5 6.8 5.9 5.2 3.4 3.8 3.8 ... ... ...
South Africa 0.3 0.8 1.6 6.3 6.2 5.5 3.3 3.4 3.6 26.7 27.4 27.7
Ghana 4.0 5.8 9.2 17.5 12.0 9.0 6.4 6.0 4.9 ... ... ...
Cte d'Ivoire 7.5 6.9 7.2 1.0 1.5 2.0 2.2 4.0 3.5 ... ... ...
Cameroon 4.4 3.7 4.3 0.9 1.0 1.4 3.6 3.1 3.0 ... ... ...
Zambia 3.0 3.5 4.0 17.9 9.0 8.0 5.5 3.2 2.5 ... ... ...
Senegal 6.6 6.8 7.0 0.9 1.9 2.0 7.1 7.8 7.7 ... ... ...
Low-Income Countries6 5.4 5.5 5.8 7.0 6.7 6.1 8.3 8.3 8.9 ... ... ...
Ethiopia 8.0 7.5 7.5 7.3 6.3 7.5 9.9 10.0 9.1 ... ... ...
Kenya 6.0 5.3 5.8 6.3 6.5 5.2 5.5 5.8 5.7 ... ... ...
Tanzania 6.6 6.8 6.9 5.2 5.1 5.0 6.3 7.2 7.0 ... ... ...
Uganda 4.7 5.0 5.8 5.5 6.3 6.0 5.9 7.0 8.1 ... ... ...
Madagascar 4.1 4.5 4.8 6.7 6.9 6.4 2.3 3.7 4.2 ... ... ...
Democratic Republic of the Congo 2.4 2.8 3.5 22.4 15.0 10.0 4.4 3.8 2.9 ... ... ...
Memorandum
Sub-Saharan Africa Excluding South
Sudan 1.5 2.7 3.5 10.5 10.3 9.4 4.0 3.8 3.7 ... ... ...
Note: Data for some countries are based on fiscal years. Please refer to Table F in the Statistical Appendix for a list of economies with exceptional reporting periods.
1Movements in consumer prices are shown as annual averages. Year-end to year-end changes can be found in Table A7 in the Statistical Appendix.
2Percent of GDP.
3Percent. National definitions of unemployment may differ.
4Includes Equatorial Guinea and South Sudan.
5Includes Botswana, Cabo Verde, Lesotho, Mauritius, Namibia, Seychelles, and Swaziland.
6Includes Benin, Burkina Faso, Burundi, the Central African Republic, Comoros, Eritrea, The Gambia, Guinea, Guinea-Bissau, Liberia, Malawi, Mali, Mozambique, Niger,

Rwanda, So Tom and Prncipe, Sierra Leone, Togo, and Zimbabwe.

International Monetary Fund | April 2017 51


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Special Feature: Commodity Market Developments and Forecasts, with a Focus on


the Role of Technology and Unconventional Sources in the Global Oil Market

Commodity prices have rallied since the release of the Figure 1.SF.1. Commodity Market Developments
October2016 World Economic Outlook (WEO). Oil
prices have increased following the announcement of the 300 1. Commodity Price Indices All commodities Energy
(2005 = 100) Food Metals
production agreement by the Organization of the Petro- 260

leum Exporting Countries (OPEC). Chinas continued 220

strength in the construction sector and the anticipated 180

possibility of a fiscal stimulus in the United States have 140

increased metal demand prospects and prices. And easing 100

of excess supply conditions has helped the recovery in food 60


prices. This special feature on commodity market develop- 20
2005 06 07 08 09 10 11 12 13 14 15 Feb.
ments includes an in-depth analysis of the role of technol- 17
ogy and unconventional sources in the global oil market. 90 2. Brent Futures Curves
1

80 (U.S. dollars a barrel; expiration dates on x-axis)


The IMFs Primary Commodities Price Index 70
has increased by 15.5percent since August2016, 60
50
the reference period for the October2016 WEO 40
April 2017 WEO
October 2016 WEO
(Figure1.SF.1, panel 1). While energy and metals have 30
April 2016 WEO
rallied, by 21.1 percent and 23.6percent, respectively, 20 October 2015 WEO
10 April 2015 WEO
food prices increased more modestly, by 4.9percent.
0
Oil prices have continued to increase, by 21.2percent, Apr. 2015 Dec. 16 Aug. 18 Apr. 20 Dec. 21
following the agreement by OPEC members to cut
oil production. Natural gas prices have increased in 200 3. Brent Price Prospects 2 Futures
(U.S. dollars a barrel) 68 percent condence interval
Europe on account of supply tightening and higher 160 86 percent condence interval
oil prices. Coal prices have rallied, by 21.0percent, 95 percent condence interval
120
following government-led reductions in coal produc-
tion in China and outages in Australia that affected 80
production and shipment.
40
On November30,2016, members of OPEC agreed
to reduce crude oil output to 32.5million barrels a 0
2010 11 12 13 14 15 16 17 18 19 20
day (mbd), effective January2017 and for a duration
of six months, extendable for another six months. 180 4. Metal Price Indices
That agreement would suggest a cut of 1.2 mbd from 160 (Jan. 2, 2014 = 100)
Aluminum Copper
production levels in October2016. Iraq, Kuwait, Saudi 140 Iron ore Nickel
Arabia, and the United Arab Emirates are bearing the 120
100
brunt of the cuts, alongside other member countries.
80
Libya and Nigeria are exempt.1 Participants at an
60
OPEC and non-OPEC meeting in Vienna on Decem- 40
20
Jan. 2014 July 14 Jan. 15 July 15 Jan. 16 July 16 Jan. 17
The authors of this feature are Rabah Arezki (team leader), Clau-
dia Berg, Christian Bogmans, and Akito Matsumoto (team coleader),
Sources: Bloomberg L.P.; IMF, Primary Commodity Price System; Thomson
with research assistance from Clara Galeazzi, Lama Kiyasseh, and Reuters Datastream; and IMF staff estimates.
Rachel Yuting Fan. The authors also thank Rystad Energy, and Per Note: WEO = World Economic Outlook.
Magnus Nysveen in particular, for very useful discussions and for 1WEO future prices are baseline assumptions for each WEO and derived from
kindly providing proprietary data on capital expenditures and cost future prices. April 2017 WEO prices are based on February 28, 2017 closing.
structures. 2Derived from prices of futures options on February 28, 2017.
1Indonesia, which accounted for 0.75 mbd of production, has

been suspended from OPEC.

52 International Monetary Fund | April 2017


Gaining Momentum? COMMODITY MARKET DEVELOPMENTS AND FORECASTS

ber10,2016, agreed to additional cuts amounting to soften rising prices, China has recently sought to relax
about 0.6 mbd. Russia, a country that is not a member restrictions on the number of days coal miners may
of OPEC, has committed to reducing production by work in a year. Growing environmental and health
0.3 mbd, and 10 other non-OPEC countries agreed concerns are expected to lead to a reduction in the
to contribute the remainder. Following these produc- share of coal in primary energy, accentuating excess
tion agreements, Saudi Arabia indicated it could cut capacity in that sector, especially in China.
production beyond its initial commitment in a bid to Oil futures contracts point to stable prices of about
enhance the credibility of the agreement. $55 a barrel (Figure1.SF.1, panel 2). Baseline assump-
In response to these agreements, spot oil prices tions for the IMFs average petroleum spot prices,
increased to more than $50 a barrel. Oil prices beyond which are based on futures prices, suggest average
that level will stimulate investment, which is expected annual prices of $55.2 a barrel in2017an increase of
to increase in2017 after two consecutive years of 28.9percent fromthe2016 averageand $55.1 a bar-
significant decline. The effectiveness of the produc- rel in2018 (Figure1.SF.1, panel 3). The response of
tion agreements could thus be partially offset by an futures prices over a three-year horizon has been more
increase in U.S.shale oil production, which, unlike muted, suggesting that the production agreements are
conventional oil, can commence within a year of initial expected to have a limited effect in the medium term.
investment. Production data from the International Uncertainty remains around the baseline assumptions
Energy Agency (IEA) for January2017 indicate that for oil prices, although risks are balanced. Upside risks
only a few OPEC members fully complied with the include unscheduled outages and geopolitical events,
agreement, although Saudi Arabia has cut more than especially in the Middle East. Although these occur-
initially agreed on. In addition, Libya, which is exempt rences could cause oil market disruptions, high inven-
from the production agreement, increased production. tory levels and a rapid response by shale production
Oil demand grew at 1.6 mbd in2016, which is should prevent a sharp rise in prices in the near future.
lower than during2015. The IEA expects demand Metal prices have increased by 23.6percent since
growth to slow further to 1.4 mbd in2017still August2016 (Figure1.SF.1, panel 4). Iron ore was
above trend growth, estimated at 1.2 mbd. Amid a one of the best performing metals in2016, almost
significant cutback in production, fairly robust demand doubling in price to $80 a metric ton. On the demand
could move the oil market from surplus to deficit in side, metal consumption in China, which accounts for
the first half of2017, in turn reducing oil inventory half of global demand, rebounded in2016 in response
levels. However, rapid investment recovery in the to the authorities policies in support of credit growth.
U.S.shale sector could tip the market back into sur- In turn, these policies have stimulated construction,
plus as early as the second half of2017. which uses metals intensively. The Chinese authorities
The natural gas price indexan average for Europe, have also addressed issues of excess capacity in the steel
Japan, and the United Stateshas increased by sector by cutting production of outdated factories,
18.7percent since August2016. Although prices in including to reduce pollution. Steel mills in mainland
Asia and the United States initially rose on expecta- China have increased their use of imported high-
tions of strong winter demand, a fairly mild winter led er-grade iron ore, which has helped increase iron ore
to subdued demand for gas-fired power generation and prices. Amid speculation over the increase in demand
contained prices. In Europe, prices rose 38.4percent, for cobalt, a key battery input, spot prices have almost
reflecting higher oil prices and a cold winter. Natu- doubled since August2016.
ral gas prices are expected to stay low because ample Announcement following the U.S.election of a
supply from the United States and Russia will meet $1trillion infrastructure plan (over 10 years) pro-
strong natural gas demand growthwhich is expected vided a further boost to metal prices. However, in the
to exceed oil demand growth. global context, the impact of this potential infrastruc-
The coal price indexan average of Australian and ture spending on world metal demand is likely to be
South African priceshas increased by 21.0percent modest. Indeed, in2015 the United States accounted
since August2016. The rally in coal prices reflects for only 8percent of global refined copper demand
a continued effort by Chinese authorities to reduce according to the World Bureau of Metal Statistics and
coal mining capacity substantially as part of a broader 3percent of iron ore demand according to the World
reform agenda to restructure its economy. To help Steel Association.

International Monetary Fund | April 2017 53


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

On the supply side, the declining investment in, and States. Annual food prices are now expected to increase
closure of, high-cost and high-polluting mining oper- by 3.0percent in2017, drop by 0.5 percent in2018,
ations have driven price increases in iron ore, nickel, and remain broadly unchanged thereafter. Rising costs
tin, zinc, and copper. However, overall excess capacity of energy and weather variability, including concerns
will probably put downward pressure on prices in about La Nia, constitute upside risks to the price
many base metals. In January2017, Indonesiaone of forecast. Downside risks may arise from China disman-
the worlds largest nickel producersrelaxed its export tling its price floor systems.
ban on ores. This action partly offsets the drop in sup-
ply caused by the Philippines closure of its mines over
environmental concerns. The Role of Technology and Unconventional
Most metal prices are expected to stay near their Sources in the Global Oil Market
current levels, except iron ore prices, which are Technological factors have played an important role
expected to decline sharply. The IMF metal price index in explaining the collapse in oil prices that started in
is projected to decline from the current level, but June2014. Although technological innovation is often
its2017 average is expected to increase by 23.2per- regarded as exogenous, it is endogenous to the level of oil
cent from the average in2016, reflecting the surge prices. Indeed, high oil prices, prompting breakthroughs
during late2016. The index is expected to decrease by in technology in extractive industries, led to the emer-
4.0percent in2018 from2017. There are downside gence of new sources known as unconventional oil.
risks to the outlook for metal prices, including from Shale, in particular, will have important consequences
the waning policy support and real estate investment for the oil market outlook in that it will help lead to
in China, from a faster rebalancing from investment to more limited and shorter production and price cycles.
consumption in the medium term, or from a disor- This special feature documents the endogenous response
derly adjustment in Chinas corporate debt market. of technology to oil prices and institutional factors.
The agriculture index, which consists of food,
beverages, and agricultural raw materials prices, has Although the OPEC production agreement has cap-
increased by 4.3percent since August2016. Although tured the publics attention, technological forces affect-
prices of palm oil, tea, and rubber have increased ing oil markets over the medium term have received
significantly, prices of rice and cocoa beans have less attention. Technology has indeed transformed the
decreased. Wheat prices reached an 11-year low in oil market in powerful ways. Technological innovation
December2016, but have since somewhat recovered. and subsequent adoption of new recovery techniques
Overall, wheat prices have increased by 15.2per- including drilling and processinghave given rise to
cent since August2016. Maize prices have increased, new sources known as unconventional oil. One recent
although they remain near historical lows. The global example of a new source is shale oil, which has become
stock-to-use ratios of wheat and maize remain sig- a major contributor to global oil supply. Provided
nificantly above the 10-year average, indicating that they pan out and diffuse, improvements in recovery
markets are well supplied. techniques mechanically increase the size of technically
Soybean prices have remained broadly unchanged recoverable oil reserves. This increase, in turn, changes
on account of continued strength in animal protein the outlook for oil supply, with potentially large
demand countering favorable supply conditions. Palm immediate implications for oil pricesacting through
oil prices climbed more than 36.7percent through- the expectation channel associated with the future path
out2016 and increased19percent year over year. of oil production. Although the feedback effect from
This rise is associated with plantations in Indonesia lower oil prices reduces investment and hence produc-
and Malaysia facing the aftereffects of the El Nio tion, the industry is forced to become more efficient,
weather system and the reduction in palm oil invento- unleashing automatic stabilization forces.
ries. The annual price of cocoa has fallen for the first Innovation in recovery techniques typically fol-
time in five years, as harvests in West Africa have been lows periods of prolonged high prices or changes in
favorable. regulations rendering new techniques economical. New
Projections for prices of most agricultural com- oil sources often come onstream in times of need
modities have been revised upward on account of less because of, say, depletion of existing conventional
favorable weather conditions, including in the United sourcesin places that have economic and institu-

54 International Monetary Fund | April 2017


Gaining Momentum? COMMODITY MARKET DEVELOPMENTS AND FORECASTS

tional systems more favorable to innovation and adop- mixture of sand, clay, and water, saturated with a dense
tion of new recovery techniques. The way drilling is and extremely viscous form of petroleum technically
performed has significantly evolved since the inception referred to as bitumen (or colloquially as tar because of
of the oil market, and in addition to improvements its superficially similar appearance). Heavy and extra
in drilling techniques that gave rise to shale and tight heavy oil are characterized by high viscosity, high den-
oil production, successive improvements in techniques sity, and high concentrations of nitrogen, oxygen, sulfur,
for offshore drilling have led to a significant increase and heavy metals. These characteristics result in higher
in new sources of oil. In the1970s production in the costs of extraction, transportation, and refining than are
North Sea and the surge of production in the Gulf incurred with conventional oil. In spite of their cost and
of Mexico were made possible by deepwater drilling technical difficulties, major oil corporations regard these
and higher oil prices after the two oil shocks during resources as providing reliable long-term flows of liquid
the1970s. Such a developmenta relatively high-cost hydrocarbons and substantial payoffs for any incremen-
producer that emerges with new oil sourcesoften tal improvements in recovery. However, environmental
gives rise to tensions with low-cost OPEC producers, concerns have often surfaced, considering the potential
who in the1980s and more recently responded strate- damage these extraction and refining activities may
gically by moderating their production levels. cause. Such concerns surrounding these new oil sources
The following discussion address four questions have often been met with specific safety regulations and
about the role of technology and unconventional oil standards to help limit the risks.
sources in the global oil market:2 Shale oil (also known as tight oil) is petroleum that
What are unconventional oil sources? consists of light crude oil contained in petroleum-bear-
Where are the production and reserve centers? ing formations of low permeability, often shale or tight
How have investment and production evolved? sandstone. Exploitation of shale oil began with the
What lies ahead? development of shale gas extraction using a combina-
tion of hydraulic fracturing (or fracking, a well-stim-
ulation technique in which rock is fractured by a
What Are Unconventional Oil Sources? hydraulically pressurized liquid) and directional drilling
Todays version of unconventional oil consists of (the practice of drilling nonvertical wells). These tech-
oil sands, extra heavy oil, shale and tight oil, and niques were later widely adopted by the oil industry,
ultradeepwater oil.3 Unconventional oil is typically primarily in the United States. Shale oil sources are
more difficult and more expensive to extract and developed by relatively smaller corporations and have a
process than conventional oil. The categorization as cost structure different from those of other oil sources.
unconventional is, of course, time specific. Before Shale oil requires lower sunk costs than conventional
being included in what is now known as conventional oil, and the lag between initial investment and produc-
sources, heavy oil and deepwater oil were considered tion is much shorter.
unconventional sources. New sources of oil are part of Deepwater and ultradeepwater oil result from off-
a continuum of oil sources that is evolving thanks to shore production activities that take place at depths of
improvements in recovery techniques. For this reason, more than 125 meters and 1,500 meters, respectively.
and to give a historical perspective on how these new As mentioned, successive improvements in drilling
sources have evolved and contributed to the transfor- techniques have allowed for drilling much farther from
mation of the oil market, this feature adopts an all- coastlines and much deeper. The type of offshore rig
encompassing definition of unconventional sources.4 used for ultradeepwater oil drilling activities is very
Oil sands are either loose sands or partially con- different from the type used for deepwater drilling.
solidated sandstone containing a naturally occurring Ultradeepwater rigs are partially submerged in water
and can involve dynamic positioning systems or can
2The be drill shipsself-propelled offshore drilling rigs that
focus of this feature is on oil, here referring to liquids includ-
ing crude oil, condensate, and natural gas liquids. can work beyond a depth of 3,000 meters. Although it
3See Kleinberg (forthcoming) for a discussion of unconventional
is a high-fixed-cost activity, ultradeepwater drilling can
sources. deliver a steady stream of oil for a very long period,
4Unless indicated otherwise, unconventional oil sources refer to

the broader definition rather than the narrower (contemporaneous) which makes these assets attractive to major interna-
definition of unconventional oil sources. tional oil corporations.

International Monetary Fund | April 2017 55


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.SF.2. Unconventional Oil, Proven Reserves, and Production, 2016

Production Reserves
(Thousands of (Billions of barrels)
barrels a day) No known reserves
Less than 50 Less than 0.5
50100 0.52.0
More than 100 2.05.0
5.010.0
More than 10.0

Sources: Rystad Energy research and analysis; and IMF staff calculations.
Note: Production and reserves include oil sands, heavy, extra heavy, tight and shale, deepwater, and ultradeepwater oil. A proven reserve is one with a greater-than-
90 percent probability that the resource is recoverable and economically protable. Deepwater is dened at 1251,500 meters. Ultradeepwater is dened at 1,500
meters and above. When deepwater (or ultradeepwater) production was also categorized as heavy (or extra heavy) oil, the production was counted once, as
deepwater (or ultradeepwater). Oil refers to crude oil, condensate, and natural gas liquids.

Where Are the Production and Reserve Centers? techniques in the form of investment in exploration
Production and reserve centers for unconventional and extraction. Conceptually, resource economists have
sources are concentrated in a few countries. North long argued that the resource base is endogenous to
America has the highest concentration of economi- how much effort is applied to exploring resources.5
cally recoverable proven reserves and production in Knowledge about the actual geology is gained through
unconventional sources (Figure1.SF.2; Table1.SF.1). exploration efforts and constantly evolves with tech-
These consist of shale oil in the United States and nological improvements. Thus, proven reserves and
oil sands in Canada. Central and South America production are governed as much by economic and
also host significant reserves and production centers, institutional factors (above-ground factors) as by actual
comprising heavy and extra heavy oil and deep- geology (below-ground factors).
water and ultradeepwater oil resources in Brazil, Economic factors affecting the geography of
Colombia, Ecuador, and Venezuela. The remainder exploration and production include proximity to
of world reserves and production of unconventional markets and complementarities with available infra-
sources are scattered and consist mostly of heavy oil structure. These factors often lead to agglomeration
in Europe and deepwater and ultradeepwater oil in in production and in proven reserves.6 Institutional
the North Sea and West Africa. It is noteworthy that
the Middle East has the highest concentration of 5The canonical model is the exploration model developed by Pin-

conventional oil reserves and production, but has a dyck (1978) in which a social planner maximizes the present value of
the social net benefits from consumption of oil, and the reserve base
relative low level of proven reserves and production
can be replenished through exploration and discovery of new fields.
in unconventional oil. Resource exploration and discovery has been investigated either as
In addition to the actual, hard-to-observe geology, a deterministic or a stochastic process (for example, Pindyck 1978;
the high concentration of unconventional proven Arrow and Chang 1982; Devarajan and Fisher 1982).
6Moreno-Cruz and Taylor (2016) propose a spatial model of
reserves and production reflects the geography of energy exploitation that determines how the location and produc-
innovation and subsequent adoption of new recovery tivity of energy resources affect the distribution of economic activity

56 International Monetary Fund | April 2017


Gaining Momentum? COMMODITY MARKET DEVELOPMENTS AND FORECASTS

Table 1.SF.1. Unconventional Oil Production, 2016


(Million barrels a day)
Oil Sands and Shale and Tight
Country Heavy Oil Extra Heavy Oil Deepwater Ultradeepwater Oil Total
United States 0.07 0.40 0.77 0.79 7.25 9.28
Canada 0.08 2.60 - - 0.60 3.28
Brazil 0.03 0.09 1.09 1.18 - 2.39
Angola 0.00 - 1.34 0.16 - 1.50
Norway 0.02 - 1.36 - - 1.39
China 0.73 0.36 0.08 0.01 0.03 1.21
Venezuela 0.18 1.00 - - - 1.18
Nigeria 0.08 0.00 0.83 - - 0.91
Mexico 0.31 0.48 0.01 - 0.00 0.80
Azerbaijan 0.01 0.00 0.72 - - 0.74
Colombia 0.13 0.50 - - 0.00 0.63
Oman 0.12 0.30 - - 0.01 0.43
United Kingdom 0.05 - 0.29 - - 0.34
Russia 0.19 0.10 - - - 0.30
Ecuador 0.20 0.01 - - - 0.21
Malaysia 0.01 0.01 0.16 - - 0.19
Australia - 0.01 0.16 - 0.00 0.17
Equatorial Guinea - - 0.17 - - 0.17
Congo, Republic of - 0.01 0.16 - - 0.17
Indonesia 0.01 0.14 0.00 - - 0.15
Kazakhstan 0.06 0.09 - - - 0.15
Argentina 0.08 0.01 - - 0.04 0.13
Sources: Rystad Energy research and analysis; and IMF staff calculations.
Note: Deepwater is defined at 1251,500 meters. Ultradeepwater is defined at 1,500 meters and above. When deepwater (or ultradeepwater) production was also
categorized as heavy (or extra heavy) oil, the production was counted once, as deepwater (or ultradeepwater). Oilrefers to crude oil, condensate, and natural
gas liquids. Dash denotes zero production in record.

factors affecting exploration and production include state-owned ones. Difficulties with production systems
openness to foreign investment and the strength of can lead to a low propensity to produce from existing
property rights, including in subsoil assets. Arezki, reserves. To exploit unconventional sources, oil compa-
van der Ploeg, and Toscani (2016) provide empirical nies need to be able to innovate or to implement new
evidence of a causaland economically significant techniques.
relationship running from changes in market orienta- Regulatory changes also play a central role in deter-
tion to discoveries of major hydrocarbon and mineral mining whether innovation and subsequent adoption
deposits, over and above increases in resource prices of recovery techniques occur. Consider shale oil in the
and depletion. United States. Most large reserves of oiland gasin
The observed differences between known reserves shale rock in the United States have been known for
and production across countries reflect differences a long timesince as early as the1920s according to
in production efficiency. These differences can be some. Until the mid-2000s, oil extraction from shale
explained by institutional factors emanating from rock formations was thought to be too costly, if not
the ownership structure of the industry. For instance, technologically impossible. In addition to high prices
Wolf (2009) provides evidence that the structure of driven by the rapid increase in demand from emerg-
ownership in the oil sectorthat is, whether it is ing economic giants, such as China and India, the
state ownedplays a key role in determining relative advent of shale oil can also be seen as the consequence
efficiency. He finds that, everything else equal, non- of a regulatory shock in the United States. This is
state-owned oil corporations significantly outperform clear from the published forecasts of the U.S.Energy
Information Administration. The expansion of shale oil
across geographic space. They find that a novel scaling law links the extraction was aided by a landmark study conducted
productivity of energy resources to population size, while rivers and by the U.S.Environmental Protection Agency in2004,
roads effectively magnify productivity. Arezki and Bogmans (2017)
provide evidence for the role of proximity to major markets and state which found that hydraulic fracturing posed no threat
capacity in the production of fossil fuels. to underground drinking water supplies.Shortly

International Monetary Fund | April 2017 57


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.SF.3. Evolution of Research and Development uncertain, contributing to broader uncertainty about
Expenditure in Select Integrated Oil and Service Companies the global oil supply outlook.
(Billions of U.S. dollars, unless noted otherwise)

8 Research and development expenditure 120 How Have Investment and Production Evolved?
Oil price APSP (U.S. dollars a barrel, right scale)
The adage necessity is the mother of invention
7 100 illustrates the cyclical nature of technological change
(Hanlon2015). The direction of technical change has
6 80 been shown to be biased toward specific needs, depend-
ing on prevailing forces (see Acemoglu2002). In the
particular case of the oil sector, the need to address the
5 60
rapid depletion of conventional oil reserves in certain
locations, resulting in periods of high oil prices, has
4 40 fostered improvements in recovery techniques. These
episodes of high prices have been accompanied by sig-
3 20 nificant increases in research and development expen-
diture, mostly on the part of major corporationsand
at times smaller corporationsoperating in the oil
2 0
1988 92 96 2000 04 08 12 16 and gas sectors (Figure1.SF.3). The current low-price
environment provides scant incentive for research in
Sources: IMF, Primary Commodity Price System; Bloomberg L.P.; and IMF staff oil-recovery techniques. Lindholt (2015) finds that
calculations.
Note: APSP = average petroleum spot priceaverage of U.K. Brent, Dubai, and
technological improvements through research and
West Texas Intermediate, equally weighted. The companies included are Baker development activity have offset the effect of ongoing
Hughes, BP P.L.C., Chevron Corporation, ExxonMobil Corporation, The Halliburton depletion on the cost of finding and developing addi-
Company, Royal Dutch Shell plc, Total S.A., and Schlumberger Limited.
tional reserves of oil around the world. However, he
finds that when considering a longer period, depletion
generally outweighs technological progress. That result
afterward, the George W. Bush administrations could stem from the fact that technical improvements
Energy Policy Act of2005 exempted chemicals used in are cyclical while depletion is not.9
hydraulic fracturing from the Safe Drinking Water Act The so-called peak-oil hypothesis posited that oil
regulations (see Gilje, Loutskina, and Strahan2016). supply would top out in the mid-2000s, precisely the
Shale oil deposits have been identified in several moment at which the shale revolution started. In many
other countries (for example, Argentina, Australia, respects, that revolution can be viewed as an endog-
Canada, China, Mexico, Russia). However, except for enous supply response to high prices in the2000s,
Argentina and Canada, where shale oil production is hence challenging the overly pessimistic view that geo-
gearing up, regulatory obstacles and technological chal- logical factors limit supply (Arezki and others2017).10
lenges, as well as recent low oil prices, have delayed or
discouraged extraction.7 Specifically, regulatory obsta- 9For the Gulf of Mexico, Managi and others (2004, 2005, 2006),
cles are related to environmental concerns, including using microlevel data from 194798, find empirical support for
water supply quality and the need for costly tailoring the hypothesis that technological change has offset depletion for
of fracking to the more complex nature of rock in offshore oil and gas production. For the United States, Cuddington
and Moss (2001) present evidence that technological improvements
some places.8 Some countries have gone so far as to respond to instances of scarcity by analyzing the determinants of the
ban all exploration and production of shale oil. All in average finding cost for additional petroleum reserves over the period
all, the global diffusion of shale oil production remains 196790.
10High oil prices also stimulate technological change in the

energy-using sector. Aghion and others (2016) provide evidence that


firms in the auto industry tend to innovate more in clean (and
7Although the prospects for shale oil diffusing beyond the United less in dirty) technologies when they face higher fuel prices. The
States have been limited so far, shale gas production is under way in lower-for-longer oil price environment could, however, delay the
a number of countries, such as Argentina, China, and Russia. energy transition by slowing technological changeand subsequent
8See Nature Climate Change (2013) for a discussion of the pros adoptiondirected toward moving away from fossil fuel use (Arezki
and cons of fracking. and Obstfeld 2015).

58 International Monetary Fund | April 2017


Gaining Momentum? COMMODITY MARKET DEVELOPMENTS AND FORECASTS

Historically, global investment and operational Figure 1.SF.4. Historical Evolution of Global Capital and
expenditures in unconventional oil have closely Operational Expenditures
(Billions of 2016 U.S. dollars, unless noted otherwise)
followed oil price developments (Figure1.SF.4).11
During episodes of dramatic price movements, as in
thelate1970s, investment in the oil sector responded Tight/Shale Ultradeepwater
Oil sands and extra heavy oil Deepwater
promptly. In late2008 during the global financial Heavy oil Conventional
crisis, oil investment plummeted but then rebounded Oil price APSP (2016 U.S. dollars
a barrel, right scale)
in 2009 following the sharp but temporary drop in
oil prices. The2000s episode marks the most unprec-
edented increase in global capital expenditure and 800 1. Capital Expenditures 150
reflects a prolonged era of high oil prices. The rapid 700
120
increase in oil demand, especially from large emerg- 600
ing market economies, such as China and India, has 500 90
driven oil prices up and encouraged further investment 400
in tight oil formations, ultradeepwater oil, and extra 300 60
heavy oil, which were previously uneconomical at 200
30
lower oil prices. While comovement between oil prices 100
and capital expenditure in unconventional sources is 0 0
1970 75 80 85 90 95 2000 05 10 15
akin to what it is in conventional sources, expenditure
in unconventional sources embodies technological
600 2. Operational Expenditures 150
changes that contribute to changing the response of
global oil production. Shale oil requires a lower level of 500 120
sunk costs than conventional oil, and the lag between 400
initial investment and production is much shorter. 90
300
Shale oil is thus contributing to shorter and more lim-
60
ited oil price cycles (Arezki and Matsumoto2016). 200
The unprecedented increase in capital expenditure in 100 30
unconventional sources in the2000s has contributed to
these sources centrality in the global oil market. In partic- 0 0
1970 75 80 85 90 95 2000 05 10 15
ular, shale oil production growth has emerged as a major
contributor to global supply growth (Figure1.SF.5).12 Sources: IMF, Primary Commodity Price System; IMF International Financial
The rapid increase in unconventional sources also Statistics database; Rystad Energy research and analysis; and IMF staff
calculations.
contributed to the change in OPECs strategic behavior, Note: APSP = average petroleum spot priceaverage of U.K. Brent, Dubai, and
leading to the dramatic collapse in oil prices (Arezki and West Texas Intermediate, equally weighted. Capital expenditure includes
Blanchard2014). Although that abrupt decline in prices exploration costs associated with seismic and drilling wildcats or appraisal wells
to discover and delineate oil and gas elds, and all development costs related to
led to a reduction in investment and expenditure, large facilities and drilling of wells. Operational expenditure includes operational
operational efficiency gains acted as automatic stabilizers. expenses directly related to oil and gas activities. The costs are estimated at
asset level and calibrated against company reported values. Deepwater is dened
The downward shift in the cost structure induced by at 1251,500 meters. Ultradeepwater is dened at 1,500 meters and above.
lower oil prices is partly temporary. A commonly held When deepwater (or ultradeepwater) production was also categorized as heavy
(or extra heavy) oil, the production was counted once, as deepwater (or
belief is that the cost structurewhich is often proxied ultradeepwater). Oil refers to crude oil, condensate, and natural gas liquids.
by the break-even price (the price at which it is eco-
nomical to produce a barrel of oil)is constant and
driven by immutable factors, such as the nature of the
oil extracted and the associated geology (Figure1.SF.6). In practice, the cost structure depends on a host of
factors, including technological improvements and the
11Investment and oil price series are deflated using a price index extent of learning by doing, which will reduce costs
of private fixed investment in mining and oilfield machinery in the permanently. In instances such as the recent dramatic
United States obtained from the Bureau of Economic Analysis website. drop in prices, break-even prices have moved down-
12In 2016, shale oil added 7.9 mbd in a market of 96 mbdthat

is, 4.4 mbd in crude oil, 2.7 mbd in natural gas liquids, and 0.8
ward in sync with oil prices. That shift is explained
mbd in condensate. by the operational efficiency gains stemming from the

International Monetary Fund | April 2017 59


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.SF.5. Growth in Unconventional World Oil Production service industrys significant reduction in margins to
and Real Oil Prices support the upstream sector. In shale oil specifically,
(Million barrels a day, unless otherwise noted)
the extraordinary resilience to the drop in oil prices
can be explained by important efficiency gains com-
4 Heavy oil Ultradeepwater 150
pounded by the fact that shale came on the scene at
Deepwater Tight/Shale
Oil sands and extra Oil price APSP (2016 the onset of an investment cycle in which learning by
heavy oil U.S. dollars a barrel, doing was important (Figure1.SF.7).13 The shale cost
3 120
right scale)
structure is likely to shift back up somewhat because
some of the efficiency gains cannot be sustained under
2 90 an expansion of oil production, while the cost of capi-
tal is expected to increase as U.S.interest rates rise.
1 60
The shift in cost structure has not been uniform
across unconventional sources. Oil sand production,
which is subject to high decommissioning costs, has
0 30 displayed continued high growth rates. However, the
lower investment in exploring new fields is expected to
affect production of oil sands down the line. Deepwa-
1 0
1970 75 80 85 90 95 2000 05 10 15 ter and ultradeepwater oil production has been subject
to active upgrading, which has made it somewhat resil-
Sources: IMF, Primary Commodity Price System; Bureau of Economic Analysis;
Rystad Energy research and analysis; and IMF staff calculations.
Note: APSP = average petroleum spot priceaverage of U.K. Brent, Dubai, and 13Figure1.SF.7 indicates that under a scenario of no cost defla-

West Texas Intermediate, equally weighted. Total world production in 2016 was tion, the oil price level required to keep shale production constant
estimated at 96.5 mbd (million barrels a day). Deepwater is dened at 1251,500 is higher than $80 a barrel. With cost deflation of about 40 percent,
meters. Ultradeepwater is dened at 1,500 meters and above. When deepwater akin to what has been observed in the recent past, the required price
(or ultradeepwater) production was also categorized as heavy (or extra heavy) oil,
level is only $40 a barrel. After having weakened production, the
the production was counted once, as deepwater (or ultradeepwater). Oil refers to
crude oil, condensate, and natural gas liquids. recent rally in oil prices has been followed by signs of recovery in
investment and production.

Figure 1.SF.6. Global Oil Supply Cost Curve and Breakeven Prices
(U.S. dollars a barrel)

120

100 Breakeven average


Cost curve 75 percent
breakeven
Onshore price
80 Onshore rest of world 74
Deepwater Russia condence
Offshore shelf interval for
62 each
60 55 57
53 54 category
49 Oil
43 sands
Onshore Middle East North American
40
Ultra shale
29
Extra heavy deepwater
Total production in 2020 from
20 oil
category as share of the total
cumulative production

0
0 10 20 30 40 50 60 70 80 90 100
Oil production (million barrels a day)

Source: Rystad Energy research and analysis.


Note: The breakeven price is the Brent oil price at which net present value equals zero, considering all future cash ows using a real discount rate of
7.5 percent. Oil refers to crude oil, condensate, and natural gas liquids.

60 International Monetary Fund | April 2017


Gaining Momentum? COMMODITY MARKET DEVELOPMENTS AND FORECASTS

Figure 1.SF.7. North American Shale Oil Wells at Different West Figure 1.SF.8. Unconventional Oil Production Outlook
Texas Intermediate Oil Prices and Cost Deation Scenarios Vintages
(Annual number of wells) (Million barrels a day, logarithmic scale)

Actual 2016 2015 2014 2013 2012


22,000 Cost deation 40 percent
30 percent 2011 2010 2009 2008 2007 2006
20,000 20 percent
18,000 10 percent Number of 2005 2004 2003 2002 2001 2000
0 percent wells needed
16,000 to achieve a
yearly growth 100 -
14,000
in production
12,000 of 1 MBD
Expected 2017 level
10,000 Non-OPEC conventional
Number of 25 -
8,000 wells needed
2016 Level to keep
6,000 production at
4,000
2,000 5-
World unconventional
0
20 30 40 50 60 70 80 90
WTI oil price, U.S. dollars a barrel

1
Source: Rystad Energy research and analysis. 2000 05 10 15 20 25 30 35 40
Note: Refers to spudded wells, dened as wells that are drilled but not extracted. At
$60/barrel, approximately 8,000 shale wells have to be drilled, with 10 percent
cost deation, to keep production at. MBD = million barrels a day; WTI = West Source: International Energy Agency.
Texas Intermediate. Note: OPEC = Organization of the Petroleum Exporting Countries. Replicated from
Wachtmeister, Henke, and Hk (2017). Dates correspond to vintages from
forecast.

ient. But again, lower investment in new fields will also nature of the processes of innovation and adoption,
tend to affect deepwater and ultradeepwater oil further owing to an interaction between below- and above-
in the future, albeit with different patterns across ground factors. All in all, the rising importance of
regions owing to below- and above-ground factors. unconventional sources in global supply is not only
changing the dynamic response of production to
prices, but also results in more uncertainty over the
What Lies Ahead? medium term.
The development of unconventional sources is Despite uncertainty about technological improve-
inherently uncertain. Uncertainty is apparent when ments and the recent OPEC agreement, rebalancing
comparing the ability to forecast unconventional oil supply in line with demand accompanied by stable
relative to conventional production (Figure1.SF.8).14,15 prices, will hinge on the prospects for unconven-
Technological improvements and their subsequent tional sources (Figure1.SF.9). The negotiated reduc-
adoptionincluding the extent of learning and spatial tion in oil production by 1.8 mbd for six months
diffusionare hard to predict. As mentioned earlier, will, in principle, help rebalance the market by the
uncertainty surrounding the development of uncon- end of2017, eliminating an excess supply currently
ventional sources is governed by the very uncertain estimated to be a little less than 1 mbd. Annual oil
demand growth, commonly projected at about 1.2
14The IEA does not provide specific forecasts for oil production
mbd, will be met by unconventional sources over
by OPEC. the next few years, mainly through resources under
15Wachtmeister, Henke, and Hk (2017) present a detailed

assessment of the production forecast prepared by the IEA using a development for deepwater and ultradeepwater oil, oil
narrower definition of unconventional oil sources. Leduc, Moran, sands, and heavy and extra heavy oil. In the absence
and Vigfusson (2013) present evidence of the rather gradual learning of shale, depletion forces and the legacy of low invest-
in futures markets.

International Monetary Fund | April 2017 61


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 1.SF.9. Unconventional Oil Outlook ment would start to kick in and push prices up sig-
(Million barrels a day) nificantly after a few years. Instead, in the new normal
for the oil market, shale oil production will be further
16 Producing Under development stimulated by a moderate price increase (Arezki and
14 Discovered Undiscovered Matsumoto2016). As a result, supply from shale
will help somewhat tame the otherwise sharp upward
12 swing in oil prices. Over the medium term, as prices
increase further, technical improvements in uncon-
10
ventional oil recovery will be reactivated, which will
8 eventually set off another price cycle.

0
2010 15 20 25 2010 15 20 25 2010 15 20 25 30
Deepwater and Oil sands, heavy oil,
Shale and tight oil
ultradeepwater extra heavy oil

Sources: Rystad Energy research and analysis; and IMF staff calculations.
Note: Deepwater is dened at 1251,500 meters. Ultradeepwater is dened at
1,500 meters and above. When deepwater (or ultradeepwater) production was
also categorized as heavy (or extra heavy) oil, the production was counted once,
as deepwater (or ultradeepwater). Oil refers to crude oil, condensate, and natural
gas liquids.

62 International Monetary Fund | April 2017


CHAPTER 1 Global Prospects and Policies

References Costinot, Arnaud, and Andrs Rodrguez-Clare.2013. Trade


Theory with Numbers: Quantifying the Consequences of
Acemoglu, Daron.2002. Directed Technical Change. Review of
Globalization. NBER Working Paper 18896, National
Economic Studies 69 (4): 781809.
Bureau of Economic Research, Cambridge, MA.
Adler, Gustavo, Romain Duval, Davide Furceri, Sinem Kili
Cuddington, John T., and Diana L. Moss.2001. Technological
elik, Ksenia Koloskova, and Marcos Poplawski-
Change, Depletion, and the U.S.Petroleum Industry. Ameri-
Ribeiro.2017. Gone with the Headwinds: Global Pro-
can Economic Review 91 (4): 113548.
ductivity. IMF Staff Discussion Note 17/04, International
Devarajan, Shantaynan, and Anthony C. Fisher.1982. Explora-
Monetary Fund, Washington, DC.
tion and Scarcity. Journal of Political Economy 90 (6): 127990.
Aghion, Philippe, Antoine Dechezlepretre, David Hemous, Ralf
Fajgelbaum, Pablo, and Amit Khandelwal.2016. Measuring the
Martin, and John Van Reenen.2016. Carbon Taxes, Path
Unequal Gains from Trade. Quarterly Journal of Economics
Dependency, and Directed Technical Change: Evidence from 131 (3): 111380.
the Auto Industry. Journal of Political Economy 124 (1): 151. Fernald, John.2014. Productivity and Potential Output before,
Ahn, JaeBin, and Romain Duval. Forthcoming. Trading with during, and after the Great Recession. NBER Working
China: New Sector-Level Evidence from Advanced Econo- Paper20248, National Bureau of Economic Research, Cam-
mies. IMF Working Paper, International Monetary Fund, bridge, MA.
Washington, DC. Feyrer, James.2007. Demographics and Productivity. Review of
Arezki, Rabah, and Olivier Blanchard.2014. Seven Ques- Economics and Statistics 89 (1): 10009.
tions about the Recent Oil PriceSlump. iMFdirect blog, Gilje, Erik P., Elena Loutskina, and Philip E. Strahan.2016.
December22. Exporting Liquidity: Branch Banking and Financial Integra-
Arezki, Rabah, and Akito Matsumoto.2016. A New Normal tion. Journal of Finance 71 (3): 115984.
for the Oil Market. iMFdirect blog, October27. Gruss, Bertrand.2014. After the BoomCommodity Prices
Arezki, Rabah and Christian Bogmans.2017. Proximity to and Economic Growth in Latin America and the Caribbean.
Markets, State Capacity and the Production of Fossil Fuels. IMF Working Paper 14/154, International Monetary Fund,
Unpublished. Washington, DC.
Arezki, Rabah, Zoltan Jakab, Douglas Laxton, Akito Matsumoto, Hanlon, W. Walker.2015. Necessity Is the Mother of Inven-
Armen Nurbekyan, Hou Wang, and Jiaxiong Yao.2017. Oil tion: Input Supplies and Directed Technical Change. Econo-
Prices and the Global Economy. IMF Working Paper 17/15, metrica 83 (1): 67100.
International Monetary Fund, Washington, DC. International Monetary Fund (IMF).2016. Small States Resil-
Arezki, Rabah and Maurice Obstfeld.2015. The Price of Oil ience to Natural Disasters and Climate ChangeRole for the
and the Price of Carbon. iMFdirect blog, December2, and IMF. IMF Policy Paper, Washington, DC.
VoxEU.org, December3. Jorda, Oscar.2005. Estimation and Inference of Impulse
Arezki, Rabah, Frederick van der Ploeg, and Frederik Tos- Responses by Local Projections. American Economic Review,
cani.2016. Shifting Frontiers in Global Resource Extraction: 95(1): 161182.
The Role of Institutions. CEPR Discussion Paper 11553, Justino, Patricia.2007. On the Links between Violent Conflict
Centre for Economic Policy Research, London. and Household Poverty: How Much Do We Really Know?
Arrow, Kenneth J., and Sheldon Chang.1982. Optimal Pricing, MICROCON Research Working Paper 1, Brighton, United
Use and Exploration of Uncertain Natural Resource Stocks. Kingdom.
Journal of Environmental Economics and Management 9 (1): .2009. Poverty and Violent Conflict: A Micro-Level
110. Perspective on the Causes and Duration of Warfare. Journal
Baldwin, Richard.2016. The Great Convergence: Information of Peace Research 46 (3): 31533.
Technology and the New Globalization. Cambridge, MA: Kleinberg, Robert L. Forthcoming. Unconventional Fossil
Belknap Press. Fuels. In M.J. Aziz and A.C. Johnson, Introduction to Energy
Besley, Timothy, and Torsten Persson.2008. Wars and State Technology: Depletable and Renewable, Wiley-VCH.
Capacity. Journal of the European Economic Association 6 Koopman, Robert, Zhi Wang, and Shang-Jin Wei.2014. Trac-
(23): 52230. ing Value-Added and Double Counting in Gross Exports.
Blattman, Christopher, and Edward Miguel.2010. Civil War. American Economic Review 104 (2): 45994.
Journal of Economic Literature 48 (1): 357. Leduc, Sylvain, Kevin Moran, and Robert Vigfusson.2013.
Cerra, Valerie, and Sweta Chaman Saxena.2008. Growth Learning about Future Oil Prices. Unpublished.
Dynamics: The Myth of Economic Recovery. American Lindholt, Lars.2015. The Tug-of-War between Resource Deple-
Economic Review 98 (1): 43957. tion and Technological Change in the Global Oil Indus-
Collier, Paul, Anke Hoeffler, and Catherine Pattillo.2004. Africas try19812009. Energy 93: 160716.
Exodus: Capital Flight and the Brain Drain as Portfolio Deci- Managi, Shunsuke, James J. Opaluch, Di Jin, and Thomas A.
sions. Journal of African Economies 13 (Supplement 2): 1554. Grigalunas.2004. Technological Change and Depletion in

International Monetary Fund | April 2017 63


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Offshore Oil and Gas. Journal of Environmental Economics United Nations (UN).2016. World Population Prospects,
and Management 47 (2): 388409. the2015 Revision. Department of Economic and Social
.2005. Technological Change and Petroleum Explora- Affairs, Population Division, New York, NY.
tion in the Gulf of Mexico. Energy Policy 33 (5): 61932. United Nations High Commissioner for Refugees
.2006. Stochastic Frontier Analysis of Total Factor (UNHCR).2016. Population Statistics Reference Database.
Productivity in the Offshore Oil and Gas Industry. Ecological New York, NY.
Economics 60 (1):20415. Voors, Maarten J., Eleonora E. M. Nillesen, Philip Verwimp,
Moreno-Cruz, Juan, and M. Scott Taylor.2016. An Energy- Erwin H. Bulte, Robert Lensink, and Daan P. Van Soest.2012.
Centric Theory of Agglomeration. NBER Working Violent Conflict and Behavior: A Field Experiment in
Paper 22964, National Bureau of Economic Research, Burundi. American Economic Review 102 (2): 94164.
Cambridge, MA. Wachtmeister, Henrik, Petter Henke, and Mikael Hk.2017.
Mueller, Hannes.2013. The Economic Costs of Conflict. Oil Projections in IEA World Energy Outlooks: Revisions,
International Growth Centre Working Paper, London. Accuracy and Uncertainty. Uppsala University Working
.2016. Growth and Violence: Argument for a per Cap- Paper, Uppsala, Sweden.
ita Measure of Civil War. Economica 83 (331): 47397. Wacziarg, Romain, and Karen Horn Welch.2008. Trade Liber-
Nature Climate Change.2013. Fracking Fracas. April25. alization and Growth: New Evidence. World Bank Economic
Pindyck, Robert S.1978. The Optimal Exploration and Review 22 (2): 187231.
Production of Nonrenewable Resources. Journal of Political Wolf, Christian.2009. Does Ownership Matter? The Perfor-
Economy 86 (5): 84161. mance and Efficiency of State Oil vs. Private Oil (1987
Rodrik, Dani.1999. Where Did All the Growth Go? External 2006). Energy Policy 37 (7): 264252.
Shocks, Social Conflict, and Growth Collapses. Journal of Yi, Kei-Mu.2003. Can Vertical Specialization Explain the Growth
Economic Growth 4 (4): 385412. of World Trade? Journal of Political Economy 111 (1): 52102.
Teulings, Coen N., and Nikolay Zubanov.2014. Is Economic .2010. Can Multistage Production Explain the
Recovery a Myth? Robust Estimation of Impulse Responses. Home Bias in Trade? American Economic Review 100
Journal of Applied Econometrics 29 (3): 497514. (1): 36493.

64 International Monetary Fund | April 2017


2
CHAPTER
ROADS LESS TRAVELED: GROWTH IN EMERGING MARKET AND
DEVELOPING ECONOMIES IN A COMPLICATED EXTERNAL ENVIRONMENT

Emerging market and developing economies have become cases. This change has taken place against a backdrop
increasingly important in the global economy in recent of fading external tailwinds, including waning potential
years. They now account for more than 75 percent of growth in advanced economies, slowdown and rebalanc-
global growth in output and consumption, almost double ing in China, and a shift in the commodity cycle that
the share of just two decades ago. The external environ- has affected commodity exporters. Together with a risk
ment has been important for this transformation. As of protectionism in advanced economies and tighter
these economies have integrated into the global economy, financial conditions as U.S.monetary policy normalizes,
terms of trade, external demand, and, in particular, these changes make for a more challenging external
external financial conditions have become increasingly environment for emerging market and developing econ-
influential determinants of their medium-term growth. omies going forward.
With potentially persistent structural shifts occurring in What are the implications of this environment for
the global economy, emerging market and developing medium-term growth in emerging market and devel-
economies may face a less supportive external environment oping economies? The still-considerable income gaps in
going forward than they experienced for long stretches of these economies vis--vis those in advanced economies
the post-2000 period. The still-considerable income gaps suggest room for catch-up and thus favorable prospects
in these economies vis--vis those in advanced economies for maintaining relatively strong potential growth in
suggest further room for catch-up, favoring their prospects emerging market and developing economies over the
of maintaining relatively strong potential growth over medium term, even if there is a persistent shift in some
the medium term. However, steady catch-up growth has key external conditions.
not been automatic in the past. Emerging market and The historical record suggests, however, that
developing economy growth has exhibited episodes of steady, sustained catch-up growth spurred by income
accelerations and reversals over time. Nevertheless, these gaps relative to advanced economies is not auto-
economies can still get the most out of a weaker growth matic (Pritchett2000; Hausmann, Pritchett, and
impulse from external conditions by strengthening their Rodrik2005; Jones and Olken2008; Berg, Ostry, and
institutional frameworks, protecting trade integration, Zettelmeyer2012). Growth across emerging market and
permitting exchange rate flexibility, and containing developing economies over time instead exhibits episodes
vulnerabilities arising from high current account deficits of accelerations and reversals, with a possible role for
and external borrowing, as well as large public debt. external conditions in influencing the patterns. Under-
standing which policies emerging market and developing
economies can deploy to maintain steady growth and
avoid reversals as the external environment becomes less
Introduction supportive is critical for improving living standards in
After a remarkable period of synchronized acceleration those economies and for lifting global growth.
in the early2000s and broad resilience immediately fol- Against this backdrop, the chapter studies how
lowing the global financial crisis, growth across emerg- country-specific external conditions affect emerging
ing market and developing economies in recent years market and developing economies medium-term
once again displays heterogeneitya mix of tapering, growth prospects (that is, over five-year horizons that
standstills, reversals, and continued strength in some smooth the influence of business cycle fluctuations)
and their likelihood of experiencing persistent accel-
The authors of this chapter are Bertrand Gruss, Malhar Nabar eration and reversal episodes. It also explores how
(team leader), and Marcos Poplawski-Ribeiro, with support from domestic policies and structural attributes influence
Felicia Belostecinic, Mitko Grigorov, Ava Yeabin Hong, and Jungjin
the impact of external conditions on the propensity
Lee, and with contributions from Patrick Blagrave, Emine Boz, Luis
Cubeddu, and Deniz Igan. to experience these episodes. After taking stock of

International Monetary Fund | April 2017 65


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

emerging market and developing economy growth external demand conditions as a whole appears
performance in recent decades and how much these to have remained broadly stable over this period,
economies income gaps have narrowed compared with demand among emerging market and developing
advanced economies, the chapter examines the follow- economies has played an increasingly powerful role.
ing main questions: Certain domestic policy settings and structural attri-
How do country-specific external demand con- butes can to some extent help offset a diminishing
ditions, external financial conditions, and terms growth impulse from less supportive external condi-
of trade influence growth patterns in emerging tions. The chapter confirms previous findings in the
market and developing economies, their likeli- literature that higher-quality legal systems and stronger
hood of experiencing accelerations or reversals, protection of property rights are associated with better
and thus how quickly they narrow income gaps medium-term growth outcomes (see Jones2016 and
vis--vis advanced economies? As emerging market references therein). Sound monetary frameworks,
and developing economies have become more financial depth, and exchange rate flexibility also
integrated into the global economy, have external enhance medium-term growth. But, crucially, the
factors become more important in shaping their chapter points to an additional role for some attri-
growth patterns over time? butes: trade integration, exchange rate flexibility, and
Which domestic policies and structural attributes strong institutions help emerging market and devel-
can help emerging market and developing econo- oping economies enhance the growth impulse from
mies get the most out of external conditions? external conditions either by increasing the likelihood
What does the current constellation of external of accelerations or by decreasing that of reversals.
conditions imply for emerging market and develop-
ing economies medium-term growth prospects and The analysis presented in this chapter focuses spe-
their ability to continue to contribute significantly cifically on the impact of the external environment on
to global growth? emerging market and developing economies medi-
um-term growth in income per capita. The external
The chapters main findings are: environment can also influence other important aspects
Country-specific external conditions have a signif- of these economies and raise associated policy chal-
icant effect on medium-term growth of emerging lenges not considered in this chapter. As documented
market and developing economies. The analysis in Chapter4 of the April2014 World Economic
establishes that variation at the country level in Outlook (WEO), external demand and financial shocks
external conditions, as well as global factors that have a quantitatively significant impact on short-term
affect all economies in a common manner during growth fluctuations in emerging market and devel-
particular intervals, matter for medium-term growth oping economies. Exposure to short-term speculative
outcomes of individual emerging market and devel- capital flows can impose costs in the form of higher
oping economies. volatility (Ostry, Loungani, and Furceri2016). Integra-
Country-specific external conditions also help tion into the global trading system also affects the way
explain the occurrence of growth accelerations and rewards of economic growth are divided across domes-
reversalsepisodes that appear to have persistent tic factors of production. As shown in Chapter3,
effects on growth outcomes in emerging market and emerging market and developing economies partici-
developing economies and their relative income gaps pation in global value chains may have contributed to
vis--vis advanced economies. lower labor income shares in these economies.
The importance of country-specific external condi- The rest of the chapter is structured as follows. It
tions for emerging market and developing econo- starts with an overview of emerging market and develop-
mies medium-term growth has increased over time, ing economy growth performance in recent decades and
particularly in the case of external financial condi- examines the role of country-specific external conditions
tions. For instance, their contribution to medium- in shaping growth patterns observed across countries
term growth has increased by about percentage and over time. It then zooms in on episodic patterns of
pointor one-third of the increase in average per emerging market and developing economy growth and
capita income growthbetween the19952004 explores the role of external conditions in affecting the
and200514 periods. While the contribution of likelihood of accelerations and reversals. The analysis

66 International Monetary Fund | April 2017


CHAPTER 2 ROaDs LEss TRavELED: GROWTh IN EMERGING MaRKET aND DEvELOpING ECONOMIEs IN a COMpLICaTED ExTERNaL ENvIRONMENT

examines how domestic policies and attributes influ- Figure 2.1. Contribution to Global Output and Consumption
ence the effect of external conditions on the likelihood Growth
(Percent)
of experiencing accelerations and reversals. Finally, the
chapter looks at the prospects for emerging market and Emerging market and developing economy growth prospects are increasingly
relevant for the global economy.
developing economy growth in the external environ-
ment they are likely to face over the medium term. Advanced economies Emerging market and
developing economies
100
Emerging Market and Developing Economy 90
Growth Performance over Time
80
In recent decades, the contribution of emerging
market and developing economies to global growth 70

of output and consumption has increased rapidly, 60


and these economies growth prospects have become
50
increasingly relevant for the entire global economy.1
Understanding how the complicated external environ- 40
ment may affect their growth prospects is therefore 30
important not just for the quest to sustain improve-
20
ments in these economies living standards, but also to
assess the overall outlook for the global economy. 10
During200008, emerging market and developing
0
economies, on average, accounted for 70percent of 197679 8089 9099 200009 1015 1621
global growth in output and consumption in purchas-
Source: IMF staff calculations.
ing-power-parity terms, nearly double their contribu- Note: Weighted averages are calculated using market exchange rates. Colored bars
tion during the1980s. After the global financial crisis, show percentage of contribution to output growth; black squares show percentage
of contribution to consumption growth.
with advanced economies experiencing a slow recovery,
emerging market and developing economies contri-
bution to global growth rose to about 80percent of
output growth and 85percent of consumption growth emerging market and developing economies, current
(see also Box1.1 of the April 2016 WEO). In market real income per capita (converted at purchasing-
exchange-rate terms, emerging market and developing power-parity exchange rates that more accurately reflect
economies accounted for close to 70 percent of global differences in the cost of living across countries) is less
output growth and just over 70 percent of global con- than half what it is in the United States. In 85percent
sumption growth during 201015 (Figure2.1). of emerging market and developing economies, real
Despite emerging market and developing economies income per worker is less than half that in the United
increasing overall importance in the global economy, States(Figure2.2).3
particularly in the2000s, income levels of individual To the extent that labor productivity growth in
countries within the group are still relatively low vis- emerging market and developing economies is in
-vis those of advanced economies.2 In 90percent of part a function of the relative productivity gap with
advanced economies (proxied by the United States),
1In this chapter, the emerging market and developing economy these large gaps in output per worker suggest that
group comprises all economies currently classified as such by the WEO there may still be significant room for catch-up
as well as those that have been reclassified as advanced since 1996
(Cyprus, Czech Republic, Estonia, Hong Kong SAR, Israel, Korea,
Latvia, Lithuania, Macao SAR, Malta, Puerto Rico, San Marino, Singa- underpins the calculations of purchasing-power-parity real income in
pore, Slovak Republic, Slovenia, Taiwan Province of China). Economies the PWT 9.0.
with populations in 2010 below 1 million according to the Penn World 3The ratios are calculated based on average real income per capita

Tables 9.0 vintage are excluded from the sample. over a five-year window, 201014, to smooth out business cycle and
2The chapter uses data on cross-country real income, factors of commodity price fluctuations that may affect the relative income lev-
production (physical and human capital; labor input), and popula- els. An important caveat is that some emerging market economies use
tion from the Penn World Tables (PWT) 9.0 vintage. See Deaton the single-deflation method to calculate real GDP, but this approach
and Aten (2017), and Inklaar and Rao (2017) for discussions on the may not fully capture relative price changes and may therefore affect
methodology of the 2011 International Comparison of Prices, which the accuracy of the calculation (Alexander and others 2017).

International Monetary Fund | April 2017 67


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 2.2. Emerging Market and Developing Economies, Figure 2.3. Distribution of Income per Capita in EMDEs in the
Relative Income in Purchasing-Power-Parity Terms 1970s and the 2010s
(Number of economies per interval) (Income per capita in PPP terms relative to the United States, percent)

Large gaps in income per worker vis--vis that of advanced economies suggest The narrowing of gaps in income per capita in EMDEs vis--vis that of advanced
there may still be signicant room for catch-up growth in EMDEs. economies has not been automatic in the past. Income gaps of several EMDEs
actually widened during 19702014.
60 1. EMDE Relative Income per Capita
30 1. Bottom Three Quintiles of EMDEs in the 1970s
50

40
20
30

20
10
10

0
10 20 30 40 50 60 70 80 90 100
0
Real income per capita relative to the U.S., 201014 average, percent 1970s 2010s

60 2. EMDE Relative Income per Worker 140 2. Top Two Quintiles of EMDEs in the 1970s

120
50
100
40
80
30
60
20 40
10 20

0 0
10 20 30 40 50 60 70 80 90 100 1970s 2010s

Real income per worker relative to the U.S., 201014 average, percent
Sources: Penn World Tables 9.0; and IMF staff calculations.
Note: The horizontal line inside each box represents the median; the upper and
Sources: Penn World Tables 9.0; and IMF staff calculations. lower edges of each box show the top and bottom quartiles; and the red markers
Note: EMDE = emerging market and developing economies; U.S. = United States. denote the top and bottom deciles of the average relative income during the
decade. EMDEs = emerging market and developing economies; PPP = purchasing
power parity.

(although some countries may be close to their own


steady-state levels and unlikely to experience further
catch-up growth).4 during the1970s (Figure2.3, panel 1). Convergence
In the past, the narrowing of income gaps has not and the narrowing of relative income gaps would have
been automatic. Other forces beyond the gap in produc- been expected to be greatest among economies in this
tivity have shaped the pattern of emerging market and group; indeed, the best performers in this group (econ-
developing economies growth. For example, consider omies in the top decile) have seen some narrowing in
the bottom three quintiles of the income distribution of income levels relative to the United States (from about
these economies in the1970sthat is, those with rela- 11percent in the1970s to about 21percent in recent
tive income per capita vis--vis the United Statesbelow years). However, the median relative income level for
the 60thpercentile of the cross-country distribution that group has in fact declined over the past four decades.
of the period-average relative income per capita levels By way of comparison, within the top two quintiles of
emerging market and developing economies relative
4Some emerging market and developing economies have been income distribution in the1970s, the median relative
experiencing a protracted slowdown in labor productivity growth income for the group has increased (Figure2.3, panel 2).
in recent years (Adler and others 2017), which would be consistent
with these economies having reached per capita income levels close The uneven record on convergence reflects time
to their steady states. variation in the speed at which emerging market

68 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

and developing economies income gaps vis--vis the Figure 2.4. Change in Real Income per Capita in EMDEs
United Stateshave narrowed over the decades (Fig- Relative to the United States over Decades
(Percentage points)
ure2.4). During the 1970s the median income gap
remained broadly unchanged as the two oil shocks Across decades there is wide variation in the pace at which EMDEs income gaps
hurt oil-importing emerging market and developing vis--vis the United States have narrowed.
economies while they lifted oil exporters income lev-
els. During the 1980s and 1990s income gaps widened 15

(that is, the median income level declined relative to


that of the United States) as emerging market and 10
developing economies suffered a lost decade (Latin
America and the Caribbean) and financial crises (Asia
and Latin America and the Caribbean). Income gaps 5

subsequently narrowed in the wake of the commodity


boom and other tailwinds in the2000s and2010s 0
(IMF2014; Chapter2 of the October2015 WEO).
However, as Box2.1 documents, regional disparities
5
remained large in some of the economies that experi-
enced relatively fast growth during that period.
It is important to note that the narrowing of emerg- 10
197079 8089 9099 200009 1014
ing market and developing economies relative income
gap with the United States during the recent period does
not reflect convergence from above: except during Sources: Penn World Tables 9.0; and IMF staff calculations.
Note: The horizontal line inside each box represents the median; the upper and
the global financial crisis, real GDP per capita in the lower edges of each box show the top and bottom quartiles; and the red markers
United States did not decline in absolute terms during denote the top and bottom deciles of the average change in relative income during
the decade. X-axis labels denote decades. EMDEs = emerging market and
the2000s and2010s. While the relatively slow growth
developing economies.
in the United States following the crisis has mechani-
cally helped faster-growing emerging market and devel-
oping economies narrow their income gaps relative to
the United States, for most of the period, this narrowing How Important Are External Conditions?
occurred in part because of exceptional tailwinds that The empirical exercise in this section defines and
supported synchronized accelerations (IMF2014). And, describes a set of external conditions for emerging mar-
in earlier periods when gaps widened, growth reversals ket and developing economies, assesses their relevance
in emerging market and developing economies appear to for medium-term growth performance in those econ-
have played an important role. The time variation in the omies, and explores how the importance of external
pace at which relative income gaps narrow and widen conditions varies across economies and over time.
therefore reflects in part the episodic nature of growth
in emerging market and developing economies, with a Country-Specific External Conditions Measures
recurrence of accelerations and reversals.
The external conditions that emerging market and
The rest of the chapter explores the role of external
developing economies face comprise a complex mix of
conditions in accounting for these patterns, building
factors that do not always move in the same direction.
on previous research that has documented the impor-
For instance, weak external demand associated with low
tance of certain aspects of external conditions for
growth in key trading partners may go hand in hand
emerging market and developing economies growth.5
with loose monetary conditions, low global interest
rates, and strong capital flows to emerging market and
5IMF (2014) demonstrates the importance of external demand
developing economies. To take this potential divergence
and terms of trade for medium-term growth in emerging market and
into account, the chapter focuses on three sets of exter-
developing economies. Jones and Olken (2008) show that growth
accelerations (upbreaks in their terminology) are associated with nal conditionsexternal demand conditions, external
increases in the trade share of GDP. Berg, Ostry, and Zettelmeyer
(2012) document a positive association between terms-of-trade ett, and Rodrik (2005) establish that very strong terms-of-trade
shocks and the duration of growth spells, while Hausmann, Pritch- realizations are associated with the onset of growth accelerations.

International Monetary Fund | April 2017 69


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

financial conditions, and terms of tradeeach of which (AnnexTable2.1.3), indicating that each dimension
can manifest itself differently for individual countries. potentially exerts an influence separate from the other
Country-specific metrics of these external conditions are two. Moreover, the country-specific measures of external
constructed to capture the specificities of the global con- conditions often deviate considerably from their corre-
text for each emerging market and developing economy, sponding global variables, suggesting that idiosyncratic
while at the same time being largely exogenous from the variation is an important driver of the variability in
point of view of each individual economy.6 external conditions at the level of individual economies
Country-specific external demand conditions are mea- (AnnexFigure2.1.1). For instance, the time-varying
sured by the export-weighted growth rate of domes- correlation of individual economies external demand
tic absorption of trading partners, along the lines conditions with aggregate world output growth shows
of Arora and Vamvakidis(2005) and IMF(2014). that idiosyncratic external conditions often deviate
Each countrys external demand measure is further significantly from the average external conditions faced
decomposed to capture external demand conditions by all countries (AnnexFigure2.1.1, panel 1). In
by three groups of trading partnersChina, other turn, external financial conditions exhibit, not sur-
emerging market and developing economies (exclud- prisingly, a strong role for the common factor at the
ing China), and advanced economies. regional level.8 By restricting the set of related econo-
Country-specific external financial conditions are prox- mies to those within the same geographical region, the
ied by a quantity-based measure of capital flows to country-specific measure nonetheless shows substantial
peer economies (other emerging market and develop- variability. This is evident in the relatively wide varia-
ing economies within the same region) as a share of tion in the correlation of individual economies external
their aggregate GDP (constructed to be exogenous to financial conditions with aggregate capital flows to
each country along the lines of Blanchard, Adler, and emerging market and developing economies (AnnexFig-
de Carvalho Filho2015). A quantity-based metric is ure2.1.1, panel 2). The correlation of changes in
used to better capture the fluctuations in availability commodity terms of trade with that of oil prices or
of diverse financial flows ranging from direct invest- aggregate commodity prices also varies substantially
ment to cross-border bank lending. These fluctua- across countries (AnnexFigure2.1.1, panel 3).
tions may be missed if price-based proxies for external
financial conditions are used, such as those calculated
from a narrow set of global interest rates. Establishing the Importance of External Conditions
Country-specific changes in the terms of trade are Have external conditions had a persistent, medium-
based on international commodity prices as in term impact on income per capita growth in emerging
Gruss2014 and Chapter2 of the October2015 market and developing economies? And how has the
WEO to ensure that they are exogenous from the importance of external conditions as a whole, and each
perspective of each economy. The country-specific one in particular, evolved over time and across groups
commodity terms of trade index is constructed by of countries?
weighting international prices of individual com- To answer these questions, this section follows the
modities according to the share of net exports of approach of Arora and Vamvakidis (2005); Caldern,
each commodity in GDP. This index provides an Loayza, and Schmidt-Hebbel (2006); and Box4.1
indication of the income windfall gains and losses in Chapter4 of the April2014 WEO to estimate
(as a share of GDP) associated with changes in those a standard growth regression over19702014 for a
prices for both commodity exporters and importers.7 broad sample of more than 80 emerging market and
developing economies (Annex2.3). The dependent
The cross-correlation between these country- variable is the growth rate of GDP per capita in
specific measures of external conditions is low purchasing-power-parity terms averaged over nonover-
6See lapping five-year windows (to smooth the influence of
Annex 2.1 for details on the construction of these measures
of external conditions. business cycles).
7The country-specific weights capture differences across countries

in the composition of commodity export and import baskets and in


the importance of commodities in the overall economy. The weights 8This is consistent with the findings of Chapter2 of the April
are predetermined, so that movements in commodity terms of trade 2016 WEO, which establish the importance of a common global
reflect exogenous changes in international prices (see Annex 2.1). factor in driving capital flows and their cycles.

70 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

The explanatory variables of interest are the coun- Figure 2.5. Elasticity of Medium-Term GDP per Capita Growth
try-specific measures of external demand conditions, in EMDEs with Respect to External Conditions
external financial conditions, and terms of trade, defined (Percentage points)
in the previous section. While the construction of the
External conditions have a signicant effect on EMDE medium-term growth. The
country-specific measures described above aims to sensitivity to external conditions has risen as EMDEs have become more integrated
capture aspects of the external environment that are into the global economy.
exogenous to the economy, for some individual cases the
measures may nevertheless be affected by growth out- 1.0 1. Full Sample Estimates
comes of the economy in question or respond to other
variables that also affect medium-term growth. A priori, 0.8

across the entire sample, there is no reason to expect 0.6


that the external conditions measures are systematically
affected by growth outcomes or by other variables that 0.4
also directly affect growth in ways that would intro-
0.2
duce reverse causality or omitted variables bias in the
estimations presented below. Nevertheless, the analysis 0.0
presented here attempts to mitigate these concerns by Demand Financial Commodity terms of trade
simultaneously including all three external conditions in
1.5 2. Rolling Window Estimates
the specifications, together with time fixed effects that
capture unobservable common factors.9 The regression 1.0
also allows for unobserved country fixed effects and
includes initial real income per capita at the start of the 0.5
period and a set of domestic variables found in the liter-
0.0
ature to be associated with medium-term growth.10
For the period as a whole, all three external con- 0.5
ditions have economically and statistically significant
effects on emerging market and developing econ- 1.0
19751994

19801999

19902009

19952014
19801999
19852004
19902009
19952014

19751994

19852004

19952014

19751994
19801999
19852004
19902009
omies medium-term growth. The coefficients are
statistically significant at the 10percent level, even
after controlling for common global factors captured Demand Financial Commodity
by the time fixed effect (Figure2.5, panel 1, and terms of trade

AnnexTable2.3.1). Specifically, Source: IMF staff calculations.


A 1percentage point increase in the growth rate Note: The gure shows the coefcient estimates from the baseline growth
of domestic absorption in trading partners is regression described in Annex 2.3. The specication includes all three external
conditions variables, country and time xed effects, and additional control
variables. The rst panel corresponds to the estimates for the whole sample period
9Additional exercises show that the results are robust to excluding (19702014), while the second panel shows estimates over subintervals. The
vertical lines denote 90 percent condence intervals. EMDEs = emerging market
key large emerging market and developing economies, using alterna-
and developing economies.
tive measures of external conditions that are less likely to be affected
by growth outcomes of the economy in question, and instru-
menting some of the external conditions variables with exogenous
variables such as interest rates from a few large advanced economies
(Annex 2.3). associated with a 0.4percentage point increase in
10Because the interest is in exploring the role of external condi-
medium-term growth, equivalent to about one-
tions rather than on assessing the contribution of all factors that
may affect medium-term growth, the domestic covariates included
fifth of the average annual growth rate of GDP per
in the regression are aimed at attenuating potential omitted variable capita in the sample. This strong effect may reflect,
bias (rather than at maximizing the share of variance explained by for example, persistent productivity gains from
the model). The country-specific measures of external conditions
economies of scale associated with a larger market
are derived from demand or financial conditions in trading partners
and from global commodity prices, so there is less of a concern of size via trade.11
omitted variable bias or endogeneity than would be the case if the
analysis were using measures of export growth or openness (which 11See Grossman and Helpman (2015) for a discussion of the

could be affected by domestic factors that directly affect per capita various potential links between integration and growth, and Ahn and
income growth). Duval (forthcoming) on the productivity gains from trade.

International Monetary Fund | April 2017 71


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

A 1percentage point increase in the ratio of capital account for less than 5percent of emerging market and
flows to GDP of emerging market and developing developing economies aggregate GDP, effectively reduc-
economies within the region raises medium-term ing the sample by about half. The coefficient on terms of
growth by 0.2percentage point. A larger volume trade is about twice as large and strongly significant when
of inflows can raise growth by, for example, easing the smallest economies are excluded (AnnexTable2.3.2),
credit rationing and reducing borrowing costs in while the coefficient on external financial conditions is
recipient economies (Box2.2).12 similar to the estimate based on the full sample. In turn,
A 1percentage point increase in commodity terms the coefficient on external demand conditions is smaller
of trade increases medium-term growth by almost and statistically insignificant in the reduced sample.
percentage point, reflecting the comovement The importance of external conditions may also
of actual and potential output with commodity change over time as, for instance, countries become
terms-of-trade windfalls (see Chapter2 of the Octo- more open to international trade (and, more recently,
ber2015 WEO).13 become more integrated with global supply chains) as
well as to cross-border capital flows (Chapter3 of the
October2015 WEO; Leigh and others2017). To trace
Has the Role of External Conditions Evolved across this evolution over time, the analysis is repeated within
Groups of Economies and over Time? subsamples. Specifically, rolling regressions are esti-
The universe of emerging market and developing mated over20-year horizons (such that each regression
economies is heterogeneous in terms of income levels, has four nonoverlapping five-year windows).14
economic size, and degree of integration with the The results of the rolling regressions indicate that
global economy. Looking within subsamples of econ- the coefficients generally increase over time as coun-
omies could shed light on whether the overall results tries become more integrated into the global economy
are affected by particular economies (for example, very (Figure2.5, panel 2). The elasticity is almost four times
large emerging market and developing economies). as large over19952014 compared with198099 in the
A first exercise along these lines examines whether case of external demand and more than twice as large in
the results reported above are driven by large emerg- the case of commodity terms of trade. The elasticity with
ing market and developing economies. The estima- respect to external financial conditions varies much less.
tion is repeated on a sample that excludes China.
Subsequently, all economies in the sample that are
members of the Group of Twenty (Argentina, Brazil, Contribution of Country-Specific External Conditions to
China, India, Indonesia, Korea, Mexico, Russia, Saudi per Capita Income Growth
Arabia, South Africa, and Turkey) are excluded from The full sample results indicate that the three external
the estimation. The coefficients for these alternative conditions considered in this chapter have collectively
samples are very similar to those for the overall sample contributed, on average, almost 2percentage points to
(AnnexTable2.3.2), suggesting that the large econo- income per capita growth over19752014 (Figure2.6,
mies are not driving the results for the entire set. panel 1). Their contribution increased from about
The baseline sample includes several very small econ- 1.7percentage points over197594 to about 2per-
omies. A natural question is to what extent the baseline centage points during the past two decades, accounting
result is representative of aggregate growth dynamics in for more than half of medium-term growth, on average,
emerging market and developing economies. A second across emerging market and developing economies
exercise repeats the estimation on a reduced sample during this latter period. In general, external conditions
that excludes the smallest economies, which collectively have been very important for growth in Latin America
and the Caribbean; the Middle East, North Africa,
12Exposureto external financial conditions does not necessarily Afghanistan, and Pakistan; and sub-Saharan Africa;
imply a loss of control over domestic financial conditions, as doc-
whereas for Asia and European emerging market and
umented in Chapter3 of the April 2017 Global Financial Stability
Report. developing economies, domestic and unaccounted-for
13A 1 percentage point change in the commodity terms of trade

index is akin to a windfall income gain of 1 percent of GDPa rela-


tively large amount. The interquartile range for the average annual 14This naturally comes at the cost of having fewer observations per

change in the commodity terms of trade index across all countries estimation, resulting in less precisely estimated coefficients, so the
and periods is 0.4 to 0.3 percent. focus of the narrative here is on comparing point estimates.

72 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Figure 2.6. Average Contribution to GDP per Capita Growth Figure 2.7. Relative Average Contribution to GDP per Capita
(Percentage points) Growth among External Conditions Variables
(Percent)
The contribution of external conditions to income per capita growth in EMDEs is
important throughout the sample period and increased somewhat during the past
External nancial conditions and intra-EMDE trade have become increasingly
two decades. External conditions appear to have been particularly important for
inuential drivers of medium-term growth in EMDEs over time.
growth in the LAC, MENAP, and SSA regions.

160 1. Contribution of Each External Variable


Demand Other common factors (Share of the contribution of all external variables, percent)
Financial Domestic and residual 140
Terms of trade Total 120 Demand Financial CTOT
100
8 1. All EMDEs 2. Asia 8 80
6 60
6
40
4 4
20
2 2 0
0 0 20
1975 80 85 90 95 2000 05 10
2 2
160 2. Contribution of External Demand by Group of Trading Partners
4 4
(Share of the contribution of the external demand variable,
140
6 6 percent)
1975 85 95 05 1975 85 95 05 120 AEs China Other EMDEs
80 90 2000 10 80 90 2000 10
100
80
8 3. LAC 4. SSA 8
60
6 6
40
4 4 20
2 2 0
1975 80 85 90 95 2000 05 10
0 0
2 2 Source: IMF staff calculations.
Note: The results are based on the estimates from the baseline growth regression
4 4 for the whole sample (see Annex 2.3). X-axis labels indicate starting year of ve-
year period. AEs = advanced economies; CTOT = commodity terms of trade; EMDEs
6 6
1975 85 95 05 1975 85 95 05 = emerging market and developing economies.
80 90 2000 10 80 90 2000 10

8 5. Emerging Europe 6. MENAP 8 factors appear to be just as important as external condi-


6 6 tions in terms of their contributions to growth.
4 4 Zooming in on the role of each external condi-
2 2
tion suggests that financial conditions, as proxied by
the intensity of gross capital inflows, are becoming
0 0
increasingly important over time. Their contribution
2 2 to medium-term growth has increased by about per-
4 4 centage pointor one-third of the increase in average
6 6 income per capita growthbetween the19952004
1975 85 95 05 1975 85 95 05
80 90 2000 10 80 90 2000 10 and200514 periods. This represents about half of the
contribution from external factors since2005up from
Source: IMF staff calculations. about one-third during19952004 (Figure2.7, panel 1).
Note: The gure shows, for each variable and period, the average contribution to Another important question regarding the shifting
tted GDP per capita growth across economies. The results are based on the
coefcient estimates from the baseline growth regression for the whole sample
role among external conditions is how Chinas growing
(see Annex 2.3). Other common factors corresponds to the estimated time xed influence in the global economy and, more generally,
effects (de-meaned). X-axis labels indicate start of a ve-year period. EMDEs = the expansion of trade among emerging market and
emerging market and developing economies; LAC = Latin America and the
Caribbean; MENAP = Middle East, North Africa, Afghanistan, and Pakistan; SSA = developing economies have affected these economies
sub-Saharan Africa.

International Monetary Fund | April 2017 73


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 2.8. Average Contribution of Terms of Trade to GDP per of external demand to GDP per capita growth in other
Capita Growth, by Groups of Economies emerging market and developing economies (up from
(Percentage points)
36percent in the late1990s).
The contribution of commodity terms of trade to medium-term growth is
While the contribution of commodity terms of trade
substantially larger for commodity exporters than for the average EMDE. to medium-term growth for the average economy in
the sample appears to be relatively small, this reflects
Commodity exporters Average EMDE the fact that the beneficial impact from higher prices
for commodity exporters is weighed down in the aver-
1.2
age by its negative impact on economies that rely on
1.0 imported commodities. The contribution of commod-
ity terms of trade to annual GDP per capita growth
0.8
is substantially larger for commodity exporters than
0.6 for the average country in the sample. It fluctuates
from about 1percentage point around the time of the
0.4
oil price shock in the late1970s and the commodity
0.2 boom in the early2000s to 0.6percentage point in
the mid-1980s (Figure2.8).
0.0
Moreover, a breakdown of the variance explained
0.2 jointly by all three external conditions suggests that,
in fact, commodity terms of trade account for a large
0.4 fraction (Figure2.9). Over the whole sample, com-
0.6
modity terms of trade account for almost 40percent of
1975 80 85 90 95 2000 05 10 the variance attributable to the three external factors,
external demand about 35percent, and external
Source: IMF staff calculations. financial conditions the remaining 25percent. The
Note: The gure shows the average contribution of commodity terms of trade to
tted GDP per capita growth across economies. Commodity exporters are dened relative contributions of each external condition to the
in Annex 2.1. The results are based on the coefcient estimates from the baseline variance of output per capita vary substantially over
growth regression for the whole sample (see Annex 2.3). X-axis labels indicate
starting year of ve-year period. EMDE = emerging market and developing time, however. The share of variance attributable to
economy. commodity terms of trade among all three external
variables over197580 was as large as 80percent, but
only about 10percent in199094.
growth outcomes. To explore how these developments In sum, the analysis in this subsection points to the
have influenced medium-term growth in emerging importance of country-specific external conditions in
market and developing economies, the decomposition influencing medium-term growth in emerging market
of the external demand measure by trading group and developing economies. These conditions have
is used instead of the aggregate external demand become more important over time as economies have
measure.15 The results show that Chinas domestic opened up to trade and became more financially inte-
absorption from2000 onward has become increasingly grated into international capital markets.
important in accounting for growth in other emerging
market and developing economies (Figure2.7, panel
2). Furthermore, the combined demand from China The Role of Common Factors
and other emerging market and developing economies Above and beyond the influence of country-specific
accounts for more than 80percent of the contribution external conditions, the shift in the contribution of
other common factors may be capturing to some extent
15While this breakdown does not separate out the role of global the influence of external conditions that are common
value chains and trade in intermediate goods (thus some of the across economies. The estimates presented above on
demand attributed to China may in fact reflect final demand from the contribution of country-specific external conditions
another country), the use of trading partner domestic absorption in
to emerging market and developing economies medi-
the construction of the external demand measure allows for a closer
mapping into final demand from the individual regions than would um-term growth could therefore be interpreted as a
have been the case had aggregate GDP been used in the calculation. lower bound on the impact of external conditions.

74 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Figure 2.9. Variance of GDP per Capita Growth Accounted for Figure 2.10. Contribution of Other Common Factors to GDP
by Each External Conditions Variable per Capita Growth and Selected Global Variables
(Share of the variance accounted for by all external variables, percent) (Percent, unless noted otherwise)

The relative importance of each country-specic external conditions variable in The increase in the contribution from other common factors to EMDE growth since
explaining growth variability across economies has varied signicantly over the early 2000s may reect a larger role of external nancial conditions and the
decades. On average, commodity terms of trade and external demand each changing nature of intra-EMDE trade linkages.
account for almost 40 percent of the variability.

8 4
Demand Financial CTOT Estimated common Aggregate capital ows
factors (right scale, to EMDEs1
100 percentage points)
6 3
Global output growth

80 4 2

2 1
60

0 0
40
2 1

20
4 2
1975 80 85 90 95 2000 05 10

0 Source: IMF staff calculations.


1975 80 85 90 95 2000 05 10 Note: Estimated common factors correspond to the estimated time xed effects
(de-meaned) from the baseline growth regression for the whole sample (see
Source: IMF staff calculations. Annex 2.3). X-axis labels indicate start year of ve-year period. EMDEs =
Note: The results are based on the estimates from the baseline growth regression emerging market and developing economies.
for the whole sample (see Annex 2.3). X-axis labels indicate starting year of ve- 1Aggregate capital ows to EMDEs are expressed as a share of their aggregate

year period. CTOT = commodity terms of trade. GDP.

The contribution of other common factors captured the association between estimated common factors and
by the time fixed effects (which includes the influence global economic activity is less clear. Economic activity
of external conditions common across economies) in advanced economies slowed during200014, largely
appears to have been relatively stable during197599, offsetting the faster growth and higher influence in the
but has increased sharply since the early2000s. global economy of large emerging market and devel-
Comparing the estimated role of common factors with oping economies. The demand implications from these
global activity and financial variables suggests that the developments are likely to be adequately captured by the
overall contribution of external conditionsand, in country-specific external demand variable. But the trans-
particular, external financial conditionsto medium- formation in trade linkages between emerging market and
term growth over the past 15 years may have been developing economies over the past few decades may have
larger than what is captured by the country-specific affected their growth through channels beyond external
external conditions variables (Figure2.10). demand. The share of value added from many emerging
The shift in the contribution of other common market and developing economies absorbed by Chinas
factors over the past few decades may reflect, in part, the final demand during the2000s increased faster than
synchronized increase of gross capital inflows to emerg- can be explained by Chinas economic growth during
ing market and developing economies.16 By contrast, that period (Box2.3). Emerging market and developing
16Given that global asset prices and capital flows to emerging growth may also indicate a change in how advanced economies
market and developing economies are affected by portfolio shifts in influence emerging market and developing economies growth, with
advanced economies, the rising importance of external financial con- the relative importance of the financial channel rising and that of the
ditions in emerging market and developing economies medium-term demand channel declining.

International Monetary Fund | April 2017 75


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

economies participation in global value chains has also Trend growth increases by at least 2percentage
increased significantly since the mid-1990s (Chapter2 points.21
of the October2016 WEO), which may have affected The level of real GDP per capita at the end of the
the efficiency of resource use and productivity growth. episode is at least as large as the maximum level
The increasing contribution of estimated common factors recorded prior to the onset of the episode (to rule
during200014 may therefore also reflect in part the out capturing the rebound from a collapse).
growth effects of changes in trade linkages among emerg-
ing market and developing economies. A fourth criterion is applied to distinguish between
persistent accelerations and those that end in a banking
How Do External Conditions Influence the crisis or growth reversal. Accelerations associated with
Occurrence of Growth Episodes? either a reversal that starts within three years of the
end of the episode, or a banking crisis (as identified
With the importance of external conditions for by Laeven and Valencia2013) that starts three years
emerging market and developing economies medi- before or after the end of the episode are labeled as
um-term growth established, this section takes a nonpersistent accelerations.
closer look at their influence on the occurrence of
In turn, a growth reversal episode is defined as an
growth accelerations and reversalsa key feature of
interval spanning five years during which the following
the growth process in several emerging market and
occurs:
developing economies.17
There is a discrete drop in the trend growth rate
such that it is at least 2percentage points lower than
Identifying Persistent Growth Acceleration and Reversal during the preceding five-year interval.
Episodes The level of real GDP per capita declines such that
Growth acceleration and reversal episodes are identi- its average during the five-year episode is lower than
fied using statistical methods similar to those employed the average during the five-year period immediately
in the literature. Along the lines of Hausmann, Pritch- preceding the episode.
ett, and Rodrik(2005), a growth acceleration episode is
defined as an interval spanning five years during which The History and Geography of Episodes
the following occur (see also Annex2.4):18,19
These filters pick up substantial variation over time
The trend growth rate of real GDP per capita
in the occurrence of growth episodes (Figure2.11).
during the period is relatively strong (at least
In total, there are 127 growth acceleration episodes in
3.5percent a year).20
the sample during19702014. Of these, 95 represent
17A persistent accelerations, and 32 represent nonper-
large volume of work has studied the occurrence and deter-
minants of episodes and structural breaks (or, alternatively, growth sistent accelerations (see AnnexTable2.4.1 for a list
regimes and spells) in the long-term time series of emerging market of country-year persistent acceleration episodes). Of
and developing economies growth. See, for example, Ben-David and the 32 nonpersistent accelerations, 12 are associated
Papell (1998); Pritchett (2000); Hausmann, Pritchett, and Rodrik
(2005); Pattillo, Poirson, and Ricci (2004); Hausmann, Rodrigues, with subsequent reversals, 18 with banking crises, and
and Wagner (2006); Jerzmanowski (2006); Jones and Olken (2008); 2 with both. The filter for reversals identifies 125 such
Reddy and Minoiu (2010); Berg, Ostry, and Zettelmeyer (2012); the episodes during19702014. (AnnexTable2.4.2 lists
April 2012 WEO; and Eichengreen, Park, and Shin (2013). The lack
of persistence in emerging market and developing economies medi- the country-year reversal episodes.)
um-term growth rates was documented by Easterly and others (1993) A closer look at the occurrence of the episodes over
and recently revisited by Pritchett and Summers (2014). time shows that accelerations picked up inthe2000s,
18Jones and Olken (2008); Berg, Ostry, and Zettelmeyer (2012); and
but were relatively rare during other decades. More
Tsangarides (2012) use an alternative statistical approach. In particular,
the latter two papers use a variant of the procedure proposed by Bai and recent decades have also seen the balance of accelerations
Perron (1998, 2003) to test for multiple structural breaks in time series shift increasingly toward the persistent kind. There was
when both the total number and the location of breaks are unknown.
19As a robustness test, an alternative interval spanning seven years
a large number of reversals in the1970s and1980s as
is used to identify episodes (Annex 2.5). oil-importing emerging market and developing econo-
20A trend growth rate of 3.5 percent a year is slightly above the

60th percentile of the distribution of the annual growth rates for the 21An increase in trend growth of 2 percentage points is about the

full sample and about the 75th percentile of the trend growth rates 75th percentile of the difference in trend growth rates between two
over five-year intervals. periods in the sample.

76 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Figure 2.11. Growth Episodes in Emerging Market and Figure 2.12. Cumulative Growth during Episodes, 19702015
Developing Economies, 19702015 (Percent change relative to the initial level of income per capita)
(Number of episodes)
While the cumulative change in income per capita during episodes is large, there is
Over time, the balance of acceleration episodes has tilted toward the persistent considerable variation across countries.
kind, and the number of reversal episodes has declined.
80 1. Persistent Accelerations
70 1. Accelerations 70 Interquartile range
Chile 2002
60 60
China 2000
Persistent Nonpersistent 50 Korea 1982
50
accelerations accelerations 40 Oman 1975
40 Slovenia 1995
30
30 20
20 10
0
10
10
0 1 0 1 2 3 4
197079 8089 90-99 200009 1015 Years from start of episode
10 2. Reversals
70 2. Reversals 0
10
60
20 Interquartile range
50 30 Hungary 1988
40 40 Mexico 1983
Philippines 1981
50
30 Qatar 1979
60 Sierra Leone 1994
20
70
10 80
1 0 1 2 3 4
0 Years from start of episode
197079 8089 90-99 200009 1015
Source: IMF staff calculations.
Source: IMF staff calculations. Note: Growth episodes are identied according to the criteria described in Annex
Note: Growth episodes are identied according to the criteria described in Annex 2.4. 2.4. For the full list of episodes, see Annex Tables 2.4.1 and 2.4.2. Interquartile
range denotes the interquartile range of the distribution of cumulative growth for
all country-year episodes.

mies suffered during the decade of high oil prices, and Middle East, North Africa, Afghanistan, and Pakistan;
other economies, particularly in Latin America and the Latin America and the Caribbean; and sub-Saharan
Caribbean, experienced severe financial crises with per- Africa (for instance, Qatar in1979, Mexico in1983,
sistent negative effects on income per capita. Reversals and Sierra Leone in1994). Asia and Europe have seen
have declined in number since then. fewer of these episodes.
Across regions, accelerations have been relatively
steady in Asia over time (including, for example, the
persistent acceleration in Korea in the beginning of Do Episodes Have Persistent Effects on Growth
the1980s and in China in the2000s; [Figure2.12, Trajectories?
panel 1]), but they have been more variable elsewhere The cumulative impact of episodes on per capita
(AnnexFigure2.4.2). It is important to note, though, income levels appears to be large, with considerable
that growth accelerations occur in all regions and are variation across country experiences. Persistent acceler-
not largely restricted to emerging market and devel- ations are associated with increases in real income per
oping economies in one or two regions of the world. capita typically ranging from 1540percent above the
Some examples include Oman in1975, Slovenia starting level before the episode (Figure2.12, panel
in1995, and Chile in2002 (Figure2.12, panel 1). 1). During reversals, real income per capita typically
Reversals, on the other hand, are more concentrated declines 530percent relative to the initial starting
geographically. They tend to occur mostly in the levelwith income drops as large as 50percent in

International Monetary Fund | April 2017 77


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 2.13. Normalized GDP per Capita during Growth some cases, such as Sierra Leone in the mid-1990s
Episodes and Their Aftermath, 19702015 (Figure2.12, panel 2).
(Scalar) Persistent accelerations and reversals also appear to
have long-lasting effects on the level of real income per
Persistent accelerations and reversals seem to have long-lasting effects on the capita beyond the span of the episode. Persistent accel-
level of income per capita beyond the span of the episode.
erations, for example, are associated with permanent
350 1. Economies Experiencing Persistent Accelerations and increases in income levels: during the two decades after
Benchmark Economies the onset of a persistent acceleration, the median level of
300
income per capita increases nearly twice as much as the
Persistent accelerations
250
Benchmark economies
median level of income per capita for economies that do
200
not experience accelerations (Figure2.13, panel 1).
Moreover, comparing persistent with nonpersistent
150 accelerations (Figure2.13, panel 2), the level of real
100 GDP per capita increases in similar fashion during
the first five years of both sets of episodes. The level
50
1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 of real GDP per capita then increases at a slower rate
Years since start of episode in the case of nonpersistent accelerations, leading to
a lower level eight years after the onset of the episode
250 2. Economies Experiencing Persistent or Nonpersistent
Accelerations
compared with that seen in the group of persistent
accelerations.
200 Persistent accelerations Reversals also appear to have persistent negative
Nonpersistent accelerations
effects on real GDP per capita, with the level not
150 returning to that attained at the start of the episode
until about 15 years after the start of the episode (Fig-
100 ure2.13, panel 3).
The persistent effects of episodes are also seen in the
50 association between cumulative income gains during
1 0 1 2 3 4 5 6 7 8 9 10 11 12
Years since start of episode
accelerations (or losses during reversals) and long-term
average growth rates (Figure2.14). Economies with
3. Economies Experiencing Reversals and Benchmark larger increases in levels of per capita income during
250
Economies persistent accelerations tend to grow faster, on average,
over the long term, while those with bigger decreases
200
in income levels during reversals tend also to witness
Reversals
Benchmark economies lower long-term average growth rates.
150

100 External Conditions during Episodes: How Different?


Before estimating the effect of external conditions
50 on the likelihood of accelerations and reversals, the
1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Years since start of episode data are examined to explore how attributes of episodes
differ from those of comparators spanning the same
Source: IMF staff calculations. time interval.22
Note: Growth episodes are identied according to the criteria described in Annex The median annual growth rate during persistent
2.4. For the full list of episodes, see Annex Tables 2.4.1 and 2.4.2. Y-axis labels
represent GDP per capita levels, normalized to be equal to 100 at t = 1. Solid
acceleration episodes in the sample is about 5.5percent
lines denote the medians. Dashed lines denote the interquartile range of the (compared with 1.7percent for comparator econo-
distribution of the normalized GDP per capita for all country-years episodes. mies not in an episode over the same period), while

22The comparison is based on a test of equality of medians, and

the results are robust to a Kolmogorov-Smirnov test of congruence


of the distribution of the variable (Chakravarti, Laha, and Roy 1967)
for the two sets of countries.

78 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Figure 2.14. Cumulative Growth Rate of Real Income per Figure 2.15. Event Analysis: Persistent Accelerations and
Capita during Episodes versus Average Growth Rate of Reversals, 19702015
Real Income per Capita during 19702015 (Percent, unless noted otherwise)
(Percent)
Differences in external conditions between economies that experience growth
The occurrence of growth episodes seems to have persistent effects on long-term episodes and those that do not suggest these conditions may play a relevant role in
income per capita growth rates. the occurrence of growth episodes.

7 1. Persistent Accelerations
Economies experiencing an episode
Average growth rate, 19702015

6 Economies not experiencing an episode


BIH
MMR
5 SGP
LAO BWA 7 1. External Demand: 2. Gross Capital Flows by 7
KOR HKG
4 Full Sample Region: Full Sample
6 (Percent of GDP) 6
3
TTO 5 *** 5
2 EST MUS
AZE 4 *** 4
1 JOR SYR
TCD 3 *** *** 3
0
0 20 40 60 80 100 120 140 160 2 2
Change in the level of income per capita (log point difference), cumulated 1 1
over the duration of acceleration episodes
0 0
Persistent Reversals Persistent Reversals
7 2. Reversals MMR accelerations accelerations
Average growth rate, 19702015

6
5 1.2 3. Change in CTOT: 4. Change in CTOT: 1.4
4 BGR
PAN Full Sample Commodity Exporters
0.9 1.0
3 ***
GAB ALB
2 0.6
IRN 0.6
1 IRQ
0.3
0 0.2
LBR COD GNB
1 MDG 0.0
KWT CAF
2 0.2
ARE 0.3 ***
3
220 200 180 160 140 120 100 80 60 40 20 0 0.6 0.6
Change in the level of income per capita (log point difference), ***
cumulated over the duration of reversal episodes 0.9 1.0
Persistent Reversals Persistent Reversals
accelerations accelerations
Source: IMF staff calculations.
Note: Growth episodes are identied according to the criteria described in Annex Source: IMF staff calculations.
2.4. For the full list of episodes, see Annex Tables 2.4.1 and 2.4.2. Data labels in Note: Each variable is measured as the average between t + 1 and t + 5, where t
the gure use International Organization for Standardization (ISO) country codes. corresponds to the onset of the episode. ***, **, and * denote signicance of an
equality test of medians at the 1, 5, and 10 percent level, respectively. The results
are robust to a Kolmogorov-Smirnov test of congruence of the distribution of the
variable for the two sets of economies. CTOT = commodity terms of trade.
the median growth rate during reversals is 3percent
(compared with 2.6percent for comparators over the
same period). points higher than for comparator economies (Fig-
External conditions during the episodes evolve ure2.15, panel 2).
differently from the comparator set not experiencing The median change in commodity terms of trade is
an episode (Figure2.15) as well as across persistent very close to zero and only marginally different between
and nonpersistent accelerations (Figure2.16). For the two sets of economies (0.2percent for persistent
persistent acceleration episodes, the median of accelerations episodes versus about 0.1percent for
trading partner growth is just above half apercent- the comparator countries), given that the full sample
age point higher than the median trading partner includes both commodity importers and exporters (Fig-
growth for comparator economies not in an episode ure2.15, panel 3). However, for commodity exporters
(Figure2.15, panel 1). The difference in medians is only (Figure2.15, panel 4), the median change in terms
statistically significant. External financingthe gross of trade is positive and significantly higher for those
capital flow into the regionis about 1.5percentage among them that experienced persistent accelerations

International Monetary Fund | April 2017 79


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Figure 2.16. Event Analysis: Persistent and Nonpersistent than for the comparator group of commodity exporters
Accelerations, 19702015 (0.9 percent and 0.1percent, respectively). The median
(Percent, unless noted otherwise) change in terms of trade is also positive and significantly
higher for those that experienced nonpersistent accelera-
While external demand and nancial conditions evolve in a similar manner in
economies experiencing persistent accelerations compared with those experiencing
tions (Figure2.16, panel 3).
nonpersistent accelerations, commodity terms of trade behave differently. For reversal episodes, trading partner growth is
almost 0.7percentage point lower than for nonepisodes
Economies experiencing an episode spanning the same time interval (Figure2.15). Capital
Economies not experiencing an episode
flows to the region for reversal episodes are also roughly
6 1. External Demand Growth 0.7percentage point lower compared with nonepisode
countries over the same period. The median change in
terms of trade for reversals is again very close to zero
4 *** *** and with no statistically significant difference between
the episode and nonepisode samples (0.10 percent and
0.08percent, respectively). However, among commod-
2 ity exporters alone, that difference becomes significant,
with commodity exporters in reversal episodes expe-
riencing a decline of about 0.75 percentage point in
0
Persistent Nonpersistent their terms of trade versus an increase of about 0.3per-
accelerations accelerations
centage point for commodity exporters that did not
2. Gross Capital Flows by Region
experience a reversal during the same period.
6
(Percent of GDP)

***
*** The Tipping Point: Do External Conditions Influence the
4
Likelihood of Experiencing Accelerations and Reversals?
To assess how external conditions affect the likeli-
2 hood of accelerations and reversals, this section reports
estimates from logit regressions (along the lines of
Hausmann, Pritchett, and Rodrik2005). The regres-
0
Persistent Nonpersistent sions are estimated with a dummy for the onset of
accelerations accelerations the identified episodes as dependent variable. Given
the challenge of accurately dating the beginning of
0.2 3. Change in Commodity Terms of Trade
the episodes, the dummy assumes the value 1 for the
periods t, t1, and t+1 of the episode (see Hausmann,
0.1
Pritchett, and Rodrik2005).
** The specifications include as independent vari-
0.0
ables the moving average of each of the three external
conditions variables between periods t and t+5.23 As an
0.1
additional control, the logit specification also includes
* country fixed effects in the baseline estimations. As
0.2 shown in Annex2.6, however, the pattern of significance
Persistent Nonpersistent
accelerations accelerations across coefficients is robust to the inclusion of additional

Source: IMF staff calculations. 23Using leading moving averages implies that the external con-
Note: Each variable is measured as the average between t + 1 and t + 5, where t
ditions variables are contemporary to the output outcome used to
corresponds to the onset of the episode. ***, **, and * denote signicance of an
equality test of medians at the 1, 5, and 10 percent level, respectively. The results identify episodes in the economy in question, raising concerns of
are robust to a Kolmogorov-Smirnov test of congruence of the distribution of the potential endogeneity. However, these variables are based on mea-
variable for the two sets of economies. sures of the external environment that are expected to be exogenous
to the economy in question. The results of the baseline and robust-
ness exercises from the linear growth model (Annex 2.3) further
suggest that the potential endogeneity of the external conditions
variables in the sample is not a serious concern.

80 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Figure 2.17. Change in the Probability of Occurrence of controls, including time fixed effects and measures of de
Growth Episodes, 19702015 jure integration and institutional variables, and controls
(Percentage points)
for the quality of the policy framework.
Figure2.17 shows the impact of a one-unit increase
External conditions inuence the growth process in EMDEs by signicantly
affecting the probability of persistent acceleration and reversal episodes. in the external conditions variable on the likelihood of
experiencing persistent accelerations, nonpersistent accel-
10 1. Persistent Accelerations and Reversals erations, and reversals. These marginal effects are derived
8
from the logit estimations presented in Annex2.5 and
6
with the external conditions evaluated at their means.
4 Accelerations
2 Reversals
In the case of accelerations, a 1percentage point
0
increase in trading partner demand evaluated at the
2 Persistent
accelerations
mean of all external conditions significantly raises the
4
probability of acceleration by 3.9percentage points
6
Demand Financial CTOT Demand Financial CTOT (Figure2.17, panel 1). Compared with the uncondi-
tional probability, this represents a near-doublingto
10 2. Persistent Acceleration and Reversal Episodes for 9.7percentof the probability of acceleration. The
Commodity Exporters
8 persistent effect of external demand conditions in
6 this instance may reflect the favorable impact of
4 higher exports on productivity growth via technol-
2 Reversals ogy upgrades and scale efficiencies associated with an
0 expansion in production.
2 Persistent In turn, a 1percentage point of GDP increase
4 accelerations in regional capital flows raises the probability of
6 persistent acceleration by 2.6percentage points,
Demand Financial CTOT Demand Financial CTOT
possibly reflecting that greater availability of funding
10 3. Persistent and Nonpersistent Accelerations facilitates investment and capital deepening (see also
8 Annex2.5).
6
Finally, an improvement in the terms of trade is not
4
significantly associated with a change in the likelihood
of persistent accelerations in the entire sample of
2
emerging market and developing economies. However,
0
there are two exceptions. First, for commodity export-
2 Persistent Nonpersistent
accelerations accelerations
ers (Figure2.17, panel 2), the increase in the terms of
4
trade is significantly associated with an increase in the
6
Demand Financial CTOT Demand Financial CTOT likelihood of persistent accelerations. This is in line
with Chapter2 of the October2015 WEO and Aslam
Source: IMF staff calculations.
Note: Estimates are from a logistic regression with a dummy for the identied
and others(2016), which find a significant effect of
episodes as dependent variable and including country xed effects and the three changes in the terms of trade on potential output.
external conditions variables. No additional controls are included in the estimates Second, for the subset of 32 nonpersistent accelerations
(see Annex 2.5). The vertical lines denote 90 percent condence intervals. EMDEs =
emerging market and developing economies. CTOT = commodity terms of trade. (Figure2.17 panel 3), the increase in the terms of
trade is significantly associated with the occurrence of
such episodes, reflecting that terms-of-trade wind-
falls may trigger accelerations with an initial surge in
growth that is not sustained over a longer horizon.24

24This finding is consistent with Collier and Goderis (2012), who

find that commodity price booms do not necessarily have positive


effects on output growth over long-term horizons.

International Monetary Fund | April 2017 81


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Reversals integrated into the global economy would be more


Turning to reversals, all three external conditions sensitive to external conditions than those that are
have a statistically significant effect on the probability relatively closed.
of a reversal (Figure2.17, panel 1). The second category includes initial conditions,
With all external conditions evaluated at the mean, such as the level of external debt and the current
a 1percentage point increase in external demand low- account balance, at the onset of the episode. A low
ers the probability of a reversal by 4percentage points level of external debt, for instance, may be associated
(about 50percent of the unconditional probability). with stronger confidence effects and thus a more
Similar patterns emerge for external financial condi- forceful response of domestic economic activity to
tions: a 1percentage point of GDP increase in capital favorable shifts in the external environment, as well
flows to the region is associated with a 2.4percentage as with stronger buffers that can smooth the impact
point decrease in the probability of a reversal. The of worsening global financial conditions (Chapter2
change in terms of trade is associated with a statisti- of the April2016 WEO).
cally significant reduction in the likelihood of reversals The third category covers aspects of the macroeco-
of 0.6percentage point. nomic policy framework, such as the exchange rate
regime, extent of monetary stability, and level of pub-
lic debt. The policy framework affects expectations of
The Role of Policies and Structural Attributes in future fundamentals, borrowing costs, and the overall
Mediating the Impact of External Conditions predictability of the economic environment. In
Although external conditions affect the likelihood turn, these factors shape firms investment decisions
of accelerations and reversals, domestic policies and and households spending on durable goodsboth
structural attributes could amplify or mitigate the critical channels that determine the persistence of the
persistence of the response of domestic activity to shifts response of domestic activity to shifts in the exter-
in external conditions. nal environment. Prudent fiscal policy, for example,
Previous research on emerging market and devel- may be associated with less crowding out of private
oping economies growth episodes has found evidence investment as public debt remains contained (Chap-
of a positive association between the duration of an ter2 of the April2016 Fiscal Monitor). It could also
episode and such attributes as macroeconomic stabil- imply larger buffers and fiscal space for a counter-
ity, quality of domestic institutions, and integration cyclical policy response to reduce the probability of
with the global economy (for example, Berg, Ostry, a persistent reversal. In addition, a flexible exchange
and Zettelmeyer2012). Greater resilience in emerg- rate regime can play an important role in adjusting to
ing market and developing economies has also been shifting external conditions by mitigating persistent
linked to improvements in policy frameworks and deviations in the real exchange rate from its equilib-
augmented policy spaceseen, for instance, in low rium level and facilitating price signals that ensure an
inflation and low public debt (Chapter4 of the Octo- efficient allocation of resources.
ber2012 WEO). Conversely, persistent declines in The fourth category represents structural factors
emerging market and developing economies growth and institutions, such as quality of governance, legal
rates (downbreaks) have been found to be associated and regulatory environment, availability of public
with increases in inflation and possibly diminished services, and level of education. These elements
monetary policy control (Jones and Olken2008). have an important bearing on long-term growth
In line with the approaches adopted in the litera- outcomes (Acemoglu, Johnson, and Robinson2001)
ture, four broad categories of domestic attributes are and could also influence, for example, how econ-
studied to examine how they influence the impact of omies respond to changes in external factors
external conditions on the likelihood of accelerations (Rodrik1999).
and reversals.
The first category of domestic attributes includes the An initial inspection of the domestic attributes
degree of de jure trade and financial integration, as comparing episodes with nonepisodes (Figure2.18)
well as domestic financial depth (as a proxy for the indicates that de jure trade integration, financial depth,
capacity to intermediate cross-border capital flows and institutional quality are significantly different
and allocate them domestically). Economies more across growth episodes and nonepisode comparators

82 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

over the same period. For example, economies experi- Figure 2.18. Domestic Attributes across Persistent
encing accelerations (reversals) have a larger (smaller) Accelerations and Reversals, 19702015
number of free trade agreements than comparator
Domestic attributes are signicantly different between economies that
economies not experiencing accelerations (reversals) experience a persistent acceleration or reversal and economies that do not.
over the same period. Similarly, economies experi-
Economies experiencing an episode
encing accelerations (reversals) have higher (lower) Economies not experiencing an episode
financial depthmeasured as the ratio of bank assets
to GDPthan comparators not experiencing accelera- 16 1. Trade Agreements 2. Bank Assets 40
14 (Number of partners) (Percent of GDP) 35
tions (reversals) over the same period.
12 *** *** 30
Some of these domestic attributes, in particular
10 *** 25
those associated with policy frameworks and struc- 8 20
tural characteristics, are likely to affect medium-term 6 15
growth outcomes in and of themselvesthat is, inde- 4 10
pendently of their effect through the impact of exter- 2 *** 5
nal conditions. Including these domestic attributes in 0 0
Persistent Reversals Persistent Reversals
the logit regressions discussed in the previous section accelerations accelerations
suggests that this is indeed the case (Annex2.6). In
0.8 3. Capital Openness 4. Current Account Balance 1.0
particular, the analysis suggests that economies with (Index number 01) (Percent of GDP)
stronger institutionsproxied by higher-quality legal 0.0
0.6
systems and better protection of property rights ***
1.0
are significantly more likely to experience persistent 0.4 ***
acceleration episodes (AnnexFigure2.6.1). The 2.0
likelihood of experiencing growth reversal episodes, 0.2
3.0
in turn, significantly decreases with the extent of
0.0 4.0
exchange rate flexibility. A sound monetary frame- Persistent Reversals Persistent Reversals
work and domestic financial depth are significantly accelerations accelerations
associated with a higher likelihood of persistent
1.2 5. Exchange Rate 6. Public Debt 70
acceleration episodes and lower likelihood of growth (Index number 01) (Percent of GDP)
1.0 60
reversal episodes. Trade and financial openness and
*** 50
initial conditions in themselves are not found to 0.8
*** 40
significantly affect the probability of experiencing a 0.6 ** 30
sustained shift in growthalthough they may affect 0.4 20
how external conditions influence the occurrence of 0.2 10
episodes, as explored below.
0.0 0
Persistent Reversals Persistent Reversals
accelerations accelerations
How Do Domestic Attributes Affect the Influence of 7. Regulation 8. Legal System,
10 10
External Conditions on Growth Episodes? (Index number 010) Property Rights
8 (Index number 010) 8
As already established in the previous section, exter-
***
nal conditions influence the likelihood of accelerations 6 *** *** 6
and reversals. This section examines whether this sen- 4
***
4
sitivity depends on domestic attributes. More precisely,
it explores whether a change in each domestic attribute 2 2

leads to an additional increase in the likelihood of an 0 0


Persistent Reversals Persistent Reversals
acceleration for a given impulse from external con- accelerations accelerations
ditions, an additional decrease in the likelihood of a
reversal, or both. Source: IMF staff calculations.
Note: Each variable is measured as the average between t 3 and t 1, where t
Results from the logit regressions confirm the corresponds to the onset of the episode. ***, **, and * denote signicance of an
role played by several of these domestic attributes in equality test of medians at the 1, 5, and 10 percent level, respectively. The results
are robust to a Kolmogorov-Smirnov test of congruence of the distribution of the
influencing the marginal effect of external conditions variable for the two sets of economies.

International Monetary Fund | April 2017 83


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

on episode probabilities. The exercise examines how increases from the 25th to the 75thpercentile in
shifting each domestic attribute from its 25thpercen- the sample, a 1percentage point increase in external
tile (low quality) to its 75thpercentile (high quality) demand raises the probability of an acceleration by 3
within the estimation sample changes the marginal additionalpercentage points.
effect of external conditions, which are evaluated at Financial development helps emerging market and
their medians.25 Each domestic attribute is measured developing economies benefit from favorable financial
as the moving average of the variable during the three conditions. For instance, supportive external financial
years preceding the onset of the episode to minimize conditions (an increase in capital inflows to the region
concerns that the attributes are responding to changes of 1percentage point of GDP) raise the probability
in growth rates during the episode.26 of accelerations by 6.6percent in economies at the
75thpercentile of financial development compared
Integration and Domestic Absorptive Capacity with 4.5percent in economies at the 25thpercen-
The analysis suggests that demand from trading tile, and the difference is statistically significant.28
partners has a stronger growth impact in emerging Deeper financial systems also further reduce, for a
market and developing economies that are de jure given impulse from external financial conditions, the
more open to international trade. Likewise, a given probability of reversals, although by only percent-
loosening of external financial conditions is more age point.
likely to result in sustained growth when these econ- Sound credit growththat is, avoiding credit
omies impose fewer restrictions on capital mobility boomsis associated with stronger growth outcomes
and the domestic financial system is sufficiently under favorable external financial conditions.29 The
developed and sound. In other words, it channels probability of a persistent acceleration when external
external financing to financially constrained agents financial conditions are supportive is about 7per-
while maintaining relatively robust risk management centage points higher when domestic credit has been
and origination standards that minimize the pitfalls growing at a healthy pace as opposed to under cred-
from excessive credit growth. More specifically (Fig- it-boom conditions. The marginal effect of external
ure2.19, panel 1): financial conditions on reversals also improves (that
Deeper de jure trade integration as captured by the is, the probability of the episode decreases further) by
coverage of trade agreements increases the likelihood 2percentage points for economies that avoid exces-
that supportive external conditions lead to growth sive credit growth.
accelerations in emerging market and developing econ- Capital account openness enhances the supportive role
omies.27 For instance, when the number of partners of external financial conditions in avoiding reversals:
with which an economy has free trade agreements in more open economies, favorable external financial
conditions lower the probability of reversals 2per-
centage points more than under restrictive capital
25The logit model specification for the purpose of evaluating the

impact of domestic attributes includes one external condition at a


account settings.30 There is a trade-off, though, as the
time, the relevant domestic attribute variable (constructed as the probability of an acceleration increases less for econo-
moving average during the three years preceding the episode), the mies with more open capital accountsalthough the
interaction of these two, and country fixed effects (see Annex 2.6
change in the marginal effect is small and not statisti-
for more details). In all estimation results discussed in this section,
the marginal effects of the external conditions on the probability of cally significant.
experiencing growth episodes evaluated at the median of the external
condition and the 75th percentile of the domestic attribute are statis-
tically significant. For a discussion on how to calculate and interpret
interaction terms and their marginal effects in a logit model see, for 28Financial depth is proxied by the ratio of bank assets to GDP
example, Ai and Norton (2003). from the World Bank World Development Indicators database
26The results discussed below are those for which the marginal (Annex 2.6).
effects of the external conditions on the probability of experiencing 29An economy is considered to have sound credit growth if it has

growth episodes (evaluated at the median of the external condition not experienced credit-boom conditions, as defined in DellAriccia
and the 75thpercentile of the domestic attribute) are statistically and others (2016), during the four years preceding the episode
significant. (Annex 2.6). As noted in Sahay and others (2015), if financial deep-
27De jure trade integration is proxied by the number of trading ening proceeds too fast and is poorly regulated and supervised, it
partners with which a country has a trade agreement according to can trigger instability by encouraging excessive risk taking.
the Design of Trade Agreements database (Chapter2 of the April 30Capital account openness is based on the Quinn (1997) measure

2016 WEO and Annex 2.6). of capital account liberalization (Annex 2.6).

84 International Monetary Fund | April 2017


CHAPTER 2 ROaDs LEss TRavELED: GROWTh IN EMERGING MaRKET aND DEvELOpING ECONOMIEs IN a COMpLICaTED ExTERNaL ENvIRONMENT

Figure 2.19. Change in Marginal Effect of External Conditions When Domestic Attributes Improve
(Percentage points)

The impact of external conditions on the likelihood of growth outcomes is signicantly affected by domestic attributes. A mix of policies that protect trade
integration, permit exchange rate exibility, and reduce vulnerabilities associated with external imbalances and high levels of debt can help emerging
market and developing economies extract the most out of external conditions.

Persistent Accelerations Reversals


1. Openness and Depth
Trade Agreements Demand

Bank Assets Financial

Sound Credit Growth Financial


Capital Account
Openness Financial

3 0 3 6 9 6 3 0 3 6

2. Initial Conditions
Demand
Current Account Balance
Financial

Demand

External Debt Financial

CTOT
3 0 3 6 9 6 3 0 3 6

3. Policy Frameworks
Demand
Exchange Rate Flexibility
Financial

Demand
Public Debt
CTOT

3 0 3 6 9 6 3 0 3 6

4. Structural Characteristics

Regulation Demand

Legal Systems and Demand


Property Rights

6 0 6 12 18 12 6 0 6 12

Source: IMF staff calculations.


Note: The gure shows the change in the marginal effect of each external condition when the domestic attribute variable is evaluated at the 75th versus at
the 25th percentile of its distribution (while holding the external condition variable at its median value). Estimation results have been transformed such that
the 75th percentile represents more openness, lower levels of external and public debt, and higher exchange rate exibility. A favorable effect from the
change in the domestic attribute is represented by a positive (negative) value in the case of persistent acceleration (reversal) episodes. Solid bars denote
difference in marginal effects signicant at the 10 percent level. CTOT = commodity terms of trade.

International Monetary Fund | April 2017 85


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Initial Conditions the likelihood of episodes of sustained growth sig-


The results point to the importance of low external nificantly improvesby 3percentage pointswith
imbalances for translating favorable external conditions exchange rate flexibility. The lower impact of posi-
into positive growth outcomes (Figure2.19, panel 2): tive external demand conditions on the likelihood of
A small current account deficit significantly increases sustained growth episodes under less flexible exchange
the marginal effect of external financial conditions on rates could reflect inefficient allocation of resources and
the probability of accelerations by percentage point, low productivity growth as price signals are distorted.
while it has a negligible and statistically insignificant The trade-off is that the effect of external demand on
impact on the probability of reversals. The marginal the probability of reversals decreases less for economies
effect of better external demand conditions on the with more flexible exchange rate regimesalthough the
likelihood of an acceleration also improves signifi- change is not statistically significantpossibly reflecting
cantlyby 1percentage pointwhen the initial cur- that steeper real appreciation under favorable external
rent account deficit is small. This finding is consistent demand growth already exerts a countervailing force on
with the idea that large current account deficits are activity. Turning to financial conditions, the effect of
often associated with overheating and thus diminished exchange rate flexibility on growth outcomes is unam-
capacity for further sustained acceleration in growth biguously positive. The effect of external financial con-
as external conditions improve. The effect of demand ditions on the probability of experiencing a period of
conditions on the probability of reversals also signifi- sustained growth is about 1percentage points larger
cantly decreasesby 1percentage pointswhen the under a more flexible exchange rate regime than other-
initial current account deficit is small. wise, while the probability of a reversal decreases further
A lower level of external debt increases the likeli- and significantlyby about 2percentage points.
hood of accelerations when external demand condi- Prudent fiscal policy, as proxied by the level of public
tions, terms of trade, or external financial conditions debt to GDP, also influences the impact of external
improveby about 1 percentage points, 1percent- demand conditions on the probability of growth
age point, and percentage point, respectively.31 It episodes. The marginal effect of external demand
also increases the extent to which improvements in conditions on the likelihood of persistent accelera-
terms of trade reduce the probability of reversals. tions significantly improvesby about 1.8percentage
pointswhen public debt is low.
Policies
Structural Characteristics
The results suggest that certain policy characteris-
tics help emerging market and developing economies Other structural characteristics that have been iden-
tified in the literature as important for medium-term
experience better growth outcomes for a given impulse
growth, such as the quality of institutions and property
from external conditions. In particular, exchange rate
rights (Hall and Jones1999; Acemoglu, Johnson, and
flexibility and fiscal discipline appear to have a broadly
Robinson2001; Acemoglu and Robinson2014), are
positive influence on growth outturns, although their
also found to influence the effect of external condi-
influences vary across specific external conditions and
tions on the likelihood of favorable growth outturns
by growth episode (Figure2.19, panel 3):32
(Figure2.19, panel 4):
The exchange rate regime plays an important role in
The quality of regulation improves the impact of
influencing the impact of external demand and finan-
external demand conditions. The marginal effect of
cial conditions on the probability of growth episodes.33
external demand on accelerations increases signifi-
The marginal effect of external demand conditions on
cantlyby 8percentage pointswhen the quality of
31The measure of external debt corresponds to the stock of exter- regulation improves.34
nal debt liabilities (updated from Lane and Milesi-Ferretti 2007) as
a share of GDP. 34The indices of quality of regulation, strength of the legal system,
32While a sound monetary framework in itself has a significant and property rights protection are from Gwartney, Lawson, and
favorable effect on the likelihood of persistent acceleration and Hall (2016). Each index is based on indicators from several sources,
reversal episodes (Annex Figure2.6.1), the exercise in this section including the Global Competitiveness Report (World Economic
suggests that it does not meaningfully influence the marginal effect Forum), International Country Risk Guide (Political Risk Services
of external conditions on episode probabilities. Group), Doing Business and World Development Indicators (World
33The exchange rate regime flexibility index is a de facto index Bank), and International Financial Statistics (IMF). See Annex 2.6
based on Aizenman, Chinn, and Ito (2010). for further details.

86 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

An improvement in the quality of the legal system Figure 2.20. Actual and Projected External Conditions for
and property rights further increases the marginal effect Emerging Market and Developing Economies
(Percentage points; difference relative to the average in 201516)
of external demand on accelerations by 9percentage
points and further decreases the probability of reversals The impulse from the external environment for EMDEs is expected to be weaker, on
by 3percentage points.35 average, over the medium term compared with what they enjoyed over long
In sum, improvements in all four categories of stretches during the post-2000 period.

domestic attributes considered are typically associated


with a better growth outturn for a given impulse from Growth of trading partners Capital ows
CTOT: Exporters CTOT: Non-exporters
external conditions. The exercise in this subsection
assumes neutral external conditionsthat is, external 10
conditions variables evaluated at their sample medi-
ans. Additional analysis suggests that the beneficial 8
impact of sound domestic attributes is even larger in
a relatively worse external environment. For instance, 6
the effect that eachpercentage point of capital flows to
the region has on reducing the likelihood of a rever- 4
sal when the financial system is deep and sound and
the exchange rate is flexible, is larger when external 2
financing is scarce than when it is abundant (see
AnnexFigure2.6.2). 0

2
Taking Stock: What Does the Current
Environment Imply for Growth Prospects in 4
Emerging Market and Developing Economies? 200316 201722

The external environment has been getting more Source: IMF staff calculations.
complicated for emerging market and developing econ- Note: Projected values for country-specic external conditions variables are
constructed based on forecasts of domestic demand, gross capital inows, and
omies over the past few years. Some conditions may be commodity prices from the IMF World Economic Outlook. CTOT = commodity
less supportive in the near future, while others remain terms of trade; EMDEs = emerging market and developing economies.
highly uncertain.
On the external demand front, some of the excep-
tionally favorable conditions that emerging market with the projected pickup in growth for this group
and developing economies enjoyed over long stretches over the medium term (see Chapter1 of this WEO
during the post-2000 period are not likely to return and Box1.1 of the October2016 WEO), growth in
soon. Waning potential output growth in advanced external demand, on average, is expected to be weaker
economies will lead to weaker demand growth for during201722 than in the past (Figure2.20).
emerging market and developing economies. WEO As discussed in Chapter1, external financial
projections for advanced economy potential output conditions facing emerging market and develop-
growth have been reduced from close to 2percent ing economies are expected to gradually tighten as
(October2014 WEO) to just over 1percent (Octo- U.S.monetary policy normalizes. However, this
ber2016 WEO). An additional complication is the generalized tightening will likely be accompanied by a
risk of protectionism in some advanced economies and continued search for yield in emerging market invest-
a less favorable view of integration, as documented in ment opportunities as long as returns remain modest
Chapter2 of the October2016 WEO. While some in a low-growth environment in advanced economies.
of these effects may be offset by rising demand among Investors may therefore discriminate across emerging
emerging market and developing economies, consistent market and developing economies based on fundamen-
tals. Those with relatively stronger fundamentals may
35These
stand to benefit from capital inflows, provided that
effects possibly reflect that better institutions are also asso-
ciated with better (fiscal) policy frameworks (Rajkumar and Swaroop capital is absorbed into productive uses that sustain
2008; Lled and Poplawski-Ribeiro 2013). growth (Box2.4).

International Monetary Fund | April 2017 87


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

The third aspect of the external environment studied result largely reflects the increasingly important role
in this chaptercommodity terms of trademay played by external financial conditions. Comparing the
improve for a subset of emerging market and devel- post-2005 period with19952004, for instance, their
oping economies as commodity prices recover, but contribution to emerging market and developing econ-
the outlook remains subdued compared with the past: omies medium-term growth has increased by about
prices are expected to approach a fraction of those percentage pointor one-third of the increase in
prevailing during the boom years. average income per capita growth for the group over
Although this constellation of external conditions is this time. Furthermore, demand among emerging mar-
not necessarily adverse for emerging market and devel- ket and developing economies has exerted an increas-
oping economies, it does point to a less buoyant exter- ingly powerful force on these economies medium-term
nal environment than a few years ago. In this context, growth outcomes (even though the contribution of
the results of this chapter suggest that emerging market external demand conditions as a whole appears to have
and developing economies should expect a weaker remained broadly stable over this period).
growth impulse from external conditions. Nevertheless, External conditions also influence the growth
as the analysis demonstrates, domestic policies and process in emerging market and developing economies
structural attributes in emerging market and develop- through their effect on the probability of persistent
ing economies matter for mediating the impact of this growth acceleration and reversal episodes. In particular,
broad constellation of external forces. In particular, the a favorable impulse from external demand and finan-
results indicate that for a given impulse from external cial conditions helps medium-term growth outcomes
conditions, certain domestic policies and reforms can by making growth accelerations more likely. It also
help these economies obtain a more favorable growth reduces the likelihood of growth reversals. The impact
outturn.36 varies across groups of economies: terms-of-trade wind-
falls are particularly influential for the medium-term
growth outcomes of commodity exporters, but less so
Conclusion for the broader sample of emerging market and devel-
Emerging market and developing economies have oping economies. These ruptures matter for growth
become increasingly important in the global economy, outcomes and the evolution of living standards over
not just as centers of production but also as final des- horizons beyond the medium-term focus of this chap-
tinations for consumer goods and services. They now ter. As far as two decades after the onset of acceleration
account for more than three-fourths of global growth or reversal episodes, real income per capita still appears
in output and consumption, almost double the share to diverge from a benchmark path of economies that
of just two decades ago. Although domestic elements do not experience the episodes.
(changes to policy frameworks, structural reforms, and Although external conditions have an impact on
accumulation of factors of production) have no doubt the likelihood of accelerations and reversals, certain
been crucial for this transformation, the external envi- domestic policies and structural attributes can affect
ronment has also played an important role in shaping the response of domestic activity to shifts in exter-
these economies medium-term growth. nal conditions (in addition to directly affecting the
The evidence presented in this chapter highlights probability of growth episodes). Faced with a poten-
that country-specific externaldemand, financial, and tially less supportive external environment than in the
terms of tradeconditions are increasingly influen- past, emerging market and developing economies can
tial determinants of emerging market and developing get the most out of a weaker growth impulse from
economies growth over time as these economies external conditions by strengthening their institutional
become more integrated into the global economy. This frameworks and adopting a policy mix that protects
trade integration; permits exchange rate flexibility; and
36For instance, the impact on the probability of an acceleration
ensures that vulnerabilities stemming from high cur-
episode of trading partners demand growing by 1 percentage point
less would be almost entirely offset by opening up to trade or allow- rent account deficits and external debt, as well as high
ing the exchange rate to fluctuate more. public debt, are contained.

88 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Box 2.1. Within-Country Trends in Income per Capita: The Cases of Brazil, Russia, India, China, and
South Africa
This box examines the province-level distribution Figure 2.1.1. Decomposition of Selected
of real purchasing power parity GDP per capita in Emerging Market Economies by Province
Brazil, Russia, India, China, and South Africa, the (Real PPP GDP per capita, thousands of PPP U.S.
BRICS economies.1 Within these emerging markets, dollars)
large regional disparities remain, with some provinces
of these economies operating at per capita levels close 60 1. Brazil
to those of upper-middle- and high-income countries, 50
whereas other provinces continue to lag. 40 So Paulo Rondnia Maranho
All BRICS economies enjoyed a period of strong Brazil (average) United States
30
income growth in the early 2000s due to a period
of favorable external tailwinds (as discussed in the 20
chapter) and as some of them exited from crises. 10
The gap between their average income per capita 0
(in purchasing-power-parity adjusted U.S. dollars) 1999 2001 03 05 07 09 11 13 14
and that of the United States narrowed significantly
100 2. Russia
between 2002 and 2014. For instance, in China
Moscow Northwestern Federal District
and Russia, per capita income as a share of that in
80 North Caucasian Federal District
the United States increased by about 13 percentage Russia (average) United States
points and 26 percentage points, respectively, during 60
that period.
40
Zooming in on developments at the national level,
the analysis shows important differences in the level 20
of real income per capita across provinces within a
0
country (Figure2.1.1). The time series on individual 1999 2001 03 05 07 09 11 13 14
province-level real GDP and population data are gath-
ered from national sources. The IMF World Economic 60 3. India
Outlook purchasing-power-parity exchange rate indica-
tor is used to convert real GDP per capita in national
40
currencies to purchasing-power-parity adjusted real Goa Karnataka Bihar
GDP per capita. The transformation allows for a India (average) United States
cross-country comparison of living standards at the 20
provincial level, after adjusting for average differences
in the cost of living across countries. However, it is
0
important to bear in mind that using national averages 1999 2001 03 05 07 09 11 13 14
may overestimate the real income level in rich prov-
inces and underestimate it in poor provinces, to the Source: IMF staff calculations.
extent that there is substantial variation in prices across Note: The selected provinces within each country denote the
top, median, and bottom provinces, ordered by real PPP GDP
provinces at times. per capita for 2014. PPP adjustment is calculated using the
While income per capita in the richest provinces in base year 2010. PPP = purchasing power parity.
some BRICS economies has risen to more than half of
that in the United States (notably in Moscow, Russia,
and, to a lesser extent, So Paulo, Brazil), the poorest
The author of this box is Felicia Belostecinic. provinces are still lagging behind. In Russia, incomes
1The box uses the term province to refer to subnational
are close to seven times higher in the richest than in
administrative units immediately below the federal government,
the poorest province; in India they are 10 times higher
as is the case in China and South Africa. In Brazil and India
these units are referred to as states, and in Russia these units are in the richest than in the poorest province (also see
federal districts. Sodsriwiboon and Cashin 2017). In Brazil and China,

International Monetary Fund | April 2017 89


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 2.1 (continued)


Figure 2.1.1. Decomposition of Selected the richest province is approximately four times better
Emerging Market Economies by Province off than the poorest one.2 In South Africa, this gap is
(contd) narrowerwith the richest province two-and-a-half
(Real PPP GDP per capita, thousands of PPP U.S. times better off than the poorest.
dollars)

60 1. China

40 Tianjin Xinjiang Gansu


China (average) United States

20

0
1999 2001 03 05 07 09 11 13 14

60 2. South Africa

40
Gauteng Mpumalanga Eastern Cape
South Africa (average) United States
20

0
1999 2001 03 05 07 09 11 13 14
2So Paulo is Brazils second-richest state (after Distrito

Source: IMF staff calculations. Federal, which includes Brasilia, the nations capital). However,
Note: The selected provinces within each country denote the given that Distrito Federal is a relatively small administrative
top, median, and bottom provinces, ordered by real PPP GDP jurisdiction with a very large fraction of its population related to
per capita for 2014. PPP adjustment is calculated using the the federal government, So Paulo was used for the purpose of
base year 2010. PPP = purchasing power parity. this analysis.

90 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Box 2.2. Growing with Flows: Evidence from Industry-Level Data


Capital inflows can enhance growth in emerging the percent change in the real output of an industry
market and developing economies through various in a given country.5 Dependence on external finance is
channels: augmentation of funds available for invest- determined following Rajan and Zingales (1998).6 The
ment, transmission of crucial know-how and techno- empirical specification is:
logical diffusion, and adoption of market discipline
= + 1Sic,t1
Gict + 2CIct + 3CIct *Di
and better governance practices. Cross-country aggre-
gate data often do not allow for a clean identification + i + c + ic + t + ict. (2.2.1)
of the causal impact of capital flows on growth because
Gict is the growth of industry i in country c in
of endogeneity and reverse causality concerns. This
period t. Sic,t1 is the share of value added by each
box uses industry-level data that permit a more reliable
industry to total value added by all industries in a
identification of causal impacts of capital inflows on
country, and comes in with a one-period lag, capturing
growth. The analysis sheds light on the role played
the heterogeneous degree of importance and devel-
by the first channel, when capital inflows relax credit
opment across industries within a country over time.
constraints and reduce borrowing costs and thereby
CIct and Di denote capital inflows and external finance
stimulate growth.1
dependence. The interaction term, CI ct *Di , is the main
The empirical strategy relies on a panel-based
variable of interest in detecting whether capital inflows
fixed-effects approach that investigates whether capital
affect growth in industries that are more dependent
inflows affect growth differentially in industries that
on external finance than those that are not. Also
are more dependent on external finance. Industries
included is an expansive set of fixed effects to capture
that depend more on external finance in countries
time-invariant industry-, country-, and cross-industry
that host more capital inflows are expected to grow
cross-country factors and time-varying global factors.
disproportionately faster; relaxation of constraints
Standard errors are clustered by industry-country.7
would benefit these firms more. The analysis uses a
Integration of emerging markets into global finan-
data set covering 28 manufacturing industries in 22
cial markets has gone hand in hand with a rapid pro-
emerging market economies during 19982007.2,3
cess of industrialization in these economies, supporting
Data on total gross private capital inflows come from
the argument that international capital is important
the Institute of International Finance and are expressed
for industrialization (for example, Markusen and Ven-
in percent of GDP.4 Industry growth is computed as
ables 1999). Indeed, aggregate industry growth moves
closely with capital inflows (Figure2.2.1).
The author of this box is Deniz Igan. The analysis is based Looking across industries distinguished by their
primarily on that in Igan, Kutan, and Mirzaei (2016).
1The analysis here uses a reduced-form specification and need for external finance and their peers located
investigates the association between capital inflows and growth. in countries receiving different amounts of capital
Evidence on the intermediate step of capital inflows relaxing inflows, it is evident that industries that are more
constraints and reducing the cost of capital has been presented, dependent on external finance grow disproportionately
for instance, in Henry (2000); Harrison, Love, and McMillan faster if they are located in countries hosting more
(2004); and Bekaert, Harvey, and Lundblad (2005).
2The countries in the sample are Argentina, Brazil, Bulgaria,

Chile, China, Colombia, the Czech Republic, Ecuador, Egypt, 5Industry output data are reported in nominal U.S. dollars.

Hungary, India, Indonesia, Korea, Malaysia, Mexico, Morocco, For the analysis, the series are deflated using the producer price
Peru, Poland, Romania, Russia, South Africa, and Turkey. The index for finished goods.
results are robust to excluding China, which stands out for its 6External finance dependence aims to capture the ability of

size and transformation experience during the sample period. internally generated funds to meet investment demand, as deter-
3Industry-level data come from the United Nations Industrial mined by an industrys intrinsic technological characteristics.
Development Organizations Industrial Statistics database. The It is computed as the ratio of capital expenditures net of cash
data can be extended to 2010 with the currently available data. flow from operations to total capital expenditures using U.S.
The box focuses on the period before the global financial crisis data (based on the observation that given the relatively advanced
given that the relationship between capital inflows and industry capital markets in the United States, U.S. firms dependence
growth is markedly different during the crisis and its immediate on external funds reflects demand factors rather than supply
aftermath. See Igan, Kutan, and Mirzaei (2016) for more details. constraints).
4The results are robust to using net inflows and capital inflows 7Given that the identification strategy aims to exploit

data from the IMFs International Financial Statistics database. cross-industry differences on external finance dependence, the
The data sources are used as alternatives, with no implications for specification focuses on financial conditions and not on external
the conclusions of the analysis. demand and terms of trade.

International Monetary Fund | April 2017 91


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 2.2 (continued)


Figure 2.2.1. Capital Inows and Industry inflows (Table2.2.1). This relationship is statistically
Growth, 19982010 significant, even after purging industry and country
effects, and holds both for annual growth rates and
for growth rates calculated over three-year windows
Industry output growth (percent, right scale)
Capital inows
(Table2.2.2).
The differential effects of capital inflows on industry
growth are economically relevant. Based on the results
0.3 12
using annual growth rates, relative to less dependent
industries (at the 25th percentile level), industries
10 dependent on external finance (at the 75th percentile
0.2
level) grow about 1.58 percent faster in a country
that receives significant capital inflows (in the 75th
8
Percent of GDP

percentile) than in a country that receives only limited


0.1
foreign capital (in the 25th percentile). This accounts
6 for approximately 14 percent of the observed sample
mean of 11 percent. This relationship is driven mainly
0.0
by, and is slightly stronger for, debt flows. An industry
4
at the 75th percentile of external finance depen-
dence grows 1.71 percent faster than one at the 25th
0.1
2 percentile if it is domiciled in a country at the 75th
percentile of debt capital inflows rather than in one
at the 25th percentile. This translates to 16 percent of
0.2 0
1998 2000 02 04 06 08 10 the observed sample mean.

Sources: Institute of International Finance; United Nations


Industrial Development Organization; and IMF staff
calculations.

Table2.2.1. Industry Growth with Low versus High Levels of Capital Inflows
Economies with Low Economies with High
Capital Inflows (25th Capital Inflows (75th
percentile) percentile) Difference
Highly Dependent Industries (75th percentile) 0.08 0.12 0.04
Less Dependent Industries (25th percentile) 0.06 0.09 0.03
Difference-in-Difference 0.02 0.03 0.01
Source: IMF staff calculations.

92 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Box 2.2 (continued)

Table2.2.2. Capital Inflows and Industry Growth


Total Inflows Equity Inflows Debt Inflows
(1) (3) (5)
Annual Growth Rates, 19982007
Share (t-1) 5.002*** 5.018*** 5.009***
(5.33) (5.40) (5.33)
Capital Inflow 0.004** 0.003 0.005**
(2.52) (1.03) (2.51)
Capital Inflow * Dependence 0.008** 0.004 0.013***
(2.34) (0.73) (2.93)
Constant 0.856*** 0.853*** 0.867***
(3.75) (3.76) (3.79)
Number of Observations 4,396 4,396 4,396
R2 0.257 0.252 0.259

Growth over Three-Year Windows, 19992007


Share (t-1) 0.951* 0.956* 0.971*
(1.89) (1.90) (1.90)
Capital Inflow 0.003 0.005 0.002
(1.32) (1.42) (0.78)
Capital Inflow * Dependence 0.006* 0.004 0.011*
(1.87) (0.47) (1.93)
Constant 0.065 0.068 0.052
(0.55) (0.57) (0.42)

Number of Observations 1,570 1,570 1,570


R2 0.548 0.546 0.547
Industry Fixed Effects Yes Yes Yes
Country Fixed Effects Yes Yes Yes
Industry * Country Fixed Effects Yes Yes Yes
Period Fixed Effects Yes Yes Yes
Number of Economies 22 22 22
Number of Industries 28 28 28
Source: IMF staff calculations.
Note: ***,**, and * denote significance at the 1, 5, and 10 percent level, respectively. t-statistics are reported in parentheses.

International Monetary Fund | April 2017 93


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 2.3. The Evolution of Emerging Market and Developing Economies Trade Integration with Chinas
Final Demand
The implications for the global economy of Chinas Figure 2.3.1. Value Added in Chinas Final
rapid growth have been studied extensively in recent Demand
years (see Chapter4 of the October 2016 World (Share of country groups world total, percent)
Economic Outlook [WEO], among others). This box
explores the evolution of various emerging market and 16
developing economies integration with China over the 1995 2011
past two decades, using data on countries value added 14
in Chinas final demand.1
As a result of many years of strong growth, China 12
has accounted for a rapidly increasing share of global
10
demandthis growth alone suggests countries
exposures to China should be increasing. As such,
8
it is not surprising that the analysis indicates that
all emerging market and developing economies have 6
become more integrated with China over time (Fig-
ure2.3.1). More interesting, commodity exporters 4
and countries outside Asia have seen more substantial
gains in recent years, outpacing the gains predicted 2
by Chinas growth alone (Figure2.3.2). In addition,
the sectors of Chinas economy to which countries 0
EM Asia EM LAC EM Europe EM Other
are linked have been relatively stable over time, with
the exception of commodity-exporting countries that
Sources: Organisation for Economic Co-operation and
benefited from the increase in oil and metal prices DevelopmentWorld Trade Organization, Trade in Value
during 200510, as well as rapid infrastructure devel- Added database; and IMF staff calculations.
opment in China. Note: EM = emerging market; LAC = Latin America and the
Caribbean.
To assess countries integration with China, this
box uses data on trade in value added, which captures
the marginal contribution of a countrys domes-
tic economy to the production of a given good or China consumed only 3 percent of the nondomestic
service. These data also provide a better measure of global-value-added production of these countries in
countries ties to China than do conventional bilateral 1995, but this measure has since increased rapidly, to
trade statistics because they account for exports that about 14 percent in 2011. Still, over this time, Asian
are ultimately consumed in Chinaeven if they are countries integration with Chinas final demand has
routed through other countriesand they discount in fact merely kept pace with Chinas rising share of
goods that are exported to China but are ultimately global GDPthat is, the rising exposure of countries
reexported elsewhere (and hence are not related to in Asia to Chinas final demand is as expected, given
changes in Chinas final demand). its strong growth. For countries outside Asia, however,
As Figure2.3.2, panel 1 indicates, commodity- China has become an increasingly important source
exporting countries have experienced a rapid increase of demandby considerably more than would be
in their integration with China, but only since 2005, suggested by Chinas strong demand growth alone
likely reflecting higher commodity prices as well (Figure2.3.2, panel 2). The sharp rise in integration
as rapid growth in Chinas infrastructure develop- since 2000 indicates that this was associated with
ment. Emerging market and developing economies Chinas accession to the World Trade Organization
in Asia have strong ties to Chinas final demand in 2001, which fostered stronger trade integration
between China and countries outside the region.
The authors of this box are Patrick Blagrave and Ava Yeabin Within countries, the sectoral composition of
Hong. links with China has been quite stable over time for
1Organisation for Economic Co-operation and Development
noncommodity exporters (Figure2.3.3); although
Trade in Value Added database.

94 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Box 2.3 (continued)


Figure 2.3.2. Relative Changes in Country Figure 2.3.3. Sector Composition of Value
Exposures to Chinas Final Demand Added in Chinas Final Demand
(Share of world exposure, index 1995 = 100) (Share of total, by sector of nal demand)

800 1. Commodity versus Noncommodity Exporters 1.6


Manufacturing Commodities and
700 (excluding related
Commodity exporters 1.4 commodity related) Services
600 Noncommodity exporters
Chinas share of global Other
500 GDP 1.2
400
300 1.0

200
0.8
100
0 0.6
1995 2000 05 10 11

0.4
800 2. Asia versus Non-Asian Economies
700 Asian EMDEs 0.2
Non-Asian EMDEs
600
Chinas share of global
0.0
500 GDP Noncommodity Noncommodity
400 exporters (1995) exporters (2011)
300
Sources: Organisation for Economic Co-operation and
200
DevelopmentWorld Trade Organization, Trade in Value
100 Added database; and IMF staff calculations.
0
1995 2000 05 10 11

Sources: Organisation for Economic Co-operation and the world (Figure2.3.4, second bar from right), the
DevelopmentWorld Trade Organization, Trade in Value increase in commodity-related exports to China has
Added database; and IMF staff calculations. been much sharper relative to the benchmark of these
Note: EMDEs = emerging market and developing
economies. countries commodity-related exports to the rest of
the world. Given that this rest-of-the-world bench-
mark provides a proxy for the relative price effect on
the sectoral composition of countries value-added
rapid growth fostered tighter integration with China, exports, the larger increase in these countries com-
this integration seems to have occurred broadly modity-related exports to China is plausibly due to
similarly across sectors for this group of countries. stronger demand for these types of goods, which
However, for commodity exporters, the share of fostered increased integration.3
exports relating to commodities has risen dramat- Ultimately, greater integration with Chinas final
ically in recent years.2 Although this development demand has been a boon to many countries over the
partly reflects a shift in relative pricesgiven that past two decades. As discussed in Chapter4 of the
these data are in nominal termsstronger, relatively October 2016 WEO, Chinas recent slowdown poses
commodity-intensive demand in China also played challenges for trading partners, as this long-standing
a role. Indeed, comparing the composition of these source of demand growth slows. However, some
countries exports to China (Figure2.3.4, far-right elements of Chinas economic transitionsuch as its
bar) to the composition of their exports to the rest of

2Commodity-related sectors are chemicals and nonmetal 3From 1995 to 2011, commodity-exporting countries share of

mineral products, basic metals and fabricated metal products, commodity-related exports to China increased by 20 percentage
and mining and quarrying. points, and by 12 percentage points to the rest of the world.

International Monetary Fund | April 2017 95


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 2.3 (continued)


Figure 2.3.4. Sector Composition of move up the value chain and the prospective boost to
Commodity-Exporting Economies Foreign domestic consumption growth in coming yearswill
Value Added create opportunities for some economies, notably
(Share of total, by sector of nal demand) in emerging Asia. In addition, the increase in ser-
vices trade associated with rebalancing and Chinas
1.6 increasing investment abroad are likely to continue to
Manufacturing Commodities and produce short-term benefits for some countries in the
(excluding related
1.4
commodity
years ahead.4
Services
related) Other
1.2

1.0

0.8

0.6

0.4

0.2

0.0
Commodity Commodity Commodity Commodity
exporters exporters exporters exporters
to rest of to China to rest of to China
world (1995) (1995) world (2011) (2011)

Sources: Organisation for Economic Co-operation and 4For a discussion of the short-term costs and long-term gains
DevelopmentWorld Trade Organization, Trade in Value of Chinas transition, see Chapter4 of the October 2016 WEO,
Added database; and IMF staff calculations.
and Hong and others (2016).

96 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Box 2.4. Shifts in the Global Allocation of Capital: Implications for Emerging Market and Developing
Economies
Uphill flows, or flows from poor to rich countries, Figure 2.4.1. EMDEs: Current Account
have intensified during most of the 2000s (Rajan Balance by Group and Net Capital Inows by
2006; Prasad, Rajan, and Subramanian 2007). Basic Type
economic theory suggests that saving should flow (Percent of world GDP)
from relatively wealthy, capital-rich countries to
poorer countries where capital is scarce and profitable 1.5 1. Current Account Balance
investment opportunities should therefore be abun- 1.0
dant. However, this theory is not borne out in the 0.5
data, as highlighted by Robert Lucas in his seminal 0.0
1990 paper. Measuring total inflows by the size of 0.5
the current account deficit (the difference between
1.0
national saving and investment), advanced economies EMDEs EMDEs: Commodity
1.5 exporters
as a group received persistent and sizable net inflows
2.0 EMDEs: Other China
during the decade preceding the global financial crisis.
These inflows reflected large and growing outflows 2.5
1980 84 88 92 96 2000 04 08 12 16
from China and commodity-exporting emerging
market and developing economies (especially fuel 1.5 2. Net Capital Inows
exporters). These were in turn supported by Chi-
1.0
nas integration into the global economy, low global
0.5
interest rates, and the sharp rise in commodity prices
(Figure2.4.1, panel 1). Moreover, the capital outflows 0.0
were dominated by official reserve accumulation, 0.5
which was used to back the export-oriented growth 1.0
models of some emerging market and developing 1.5
Overall ows FDI
economies, smooth the use of the commodity wind- 2.0 Non FDI Change in reserves
falls, and self-insure against external shocks.
2.5
After the global financial crisis, however, uphill flows 1980 84 88 92 96 2000 04 08 12 15
slowed and have reversed more recently (Boz, Cubeddu,
and Obstfeld 2017). Net outflows from emerging Source: IMF staff calculations.
market and developing economies fell and reversed, as Note: EMDEs = emerging market and developing economies;
FDI = foreign direct investment.
China started to rebalance its economy toward domestic
absorption and the commodity income windfall for
commodity exporters vanished (Chapter 4 of the Octo-
ber 2016 WEO). The slowdown and eventual reversal
since 2000, and foreign direct investment has flowed
in uphill flows largely reflected movements in official
in the expected direction (Figure2.4.1, panel 2;
foreign reserves, which started registering an overall
Figure2.4.2).
decline a few years ago (Figure2.4.1, panel 2). These
Across emerging market and developing economies,
declines in foreign reserves, which are official capital
about 75 percent of countries were, on average, net
inflows, imply that private net capital inflows need not
recipients of inflows after 2000; excluding com-
match the behavior of total capital inflows and, indeed,
modity exporters, this ratio increases to about 90
some emerging market and developing economies have
percent. Moreover, although these countries net
recently experienced increased total net inflows despite
capital inflows were small in relation to world GDP,
decreased private net inflows.
their unweighted average inflow ratio to domestic
Despite these shifts in the global allocation of capi-
GDP reached as high as almost 4 percent.
tal, most emerging market and developing economies
Net foreign direct investment inflows to emerging
have consistently been net recipients of capital inflows
market and developing economies have stayed
positive throughout the post-2000 period and
The authors of this box are Emine Boz and Luis Cubeddu. have displayed far more stability than other capital

International Monetary Fund | April 2017 97


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Box 2.4 (continued)


Figure 2.4.2. Distribution of EMDEs Average Figure 2.4.3. Correlation between Capital
Current Account Balances, 200016 Flows and per Capita Real GDP Growth
(Number of economies per interval) (Correlation coefcient, 20-year rolling windows)

55 Foreign direct investment


Net ows
50 0.8

45

40
0.6
35

30

25
0.4
20

15

10 0.2
5

0
30 20 10 0 10 20 30
Current account to GDP (percent) 0.0
1990 95 2000 05 10 15

Source: IMF staff calculations. Source: IMF staff calculations.


Note: EMDEs = emerging market and developing economies.

correlation between net inflows and per capita real


account components. This stability is consistent GDP growth across around 150 emerging market
with findings of other researchers (Alfaro, Kalemli- and developing economies using 20-year rolling
Ozcan, and Volosovych 2014), who have docu- window averages is, moreover, fairly stable through-
mented that sovereign-to-sovereign flows, including out the period. In other words, countries with higher
foreign reserve accumulation, accounted for a large growth rates have tended to run smaller current
share of uphill flows, and that, apart from such account surpluses and to be net capital importers.2
flows, the data are consistent with private capital The analogous correlation has been positive for net
flowing from rich to poor countries. This result foreign direct investment flows, as well, although
is also broadly consistent with the finding that the relationship appears to have weakened over time.
nonreserve capital flows respond strongly to growth Overall, capital flows seem to have discriminated
differentials (Chapter2 of the April 2016 World among potential destinations, on average favoring
Economic Outlook). countries with higher output growth.
Capital has tended to flow somewhat more to Going forward, the overall direction of flows will
countries with higher per capita output growth, depend on the relative strength of several forces. On
which is positively correlated with labor productiv- the one hand, stronger growth and infrastructure
ity growth (Figure2.4.3).1 Although it is not clear needs in emerging market and developing economies,
which way causality runs, the data suggest a slightly as well as structural changes such as population aging
positive relationship between overall net inflows and in advanced economies, could direct excess savings
per capita output growth since 1990. The positive to emerging market and developing economies. On

1A vast amount of literature studies the drivers of capital flows 2This exercise is in the spirit of Gourinchas and Jeanne (2013),

to emerging market and developing economies and was recently who calculate a similar correlation for 19802000, but only for a
surveyed by Koepke (2015). narrower set of countries.

98 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Box 2.4 (continued)


the other hand, prospects of monetary policy norm these countries further strengthen policy frameworks
alization in advanced economies could work in the to address potential capital flow reversals triggered by
opposite direction, especially if associated with a more higher U.S. interest rates and a stronger U.S. dollar.
expansionary U.S. fiscal stance or adverse balance sheet Exchange rate flexibility in particular can help insulate
effects in emerging market and developing economies. these economies from changes in global financial
Moreover, global uncertainties remain large, not least conditions, although additional tools may be needed
because of the rising risk of protectionism, which, if at times to maintain orderly market conditions (IMF
realized, could affect emerging market and developing 2016). Moreover, as highlighted in a vast literature
economies disproportionately. In sum, a large and on the topic, robust institutions and policy frame-
persistent downhill flow of capital seems unlikely to works (Obstfeld 1998; Kose and others 2006; Ghosh,
develop over the short term. Ostry, and Qureshi 2016), including well-functioning
Reaping the benefits of capital inflows remains a domestic and international financial markets (Igan,
central challenge for emerging market and developing Kutan, and Mirzaei 2016), remain crucial to harness
economies. Meeting this challenge will require that the benefits of capital inflows.

International Monetary Fund | April 2017 99


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex 2.1. Data reclassified as advanced since 1996 (Cyprus, Czech


Data Sources Republic, Estonia, Hong Kong Special Administrative
Region, Israel, Korea, Latvia, Lithuania, Macao Spe-
The primary data sources for this chapter are the cial Administrative Region, Malta, Puerto Rico, San
IMF World Economic Outlook (WEO) database, Marino, Singapore, Slovak Republic, Slovenia, Taiwan
the Penn World Tables (version 9.0), and the World Province of China), but excludes economies with a
Bank World Development Indicators database. The population of less than 1 million in 2010 (according
chapter also uses several other databases to construct to Penn World Tables 9.0 data).
the external conditions variables and policy and other
domestic attribute variables used in the empirical anal- Data Definitions
yses. Annex Table2.1.1 lists all indicators used in the
Real GDP per Capita
chapter as well as their sources.
The sample of countries included in the various Aggregate GDP and population data used to con-
analytical exercises varies due to data constraints. struct real GDP per capita at purchasing-power-parity
Annex Table2.1.2 lists the sample of all emerging adjusted U.S. dollars are from Penn World Tables 9.0.
market and developing economies used in the various The source for aggregate GDP used to construct real
analytical exercises. It includes all emerging market GDP per capita at constant national prices is also Penn
and developing economies currently classified as World Tables 9.0, to be consistent with data used on
such by the WEO as well as those that have been production factors (labor and capital).

Annex Table2.1.1. Data Sources


Indicator Source
Banking Crisis Indicator Laeven and Valencia (2013)
Bilateral Cross-Border Bank Claims Bank for International Settlements
Capital Account Openness Quinn (1997); Aizenman, Chinn, and Ito (2010)
Capital Inflows IMF, Financial Flows Analytics database
Capital Stock Penn World Tables 9.0
Commodity Terms of Trade Gruss 2014
Commodity Export Weights United Nations Commodity Trade Statistics (Comtrade) database;
IMF, World Economic Outlook database
Credit Boom Episodes DellAriccia and others (2016)
Current Account Balance IMF, World Economic Outlook database
Deposit Money Banks' Assets Ratio to GDP (percent) World Bank, World Development Indicators database
Employment Penn World Tables 9.0
Exchange Rate Stability Index Aizenman, Chinn, and Ito (2010)
Export Value of Goods (bilateral) IMF, Direction of Trade Statistics database
External Debt Liabilities as a Share of GDP Lane and Milesi-Ferretti (2007)
Free Trade Agreements by Year of Signature of Agreement DESTA, Free Trade Area database; October 2016 World Economic Outlook
Free Trade Agreements Coverage WTO Regional Trade Agreements database; October 2016 World Economic Outlook
Human Capital Penn World Tables 9.0
Legal System and Property Rights Quality Index Gwartney, Lawson, and Hall (2016)
Nominal GDP IMF, World Economic Outlook database
Nominal Interest Rate IMF, World Economic Outlook database
Oil Price in U.S. Dollars IMF, Global Assumptions database
Polity Score (combined) Polity IV/Transparency International
Population Penn World Tables 9.0; United Nations Population database
Public Debt as a Share of GDP Mauro and others (2013); IMF, World Economic Outlook database
Real GDP at Constant National Prices IMF, World Economic Outlook database; Penn World Tables 9.0
Real GDP in Purchasing Power Parity Terms Penn World Tables 9.0
Real Domestic Absorption Penn World Tables 9.0
Regulation Quality Index Gwartney, Lawson, and Hall (2016)
Sound Monetary Framework Gwartney, Lawson, and Hall (2016)
Tariffs UNCTAD, Trade Analysis Information System; WTO Tariff Download Facility;
IMF, Structural Reforms database; October 2016 World Economic Outlook
Source: IMF staff compilation.
Note: DESTA = Design of Trade Agreements database; UNCTAD = United Nations Conference on Trade and Development; WTO = World Trade Organization.

100 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Annex Table2.1.2. Sample of Emerging Market and


China, other emerging market and developing econo-
Developing Economies Included in the Analyses
mies (excluding China), and advanced economies.
Albania, Algeria*, Angola*, Argentina, Armenia, Azerbaijan*, Bahrain*, Following Blanchard, Adler, and de Carvalho Filho
Bangladesh, Belarus, Benin, Bolivia*, Bosnia and Herzegovina,
Botswana, Brazil, Bulgaria, Burkina Faso, Burundi, Cambodia, (2015), country-specific external financial conditions are
Cameroon*, Central African Republic, Chad, Chile*, China, Colombia, measured by the ratio of capital inflows to the region
Democratic Republic of the Congo, Republic of Congo*, Costa of the economy in question (excluding inflows to that
Rica, Cte dIvoire*, Croatia, Czech Republic, Dominican Republic,
Ecuador*, Egypt, El Salvador, Estonia, Ethiopia, Gabon*, The Gambia,
economy) as a share of GDP of other economies in the
Georgia, Ghana, Guatemala, Guinea*, Guinea-Bissau, Haiti, Honduras, same region. Thus, for emerging market economy j in
Hong Kong SAR, Hungary, India, Indonesia*, Islamic Republic of year t, the external financial condition is measured by
Iran*, Iraq*, Israel, Jamaica, Jordan, Kazakhstan*, Kenya, Korea,
the ratio
Kuwait*, Kyrgyz Republic, Lao P.D.R., Latvia, Lebanon, Lesotho,
Liberia, Lithuania, FYR Macedonia, Madagascar, Malawi, Malaysia, i\jK_inflowi,t
Mali, Mauritania*, Mauritius, Mexico, Moldova, Mongolia*, Morocco, _____________


, (2.2)
Mozambique, Myanmar, Namibia, Nepal, Nicaragua, Niger, Nigeria*, i\jGDPi,t1
Oman*, Pakistan, Panama, Paraguay, Peru*, Philippines, Poland,
Qatar*, Romania, Russia*, Rwanda, Saudi Arabia*, Senegal, Serbia,
in which K_inflowi,tis gross inflows to economy i,
Sierra Leone, Singapore, Slovak Republic, Slovenia, South Africa, Sri GDPi,t1is GDP of economy i measured in U.S. dol-
Lanka, Sudan, Swaziland, Syria, Taiwan Province of China, Tajikistan*, lars, and \jis the set of all related economies (within
Tanzania, Thailand, Togo, Trinidad and Tobago*, Tunisia, Turkey,
the same region) but excluding economy j. By exclud-
Turkmenistan*, Uganda, Ukraine, United Arab Emirates, Uruguay,
Uzbekistan, Venezuela*, Vietnam, Yemen*, Zambia*, Zimbabwe ing capital flows to the economy itself and aggregating
Source: IMF staff compilation. capital flows to related economies, the measure aims to
Note: The classification of emerging market and developing economies includes capture push factors that are exogenous to the econ-
economies considered emerging markets before 1996. * denotes commodity
exporters, which are economies for which commodity exports constitute the omy in focus. While economies within a comparable
main source of export earnings during the sample period (commodity exports group naturally have an important common element,
exceed 65 percent of total exports of goods, and net commodity exports account
for at least 6 percent of GDP). there is important variation across economies, as
shown in Annex Figure2.1.1, panel 2.
The change in terms of trade is analyzed in the
chapter through commodity terms of trade (CTOT)
Country-Specific External Conditions Measures indices. These are constructed for each economy as a
The country-specific external demand condition is trade-weighted average of the prices of imported and
measured as the export-weighted domestic absorption exported commodities, following Gruss (2014). The
of trading partners, as in Arora and Vamvakidis 2005 annual change in the economy is CTOT index in year
and IMF 2014. Thus, for an emerging market econ- t is given by
omy j in year t, the growth rate of external demand
logCTOTt=j=1
J i,j,t,
logPj,t (2.3)
can be represented by

i i,t *dai,t, (2.1) in which Pj,tis the relative price of commodity j at


j
time t (in U.S. dollars and divided by the IMFs unit
in which i,tis the share of economy js exports value index for manufactured exports), and denotes
accounted for by economy i (based on IMF Direction the first difference. Economy is weights for each com-
of Trade Statistics [DOTS] data); dai,t is the annual modity price, i,j,t, are given by
growth rate of real domestic absorption in economy i
xi,j,t1 mi,j,t1
(at constant national prices, from Penn World Tables i,j,t =____________
, (2.4)
GDP i,t1
9.0); j is the set of economy js trading partners for
which bilateral export data are reported in DOTS and in which xi,j,t1 (mi,j,t1) denote the average export
collectively account for at least 50 percent of total (import) value of commodity j by the economy i
exports. The time-varying correlation of individual between t1 and t3 (in U.S. dollars, from the United
country external demand conditions with aggregate Nations Comtrade database), and GDPi,t1 denotes
world output growth shows that the external conditions the average GDP of the economy i between t1
that each faces often deviate significantly from average and t3 (in U.S. dollars). An alternative index with
external conditions (Annex Figure2.1.1). The coun- (Jj=1
xi,j,t1 +Jj=1
mi,j,t1) instead of GDP i,t1
try-specific external demand series was further decom- in equation 2.4 is used in robustness exercises in
posed into three components, capturing demand from Annex 2.3.

International Monetary Fund | April 2017 101


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Figure 2.1.1. Correlation between Country-Specic Annex Table2.1.3. Pairwise Correlation between
External Conditions Variables and Global Variables over Time External Conditions Variables
(Correlation coefcient) External External Commodity
Demand Financial Terms of
Variable Conditions Conditions Trade
1. Correlation between Country-Specic External Demand External Demand Conditions 1
Conditions and World GDP Growth External Financial Conditions 0.1288 1
1.0
Commodity Terms of Trade 0.0737 0.0016 1
Source: IMF staff calculations.
0.5

0.0
The commodity price series start in 1960. Prices of 41
0.5 commodities are used, sorted into four broad categories:
1. Energy: coal, crude oil, and natural gas
1.0
197079 8089 9099 200009 1014 2. Metals: aluminum, copper, iron ore, lead, nickel,
tin, and zinc
2. Correlation between Country-Specic External Financial 3. Food: bananas, barley, beef, cocoa, coconut oil,
Conditions and Aggregate Capital Flows to EMDEs coffee, corn, fish, fish meal, groundnuts, lamb,
1.0
oranges, palm oil, poultry, rice, shrimp, soybean
0.5 meal, soybean oil, soybeans, sugar, sunflower oil,
tea, and wheat
0.0 4. Raw materials: cotton, hardwood logs and sawn
wood, hides, rubber, softwood logs and sawn wood,
0.5
soybean meal, and wool
1.0
197079 8089 9099 200009 1014 The primary source for international commodity
prices is the IMFs International Financial Statistics
3. Correlation between Change in CTOT and Change in Oil Prices database. The price of crude oil is the simple average
1.0
of three spot prices: Dated Brent, West Texas Inter-
0.5 mediate, and Dubai Fateh. The World Banks Global
Economic Monitor database is used to extend the price
0.0 series of barley, iron ore, and natural gas from the
IMFs Primary Commodity Price System back to 1960.
0.5 The price of coal is the Australian coal price, extended
back to 1960 using the World Banks Global Economic
1.0 Monitor database and U.S. coal price data from the
197079 8089 9099 200009 1014
U.S. Energy Information Administration.
Annex Table2.1.3 shows the pairwise correlation
Source: IMF staff calculations.
Note: The gure shows the rolling correlation between country-specic variables between the three external conditions variables. The
and global variables over nonoverlapping ve-year windows. The horizontal line low correlation between these variables suggests that
inside each box represents the median; the upper and lower edges of each box
show the top and bottom quartiles; and the red markers denote the top and bottom each dimension potentially exerts a separate influence
deciles. World GDP growth is the weighted average (using market exchange rates) from the other two.
of growth in individual economies. CTOT = commodity terms of trade; EMDEs =
emerging market and developing economies.
Annex 2.2. Channels through Which Emerging
Market and Developing Economies Have
Narrowed Income Differentials with Advanced
Economies
Over the medium term, once the effects of
business cycle fluctuations are smoothed out, gaps
in income per capita between countries are associ-
ated with differences in the stocks of physical and

102 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Annex Figure 2.2.1. Changes in Levels of Selected Variables human capital used in production (factor gaps) and
Relative to the United States differences in technology and efficiency (total factor
(Percentage points)
productivity [TFP] gaps). This annex examines the
variation over time in factor and TFP gaps between
15 1. GDP per Worker in Purchasing-Power-Parity Terms
emerging market and developing economies and the
10
United States.
5
Using a standard production function approach,
0
aggregate output can be expressed as
5
10 Y=A*K (hL) 1, (2.5)
15
in which Y is real output, K is the stock of physical
20
1970s 80s 90s 2000s 10s capital, h is human capital per worker, L is labor input,
A is total factor productivity, and is the capital share
10 2. Human Capital per Worker of income.
Aggregate GDP can be reexpressed to give output
5
per worker as a function of human capital per worker,
0 the capital-output ratio, and TFP37:

______
* h *(__ )
1
______ 1
5 A 1
y=Y/L= K . (2.6)
Y
10 Comparing each emerging market economy i to the
1970s 80s 90s 2000s 10s
United States, the gap in output per worker can be
decomposed into the factor gaps and the residual TFP
40 3. Capital Intensity 1
gap as follows38:

((__) )
20
______
1
1
______ K
yi ( A 1 ) i ____
_________ hi ___________ Y
____ =

*
* i
.
(2.7)
((__) )
0 yU.S.
(A 1 )
1
______
hU.S.
K
______
1
U.S.
Y U.S.
20
The decomposition reveals that, over time, the
40
relative importance of different channels through
1970s 80s 90s 2000s 10s which income gaps have narrowed and widened has
shifted (Annex Figure2.2.1). During the 1970s,
60 4. Total Factor Productivity
1980s, and 1990s, movements in income-per-worker
40 gaps mirrored movements in the TFP gap, with factor
20 accumulation often moving in the opposite direction.
0 By contrast, over the past 15 years, the relative output-
20 per-worker gap has mirrored movements in the factor
gaps more than it has TFP gaps. This suggests that the
40
channels have varied in importance: whereas the TFP
60
1970s 80s 90s 2000s 10s channel appears more important in the 1970s, 1980s,
and 1990s, factor accumulation appears to have played
Source: IMF staff calculations. a greater role in recent years.
Note: The horizontal line inside each box represents the median; the upper and
lower edges of each box show the top and bottom quartiles; and the red markers
denote the top and bottom deciles of the average change in the selected variable
during the decade.
1
Capital intensity is dened as (K/Y) /(1) as explained in Annex 2.2.

37Klenow and Rodrguez-Clare (1997); Hall and Jones (1999);

Hsieh and Klenow (2010).


38The residual also captures any measurement error in output of

any of the inputs.

International Monetary Fund | April 2017 103


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex 2.3. Estimation of the Impact of External captured by the combined Polity IV index of gover-
Conditions on Emerging Market and Developing nance characteristics).
Economy Growth The model is estimated with the generalized method
The empirical framework used to assess the role of of moments (GMM) for dynamic panel models
external conditions for medium-term growth over time developed by Arellano and Bond (1991) and Arellano
is based on a fixed-effects panel growth regression that and Bover (1995). Given that the model is dynamic,
is standard in the literature.39 The general regression estimation by ordinary least squares (OLS) may lead to
equation is given by biased estimates (lagged income can be correlated with
the fixed effects in the error term, leading to dynamic
git = i +t + Xit +Zit + it , (2.8) panel bias). Moreover, some of the control variables are
in which gitis the average annual growth rate of real potentially endogenous. The difference GMM estima-
GDP per capita in purchasing-power-parity terms tor relies on differencing and instrumentation to deal
in country iover period t; i captures time-invariant with these issues.
country fixed effects; and tis a time fixed effect that Table2.3.1 reports the estimation results. Col-
controls for common, global factors. As is common in umns (1) to (3) report the results when one external
the literature, each period corresponds to a five-year conditions variable is included at a time. The results
nonoverlapping window to smooth the influence of reported in the text of the chapter correspond to
business cycles. The period of analysis is 19702014, column (4), in which all three country-specific external
although the panel is not balanced (that is, data are variables are included jointly. The results using an OLS
not available for all countries in all periods). estimator with country fixed effects are reported in
The vector Zitincludes the main variables of interest, columns (5) through (8).
that is, the three country-specific external conditions
described in Annex 2.1. The equation also includes a Robustness Exercises
vector (Xit ) of standard covariates in long-term growth
regressions. Given that the interest is in exploring the Annex Table2.3.2 reports results from robustness
role of external conditions, rather than assessing the exercises. In all of these exercises, all external condi-
contribution of all factors that may affect medium- tions variables are included jointly and the model is
term growth, Xitis a parsimonious set of control estimated with a difference GMM method.
variables mainly aimed at attenuating potential omitted Sample of Countries
variable bias affecting the estimates (rather than at
maximizing the share of variance explained by the Some large emerging market and developing
model).40 The set of controls includes the initial level economies, notably China, have started to play a key
of income per capita (average log GDP per capita role in global activity in recent decades. To address
over the previous five-year period) to account for concerns of potential endogeneity of external demand
transitional convergence, the average rate of inflation conditions, key large emerging market and devel-
to account for macroeconomic stability, the level of oping economies are excluded from the estimation
human capital, de jure measures of trade and financial sample in the exercises reported in columns (1) and
openness (proxied by the level of average import tariffs (2). More precisely, column (1) excludes China from
and an index of restrictions to the capital account, the estimation sample, while, in column (2), all large
respectively), and deep institutional characteristics (as emerging market and developing economies (that is,
those in the sample that are members of the Group of
39See for instance Barro and Sala-i-Martin (2004). Related studies TwentyArgentina, Brazil, China, India, Indonesia,
that include measures of external conditions in standard growth Korea, Mexico, Russia, Saudi Arabia, South Africa, and
regression include, among others, Arora and Vamvakidis (2005); Turkey) are excluded from the estimation.
Caldern, Loayza, and Schmidt-Hebbel (2006); and Box4.1 in IMF
(2014).
The baseline sample includes many very small econ-
40In any case, the variables in Z are derived from demand or
it
omies (even if economies with very small population
financial conditions in trading partners and from global commodity are excluded). In this sense, the average coefficients
prices, so there is less of a concern of omitted variable bias or endog-
from the baseline sample may not be very represen-
eneity than would be the case if the analysis were using measures
of export growth or openness (which could be affected by domestic tative of aggregate growth in emerging market and
factors that directly affect per capita income growth). developing economies, which is largely driven by large

104 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Annex Table2.3.1. Estimation Results from Linear Panel Growth Regression


Dependent Variable:
GDP per Capita Growth Rate (1) (2) (3) (4) (5) (6) (7) (8)
Explanatory Variables
External Demand Conditions 0.524** 0.421** 0.331 0.243
(0.203) (0.192) (0.199) (0.189)
External Financial Conditions 0.266*** 0.186** 0.339*** 0.289***
(0.099) (0.085) (0.096) (0.086)
Commodity Terms of Trade 0.453* 0.481* 0.539** 0.538**
(0.238) (0.249) (0.220) (0.218)
Estimation Details
Estimation Method GMM GMM GMM GMM OLS OLS OLS OLS
Time Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes
Country Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes
Other Control Variables Yes Yes Yes Yes Yes Yes Yes Yes
Number of Observations 505 517 509 497 587 601 592 578
Number of Economies 81 84 83 80 82 84 83 81
R2 0.411 0.422 0.417 0.432
Specification Tests (p-values)
Second-Order Correlation Test 0.863 0.913 0.567 0.507
Hansen Test 0.149 0.173 0.197 0.201
Source: IMF staff calculations.
Note: The dependent variable is the annual growth rate of GDP per capita in purchasing-power-parity terms, averaged over nonoverlapping five-year windows.
One unit of external demand conditions corresponds to a 1 percentage point growth in domestic absorption of trading partners; one unit of external financial
conditions corresponds to 1 percentage point of GDP in capital flows to regional economies; one unit of the commodity terms of trade corresponds to a
1 percent increase in the commodity terms of trade index (akin to a windfall income gain of 1 percent of GDP). The sample period is 19702014. Robust
standard errors are reported in parentheses. GMM = generalized method of moments; OLS = ordinary least squares. ***,**, and * denote significance at the 1,
5, and 10 percent level, respectively.

Annex Table2.3.2. Estimation Results from Linear Panel Growth Regression: Robustness Exercises
Dependent Variable:
GDP per Capita Growth Rate (1) (2) (3) (4) (5) (6)
Explanatory Variables
External Demand Conditions 0.401** 0.361* 0.153 0.408** 0.400** 0.372*
(0.194) (0.204) (0.322) (0.191) (0.196) (0.214)
External Financial Conditions 0.204** 0.223** 0.194** 0.199** 0.244*** 0.330***
(0.087) (0.101) (0.089) (0.086) (0.093) (0.111)
Commodity Terms of Trade 0.502** 0.454* 1.036*** 0.195*** 0.473* 0.954***
(0.255) (0.245) (0.293) (0.053) (0.246) (0.213)
Estimation Details
Estimation Method GMM GMM GMM GMM GMM GMM
Time Fixed Effects Yes Yes Yes Yes Yes Yes
Country Fixed Effects Yes Yes Yes Yes Yes Yes
Other Control Variables Yes Yes Yes Yes Yes Yes
Number of Observations 491 441 235 497 497 413
Number of Economies 79 71 36 80 80 80
Specification Tests (p-values)
Second-Order Correlation Test 0.512 0.462 0.681 0.602 0.693 0.523
Hansen Test 0.198 0.235 1.000 0.138 0.327 0.207
Source: IMF staff calculations.
Note: The dependent variable is the annual growth rate of GDP per capita in purchasing-power-parity terms, averaged over nonoverlapping five-year windows.
One unit of external demand conditions corresponds to a 1 percentage point growth in domestic absorption of trading partners; one unit of external financial
conditions corresponds to 1 percentage point of GDP in capital flows to regional economies; one unit of the commodity terms of trade corresponds to a 1 per-
cent increase in the commodity terms of trade index (akin to a windfall income gain of 1 percent of GDP). The sample period is 19702014. Robust standard
errors are reported in parentheses. GMM = generalized method of moments. ***,**, and * denote significance at the 1, 5, and 10 percent level, respectively.

International Monetary Fund | April 2017 105


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

economies. To explore how the baseline results may weight interest rates start only in 1984), but the results
be affected by this, the exercise reported in column are qualitatively unchanged and reinforce the finding
(3) excludes the smallest economies, which collectively that external financial conditions have a significant
accounted for less than 5 percent of emerging market effect on medium-term growth in emerging market
and developing economies aggregate GDP in purchas- and developing economies.
ing-power-parity terms in 2011. To construct the CTOT index, individual commodity
price fluctuations are weighted by net exports of each
Endogeneity of External Conditions Variables commodity as a share of GDP (Annex 2.1). While the
A priori there is no reason to believe that country- weights are lagged, they could potentially be affected by
specific external conditions variables used in the growth outcomes averaged over five years in the econ-
analysis are systematically affected by growth outcomes omy in question. Even if there is some overlap between
of the economy in question or by other variables that the window over which the dependent variable and the
also directly affect medium-term growth in ways that weights of individual commodity prices are constructed,
would introduce reverse causality or omitted variable it is not clear that this implies that the aggregate CTOT
bias in the baseline estimation. Nonetheless, this exog- index is systematically affected by growth outcomes in
eneity assumption may be questioned in some individ- a way that would bias the coefficient estimates. None-
ual cases. The baseline estimation attempts to mitigate theless, an additional exercise is reported in column (6)
these concerns by simultaneously including all three based on an alternative CTOT index that uses overall
external conditions in the specifications together with commodity trade rather than GDP to weight individual
time fixed effects that capture unobservable common price fluctuations. The coefficient is larger, given that the
factors. The robustness exercise reported in column (2), alternative index has larger variability, but the results are
which excludes large emerging market and developing qualitatively unchanged.42
economies, should also alleviate these concerns.
Columns (4) to (6) report additional robustness
exercises related to potential endogeneity of the exter- Annex 2.4. Identification of Growth Episodes
nal financial conditions variable and the commodity The procedure to identify growth acceleration
terms of trade (CTOT) variable. Regarding the former, episodes follows Hausmann, Pritchett, and Rodrik
the regional criterion to select peer economies in the (2005). The trend growth rate of each economy at
construction of the external financial variable may time t over horizon h, gt,t+h, is defined as the least
introduce spatial correlation in capital flows caused by squares growth rate of real GDP per capita at con-
omitted variables, potentially biasing the estimates. The stant national prices (y) from t to t+h described by
external financial variable may also be affected by pull the following equation estimated over rolling win-
factors of other economies in the same region that are dows of six years [t,t+h]43:
unrelated to the availability of external finance for the
ln(yt+i) = +gt,t+h i,i=0,,h. (2.9)
economy in question. In a first exercise, the external
financial conditions variable was considered as poten- A growth acceleration episode is defined as a time
tially endogenous and instrumented in the difference interval spanning [t,t+h] with the following attributes
GMM estimation with its own lags. The results in (in which the horizon h is set at five years in the base-
column (4) show that the coefficient is marginally line case):
larger and even more statistically significant than in the the trend growth rate of real GDP per capita is at
baseline estimation. In a second exercise, reported in least 3.5 percent a year (gt,t+h 3.5);
column (5), a country-specific financial-flows-weighted the trend growth rate during the episode exceeds
average of interest rates in large advanced economies the trend growth rate during the preceding equal-
(France, Germany, Japan, United Kingdom, United
States) is used as an additional instrument.41 The coef- 42The interquartile range for the average annual change in the
ficient estimates are somewhat different, as the time alternative CTOT index across all countries and periods is 2.8 to
sample is also different (the financial flows data used to 3percent, while it is 0.4 to 0.3 percent in the case of the baseline
CTOT index.
43Episodes are identified up to the year 2010 using real income
41The country-specific weights are constructed from cross-border per capita from PWT 9.0 through 2014 and extended to 2015 using
flows from Bank for International Settlements data. the growth rate of real income per capita from the WEO database.

106 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Annex Table2.4.1. Persistent Acceleration Episodes


Economy Year Economy Year
Albania 1995 FYR Macedonia 2003
Algeria 2000 Malawi 2005
Argentina 2003 Malaysia 2002
Armenia 2000 Mali 1974
Azerbaijan 2003 Mauritius 1973, 1985
Belarus 1999, 2002 Mozambique 1994
Benin 1977 Myanmar 1993, 1998
Bosnia 1995 Namibia 2002
Botswana 1970, 1986, 1994, 2003 Nigeria 2000
Bulgaria 2003 Oman 1975
Burkina Faso 1994 Pakistan 2002
Cambodia 2003 Panama 2003
Cameroon 1970, 1976 Paraguay 2000, 2009
Chad 2000 Peru 2003
Chile 2002 Philippines 2003
China 1980, 2000 Poland 1995, 2003
Colombia 2004 Rwanda 1975, 2003
Costa Rica 2003 Sierra Leone 2009
Czech Republic 2003 Singapore 1977, 1986, 2003
Dominican Republic 1994, 2004 Slovak Republic 2003
Ecuador 1970 Slovenia 1995
Egypt 2004 Sri Lanka 1976, 1990, 2003
Estonia 2002, 2010 Sudan 1997
Ethiopia 2003 Swaziland 1985
Ghana 2008 Syria 1972, 1993
Honduras 2003 Taiwan Province of China 1984
Hong Kong SAR 1976, 2003 Tanzania 2000
Hungary 1997 Thailand 1986, 2002
India 1993, 2002 Trinidad and Tobago 1996, 2001
Indonesia 1988, 2002 Tunisia 1995
Jordan 1975, 2001 Turkey 2002
Korea 1982 Turkmenistan 2004
Lao P.D.R. 1979 Uzbekistan 2003
Lesotho 1987, 2005 Vietnam 1975, 1981
Lithuania 2002
Source: IMF staff calculations.

length interval by at least 2 percentage points through this procedure. Their distribution by region and
(gt,t+h
gt,th 2); and decade is shown in Annex Figure2.4.1.
the level of real GDP per capita at the end of A reversal episode, in turn, is defined as an interval
the episode is at least as large as the maximum spanning [t,t+h] during which
level recorded prior to the onset of the episode the trend growth rate during the reversal is at least
(yt,t+h
max{yi}, it). 2 percentage points lower than during the preceding
interval (gt,th
gt,t+h 2); and
The set of acceleration episodes identified is in line real GDP per capita declines such that the average
with those in Hausmann, Pritchett, and Rodrik (2005) level of real GDP per capita during the episode
for the period during which the samples overlap. [t,t+h] is lower than the average level of real GDP
Starting with the set of identified acceleration episodes, per capita during [th,t], or ( y t,t+h y
th,t).
a persistent acceleration episode is defined as an acceler-
ation that is not associated with a subsequent reversal Annex Table2.4.2 lists the 125 episodes identified
(defined below) or a banking crisis (as defined by Laeven as reversals, and Annex Figure2.4.2 shows the distri-
and Valencia 2013) within three years before or after bution of reversal episodes by region and decade.
the end of the acceleration episode. Annex Table2.4.1
lists the 95 episodes of persistent accelerations identified

International Monetary Fund | April 2017 107


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Figure 2.4.1. Persistent Acceleration Episodes by Annex Figure 2.4.2. Reversal Episodes by Region
Region (Number of episodes)
(Number of episodes)
60 1. EMDEs 2. Asia 4
50 1. EMDEs 2. Asia 12
50
10 3
40 40
8
30 30 2
6
20
20
4 1
10
10 2
0 0
1970 80 90 2000 10 1970 80 90 2000 10
0 0
1970 80 90 2000 10 1970 80 90 2000 10
16 3. LAC 4. SSA 30
12 3. LAC 4. SSA 12
14
25
10 10 12
20
8 8 10
8 15
6 6
6
10
4 4 4
2 5
2 2
0 0
0 0 1970 80 90 2000 10 1970 80 90 2000 10
1970 80 90 2000 10 1970 80 90 2000 10

5. Emerging Europe 6. MENAP 6 5. Emerging Europe 6. MENAP 12


12 12

10 10 10

4 8
8 8

6 6
6
2 4
4 4
2
2 2
0 0
0 0 1970 80 90 2000 10 1970 80 90 2000 10
1970 80 90 2000 10 1970 80 90 2000 10

Source: IMF staff calculations. Source: IMF staff calculations.


Note: X-axis labels indicate the start year of a 10-year period. EMDEs = emerging Note: X-axis labels indicate the start year of a 10-year period. EMDEs = emerging
market and developing economies; LAC = Latin America and the Caribbean; market and developing economies; LAC = Latin America and the Caribbean;
MENAP = Middle East, North Africa, Afghanistan, and Pakistan; SSA = MENAP = Middle East, North Africa, Afghanistan, and Pakistan; SSA =
sub-Saharan Africa. sub-Saharan Africa.

108 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Annex Table2.4.2. Reversal Episodes


Economy Year Economy Year
Albania 1988 Lesotho 1980
Algeria 1985 Liberia 1979, 1989, 2003
Angola 1976, 1989 Madagascar 1973, 1979, 1990, 2009
Argentina 1980, 1999 Malawi 1980, 1999
Bahrain 1981, 2006 Mauritania 1979
Bangladesh 1971 Mexico 1983
Bolivia 1981 Mongolia 1989
Brazil 1989 Mozambique 1981
Bulgaria 1989 Myanmar 1985
Burkina Faso 1981 Namibia 1981
Burundi 1992 Nicaragua 1976, 1985
Cameroon 1985 Niger 1971, 1982
Central African Republic 1970, 1978, 2000, 2010 Nigeria 1979
Chad 1977, 1991 Oman 2010
Chile 1971 Panama 1985
Democratic Republic of the Congo 1974, 1989 Paraguay 1983, 1996
Republic of Congo 1986 Peru 1980, 1987
Costa Rica 1980 Philippines 1981
Croatia 2009 Poland 1979, 1988
Cte d'Ivoire 1979, 1989, 1999 Qatar 1979
El Salvador 1978 Rwanda 1985, 1990
Ethiopia 1973, 1982, 1988 Saudi Arabia 1980, 1994
Gabon 1978, 1983, 1997 Senegal 1976, 1989
The Gambia 1984 Sierra Leone 1994
Ghana 1973, 1979 Slovenia 2009
Guatemala 1982 South Africa 1982
Guinea 1989 Sudan 1978
Guinea-Bissau 1978, 1997 Syria 1985, 2010
Haiti 1981, 1990, 2000 Tanzania 1979
Honduras 1981 Togo 1972, 1979, 1989, 1998
Hungary 1988 Trinidad and Tobago 1982
Iran 1976, 1984 Uganda 1976
Iraq 1980, 1987 United Arab Emirates 1984, 2005
Jamaica 1975, 1996, 2007 Uruguay 1981, 1999
Jordan 1986 Venezuela 1979, 1998
Kenya 1990 Zambia 1970, 1976, 1990
Kuwait 1979, 1986, 1998, 2007 Zimbabwe 1974, 1983, 2001
Lebanon 1987
Source: IMF staff calculations.

Annex 2.5. Estimation of the Influence of External 1 for the economy-years identified as reversal episodes
Conditions on the Likelihood of Experiencing (Annex Table2.4.2) and zero otherwise. Given the
Persistent Accelerations and Reversals empirical challenge of accurately dating growth episodes,
This annex provides additional details on the empir- following Hausmann, Pritchett, and Rodrik (2005) the
ical approach used to explore the influence of external dummy variables also take a value of 1 in the first lead
conditions on the occurrence of growth accelerations (t+1) and lag (t1) around each identified episode.
and reversals and reports the main results as well as Using these dummy variables, the influence of
robustness analyses. country-specific external conditions on the likelihood
of growth episodes can be tested by the following
Data and Methodology distribution function:
Two dummy variables are constructed to implement Pr(episodeit = 1) = (Zit), (2.10)
the empirical analysis on growth episodes: one dummy
takes a value of 1 for the economy-years identified as in which Zitis the vector of moving averages (between
persistent acceleration episodes (Annex Table2.4.1) and t+1 and t+h) of the three country-specific external
zero otherwise; and the other dummy takes a value of conditions variables described in Annex 2.1, and is

International Monetary Fund | April 2017 109


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Table2.5.1. Logistic Estimates of the Effects of External Conditions Variables on the Odds Ratio of
Persistent Accelerations
No Country Country Fixed Country and Probit Baseline
or Time Fixed Effects and Time Fixed Time Fixed Random Random Country Fixed
Specification Effects Other Controls Effects Only Effects Effects Effects Effects
External Demand 1.248*** 1.607*** 1.095 1.158** 1.330*** 1.165*** 1.384***
(0.087) (0.151) (0.097) (0.085) (0.119) (0.052) (0.088)
External Financial 1.209*** 1.227*** 1.103** 1.098** 1.243*** 1.123*** 1.240***
(0.045) (0.050) (0.050) (0.044) (0.049) (0.021) (0.034)
Change in Terms of Trade 0.970 1.042 0.935 1.040 1.007 1.009 1.052
(0.047) (0.091) (0.046) (0.076) (0.063) (0.030) (0.066)
Model Chi-Squared Test 43.4*** 98.2*** 31,482.8*** 245.5*** 45.8*** 51.8*** 103.6***
Country Fixed Effects No Yes No Yes No No Yes
Time Fixed Effects No No Yes Yes No No No
Other Controls No Yes No No No No No
Number of Economies1 110 110 110 110 110 116 110
Number of Observations 4,176 1,325 4,176 2,279 4,176 4,322 2,279
Source: IMF staff calculations.
Note: ***,**, and * denote significance at the 1, 5, and 10 percent level, respectively; other controls not reported include de jure measures of trade and financial
openness, the level of inflation, and deep institutional characteristics. The coefficients report changes in the odds ratio of persistent accelerations. Value greater
(smaller) than 1 indicates increase (decrease) in the odds ratio relative to the unconditional odds. Robust standard errors are reported in parentheses.
1Maximum number of economies. For estimations with country fixed effects, economies without episodes are excluded.

Annex Table2.5.2. Logistic Estimates of the Effects of External Conditions Variables on the Odds Ratio of Reversals
No Country Country Fixed Country and Baseline
or Time Fixed Effects and Time Fixed Time Fixed Random Probit Random Country Fixed
Specification Effects Other Controls Effects Only Effects Effects Effects Effects
External Demand 0.818*** 0.738*** 0.841*** 0.793*** 0.736*** 0.851*** 0.655***
(0.047) (0.067) (0.046) (0.061) (0.055) (0.033) (0.038)
External Financial 0.822*** 0.710*** 1.014 0.977 0.788*** 0.876*** 0.774***
(0.037) (0.043) (0.061) (0.055) (0.041) (0.023) (0.028)
Change in Terms of Trade 0.933* 0.851* 0.976 0.973 0.935** 0.963** 0.941**
(0.039) (0.074) (0.041) (0.028) (0.031) (0.017) (0.027)
Country Fixed Effects No Yes No Yes No No Yes
Time Fixed Effects No No Yes Yes No No No
Other Controls No Yes No No No No No
Number of Economies1 110 110 110 110 110 116 110
Number of Observations 4,176 1,184 4,176 2,835 4,135 4,322 2,835
Source: IMF staff calculations.
Note: ***,**, and * denote significance at the 1, 5, and 10 percent level, respectively; other controls not reported include de jure measures of trade and financial
openness, the level of inflation, and deep institutional characteristics. The coefficients report changes in the odds ratio of persistent accelerations. Value greater
(smaller) than 1 indicates increase (decrease) in the odds ratio relative to the unconditional odds. Robust standard errors are reported in parentheses.
1Maximum number of economies. For estimations with country fixed effects, economies without episodes are excluded.

a nonlinear function representing how Zit affects the in which Xitis a vector of controls (using moving
probability Pr(episodeit = 1). The nonlinear binary averages between t3 and t1) that includes domes-
dependent model is then empirically estimated using tic covariates associated with medium-term growth
either a probit or a logit functional form to replace (for example, de jure integration, credibility of policy
().44 To establish an appropriate baseline specification, frameworks), and i captures time-invariant country
country and time fixed effects as well as additional fixed effects.
control variables are considered. The benchmark speci-
fication is given by the following equation: Logit Estimates
The coefficient estimates of several variations of the
( 1Pr(episodeit = 1))
Pr(episode = 1)
log ________________

it
=Zit + Xit
+ i + it, model in (2.11) are reported in Annex Tables 2.5.1
(2.11) (persistent accelerations) and 2.5.2 (reversals) and in
Annex Figure2.5.1, panels 1 and 2. They indicate
44As a robustness check, the linear probability model was also tested, a robust positive association between the odds ratio
and the significance of the variables are robust to this estimation method. of persistent accelerations and external demand and

110 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

financial conditions in all specifications. In turn, the Annex Figure 2.5.1. Change in the Odds Ratio of Occurrence of
commodity terms-of-trade variable is not significant in Growth Episodes, 19702015
(Percent)
any of the specifications estimated on the full sample
of countries (including commodity exporters and non-
120 1. Persistent Accelerations
commodity exporters).
In the case of reversals, external financial con-
80
ditions are not statistically significant when time
fixed effects are included in the regression (columns
(3) and (4) in Annex Table2.5.2 and Annex Fig- 40

ure2.5.1, panel 2). This is likely due to the impor-


tance of common factors in explaining capital flows 0
to emerging markets, as documented in Chapter2 of
the April 2016 World Economic Outlook and Fig- 40
No country Country Only time Country and Country Baseline
ure2.10. The effect of commodity terms of trade on or time xed xed effects xed effects time random (country
the likelihood of reversals is also statistically insig- effects and other xed effects xed effects xed effects)
controls
nificant when time fixed effects are included, which
likely capture common drivers of commodity prices,
30 2. Reversals
while they are statistically significant in all other
20
specifications.
10
In sum, Annex Tables 2.5.1 and 2.5.2 and Annex
0
Figure2.5.1 show that the statistically significant asso-
10
ciation between external conditions and the increase in
20
the odds ratio of persistent accelerations and reversals
30
is robust to different specifications, including when
country fixed effects are not included (column (1) 40

of the tables), or estimating the model with random 50


No country Country Only time Country and Country Baseline
effects using logit or probit approaches (columns (5) or time xed xed effects xed effects time random (country
effects and other xed effects xed effects xed effects)
and (6) of the tables). The baseline specification used controls
in the analysis (equation 2.11) includes only country
fixed effects. Source: IMF staff calculations.
Note: For each estimation procedure, the rst estimation point refers to external
demand conditions; the second estimation point refers to external nancial
conditions; and the third estimation point refers to commodity terms of trade.
Marginal Effects Other controls include de jure measures of trade and nancial openness, the
level of ination, and deep institutional characteristics. Vertical lines denote 90
The logit estimates of the previous section can percent condence intervals.
be used to compute the average marginal effect of a
one-unit change in a given variable on the likelihood
of a growth episode. This is the statistic used in the
text and figures of the chapter to discuss the impact of z1,itand all other variables in equation (2.11)
of external conditions as well as domestic attributes and hence the need for parsimony in the number of
on the likelihood of growth episodes. Using equations explanatory variables. The baseline results reported in
(2.10) and (2.11), the average marginal effects can be Figure2.17 are based on a specification that includes
represented by only the external conditions variables, which are evalu-
ated at their sample means.
Pr(episodeit = 1)
_______________

= 1 ( 1 z1,it + 2 z2,it + 3 z3,it Robustness Tests
z 1,it
+ 1x1,it + + nxn,it + 1 The baseline results for the effects of external
+ + N). (2.12) conditions on the likelihood of growth episodes are
compared with those based on different country
Marginal effects in nonlinear binary dependent samples. Annex Figure2.5.2 reports the change in
models depend not only on 1, but also on the value the odds ratio (in percent) of a one-unit increase in

International Monetary Fund | April 2017 111


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

Annex Figure 2.5.2. Change in the Odds Ratio of Occurrence of Annex Figure 2.5.3. Change in the Probability of Occurrence of
Growth Episodes by Subsamples, 19702015 Growth Episodes (Marginal Effect) Using Seven-Year Durations,
(Percent) 19702015
(Percentage points)
120 1. Persistent Accelerations
8

80 6

40 4

2
0
0
40
Baseline Excluding Excluding
2
China G20

4
20 2. Reversals

10 6
0
8
10 Demand Financial CTOT Demand Financial CTOT
20 Persistent accelerations Reversals

30 Source: IMF staff calculations.


40 Note: Vertical lines denote 90 percent condence intervals. CTOT = commodity
terms of trade.
50
Baseline Excluding Excluding
China G20

Source: IMF staff calculations.


change slightly relative to those for the five-year epi-
Note: For each estimation procedure, the rst estimation point refers to external sodes, but the pattern of statistical significance of the
demand conditions; the second estimation point refers to external nancial results is unchanged.
conditions; and the third estimation point refers to commodity terms of trade.
Vertical lines denote 90 percent condence intervals. G20 = Group of Twenty.
Channels through Which External Conditions
Help Emerging Market and Developing Economies
Narrow Income Differentials with Respect to
each external condition for different sample splits. Advanced Economies
For both persistent accelerations and reversals, the
The analysis in this section aims at understand-
results of the baseline specification of Annex Tables
ing the role of external conditions in influencing
2.5.1 and 2.5.2 are robust to the sample splits (that
the channels of physical capital deepening and other
is, samples excluding China or Group of Twenty
factors (documented in Annex 2.2) through which
economies).
income gaps between emerging market and developing
A second robustness test extends the horizon of the
economies and advanced economies are narrowed. To
growth episode identification criteria of Annex 2.4 to
this end, the persistent acceleration episodes listed in
seven years (thus, h = 7instead of h = 5in equation
Annex Table2.4.1 are further split into capital-led and
(2.9) as well as in all identification criteria in that sec-
non-capital-led accelerations.
tion). The logit model (2.11) and its marginal effects
Capital-led accelerations are those in which the
represented by equation (2.12) are reestimated using
contribution to growth during the episode from capital
the seven-year span for episodes. Annex Figure2.5.3
deepening (measured as described above using the
reports the marginal effects of those reestimations. It
capital-output ratio rather than capital per worker)45
shows that the marginal effects of external conditions
are robust in terms of statistical significance to the 45See Klenow and Rodrguez-Clare (1997); Hall and Jones (1999);

change in the span of the episode. The point estimates and Jones (2016).

112 International Monetary Fund | April 2017


CHAPTER 2 Roads Less Traveled: Growth in Emerging Market and Developing Economies in a Complicated External Environment

Annex Figure 2.5.4. Change in the Probability of Occurrence of Free trade agreements: Data on flows of agreements
Persistent Accelerations (Marginal Effect) by Type of by year of signature are obtained from the October
Acceleration, 19702015 2016 World Economic Outlook (Chapter2) using the
(Percentage points)
Design of Trade Agreements database. This data set is
complemented with the stock of free trade agreements
8.0
in effect from the World Trade Organization Regional
Trade Agreements database. The former builds on the
6.0 latter, supplementing it with data from other multilat-
eral institutions and national sources.
Financial depth: Financial depth is proxied by total
4.0 assets held by deposit money banks as a share of GDP
from the World Banks Global Financial Development
database.
2.0 Sound credit growth: While a deeper financial
system is associated with increased access to finance
and greater support for economic activity, a too-rapid
0.0
expansion of credit may lead to vulnerabilities that end
up undermining growth. The identification of excessive
2.0
credit growthor credit boomsfollows DellAriccia
Demand Financial CTOT Demand Financial CTOT and others (2016).
Capital-led Non-capital-led Capital account openness: The index of de jure capital
persistent accelerations persistent accelerations
account openness is an update of the Quinn (1997)
Source: IMF staff calculations. measure of capital controls, which draws from the nar-
Note: Vertical lines denote 90 percent condence intervals. CTOT = commodity rative portion of the IMFs Annual Report on Exchange
terms of trade. Arrangements and Exchange Restrictions. A higher value
denotes fewer restrictions.
exceeds the average contribution to growth from Current account balance: The current account
capital deepening for that country in the entire sample. balance as a share of GDP is from the IMF World
The remaining acceleration episodes are classified as Economic Outlook database.
non-capital-led. Based on this criterion, there are 61 Exchange rate flexibility: The degree of exchange rate
capital-led and 34 non-capital-led acceleration episodes flexibility is based on the de facto index developed by
in the set of identified persistent accelerations. Aizenman, Chinn, and Ito (2010).
Annex Figure2.5.4 reports the marginal effect of Public debt: The ratio of public debt to GDP from
external conditions on the two episode probabilities. Mauro and others (2013) is used as a proxy for fiscal
Favorable external demand raises the probability of prudence.
non-capital-led acceleration episodes relatively more Sound monetary framework: The quality of the mone-
than the probability of capital-led episodes, whereas tary framework is proxied by the sound money index
favorable external financing raises the probability from Gwartney, Lawson, and Hall (2016). The index
of capital-led episodes more than the probability of is a standardized measure that combines indicators on
non-capital-led episodes. the growth of money supply, the level and volatility
of inflation, and the possibility of owning foreign
currency bank accounts, based on data from the World
Annex 2.6. Analysis of Domestic Attributes in Developments Indicators (World Bank), International
Mediating the Impact of External Conditions Financial Statistics and Annual Report on Exchange
This annex provides additional details on the empir- Arrangements and Exchange Restrictions (IMF), and
ical analysis carried out in the chapters section on the United Nations National Accounts.
role of policies and structural attributes. The analysis Regulation, legal system, and property rights: The
explores how policies and other domestic attributes indices on the quality of regulation, the legal system,
may influence the impact of external conditions on the and protection of property rights are from Gwartney,
likelihood of acceleration of reversal episodes. Lawson, and Hall (2016). A higher value is associated

International Monetary Fund | April 2017 113


WORLD ECONOMIC OUTLOOK: Gaining Momentum?

with better quality of institutions. Each index compiles in the likelihood of a growth episode, in percentage
indicators from several sources, including the Global points) when the policy or domestic attribute changes
Competitiveness Report (World Economic Forum), by an amount equivalent to moving from the 25th
International Country Risk Guide (Political Risk Ser- percentile to the 75th percentile of its sample distri-
vices Group), Doing Business and World Developments bution (in the case of the exchange rate regime, the
Indicators (World Bank), and International Financial 25th percentile corresponds to a fully flexible exchange
Statistics (IMF). Some individual indicators may be rate regime, while the 75th percentile corresponds to a
vulnerable to perception-based rankings and measure- fixed exchange rate regime).
ment uncertainties. However, by combining several
indicatorsincluding from international financial Exploring How the Impact of External Conditions on the
institutions that compile their data from national Likelihood of Growth Episodes Depends on Policies and
official sourcesthe constructed indices potentially Other Domestic Attributes
have more comprehensive data coverage than a single To explore how domestic attributes affect the impact
indicator and may also be less sensitive to outliers and of external conditions on the likelihood of growth
concerns about subjectivity. episodes, the baseline empirical specification (2.11) is
modified to include interaction terms as follows:
Direct Effect of Domestic Policies and Attributes on the

(1Pr(episodeit = 1))
Likelihood of Growth Episodes Pr(episodeit = 1)
log ________________

=
Before analyzing how policies and other domestic
attributes affect the impact of external conditions on zit + xit + (zit xit) + i + it,
(2.13)
the likelihood of acceleration of reversal episodes,
the direct effect of these domestic attributes on the in which zitis one of the three country-specific
likelihood of growth episodes is explored. To this end, external conditions; xitis the moving average between
a variation of the logit regression (2.11) described in t3 and t1 of the domestic policy or attribute in
Annex 2.5 is used in which Xit includes the mov- question; and icaptures time-invariant country fixed
ing average (between t3 and t1) of one domestic effects. The estimates from the logit regression with
policy or attribute at a time. This allows for testing interaction terms in (2.13) are then used to derive the
of whether the policy or domestic attribute variable marginal effects reported in Figures 2.19 and Annex
significantly affects the likelihood of growth episodes Figure2.6.2:46
once the external conditions and country fixed effects The exercise reported in Figure2.19 is based on two
are controlled for. sets of marginal effects: one in which the domestic
Annex Tables 2.6.1 and 2.6.2 report the results for attribute variable is set at a low value (the 25th
persistent acceleration and reversal episodes, respec- percentile of its sample distribution) and one in
tively. The coefficients on the domestic attribute which it is set at a high value (the 75th percentile
variables indicate their impact, in percent, on the of its sample distribution). In both cases, however,
odds ratio of experiencing a growth episode versus the external conditions variable is set at its sample
not experiencing one: values below (above) 1 indicate median (interpreted as neutral external conditions).
lower (higher) odds of experiencing an episode versus The bars in Figure2.19 correspond to the difference
not experiencing an episode for higher values of the between these two sets of marginal effects, inter-
domestic attribute variable. The results suggest that preted as the change in the marginal effect of the
more financial depth, a sound monetary framework, external conditions variable as the domestic attribute
and better quality of institutions significantly increase improves; in the case of some variables, such as the
the odds ratio of a persistent acceleration episode exchange rate stability index, the credit boom indi-
(Annex Table2.6.1). A sound monetary framework cator, and the external and public debt variables, the
and more financial depth also significantly reduce 46The use of marginal effects is particularly relevant for exploring
the odds ratio of a reversal episode, whereas lower how domestic attributes affect the impact of external conditions on
exchange rate flexibility increases the odds ratio of the likelihood of growth episodes, given that the coefficient of the
interaction term in the nonlinear logit estimation using odds ratios
experiencing a reversal (Annex Table2.6.2). To assess
(2.13) is not sufficient to infer how the effect of one independent
the economic relevance of these results, Annex Fig- variable depends on the magnitude of another independent variable
ure2.6.1 shows the marginal effect (that is, the change (Ai and Norton 2003).

114 International Monetary Fund | April 2017


Annex Table2.6.1. Logistic Estimates of the Effects of Policy Variables on the Odds Ratio of Persistent Accelerations
External Demand 1.266*** 1.296*** 1.234*** 1.382*** 1.275*** 1.285*** 1.264*** 1.268*** 1.282*** 1.352*** 1.279*** 1.293*** 1.401***
(0.088) (0.094) (0.091) (0.110) (0.088) (0.093) (0.097) (0.090) (0.104) (0.109) (0.104) (0.103) (0.143)
External Financial 1.200*** 1.217*** 1.209*** 1.193*** 1.223*** 1.213*** 1.224*** 1.195*** 1.204*** 1.218*** 1.213*** 1.215*** 1.215***
(0.041) (0.042) (0.041) (0.044) (0.040) (0.039) (0.040) (0.040) (0.040) (0.045) (0.049) (0.047) (0.055)
Change in Terms of 0.970 0.950 0.945 0.955 0.967 1.016 0.985 0.981 0.995 1.005 1.066 1.024 1.318
Trade (0.082) (0.044) (0.062) (0.049) (0.049) (0.053) (0.060) (0.051) (0.070) (0.070) (0.080) (0.081) (0.223)
Number of Trading 0.928 0.916
Partners (Log) (0.088) (0.112)
Financial Openness 0.813
Index (0.312)
Deposit Money Banks 1.007** 1.009**
Assets to GDP (0.003) (0.004)
Capital Account 1.190 0.919
Openness (0.523) (0.674)
Credit Booms 0.599 0.643
(0.308) (0.339)
Current Account Balance 0.979 0.960
to GDP (0.022) (0.024)
External Debt to GDP 1.000 1.002