Vous êtes sur la page 1sur 162

World Eco no m i c a nd F i na nci a l S u r v e y s

Global Financial Stability Report


April 2015

Navigating Monetary Policy Challenges


and Managing Risks

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
2015 International Monetary Fund

Cover and Design: Luisa Menjivar and Jorge Salazar


Composition: AGS

Cataloging-in-Publication Data

Joint Bank-Fund Library

Global financial stability report Washington, DC :


International Monetary Fund, 2002
v. ; cm. (World economic and financial surveys, 0258-7440)

Semiannual
Some issues also have thematic titles.
ISSN 1729-701X

1. Capital market Development countries Periodicals.


2. International finance Periodicals. 3. Economic stabilization
Periodicals. I. International Monetary Fund. II. Series: World economic and financial surveys.
HG4523.G563

ISBN 978-1-49837-293-0 (paper)


978-1-47554-758-0 (ePub)
978-1-47554-808-2 (Mobipocket)
978-1-47559-658-8 (PDF)

Disclaimer: The Global Financial Stability Report (GFSR) is a survey by the IMF staff published twice
a year, in the spring and fall. The report draws out the financial ramifications of economic issues high-
lighted in the IMFs World Economic Outlook (WEO). The report was prepared by IMF staff and has
benefited from comments and suggestions from Executive Directors following their discussion of the
report on April 3, 2015. 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, Global Financial Stability ReportNavigating


Monetary Policy Challenges and Managing Risks (Washington, April 2015).

Please send orders to:


International Monetary Fund, Publications 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 vii

Further Information and Data vii

Preface viii

Executive Summary ix

Chapter 1 Enhancing Policy Traction and Reducing Risks 1


Financial Stability Overview 1
Box 1.1. The Oil Price FalloutSpillovers and Implications for the Financial Sector 5
Macroeconomic Versus Balance Sheet Deleveraging: What Is in the Mix? 9
Disinflationary Risks and Financial Stability 12
When Market Liquidity Vanishes 31
Emerging Markets: Safeguarding the Financial Sector against Global Headwinds 37
Box 1.2. Russias Financial Risks and Potential Spillovers 47
Annex 1.1. Progress on the Financial Regulatory Reform Agenda 50
Annex 1.2. External Portfolio Rebalancing under Quantitative Easing in the Euro Area and Japan 51
References 53

Chapter 2 International Banking after the Crisis: Increasingly Local and Safer? 55
Summary 55
Introduction 56
What Has Changed? 57
The Drivers of the Changes in International Banking 63
Box 2.1. The International Expansion of Chinese and Japanese Banks 64
Effects on Financial Stability 69
Box 2.2. The Expansion of Pan-African Banks: Opportunities and Challenges 70
Policy Implications 74
Conclusion 76
Box 2.3. Banking Union in Europe 78
Box 2.4. Global Banks: Regulatory and Supervisory Areas in Need of Attention 79
Annex 2.1. Regression Analysis of the Drivers of the Decline in Foreign Banking Claims 81
Annex 2.2. Analysis of the Role of International Banking Linkages in Mitigating
or Amplifying Shocks 85
Annex 2.3. Analysis of the Effect of International Banking Linkages on the Probability
of a Banking Crisis 89
References 90

Chapter 3 The Asset Management Industry and Financial Stability 93


Summary 93
Introduction 94
Financial Stability Risks of Plain-Vanilla Funds: Conceptual Issues 97
Box 3.1. Possible Incentive Problems Created by Delegated Management 100
Box 3.2. Fund Share Pricing Rules and First-Mover Advantage 101

International Monetary Fund | April 2015 iii


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Financial Stability Risks of the Mutual Fund Industry: Empirical Analysis 104
Revamping the Oversight Framework to Address Financial Stability Risks 115
Conclusion 120
Annex 3.1 Primer on the Asset Management Industry 122
Annex 3.2 Empirical Framework 127
References 132

Glossary 137

IMF Executive Board Discussion Summary 145

Tables
1.1. Household Debt in the Euro Area, Japan, the United Kingdom, and the United States 13
1.2. Corporate Debt in the Euro Area, Japan, the United Kingdom, and the United States 14
1.3. Reallocating Assets: Stylized Investment Choices 20
1.4. Energys Impact on Two Barclays Corporate Credit Indices 28
1.5. Summary of Sovereign, Corporate, and Banking Indicators 45
Annex 1.2.1. Potential Portfolio Outflows by Euro Area Investors, 2015 51
Annex 1.2.2. Japan: A Potential Portfolio Rebalancing Scenario under QQE2, 201517 52
Annex 1.2.3. Potential Portfolio Outflows by Japanese Institutional Investors, 201517 53
2.1. Effects of International Banking Linkages on the Incidence of Crises 76
2.2. Main Findings of the Analysis of the Effects of International Banking Linkages
on Domestic Credit Growth 77
2.1.1. Survey on the Regulation of Banks International Operations 82
2.1.2. Definition of the Variables 83
2.1.3. Results of Country-level Regression for the Drivers of the Changes
in Foreign Banking Claims 84
2.2.1. Credit Growth Panel Regressions from the Perspective of Host Countries
of Foreign Banks 87
2.2.2. Credit Growth Panel Regressions from the Perspective of Home Countries
of Foreign Banks 87
2.2.3. Bank-Level Evidence on Foreign Bank Stabilization Role during Crises 88
2.3.1. Detailed Probit Regression Results 89
3.1. Summary Characteristics and Risk Profiles of Major Investment Vehicles 98
3.1.1. An Illustrative Example of Asset Managers Incentives for Risk Taking 100
3.2.1. Comparison of Fund Pricing Rules 101
3.2. Mutual Fund Flows and Asset Returns 105
3.3. Selected Regulations for Publicly Offered Funds 117
3.4. Summary of Analysis and Policy Implications for Mutual Funds and ETFs 119
Annex 3.1.1. Features and Risk Profiles of Key Investment Vehicles 122
Annex 3.2.1. List and Definition of Variables for Empirical Exercises 128

Figures
1.1. Global Financial Stability Map: Risks and Conditions 1
1.2. Global Financial Stability Map: Components of Risks and Conditions 2
1.3. Global Disinflationary Forces 4
1.1.1. Developments in Oil Markets 6
1.4. Asset Valuation and Volatility Heat Maps 8
1.5. Quantitative Easing Impact Channels 9
1.6. Private Nonfinancial Sector Gross Debt 10
1.7. Episodes of Private Sector Deleveraging in Selected Advanced Economies 11
1.8. Insolvency Frameworks and Nominal Deleveraging 12
1.9. Quantitative Easing and Financial Markets 15
1.10. Illustrative Baseline QE Portfolio Rebalancing Scenarios in Japan and the Euro Area 18

iv International Monetary Fund | April 2015


Contents

1.11. Bank Lending and Constraints 19


1.12. Bank Nonperforming Loans and Lending Conditions 21
1.13. Life Insurance Industry Characteristics 23
1.14. United States: Nonfinancial Corporations 26
1.15. U.S. Credit Spreads, Firm Leverage, and Interest Coverage 27
1.16. U.S. High-Yield Energy Markets 28
1.17. U.S. Interest Rates and Term Premiums 29
1.18. Global Interest-Rate Developments 30
1.19. The October 15 Flash Rally in U.S. Treasuries 32
1.20. Asset Comovements and Correlations Spillovers 34
1.21. Wide Range in the Inflation Outlook of Emerging Market Economies 36
1.22. China: Real Estate and Interest Rate Developments 38
1.23 Emerging Market Nonfinancial Corporate Investment Continues to Shrink 39
1.24. Dependence of Emerging Market Sovereigns on Commodities, and Market Reaction 40
1.25. Energy Corporate Sector Metrics 41
1.26. Large Increase in Emerging Market Debt 42
1.27. Firms in Countries with Large Currency Selloffs Also Had High Equity Volatility 43
1.28. Financial Stability of Emerging Market Banks 44
1.2.1. Russian External Debt Amoritization Schedule 47
1.2.2. Russian Banking System Key Financial Soundness Indicators 47
1.2.3. Foreign Bank Exposures 48
Annex 1.2.1. Euro Area Negative-Yielding European Government Bonds and Baseline Rebalancing 52
2.1. Developments in Foreign Banking Claims 58
2.2. Types of Claims in Bank for International Settlements Consolidated Statistics 58
2.3. Number of Branches and Subsidiaries of Foreign Banks in 2008 and 2013, by Region 59
2.4. Banking Regionalization in Asia 60
2.5. Trends in Latin America and Europe 61
2.6. Precrisis and Postcrisis Geographic Correlation Networks from Banks Stock Returns 62
2.7. Changes in Corporate Borrowing 63
2.1.1 The Internationalization of Chinese and Japanese Banks 64
2.1.2. Ratio of Foreign Interest Income to Foreign Revenue for Chinese and Japanese Banks 65
2.1.3. Funding Vulnerabilities for Chinese and Japanese Banks 66
2.8. Share of Countries that Changed Regulations on International Banking Operations
between 2006 and 2014 67
2.9. Share of Countries that Tightened Regulations on International Banking Operations
between 2006 and 2014, by Region 68
2.10. Effects of Regulations and Other Factors on International Banking Linkages 69
2.2.1 Major Pan-African Banks: Cross-Border Expansion, 200214 70
2.2.2 Major Pan-African Banks: Systemic Importance by Country, 2013 70
2.11. Changes in Capital Flows 72
2.12. Effect of International Banking Linkages on Domestic Credit Growth 73
2.13. Lending Growth by Domestic and Foreign-Owned Banks during Crises 74
2.14. Effect of Parent and Subsidiary Characteristics on Subsidiary Lending Growth 75
2.3.1. Sovereign Bond and Corporate Lending Rates in the Euro Area 78
3.1. Financial Intermediation by the Asset Management Industry Worldwide 94
3.2. Products Offered by Asset Managers and Their Recent Growth 95
3.3. Key Domiciles of Mutual Funds 96
3.4. Unleveraged Open-End Funds and Systemic Risk 99
3.2.1. Structure of Exchange-Traded Funds 102
3.2.2. Difference between NAV and ETF Share Price 103
3.5. Liquidity Mismatches 104
3.6. Growth in Bond Funds by Investment Focus 105
3.7. Bond Ownership Concentration and Its Effects on Credit Spreads 107

International Monetary Fund | April 2015 v


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

3.8. Drivers of Fund Flows from End Investors 108


3.9. Convexity of Fund FlowPerformance Relationship 109
3.10. Liquidity Risk and Fund Structures 110
3.11. Brand Name Effects 111
3.12. Funds Liquidity Risk Management 112
3.13. Herding among U.S. Mutual Funds 113
3.14. Ownership Structure of the 25 Largest Global Asset Management Companies 114
3.15. Bank Financing by Mutual Funds and Money Market Funds 115
3.16. Contribution to Systemic Risk by Mutual Funds 116
Annex 3.1.1. Investment Vehicles by Size, Domicile, and Investment Focus 124
Annex 3.1.2. Operation of a Fund 126
Annex 3.2.1. Drivers of Changes in Credit Spreads during Stress Episodes 130

Statistical Appendix
Available for download from IMF.org

Editors notes
(April 16, 2015)

Figure 2.9, panel 1 in Chapter 2 (page 68) has been corrected and replaced. The bars in panel 1 for Selected
euro area economies and Other euro area economies were transposed in the original release of the report.

Figure 3.5 in Chapter 3 (page 104) has been corrected and replaced. The circle for EM bond MF appeared
slightly too small in the original release of the report. The AE HY corporate bond circle has been renamed as
AE HY bond.

Figure 3.6 in Chapter 3 (page 105) has been corrected and replaced. The data in this figure as they originally
appeared in the report were incorrect.

Annex Figure 3.1.1, panel 3 in Chapter 3 (page 124) has been corrected and replaced. The data in panel 3 as
printed in the original report were partially incorrect.

vi International Monetary Fund | April 2015


1
CHAPTER

ASSUMPTIONS AND CONVENTIONS

The following conventions are used throughout the Global Financial Stability Report (GFSR):
. . . to indicate that data are not available or not applicable;
to indicate that the figure is zero or less than half the final digit shown, or that the item does not exist;
between years or months (for example, 201314 or JanuaryJune) to indicate the years or months covered,
including the beginning and ending years or months;
/ between years or months (for example, 2013/14) to indicate a fiscal or financial year.
Billion means a thousand million.
Trillion means a thousand billion.
Basis points refer to hundredths of 1 percentage point (for example, 25 basis points are equivalent to of 1
percentage point).
If no source is listed on tables and figures, data are based on IMF staff estimates or calculations.
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.

Further Information and Data


This version of the GFSR is available in full through the IMF eLibrary (www.elibrary.imf.org) and the IMF website
(www.imf.org).

The data and analysis appearing in the GFSR are compiled by the IMF staff at the time of publication. Every effort
is made to ensure, but not guarantee, their timeliness, accuracy, and completeness. When errors are discovered,
there is a concerted effort to correct them as appropriate and feasible. Corrections and revisions made after publica-
tion 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 contents of this publication, please refer to the IMF Copy-
right and Usage website, www.imf.org/external/terms.htm.

International Monetary Fund | April 2015 vii


PREFACE

The Global Financial Stability Report (GFSR) assesses key risks facing the global financial system. In normal times,
the report seeks to play a role in preventing crises by highlighting policies that may mitigate systemic risks, thereby
contributing to global financial stability and the sustained economic growth of the IMFs member countries.
The current report finds that, despite an improvement in economic prospects in some key advanced economies, new
challenges to global financial stability have arisen. The global financial system is being buffeted by a series of changes,
including lower oil prices and, in some cases, diverging growth patterns and monetary policies. Expectations for rising
U.S. policy rates sparked a significant appreciation of the U.S. dollar, while long-term bond yields in many advanced
economies have decreasedand have turned negative for almost a third of euro area sovereign bondson disinflation
concerns and the prospect of continued monetary accommodation. Emerging markets are caught in these global cross
currents, with some oil exporters and other facing new stability challenges, while others have gained more policy space
as a result of lower fuel prices and reduced inflationary pressures. The report also examines changes in international
banking since the global financial crisis and finds that these changes are likely to promote more stable bank lending in
host countries. Finally, the report finds that the asset management industry needs to strengthen its oversight framework
to address financial stability risks from incentive problems between end-investors and portfolio managers and the risk of
runs due to liquidity mismatches.
The analysis in this report has been coordinated by the Monetary and Capital Markets (MCM) Department under
the general direction of Jos Vials, Financial Counsellor and Director. The project has been directed by Peter Dattels
and Dong He, both Deputy Directors, as well as by Gaston Gelos and Matthew Jones, both Division Chiefs. It has
benefited from comments and suggestions from the senior staff in the MCM Department.
Individual contributors to the report are Ali Al-Eyd, Nicols Arregui, Serkan Arslanalp, Jonathan Beauchamp, Rina
Bhattacharya, John Bluedorn, Antoine Bouveret, Peter Breuer, Yingyuan Chen, Martin ihk, Fabio Cortes, Cristina
Cuervo, Pragyan Deb, Reinout De Bock, Martin Edmonds, Johannes Ehrentraud, Jennifer Elliott, Michaela Erbenova,
Brenda Gonzlez-Hermosillo, Tryggvi Gudmundsson, Sanjay Hazarika, Geoffrey Heenan, Allison Holland, Eija Holt-
tinen, Hibiki Ichiue, Bradley Jones, David Jones, William Kerry, Oksana Khadarina, Yoon Kim, Frederic Lambert,
Daniel Law, Min-Jer Lee, Peter Lindner, Andrea Maechler, Joe Maloney, Alejandro Lopez Mejia, Win Monroe, Hiroko
Oura, Evan Papageorgiou, Alexandra Peter, Vladimir Pillonca, Alvaro Piris Chavarri, Jean Portier, Gabriel Presciuttini,
Shaun Roache, Luigi Ruggerone, Martin Saldias, Luca Sanfilippo, Tsuyoshi Sasaki, Katharine Seal, Nobuyasu Sugi-
moto, Narayan Suryakumar, Shamir Tanna, Nico Valckx, Chris Walker, Jeffrey Williams, and Kai Yan. Magally Bernal,
Carol Franco, Daniela Mendoza, Juan Rigat, and Adriana Rota were responsible for word processing.
Joe Procopio from the Communications Department led the editorial team and managed the reports production
with support from Michael Harrup and Linda Kean and editorial assistance from Cathy Gagnet, Lucy Scott Morales,
Sherrie Brown, Gregg Forte, Linda Long, David Einhorn, EEI Communications, and AGS.
This particular edition of the GFSR draws in part on a series of discussions with banks, securities firms, asset man-
agement companies, hedge funds, standards setters, financial consultants, pension funds, central banks, national treasur-
ies, and academic researchers.
This GFSR reflects information available as of March 27, 2015. The report benefited from comments and sugges-
tions from staff in other IMF departments, as well as from Executive Directors following their discussion of the GFSR
on April 3, 2015. However, the analysis and policy considerations are those of the contributing staff and should not be
attributed to the IMF, its Executive Directors, or their national authorities.

viii International Monetary Fund | April 2015


EXECUTIVE SUMMARY

Macroeconomic shocks pose new challenges for global est rate environment also poses challenges for long-
markets term investors, particularly for weaker life insurance
A wide range of positive and negative macroeconomic companies in Europe.
and financial developments have occurred in the past Oil- and commodity-exporting countries and firms
six months. On a net basis, these developments have have been severely affected by falling asset valuations
increased financial stability risks. and rising credit risks. Energy and commodity firms
On the positive side, as discussed in the April 2015 in emerging markets, which account for more than a
World Economic Outlook, growth in 2015 is expected third of nonfinancial corporate bonds issued in hard
to be slightly higher than that of 2014, improving in currency since 2007, have been particularly hard hit.
advanced economies enough to offset slower growth in Strains in the debt-repayment capacity of the oil and
emerging market and developing economies. Sharply gas sector have become more evident for firms in
lower oil and commodity prices, coupled with lower Argentina, Brazil, Nigeria, and South Africa because
interest rates from expanded monetary accommodation, of low oil prices, as well as for sovereigns reliant on
are expected to support growth through 2016. Bold oil revenues such as Nigeria and Venezuela.
monetary policy actions have been taken in both the Rapidly depreciating exchange rates have increased
euro area and Japan to arrest and reverse disinflation- pressures on firms that borrowed heavily in for-
ary pressures. Quantitative easing provides a strong eign currencies and have sparked significant capital
framework for addressing deflation risks, and some key outflows for several emerging markets. These
transmission channels are already working. Spreads on developments could add stress to emerging mar-
credit have narrowed in the euro area, equity prices ket sovereigns that have increased their combined
have surged, and the euro and yen have depreciated exposure to foreign currency borrowings and foreign
significantly, helping to raise inflation expectations. investor holdings of local currency debt.
At the same time, the U.S. dollar has appreciated Volatility in major exchange rates has increased
substantially, reflecting diverging monetary policies. The by more than during any similar period since the
dollar has strengthened more against major currencies global financial crisis. Reduced liquidity in both the
during the past nine months than it has during any foreign exchange and fixed-income markets, as well
similar period since 1981. The resulting movements in as the changing composition of the investor bases
real exchange rates have broadly reflected changes in in these markets, has added frictions to portfolio
growth prospects and exposures to lower oil prices, and adjustments. The resulting tensions in global finan-
should help support the global recovery. cial markets have increased market and liquidity
However, the financial stability risks around this risks, given that sudden episodes of volatility could
baseline are rising and rotating. Although the benefits of become more common and more pronounced.
the improving baseline are widely distributed and accrue
over time, the adverse impact of recent shocks is concen- Existing legacy challenges add to these pressures, leav-
trated and is already affecting sectors and economies with ing overall financial stability risks higher.
preexisting vulnerabilities. Meanwhile, continued financial
risk taking and structural changes in credit markets are Financial stability is not fully grounded in advanced
shifting the locus of financial stability risks from advanced economies, and risks have increased in many
economies to emerging markets, from banks to shadow emerging markets
banks, and from solvency to market liquidity risks. Long-term bond yields in many advanced economies
Continued financial risk taking and search for yield have decreased on disinflation concerns and the prospect
keep stretching some asset valuations. The low inter- of continued monetary accommodation. In the euro

International Monetary Fund | April 2015 ix


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

area, almost one-third of short- and long-term sovereign Lower commodity prices and lower inflationary pressures
bonds now carry negative yields. But a prolonged low are benefiting many emerging market economies, provid-
interest rate environment will pose severe challenges for ing monetary policy space to combat slowing growth.
a number of financial institutions. Weak European mid- However, oil- and commodity-exporting countries and
sized life insurers face a high and rising risk of distress countries with high foreign indebtedness face more
stress tests (conducted by the European Insurance and formidable risks. Although the stronger dollar can help
Occupational Pensions Authority) show that 24 percent improve competitiveness in emerging market economies
of insurers may not be able to meet their solvency in general, and lead to higher growth, the dramatic
capital requirements under a prolonged low interest rate movements in commodity prices and in the exchange
scenario. The industry has a portfolio of 4.4 trillion rates of many emerging market economies during the
in assets in the European Union, with high and rising past six months have already had a significant impact
interconnectedness with the wider financial system, on firms market valuations in these economies. Many
creating a potential source of spillovers. companies borrowed heavily in international markets
High debt levels in the private sector continue to substituting international borrowings in dollars for local
hinder growth and financial stability. Accommodative currency borrowing from bankspotentially leading to
monetary policies in advanced economies have helped balance sheet pressures.
reduce private sector debt ratios by supporting inflation In turn, a retrenchment of overinvested industries,
and growth and by increasing asset prices. However, the real estate sector adjustments, and property price
assumptions for growth and inflation in this report sug- declinesespecially in Chinacould spill over to
gest that private sector debt levels in a number of major emerging markets more broadly. The broader impact of
advanced economies will remain high. This continuing a sudden deterioration in corporate health on banking
high debt calls for an additional response to address the system stability depends on credit exposures. In China,
crisis legacies and unshackle economic potential. Gross exposures to real estate (excluding mortgages) are
corporate debt in France, Italy, Portugal, and Spain is almost 20 percent of GDP, and financial stress among
expected to remain above or near 70 percent of GDP real estate firms could lead to direct cross-border spill-
by 2020, and gross household debt in Portugal and the overs, given the substantial increase in external bond
United Kingdom is projected to remain high compared issuance since 2010. In 11 of the 21 emerging market
with that of other major advanced economies. banking systems analyzed in this report, more than
At the same time there is a clear upside risk to interest half of the bank loan books consists of loans to firms,
rates in the United States. Two possible scenarios charac- rendering them more exposed to corporate weakness,
terize the future normalization of U.S. monetary policy: particularly in Nigeria, Peru, Turkey, and Ukraine.
a smooth well-telegraphed exit, or, despite clear com- In a downside risk scenario, further rapid dollar
munication, a bumpy ride with a more rapid decompres- appreciation and an abrupt rise in U.S. interest rates,
sion of term premiums leading to rapidly rising yields coupled with a rise in geopolitical risks, could put
and substantially higher volatility. Indeed, declines in added pressure on emerging market currencies and
structural liquidity in fixed-income markets in both the asset markets. After a prolonged period of inflows, for-
United States and other economies have amplified asset eign investors could abruptly reduce their holdings of
price responses to shocks, increasing potential spillovers. local currency debt, thereby adding to turbulence and
Technological change, increased regulation, and the creating debt rollover challenges. Markets also appear
shifting composition of market participants have altered complacent when it comes to geopolitical and politi-
the microstructure of fixed-income markets. Illiquidity cal risks. As noted in the April 2015 World Economic
events now spill over to other asset classes and to emerg- Outlook, ongoing events in Russia and Ukraine, the
ing markets, as witnessed in the U.S. Treasury market Middle East, and parts of Africa could lead to greater
and in policy-induced instability in foreign exchange tensions and increased disruptions to global trade and
markets following the removal of the Swiss franc floor. financial transactions. Direct financial linkages between
These developments highlight some key vulnerabilities in Russia and the rest of the world are limited, but the
capital markets and the shadow banking system. indirect connections with neighboring countries could
Emerging markets are caught in these global crosscur- raise financial stability risks. Stronger institutional
rents, as they address their own domestic challenges. frameworks in the euro area have reduced the threat

x International Monetary Fund | April 2015


Executive summary

of contagion from Greece, but risks and vulnerabilities about 36 percent of the system. A deeper and broader
remain. capital market would improve access to finance, par-
ticularly for smaller firms, and make financial markets
A range of additional policies are required to more efficient. In the euro area, encouraging the use
increase policy traction and ground stability of capital markets requires harmonization of com-
This report assesses the policy responses of cen- pany law, corporate governance, insolvency regimes,
tral banks in both advanced and emerging market and taxation, in line with the latest Capital Markets
economies. A key message is that additional policy Union proposal by the European Commission.
measuresbeyond monetary policiesare required
to make a well-grounded exit from the crisis. Poli- In Japan the effectiveness of quantitative easing
cies must address crisis legacies and facilitate sustain- depends on the policies supporting it. Steadfast imple-
able economic risk taking while containing financial mentation of Abenomics second and third arrows
excesses across global markets. (fiscal and structural reforms) is essential. If these
To maximize the impact of quantitative easing in reforms are incomplete, efforts to pull the economy
the euro area, central bank actions must be comple- out of deflation are less likely to succeed. The Bank
mented with measures to restore balance sheet health of Japan should consider strengthening the portfolio
in the private sector, unclog credit channels, enhance rebalancing effects of its asset purchases by increasing
the soundness of nonbank institutions, and promote the share of private assets in purchases and extending
structural reforms. In particular, the program to longer-maturity government bonds, as
Unclogging credit channels requires comprehensive necessary, to achieve its 2 percent inflation target. To
actions to tackle the burden of nonperforming loans. further stimulate bank lending to the private sector,
Despite improving bank resilience in the wake of the the authorities should expand special lending facili-
ECBs Comprehensive Assessment and introduction ties; jumpstart the securitization market for small and
of the Single Supervisory Mechanism, asset quality medium enterprise credits and mortgages; and enhance
continues to deteriorate, although at a slowing pace, risk capital provision, including by encouraging more
with total nonperforming loans now standing at asset-based lending and removing barriers to entry and
more than 900 billion. Banks should be encouraged exit for small and medium enterprises.
to develop and use specialized internal and external In the United States, the impact of global market
capacity for handling the stock of nonperforming forces requires appropriately balanced policies, includ-
assets, actively manage their provisions, and write ing continued clear communication of monetary poli-
off their nonperforming assets. Further efforts are cies. A smooth market adjustment will be more likely
needed to improve the effectiveness of legal frame- if there is extensive discussion and interpretation of key
works governing bankruptcy of companies and economic variables given that monetary policy is now
individuals. Without corrective policy actions, bank data dependent. Yet market expectations can differ
lending capacity could be limited to a meager 1 to 3 from the Federal Reserves guidance, leading to market
percent on average a year. tensions and raising market and liquidity risks.
The challenges facing life insurers should also be In the United States and other economies with sig-
tackled promptly. Regulators need to reassess the via- nificant nonbank financial systems, addressing illiquidity
bility of guarantee-based products and work to bring and potential spillovers by strengthening market struc-
minimum return guarantees offered to policyholders tures will help enhance stability. As noted in Chapter 3,
in line with secular trends in policy rates. Prompt the asset management industry needs stronger oversight
regulatory and supervisory actions are needed to that combines better microprudential supervision of
mitigate damaging spillovers from potential difficul- risks with the adoption of a macroprudential orientation.
ties of individual insurers. The introduction of a Policies should seek to address the mismatch between
more harmonized safety net would further increase the liquidity promised to mutual fund owners in good
the industrys resilience. times and the cost of illiquidity when redemptions must
Sources of funding need to be diversified away from be met in times of stress. Policies can help to accom-
banks and toward capital markets. Despite the surge plish this objective by reducing asset owners incentives
in capital market borrowings, they represent only to run (by aligning funds redemption terms with the

International Monetary Fund | April 2015 xi


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

underlying liquidity in the assets invested); enhancing The international financial regulatory reform
the accuracy of net asset values; increasing liquidity cash agenda has strengthened regulatory frameworks, and is
buffers in mutual funds; and improving the liquidity and helping to make financial institutions and the global
transparency of secondary markets, especially of longer- financial system more robust. Global standard setters
term debt markets. Market participants in government and national regulators now need to provide further
bond and foreign exchange markets should also have clarity about regulatory standardsand thus improve
greater incentives to provide secondary market liquid- certainty for banks adapting their business models
ity. Authorities should review current circuit breakers to by finalizing the calibration of recent requirements,
enhance their functioning. Risk management and control including the leverage ratio, the net stable funding
should be reinforced: supervisors should provide coor- ratio, and total loss-absorbing capital requirements.
dinated guidance to trading firms, allowing them to set Promptly putting in place regulations to transform
consistent and appropriate risk limits on individual retail shadow banking into a stable source of market-based
investors. Regulators and monetary authorities should finance is also a must.
consider the correlation between asset classes when evalu- At the same time, micro- and macroprudential
ating systemic risks in financial markets. policies for nonbanks should be strengthened. Exist-
Emerging marketsshould aim to cushion the ing regulatory frameworks may need to be reassessed
impact of global headwinds and safeguard the to enable the authorities to better understand the less
resilience of their financial systems through enhanced closely regulated corners of the financial sector that
surveillance of vulnerable sectors, particularly in the could cause problems for the banking system and the
following areas: broader economy, and act as needed to mitigate identi-
In China, the overall priority must be to allow an fied vulnerabilities.
orderly correction of excesses, curtailing the riskiest
parts of shadow banking. At the same time, orderly Changes in international banking models have
deleveraging requires comprehensive policies that reduced risks in host financial systems
allow credit growth to slow gradually and, where On a more positive note, Chapter 2, which examines
necessary, the mechanisms to be provided for orderly changes in international banking since the global finan-
corporate debt restructuring, and the exit of nonvi- cial crisis, finds that these changes are likely to promote
able firms. more stable bank lending in host countries. It also finds
Across emerging markets more generally, the large a need for more international cooperation for dealing
portion of debt denominated in foreign currencies with regional or global shocks to maximize the benefits
means that micro- and macroprudential measures of cross-border banking while mitigating risks.
have important roles to play in limiting the risks International banks, especially those in Europe,
from shocks. Regulators need to conduct bank stress have reduced their cross-border lending, while local
tests related to foreign currency and commodity loans by branches and subsidiaries of foreign banks
price risks and more closely and regularly monitor abroad have remained steady. Local and regional
corporate leverage and foreign currency exposures, banks have stepped in to offset, at least partially, euro
including derivatives positions. area banks reduction in exposure to some regions.
To ensure markets function properly, authorities As a result, intraregional linkages have deepened, in
need to prepare for lapses of liquidity in local cur- particular in Asia. Regulatory changes and weaknesses
rency bond markets. Country authorities might in bank balance sheets have contributed significantly
potentially use cash balances when needed, or in the past to the observed cutback in cross-border
lower the supply of long-term debt to the market lending, whereas accommodative monetary policies
to help curtail bond spread increases. Bilateral and may have slowed the cutback.
multilateral swap line agreements, by providing The relative shift from cross-border lending to more
foreign currency funding in times of stress, can local lending by affiliates should improve the finan-
enhance confidence and help reduce excess volatil- cial stability of host countries. Cross-border lending
ity in currency markets. Multilateral resources flows are more sensitive to global shocks than are local
such as IMF facilities could also provide additional lending and international portfolio flows. Cross-border
buffers. lending also tends to amplify the effect of adverse

xii International Monetary Fund | April 2015


Executive summary

domestic shocks on credit. In contrast, lending by opinions about less leveraged plain-vanilla investment
foreign subsidiaries is more resilient than lending products are divided.
by domestic banks during domestic crises when the However, even plain-vanilla products may pose
parent bank is well capitalized and less dependent on financial stability risks through two channels: (1)
nondeposit funding sources. However, restrictions on incentive problems between end investors and portfo-
cross-border lending may jeopardize other benefits not lio managers (which potentially can lead to herding,
examined in the chapter. among other things) and (2) run risk stemming from
the presence of liquidity mismatches. The empirical
Oversight of asset managers must be proportional to analysis finds evidence of many of these risk-generating
the risks they pose to the financial system mechanisms, although their importance varies across
Chapter 3 finds that the asset management industry asset markets. Without providing a verdict on whether
needs to strengthen its oversight framework in two key large asset managers should be designated as systemi-
areas: better microprudential supervision of risks and cally important, the analysis indicates that a funds
adoption of some macroprudential concerns as a stan- investment focus is relatively more important than its
dard part of its orientation. Asset management firms size when it comes to its contribution to systemic risk.
can provide credit to the real economy even when These findings suggest that securities regulators
banks are distressed, and they have certain advantages should shift to a more hands-on supervisory model,
over banks from a financial stability perspective. How- supported by global standards on supervision and bet-
ever, the sectors growth and the structural changes in ter data and risk indicators. The roles and adequacy of
financial systems have heightened stability concerns. existing risk-management tools should be reexamined
Although the risks posed by leveraged hedge and to take into account the asset management industrys
money market funds are already widely recognized, role in systemic risk and the diversity of its products.

International Monetary Fund | April 2015 xiii


1
CHAPTER

ENHANCING POLICY TRACTION AND REDUCING RISKS

Financial Stability Overview have sparked rapid appreciation of the U.S. dollar.
Emerging markets are caught in these global cross-
Developments over the past six months have
currents and face higher financial stability risks, as
increased global financial stability risks. Risks have
companies that borrowed heavily on international
also rotated from advanced economies to emerging
markets could face balance sheet strains. Additional
markets, from banks to shadow banks, and from
policy measures are needed to enhance the effective-
solvency to market liquidity risks. The global finan-
ness of monetary policies, address crisis legacies, and
cial system is being buffeted by a series of changes
facilitate sustainable economic risk taking while
in financial markets, reflecting diverging growth
containing financial excesses across global markets.
patterns and monetary policies as global growth
prospects have weakened. Disinflationary forces Financial stability risks have increased since the
have strengthened as oil and commodity prices have October 2014 Global Financial Stability Report and
dropped. Although the latter has benefited commod- are reflected in the Global Financial Stability Map
ity- and oil-importing countries and increased the (Figure 1.1) and in its components (Figure 1.2). As
room to maneuver for monetary policy in countries discussed in the April 2015 World Economic Outlook
with higher inflation, it has increased financial risks (WEO), the distribution of risks to global growth
in some exporting countries and in the oil sector. is now more balanced, but still tilted to the down-
As a result of these developments, inflation expecta- side. Weaker inflation and greater uncertainty are
tions and long-term bond yields have fallen. Bold weighing on the macroeconomic outlook. But these
monetary policy actions have been taken in both forces are broadly offset by favorable developments
the euro area and Japan to arrest and reverse this in high-frequency indicators, reflecting the expected
disinflation pressure, while the pull of expectations benefits of lower oil prices and additional monetary
for rising U.S. policy rates and the push of addi- accommodation, leaving macroeconomic risks broadly
tional monetary stimulus by other major economies unchanged since October.

Figure 1.1. Global Financial Stability Map: Risks and Conditions


Risks
Emerging market risks Credit risks

April 2015 GFSR


October 2014 GFSR Macroeconomic risks Market and liquidity risks

Away from center signies higher risks, easier


monetary and nancial conditions, or higher Monetary and nancial Risk appetite
risk appetite.
Conditions
Source: IMF staff estimates.
Note: GFSR = Global Financial Stability Report.

International Monetary Fund | April 2015 1


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.2. Global Financial Stability Map: Components of Risks and Conditions
(Notch changes since the October 2014 Global Financial Stability Report)

Macroeconomic risks are unchanged, as improved economic activity Emerging market risks have increased, driven by elevated volatility
offsets weaker ination and greater uncertainty. and worsening corporate sector and liquidity risks.
1. 2.
4 2
More risk
3 More risk
1
2

1 0

0
Unchanged Less risk 1
1 Less risk
2 2
Overall Sovereign Ination/ Economic Economic Overall Fundamentals Volatility Corporate Liquidity External
(8) credit deation risks activity uncertainty (10) (2) (2) sector (2) nancing
(1) (1) (4) (2) (2) (2)

Credit risks are unchanged, as worsening in corporate sector is offset Monetary and nancial conditions have been accommodative,
by improvements in banking and household indicators. with lending conditions easing.
3. 4.
2 2
More risk
Easier
1
1

0 Tighter
Unchanged 1
Less risk

1 2
Overall Banking Corporate Household Overall Monetary Financial Lending QE and CB
(8) sector sector sector (6) policy conditions conditions balance sheet
(3) (3) (2) conditions index (1) expansion
(3) (1) (1)
Risk appetite has declined, reecting lower relative asset returns and Market and liquidity risks have increased, reecting heightened
rapid outows from emerging markets. volatility and more stretched positioning indicators.
6.
3 4

More risk 3
2 Higher risk appetite
2
1
1
0
0
1
1
Less risk
2 Lower risk appetite 2
3 3
Overall Institutional Relative asset Emerging Overall Liquidity and Volatility Market Equity
(3) allocations returns markets (8) funding (2) positioning valuations
(1) (1) (1) (3) (2) (1)
Source: IMF staff estimates.
Note: Changes in risks and conditions are based on a range of indicators, complemented with IMF staff judgment (see Annex 1.1 in the April 2010 Global Financial
Stability Report and Dattels and others (2010) for a description of the methodology underlying the Global Financial Stability Map). Overall notch changes are the
simple average of notch changes in individual indicators. The number below each legend indicates the number of individual indicators within each subcategory of
risks and conditions. For lending conditions, positive values represent slower pace of tightening or faster easing. CB = central bank; QE = quantitative easing.

2 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

The U.S. economy is expanding, with rising ing the Financial Sector against Global Headwinds
employment and an improving investment outlook, as identifies these vulnerabilities and discusses how best
economic risk taking has taken hold. U.S. monetary to safeguard emerging markets against these forces.
authorities have clearly communicated that a process Reflecting the challenges facing emerging markets, risk
of monetary normalization could begin this year with appetite is lower as currency volatility and adjustments
an increase in policy rates. The bad news is that lower have prompted a pullback of capital flows by foreign
growth prospects elsewhere, relative to October 2014, investors. Lower allocations of global funds to risky
and disinflationary forces have continued to exert a assets and lower excess returns also point to slightly
strong influence on the global economy. The number lower risk appetite compared to October, although
of countries with low or negative rates of headline appetite remains above its historical average.
inflation, and their share of global output, increased Credit risks are broadly unchanged. Although the
significantly through 2014 (Figure 1.3, panels 1 and macroeconomic benefits of lower energy prices should
2). Falling commodity prices, particularly oil prices, have a favorable impact on household balance sheets,
amplified this disinflation pressure, and the inflation the immediate credit impact of oil and commodity
rate in many advanced economies fell below inflation price declines on firms in the energy sector is nega-
objectives. More emerging market economies than tive. Box 1.1 and the section Disinflationary Risks
advanced economies have headline inflation above and Financial Stability examine the energy segments
their inflation goals, although many major Asian of the high-yield market and highlight the potential
economies are at their inflation cycle lows (Figure 1.3, strains and exposures to the banking system. Fur-
panels 3 and 4). thermore, the fall in nominal yieldsshould it be
Central banks have responded to increased down- sustainedraises a serious threat to the life insurance
ward risks to price stability. Since October, the Bank and pension fund sectors, especially in Europe, as
of Japan (BOJ) and the European Central Bank discussed in the Disinflationary Risks and Financial
(ECB) have announced bold new monetary measures Stability section.
designed to ward off deflation pressure and move their These developments have created various tensions
economies closer to their inflation objectives (Figure in global financial markets, raising market and liquid-
1.3, panel 5). Other central banks have cut rates or ity risks. Asynchronous monetary policies have led to a
loosened their monetary policy stances, and markets sharp increase in volatility in foreign exchange markets
are generally pricing in lower policy rates by the end amid a rapid appreciation of the U.S. dollar. Despite
of 2015 for a number of countries (Figure 1.3, panel the prospect of gradual U.S. policy rate tightening,
6). The policy easing has offset modestly tighter real longer-term U.S. bond yields and term premiums
interest rates and thus loosened monetary and financial remain compressed as the ECB and BOJ ramp up their
conditions overall. This report examines some of the asset purchases. Asset valuations remain elevated rela-
financial channels through which quantitative easing tive to the past 10 years as monetary policies continue
(QE) worksand how to maximize its benefits while to exert downward pressure on spreads (Figure 1.4,
mitigating the risks to financial stability. panel 1). Market volatility (Figure 1.2, panel 6) has
Emerging market financial stability risks have increased across the asset spectrum, rising from the
increased. The easing of inflation pressure is benefiting record lows at the time of the October 2014 Global
many emerging market economies, giving them mon- Financial Stability Report (Figure1.4, panel 2). The
etary policy space to combat slowing growth. However, section When Market Liquidity Vanishes examines
recent global shocksincluding higher political risks the structural features that have contributed to reduced
leave several emerging market economies more vulner- market liquidity and warns that economic and policy
able. Oil and commodity price declines have hurt tensions leave global markets vulnerable to bouts of
commodity exporters and sectors faced with overcapac- illiquidity that could prove systemic.
ity, while companies that borrowed heavily on interna- This report takes a closer look at recent challenges to
tional markets face balance sheet strains from revalued the global economy and central banks policy responses
foreign currency liabilities. In China, the disinflation- to these challenges. The report discusses how to maxi-
ary force of property price declines could strain bank mize the effectiveness of these accommodative mon-
and shadow bank balance sheets and spill over more etary policies while minimizing the financial stability
broadly. The section Emerging Markets: Safeguard- side effects, with a particular focus on QE.

International Monetary Fund | April 2015 3


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.3. Global Disinationary Forces

1. Number of Countries with Low Ination Rates 2. Share of Countries with Low Ination Rates
(Percent of global GDP)
40 Ination rate below 2% 100
Ination rate below 2% Ination rate below 1% 90
35
Ination rate below 1% Ination rate below 0% 80
30 Ination rate below 0%
70
25 60
20 50
15 40
30
10
20
5 10
0 0
2004 06 08 10 12 14 2004 06 08 10 12 14
Sources: Bloomberg, L.P.; IMF, World Economic Outlook database; Haver Analytics; Sources: Bloomberg, L.P.; IMF, World Economic Outlook database; Haver Analytics;
and IMF staff calculations. and IMF staff calculations.
Note: Annual percent change in consumer prices includes 33 advanced economies Note: Annual percent change in consumer prices includes 33 advanced economies
and 17 emerging market economies. and 17 emerging market economies.
3. Advanced Economies Ination Phase Curve (as of February 2015) 4. Emerging Markets Ination Phase Curve (as of February 2015)
Accelerating Decelerating Accelerating Accelerating Decelerating Accelerating
ination ination ination ination ination ination
Brazil,
Neutral zone Colombia, Russia Neutral zone
Chile
Turkey Indonesia
Peru
Mexico
South Africa Philippines India
Canada
Austria
Euro area, France, Germany,
Ireland, Italy, Japan, Netherlands, Hungary,
Portugal, Spain, Switzerland, Poland, China
United Kingdom, United States Thailand

Sources: National authorities; and IMF staff estimates.


Note: In panels 3 and 4, countries are placed on the phase curve according to where their February 2015 year-over-year headline ination print is relative to (1) central
bank ination target band (where available), (2) three-month trend, and (3) the efcacy of monetary policy (using past two years ination targeting performance). The
ination measure used for Japan excludes tax effects. Data for Japan are as of January 2015. When ination is in the ination band it is placed within the dashed
lines, otherwise outside, and the distance from the closest dashed line is determined by criterion 3. For advanced economies, the ination band is taken to be 13
percent.

5. Central Bank Balance Sheets 6. Market-Implied Interest Rate Moves


(Percent of GDP) (Basis points)
80 200
Projections
70 Federal Reserve 150
Bank of Japan Since October 1, 2014
60 Next 12 months
European Central Bank 100
50 Bank of England
50
40
0
30
20 50

10 100
0 150
2008 09 10 11 12 13 14 15
Norway
Australia
Turkey
Korea
New Zealand
Hungary
Euro area
Japan
Sweden
Poland
Switzerland
Canada
United Kingdom
Brazil
United States
Indonesia
South Africa
Mexico

Sources: Bloomberg, L.P.; and IMF staff estimates. Sources: Bloomberg, L.P.; and IMF staff estimates.

4 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Box 1.1. The Oil Price FalloutSpillovers and Implications for the Financial Sector
The recent steep decline in oil prices reflects to a significant Corporate refinancing in the energy sector: Scaled-
extent supply factors, providing a net benefit to the global back energy sector exposure by banks and corporate
economy. Nevertheless, the speed and magnitude of the bond investors could amplify strains associated with
movement in oil prices raise questions about how stress falling revenue and higher funding costs. Historically,
can be transmitted through the financial sector. This box corporate defaults in the energy sector have tended
addresses several channels through which lower oil prices to pick up in response to falling oil prices, with a lag
could spawn financial vulnerabilities: a self-reinforcing of about 12 months, (Fitch 2015b) likely reflecting a
cycle of rising credit risk and deteriorating refinancing typical one-year hedging horizon by producers. Since
conditions for countries and companies, a decline in oil the downdraft in oil prices did not begin to accelerate
surplus recycling in world funding markets, and strains on until September 2014 (at which point Brent and West
the financial market infrastructures ability to accommodate Texas Intermediate prices were still higher than $100
prolonged heightened energy price volatility. a barrel), aftershocks for the corporate sector may not
yet have fully filtered through.
Background: As one of the steepest on record (Figure The outstanding worldwide notional value of bank
1.1.1, panel 1), the recent decline in oil prices appears loans and corporate debt extended to the energy sector
to reflect supply factors, a net benefit to the global amounts to about $3 trillion,2 $247 billion of which
economy over the medium term.1 Nevertheless, the is attributable to the U.S. high-yield bond market
speed and magnitude of the movement in oil prices may alone (Fitch 2015a) (see Figure 1.16 and Table 1.4 for
produce financial strains in selected areas as markets further discussion of energy and the U.S. high-yield
adjust to a new pricing environment. This box discusses sector). Global syndicated loan issuance in the oil and
three channels through which such an adjustment could gas sector has risen markedly in recent years, with 450
potentially contribute to an increase in market volatility. billion in issuance in 2014 alone, almost double that
Amplification of credit risk: Countries and of the previous cycle peak in 2007 (Figure 1.1.1, panel
companies dependent on oil revenues have already 3). In addition, the leveraged (that is, high-yield) share
been significantly repriced by investors since summer of syndicated oil and gas loan issuance has steadily
2014, as reflected in bond spreads, equity prices, and increased, from 17 percent in 2006 to 45 percent in
currency movements (Figure 1.1.1, panel 2). Although 2014. The majority of global systemically important
risk premiums have widened, however, the impact has banks have about 2 to 4 percent of their total loan
probably not yet fully hit in several areas. These effects book exposures devoted to the energy sector.3 Available
include refinancing risk for energy-producing sover- data suggest that there are higher exposures by selected
eigns and firms, and the reduction in bank funding banks in emerging markets and among some U.S.
lines to energy companies in response to breaches in regional banks (although firm estimates are difficult to
lending covenants. determine). A prolonged period of low oil prices will
Country refinancing risk: Fiscal breakeven prices vary jeopardize the debt-servicing capacity of exploration and
widely across oil-producing countries in emerging production firms that have high cost bases.4
markets, from $54 a barrel for Kuwait to as much as Oil surpluses and global liquidity: Foreign
$184 a barrel for Libya. Barring spending cuts, new exchange reserves accumulated by net oil-exporting
sources of revenue, or tapping fiscal buffers, the loss in countries have increased $1.1 trillion, or almost
oil revenue will require new sources of financing. U.S. fivefold, over the past decade (Figure 1.1.1, panel
dollarbased bond spreads for emerging market oil- 4). Accounting for about 15 percent of the cumula-
exporting countries have already doubled since sum- tive rise in world foreign exchange reserves since
mer 2014, which suggests that refinancing conditions 2004, these funds have been an important source
are now more problematic. Local currency deprecia-
tion may also put upward pressure on inflation where 2Bank for International Settlements (BIS), Dealogic.
domestic inflation expectations are not well anchored, 3Bernstein Research; Bloomberg, L.P., industry reports; and
further raising the risk premium on sovereign debt. IMF staff.
4Among U.S. energy companies, about $380 billion is owed

by firms with a ratio of debt to earnings before interest, taxes,


The authors of this box are Bradley Jones, Gabriel Presciuttini, depreciation, and amortization (EBITDA) that is negative or
Peter Breuer, Peter Lindner, Tsuyoshi Sasaki, and Fabio Cortes. with a debt-to-EBITDA ratio in excess of 5, amounting to 33
1See the April 2015 World Economic Outlook. percent of debt.

International Monetary Fund | April 2015 5


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS
Box 1.1. (continued)
Figure 1.1.1. Developments in Oil Markets
1. Largest Annual Oil Price Declines 2. Emerging Markets Bond Index Global Spreads
(Percent) (Basis points over Treasuries)
0 Oil exporters1 550
10 Oil importers2 500
Nonoil commodity exporters3
20 450
30 400

40 350

50 300

60 250

70 200
Feb. 2009

Mar. 1986

Jan. 2015

Nov. 2001

Dec. 1998

Oct. 1991

Oct. 1988

Jan. 2014
Feb. 14
Mar. 14
Apr. 14
May 14
Jun. 14
Jul. 14
Aug. 14
Sep. 14
Oct. 14
Nov. 14
Dec. 14
Jan. 15
Feb. 15
Sources: Haver Analytics; and IMF staff calculations. Sources: Bloomberg, L.P.; and IMF staff calculations.
Note: Based on monthly West Texas Intermediate oil price. End of 1
Colombia, Gabon, Kazakhstan, Nigeria, and Russia.
rolling 12-month period on horizontal axis. Figure depicts episodes 2
Argentina, Bosnia, Brazil, Bulgaria, Chile, China, Costa Rica, Cte
where the rolling 12-month fall in oil prices exceeded 30 percent, d'Ivoire, Croatia, the Dominican Republic, Egypt, El Salvador, Ghana,
broadly equivalent to a one standard deviation event. Guatemala, Hungary, Indonesia, Jamaica, Lebanon, Malaysia,
Mexico, Pakistan, Panama, Peru, the Philippines, Poland, Romania,
South Africa, Sri Lanka, Turkey, Ukraine, Uruguay, and Vietnam.
3
Chile, Cte d'Ivoire, South Africa, Uruguay, and Zambia.

3. Global Syndicated Loan Issuance from the Oil and Gas Sector 4. Oil Surplus Reserve Growth and Oil
(Billions of euros)
500 350 Annual change in Average oil price (dollars per 120
oil surplus reserves barrel; right scale)
Leveraged Investment grade 300
(billions of U.S. 100
400 250 dollars; left scale)
200 80
300 150
100 60
200 50
0 40
100 50 20
100
0 150 0
2005 06 07 08 09 10 11 12 13 14 2000 01 02 03 04 05 06 07 08 09 10 11 12 13 14
Sources: Dealogic; and IMF staff calculations. Sources: Bloomberg, L.P.; Haver Analytics; and IMF staff calculations.
Note: Oil exporter reserves include Bahrain, Canada, Colombia,
Ecuador, Gabon, Kazakhstan, Kuwait, Nigeria, Norway, Qatar, Russia,
Saudi Arabia, Trinidad and Tobago, United Arab Emirates, and
Venezuela. Oil prices are based on the average monthly West Texas
Intermediate oil price through the calendar year.
5. Foreign Holdings of U.S. Asset Classes 6. Net Positioning of Speculators in West Texas Intermediate
(Trillions of U.S. dollars) Oil Futures and Options
(Thousands of contracts)
7 600
Although there has been a material reduction
All other countries
6 in their net positioning, speculators remain
Oil-exporting countries1 500
quite long on crude oil
5
400
4
300
3
200
2
1 100

0 0
Equity Treasuries Credit Agencies 2008 09 10 11 12 13 14 15
Sources: U.S. Treasury Department; and IMF staff calculations. Sources: Commodities Futures Trading Commission; Intercontinental
1
Includes data for Algeria, Bahrain, Canada, Colombia, Ecuador, Exchange; and IMF staff calculations.
Gabon, Iran, Iraq, Kazakhstan, Kuwait, Libya, Nigeria, Norway, Oman,
Qatar, Russia, Saudi Arabia, Trinidad and Tobago, United Arab
Emirates, and Venezuela.

6 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Box 1.1. (continued)


of funding for the global banking sector and capital speculative) investors held about 45 percent of West
markets more broadly. Deposits from oil-exporting Texas Intermediate futures contracts in 2014, about
countries in Bank for International Settlements triple the level held during the 1990s. Banks have also
reporting banks have doubled to $972 billion since retreated from their market-making and structuring
2004, and this group of countries (private and public roles in energy markets, with a shift in trading activity
sector) now holds more than $2 trillion in U.S. assets to centrally cleared contracts (as desired by regulators)
(Figure 1.1.1, panel 5), spread across equities ($1.3 and physical commodity trading houses. With such
trillion), Treasuries ($580 billion), credit ($230 bil- major changes in market structure, questions have
lion), and agency debt ($21 billion).5 Following the been raised as to whether an additional wave of selling
$88 billion contraction in oil-exporter reserves in pressure might destabilize markets. There has already
2014, sensitivity analyses point to further significant been substantial sellingnet investment exposure is
declines in 2015 if oil prices follow the path implied nearly what it was at its peak in early 2014 (Figure
by futures markets. In principle, the decline in 1.1.1, panel 6), and mutual fund data suggest that
investable oil surpluses is part of global rebalancing U.S. high-yield bond funds are already underweight
and ought to be counterbalancedat least to some in energy compared with the benchmark. Assets under
extentby wealth gains on the part of oil importers. management in commodity funds, combined with
But such redistribution between agents with poten- commodity-linked exchange-traded products, are
tially varying savings and portfolio preferences may nearly half their 2010 peak. Implied volatility (a mea-
also have market repercussions, particularly if the sure of insurance value) has increased, but only to lev-
pace of adjustment creates market dislocations. els recorded in 201112 and well shy of levels reached
Strains on financial infrastructure: Oil and other in 2008. On balance, few indicators point to severe
commodity markets have attracted much greater focus dislocations in oil markets. Commodity exchanges
from the institutional investment community over have a long history of managing counterparty risk dur-
the past decade. For example, noncommercial (that is, ing heightened volatility (through changes in margin-
ing requirements and circuit breakers). Nevertheless,
5We concentrate here on assets held in U.S. dollars given this financial intermediaries should remain on the alert for
is the currency in which oil revenues are denominated. threats to efficient market functioning.

In principle, QE can durably boost inflation and if there is a strong rebalancing toward foreign assets, lift-
growth through several key transmission channels ing inflation and boosting competitiveness. QE should
(Figure 1.5). First, the QE program itselfand an then lead to greater economic risk taking, with firms
associated commitment to a significant expansion of the investing more and households increasing their con-
central banks balance sheetshould help raise expecta- sumption. This should also help improve the financial
tions of higher inflation and build confidence in the position of households and firms as a stronger economy
economy. Second, central bank purchases of government and increased asset values help improve balance sheet
bonds will lower risk-free interest rates in the economy, health.
which has a direct impact on real interest rates and trig- QE is appropriate for addressing disinflationary
gers various transmission channels to real activity (see pressures in the euro area and Japan, and some of the
also Draghi 2015 and Box 1 in ECB 2015).1 Among key transmission channels are already working. Finan-
these transmission channels, investors selling govern- cial markets have responded swiftly and positively,
ment bonds will seek to rebalance portfolios toward other appreciably lowering sovereign and private borrowing
higher-yield assets; higher asset prices and lower risk-free costs and weakening currencies. This has helped to
rates will drive down borrowing costs in capital markets. significantly reduce fragmentation and lift demand
This should, in turn, help rekindle bank lending as banks for loans in the euro area. Inflation expectations have
pass on lower funding costs by reducing interest rates improved, and strong gains in equity markets under-
on their loans. These channels, in combination, will also score further progress through portfolio rebalancing
lead to a depreciation of the exchange rate, particularly channels, laying the basis for positive wealth effects.

International Monetary Fund | April 2015 7


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.4. Asset Valuation and Volatility Heat Maps

1. Asset Valuation Heat Map

USA
Equities

EU
EMs
Sovereign bonds

USA
EU
EMs
High yield Investment grade

USA
Corporate bonds

EU
EMs
USA
EU

2004 05 06 07 08 09 10 11 12 13 14

Source: IMF staff calculations.


Note: Red, orange, yellow, and green = the four quartiles of the price (spread) distribution of equity valuations (bond
spreads), with red denoting the top (bottom) quartile of the distribution over July 2004February 2015. EM = emerging
market; EU = European Union; USA = United States.

2. Volatility Heat Map

AE equity

EM equity

AE govt bond

EM govt bond

AE FX

EM FX

AE credit

Commodities

2004 05 06 07 08 09 10 11 12 13 14

Sources: Bloomberg, L.P.; and IMF staff calculations.


Note: Red, orange, yellow, and green = the four quartiles of the volatility distribution, with red denoting the top quartile of
the distribution over 200415. Based on percentiles of three-month realized asset volatility. AE = advanced economy;
EM = emerging market; FX = foreign exchange; govt = government.

However, to maximize the benefits of QE in boosting Moreover, steps should be taken to mitigate some
real activity through higher credit growth, additional of the challenges that arise with QE. By design, QE
measures are needed to restore balance sheet health encourages greater financial risk taking, yet monitor-
in the private sector, particularly policies aimed at ing and eventually addressing any ensuing financial
comprehensively tackling the burden of nonperforming excesses and other undesirable financial side effects
assets in the euro area. is necessary. Although a wealth effect is a benefit of

8 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.5. Quantitative Easing Impact Channels debt repayment and write-offs has reduced debt levels
in a number of euro area countries, while macroeco-
Ination nomic deleveraging through growth and inflation has
Interest rates lowered Portfolios rebalanced
expectations raised
played a larger role in the United Kingdom and the
United States. But private sector leverage remains
Capital Markets Firms and Households elevated in many economies. Looking forward, expected
Banking System
Higher asset prices Lower borrowing costs growth and inflation under existing monetary policies
Higher net worth Reduced leverage Lending rekindled
will likely be insufficient to reduce debt levels signifi-
cantly. A more complete set of policy actions is required
Increasing Improving private
risk appetite balance sheets to complement accommodative monetary policies
Stretched asset Greater Greater and address the debt overhang in the private sector.
valuations nancial economic Greater
risk taking risk taking investment
Rising credit
and liquidity risks
In the years leading up to the global financial
Stronger
Life insurers consumption crisis, the private sector in many advanced economies,
under pressure Higher growth including in the euro area, the United Kingdom, and
Currency Greater ows to Higher and potential the United States, increased leverage on the strength
volatility emerging markets ination
of rising growth expectations and favorable financial
Source: IMF staff. conditions (Figure1.6). The crisis exposed the fragil-
ity of this credit-driven growth model and the risks to
growth associated with high debt. In particular, high
increased asset prices, there is also a risk of stretched private debt levels raise the sensitivity of borrowers to
asset prices. Lower interest rates also place strains on adverse shocks, reduce profitability, and put upward
the profitability of financial institutions that derive pressure on nonperforming loans and corporate
interest income by exploiting the slope of the yield bankruptcies, increasing risks to bank asset quality
curve. Life insurers with guaranteed payouts on their and broader financial stability.1 Furthermore, when
liabilities are at particular risk in a low-interest-rate highly indebted private agents are unable to benefit
environment. Low interest rates may also lead to a from lower funding rates to increase their borrowing,
search for yield by investors, prompting them to take high debt also undermines monetary policy transmis-
on greater credit and liquidity risks to generate more sion mechanisms. This hampers private balance sheet
income. A sharp depreciation of the domestic exchange cleanup and economic recovery, as is discussed in the
rate from significant portfolio rebalancing into foreign section Disinflationary Risk and Financial Stability.
assets could increase volatility in currency markets. In countries where private balance sheets remain over-
This report examines the risk landscape as the BOJ extended, debt reduction necessary to reduce financial
and ECB augment their expanded asset purchase stability risks, but debt reduction must be handled in a
programs while the Federal Reserve is expected to start way that is consistent with the recovery. The pace and
gradually raising policy rates. A key message of this composition of deleveraging have important macroeco-
report is that additional policy measures are required nomic implications.
to enhance the effectiveness of accommodative central Major advanced economies have made mixed
bank policies. These measures are needed to facilitate progress in deleveraging private nonfinancial sector
sustainable economic risk taking, contain the resulting balance sheets. Householdsespecially in the United
financial excesses, address crisis legacies, and engineer a Kingdom and the United Stateshave sharply reduced
successful exit from the global financial crisis. their gross debt as a share of GDP, but gross household
debt is still high in many countries. Although lever-
Macroeconomic Versus Balance Sheet
Deleveraging: What Is in the Mix? 1High debt can impede growth, which in turn can undermine

financial stability. Studies have shown that high debt is generally


Accommodative monetary policies in advanced econo- associated with low medium-term growth, although at different debt
mies have helped reduce private nonfinancial debt thresholds (see references in Chen and others 2015). Other studies
have shown that high private sector leverage has been detrimental
ratios by supporting inflation and growth and increas- to postcrisis economic performance (see Bornhorst and Ruiz Arranz
ing asset prices. Balance sheet deleveraging through 2013; ECB 2012).

International Monetary Fund | April 2015 9


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.6. Private Nonnancial Sector Gross Debt


(Percent of GDP)

1. Major Advanced Economies 2. Selected Euro Area Countries


Household debt Nonnancial corporate debt Household debt Nonnancial corporate debt
120 160 120 160
Greece Italy
140 Portugal Spain 140
100 100
120 120
80 80
100 100

60 80 60 80

60 60
40 40
Euro area 40 40
Japan
20 United Kingdom 20
United States 20 20

0 0 0 0
1990 96 2002 08 14 1990 96 2002 08 14 1990 96 2002 08 14 1990 96 2002 08 14
Sources: Bank for International Settlements; and IMF staff estimates. Sources: Bank for International Settlements; and IMF staff estimates.

age among nonfinancial firms is down from its peak crisis is that addressing weak balance sheets early on
in many advanced economies, the corporate sector in can improve the financial and economic responses to
some euro area countries is still highly leveraged, in unconventional monetary policies.
part because resolution of impaired assets has pro- Asset price appreciation due to accommodative
gressed slowly. In the United States, where corporate monetary policies (conventional and unconventional)
leverage is relatively low, companies have stepped up can also contribute to deleveraging. The appreciation
borrowing in recent years amid favorable financing of household and corporate financial assets can help
conditions and increased financial risk taking. reduce the net financial debt of the private sector, even
if gross debt remains unchanged. This is an important
What factors have contributed to deleveraging? channel for policy, especially for countries in which
Reductions in gross debt ratios can come from two central bank asset purchases have helped to lower the
sources: macroeconomic deleveraging (through growth risk-free rate. Asset-side deleveraging has not oper-
and inflation) and balance sheet deleveraging (through ated much in the euro area so far, but it has played an
debt repayment and write-offs). Countries that have important role in Japan, the United Kingdom, and the
been able to generate higher growth and inflation have United States. Since 2007, the net financial debt of
been able to minimize the need for balance sheet dele- households and firms in these economies has declined
veraging and the associated credit contraction (Figure by about 10 percentage points of GDP or more solely
1.7). But the deleveraging process has varied substan- as a result of asset price gains (Tables 1.1 and 1.2, asset
tially across countries. revaluation columns). In contrast, euro area countries
Write-offs can play an important role in tackling such as France, Greece, Portugal, and Spain have not
high debt burdens where efficient debt resolution benefited as much from this channel so far.
mechanisms are in place. In particular, the cleanup of
impaired assets on balance sheets can contribute to How much more deleveraging could be achieved
private sector deleveraging as long as countries have through unconventional monetary policies?
efficient mechanisms for debt restructuring (Figure Macroeconomic deleveraging through 2020 could
1.8). These mechanisms may allow countries to limit reduce corporate and household indebtedness, but in
the macroeconomic costs of debt restructuring and a number of economies it would not be sufficient to
restart credit flows more rapidly. A key lesson from the eliminate high debt loads. Although it is difficult to

10 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.7. Episodes of Private Sector Deleveraging in Selected Advanced Economies

1. Household Gross Debt: Contributions of Macroeconomic Factors 2. Nonnancial Corporate Gross Debt: Contributions of Macroeconomic
and Balance Sheet Deleveraging since 2007 Factors and Balance Sheet Deleveraging since 2007
(Percent of GDP) (Percent of GDP)
25 40
Growth Ination Debt repayment Growth Ination Debt repayment
20
Debt write-offs Change in debt Debt write-offs Change in debt 30
15
10 20
5 10
0
5 0
10 10
15
20
20
25 30

Lithuania
Lithuania

Italy

Italy
Austria
United Kingdom

Austria
United Kingdom
Portugal

Japan

Japan
Latvia

Portugal
United States
Latvia
United States

Spain
Spain

France
Slovenia

Slovenia
Germany

France
Germany
Greece

Greece
Sources: Bank of Japan; Eurostat; Federal Reserve; and IMF staff estimates.
Note: Debt write-offs reect other changes in debt unexplained by ows and may also capture revaluation of marketable debt.

3. Household Gross Debt: Range and Projection of Debt Levels 4. Nonnancial Corporate Gross Debt: Range and Projection of
(Percent of GDP) Debt Levels
(Percent of GDP)
2014 Minimum since 1999 Maximum since 2006 2014 Minimum since 1999 Maximum since 2006
2020 Average 19992006 2020 Average 19992006
120 120

100 100

80 80

60 60

40 40

20 20

0 0
GRC ITA PRT ESP AUT FRA DEU JPN GBR USA GRC ITA PRT ESP AUT FRA DEU JPN GBR USA

Sources: Bank of Japan; Eurostat; Federal Reserve; and IMF staff estimates. Sources: Bank of Japan; Eurostat; Federal Reserve; and IMF staff estimates.
Note: Expected deleveraging estimates are based on latest World Economic Note: Expected deleveraging estimates are based on latest World Economic
Outlook projections of growth and ination. See Table 1.1 for further details. Outlook projections of growth and ination. See Table 1.2 for further details.
AUT = Austria; DEU = Germany; ESP = Spain; FRA = France; GBR = United AUT = Austria; DEU = Germany; ESP = Spain; FRA = France; GBR = United
Kingdom; GRC = Greece; ITA = Italy; JPN = Japan; PRT = Portugal; USA = Kingdom; GRC = Greece; ITA = Italy; JPN = Japan; PRT = Portugal; USA =
United States. United States.

define a threshold for a safe level of debt, a number Italy, Portugal, and Spain would remain above or near
of major advanced economies whose debt increased 70 percent of GDP by 2020 under current World
sharply are still likely to have debt above their precrisis Economic Outlook projections for growth and inflation,
average.2 For example, gross corporate debt in France, higher than their precrisis averages and higher than
those of other major advanced economies (Table 1.2;
2High debt is generally associated with low medium-term growth
Figure 1.7, panel 4). Similarly, under current World
(see Cecchetti, Mohanty, and Zampolli 2011; Kumar and Woo
2010; Baum, Checherita, and Rother 2013; Reinhart and Rogoff Economic Outlook projections for growth and inflation,
2010), albeit at different thresholds (Chen and others 2015). by 2020, gross household debt in Portugal and the

International Monetary Fund | April 2015 11


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.8. Insolvency Frameworks and Macroeconomic Third, minimizing the negative impact of debt
Deleveraging restructuring on the economy requires efficient
legal and institutional mechanisms for the prompt
United
0.6
cleanup of impaired assets.
Stronger Germany Austria States
Insolvency Japan 0.4 Finally, countries with high public debt must
Frameworks Latvia 0.2 improve their fiscal frameworks, as highlighted

Legal system (z-score)


France United 0.0
Portugal Kingdom in the April 2015 Fiscal Monitor. High debt and
0.2 deleveraging in all three sectors (public, corporate,
Spain Lithuania 0.4
Slovenia 0.6
household) has been shown to be especially del-
y = 0.041x 0.2224 0.8 eterious to growth (see Bornhorst and Ruiz Arranz
R = 0.4538 1.0 2013). Fiscal frameworks with better guidance
Greece Italy 1.2 on the medium-term objectives can provide more
1.4
20 10 0 10 20 flexibility on the conduct of fiscal policy over the
Cumulative macroeconomic deleveraging, 200714 (percent of GDP)
economic cycle.

Sources: National statistics ofces; World Bank, Doing Business Survey (2014); IMF,
Financial Soundness Indicators; and IMF staff estimates.
Note: The legal system indicator is an average of z-scores from seven different
indicators of legal system strength from the Doing Business Survey, relating to
Disinflationary Risks and Financial Stability
resolving insolvency, enforcing contracts, and the strength of legal rights. Quantitative Easing in the Euro Area and Japan: What
Are the Channels and Risks?
While the U.S. Federal Reserve is expected to start
United Kingdom would remain relatively high com-
gradually raising policy rates, the euro area and Japan
pared with that of other major advanced economies
have recently embarked on further asset purchases (QE)
(Table 1.1; Figure 1.7, panel 3).3
to significantly strengthen their responses to persistent
Policies to facilitate further private sector disinflationary pressures. Some key transmission chan-
deleveraging nels of QE are already beginning to work. Financial
markets have responded swiftly and positively, appre-
High private sector debt levels can continue to pose
ciably lowering sovereign and private borrowing costs
obstacles to growth and financial stability. Contri-
and weakening currencies. To maximize the impact
butions may be needed from all three deleveraging
of QE, it is necessary to complement central bank
sources: macro deleveraging (growth and inflation),
actions with measures to restore balance sheet health in
balance sheet deleveraging (debt repayment and
the private sector, including through expeditious debt
restructuring), and asset revaluation (for net indebted-
write-downs and restructuring, enhance the sound-
ness). A complete set of policies is necessary to return
ness of nonbank institutional investors, and promote
debt to safer levels:
structural reforms. Failure to support current monetary
First, accommodative monetary policies (including
policies will leave the economy vulnerable and risks tip-
QE) should help support private sector deleveraging,
ping it into a downside scenario of increased deflation
including by boosting asset prices and generating
pressure, a still-indebted private sector, and stretched
wealth effects. But these will likely not be sufficient
bank balance sheets. Finally, QEby designentails
if potential growth remains low. In such cases,
a continued low-interest-rate environment. While this
countries need to enhance their longer-term growth
should help the macro economy, it will pose severe
potential through a comprehensive program of
challenges to institutional investors, particularly weak
structural reforms.
European life insurers, further weighing on their abil-
Second, debt restructuring and write-offs can
ity to rerisk their balance sheets in support of QE.
improve the financial and economic response to
unconventional monetary policies by unclogging the Central banks have embarked on further mon-
monetary transmission mechanism. etary easing in the euro area and Japan

3The projections for growth and inflation are based on the latest In October 2014, amid weak demand and continu-
WEO forecasts and assume no new debt and no debt write-offs. ing downward price pressures, the BOJ introduced

12 International Monetary Fund | April 2015


Table 1.1. Household Debt in the Euro Area, Japan, the United Kingdom, and the United States
(Percent of GDP)
Gross Debt Net Financial Debt Contributing Factors Expected Gross Debt
Change from Deleveraging 2020
2007 from Growth (with growth Precrisis
(percentage Net Debt Debt Asset Asset and Inflation and inflation average
2007 2014 2007 2014 points) Growth Inflation Issuance Write-Offs Accumulation Revaluation by 2020 only) (19992006)
Euro area
Austria 51.9 50.5 81.6 83.4 1.8 2.1 5.9 3.3 3.3 3.9 4.2 7.6 43.0 47.7
France 46.5 55.6 76.3 71.9 4.4 1.2 3.8 13.2 0.9 3.0 1.7 8.4 47.1 37.4
Germany 61.2 54.4 55.4 56.7 1.3 2.7 5.6 1.2 0.2 0.7 4.9 8.5 45.9 69.1
Greece 46.0 63.8 89.3 60.7 28.6 17.9 0.9 2.6 3.4 44.2 55.1 14.2 49.5 23.5
Italy 38.2 42.8 149.4 151.4 2.0 3.8 4.0 5.3 0.5 9.6 2.9 4.8 38.0 27.4
Latvia 43.6 26.9 4.0 34.7 30.6 4.6 4.5 12.1 4.7 20.4 6.5 7.5 19.5 29.4
Lithuania 26.5 22.0 25.2 39.4 14.2 0.1 4.8 1.4 1.2 2.3 11.9 6.2 15.8 14.4
Portugal 86.9 82.6 53.7 68.3 14.6 7.0 5.5 4.9 0.9 15.6 5.2 12.2 70.4 69.2
Slovenia 24.5 28.6 62.1 58.3 3.9 1.2 2.6 4.5 1.0 24.7 24.5 4.6 24.1 17.6


Spain 81.1 72.0 61.6 77.5 15.9 4.3 2.5 8.3 2.7 12.1 5.3 10.9 61.2 56.4
Japan 63.1 62.2 143.3 174.1 30.8 0.4 3.7 3.2 1.0 28.7 1.2 5.4 56.8 67.2
United Kingdom 96.2 87.1 23.8 34.9 11.1 2.9 14.3 11.3 3.2 18.5 20.5 18.1 69.0 80.0
United States 95.5 76.9 84.5 118.2 33.7 6.1 9.3 3.7 7.0 13.0 28.0 17.3 59.5 79.2
Sources: Bank of Japan flow of funds; Eurostat financial accounts and consolidated debt statistics; Federal Reserve flow of funds; and IMF staff estimates.
Note: Figures for 2014 are preliminary and as of 2014:Q3.Gross debt includes securities and loans.Net financial debt is defined as gross debt minus financial assets in the forms of cash and deposits, debt security holdings, and equity
and mutual fund shares.For euro area countries, debt figures are on a consolidated basisas of end-2013(that is,netting out intrasectoralborrowing). Growth contribution is derived as g/(1+g++g) times previous period debt ratio;
inflation contribution is derived as /(1+) times previous period debt ratio in which = growth rate of GDP deflator and g = real GDP growth rate.Net debt issuance and debt write-off contributions come from flow of funds. Debt
write-offs reflectother changes in debt unexplained by flows andmay also capturerevaluation of marketable debt. Asset accumulation indicates changes in asset-to-GDP ratios, excluding asset revaluation effects, which come from flow
of funds. Expected deleveraging estimates are based on latestWorld Economic Outlookprojections of growth and inflation.For Latvia and Lithuania, the precrisis averages are from 2004 to 2006; for Slovenia the precrisis average is from
2001 to 2006. Color coding is based on the percentile, with factors reducing (increasing) debt shown in green (orange).
CHAPTER 1 Enhancing policy traction and reducing risks

International Monetary Fund | April 2015 13


14
Table 1.2. Corporate Debt in the Euro Area, Japan, the United Kingdom, and the United States
(Percent of GDP)
Gross Debt Net Financial Debt Contributing Factors Expected Gross Debt
Change from Deleveraging 2020 (with
2007 from Growth growth Precrisis
(percentage Net Debt Debt Asset Asset and Inflation and inflation average
2007 2014 2007 2014 points) Growth Inflation Issuance Write-Offs Accumulation Revaluation by 2020 only) (19992006)
Euro area

International Monetary Fund | April 2015


Austria 75.3 73.9 21.5 7.4 14.1 2.9 8.6 9.4 0.8 7.4 5.3 11.1 62.9 75.1
France 69.1 84.4 10.0 3.7 6.4 1.8 5.6 20.0 2.7 10.2 1.3 12.8 71.6 66.4
Germany 49.7 47.3 20.5 30.6 10.1 2.2 4.7 4.7 0.3 4.5 12.1 7.4 39.9 51.2
Greece 55.9 63.0 23.3 22.5 0.8 20.3 1.6 11.2 22.7 2.0 10.0 14.1 49.0 44.9
Italy 71.5 76.7 23.2 26.0 2.7 7.0 7.2 5.3 0.1 7.7 5.2 8.6 68.1 58.1
Latvia 59.6 59.8 41.4 39.3 2.1 4.2 7.9 0.9 2.9 5.5 3.3 16.6 43.2 53.1
Lithuania 48.1 35.5 30.4 1.2 31.6 0.8 7.9 2.9 2.7 4.9 14.0 10.0 25.5 36.5
Portugal 98.2 108.5 62.7 66.0 3.3 8.7 6.8 17.7 9.3 5.3 1.7 16.0 92.5 86.2
Slovenia 71.8 69.4 46.1 50.1 4.0 4.0 7.4 8.9 7.8 1.5 7.9 11.1 58.4 51.8
Spain 110.7 94.4 43.5 19.3 24.1 6.2 3.4 10.0 9.1 7.7 0.1 14.2 80.2 71.1
Japan 100.2 103.0 3.2 22.9 26.0 0.6 6.1 2.6 0.1 20.6 8.2 9.0 94.1 113.9
United Kingdom 87.8 75.4 6.7 3.6 3.1 2.5 13.7 0.2 4.0 1.7 10.9 15.7 59.7 79.2
United States 69.8 67.5 58.7 56.9 1.8 4.9 7.2 11.1 1.3 0.6 0.1 15.2 52.3 63.3
Sources: Bank of Japan flow of funds; Eurostat financial accounts and consolidated debt statistics; Federal Reserve flow of funds; and IMF staff estimates.
GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Note: Figures for 2014 are preliminary and as of 2014:Q3.Gross debt includes securities and loans.Net financial debt is defined as gross debt minus financial assets in the forms of cash and deposits, debt security holdings, and equity
and mutual fund shares.For euro area countries, debt figures are on a consolidated basisas of end-2013(that is,netting out intrasectoralborrowing). Growth contribution is derived as g/(1+g++g) times previous period debt ratio;
inflation contribution is derived as /(1+) times previous period debt ratio in which = growth rate of GDP deflator and g = real GDP growth rate.Net debt issuance and debt write-off contributions come from flow of funds. Debt
write-offs reflectother changes in debt unexplained by flows andmay also capturerevaluation of marketable debt. Asset accumulation indicates changes in asset-to-GDP ratios, excluding asset revaluation effects, which come from flow
of funds. Expected deleveraging estimates are based on latestWorld Economic Outlookprojections of growth and inflation.For Latvia and Lithuania, the precrisis averages are from 2004 to 2006; for Slovenia the precrisis average is from
2001 to 2006. Color coding is based on the percentile, with factors reducing (increasing) debt shown in green (orange).
CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.9. Quantitative Easing and Financial Markets

QE began working well before it was announced... ...and has delivered some price improvements.
1. Changes in 10-Year Sovereign Yields 2. Changes in Equities and Exchange Rates
(Percentage points) (Percent)
Since
QQE1
First two years of program First two years of program
1.00 60
Since
Since Jackson Hole Since QE 50
QQE1
0.50 Since
40
Jackson
Hole 30
0.00
Since QE 20
0.50 10
0
1.00
10
Exchange rates (trade-weighted) Equities
1.50 20
USA USA USA DEU ESP ITA DEU ESP ITA JPN USA USA USA EA EA JPN
QE1 QE2 QE3 QE1 QE2 QE3

An increasing number of short- and long-term European government Negative policy rates in some European countries have reinforced
bonds carry a negative yield. negative yield dynamics.
3. European Government Bonds with Negative Yields 4. Policy Rate and Two-Year Government Bond Yields
(Years to maturity) (Percent)

1 2 3 4 5 6 7 8 9 10 Switzerland Denmark Sweden Euro area


0.0
Germany 0.1
Finland
0.2
Netherlands
Austria 0.3
Belgium 0.4
France
Policy rate 0.5
Spain
Two-year bond 0.6
Italy
Portugal 0.7
0.8
Sources: Bloomberg, L.P.; and IMF staff calculations.
Note: Jackson Hole refers to ECB President Draghis speech in August 2014 at the U.S. Federal Reserves Economic Policy Symposium in Wyoming. ECB QE
was announced in January 2015. Japans QQE1 is taken from April 2013. U.S. Fed programs QE1 November 2008; QE2 November 2010; QE3 September
2012. DEU = Germany; EA = Euro area; ESP = Spain ITA = Italy; JPN = Japan; QE = quantitative easing; QQE = quantitative and qualitative easing; USA = United
States. In Figure 1.9.4, policy rates comprise the ECB deposit rate, Riksbank repo rate, Swiss National Bank Libor target, and Danmarks National Bank CD rate. The
two-year bond for the euro area is a debt-weighted yield of negative yielding two-year government bonds.

an expanded program of quantitative and qualita- largely be accommodated in sovereign markets, with a
tive easing (QQE2). The BOJ announced that it was small portion also coming from European Union (EU)
accelerating the pace of Japanese government bond institutions. If fulfilled, QE will take the ECBs bal-
purchases from an annual pace of 50 trillion to about ance sheet from an estimated 22 percent to 31 percent
80 trillion, and extending the average remaining of GDP, in line with the initial QE programs of the
maturity of government bond purchases to about 7 to Federal Reserve and the BOJ, which each subsequently
10 years. The BOJs balance sheet is expected to exceed increased their programs to about 20 percent and 45
70 percent of GDP by the end of 2015. percent of respective GDP.
Similarly, the ECB increased its monthly asset pur- Although at different stages, QE programs in Japan
chases to 60 billion, after averaging about 12 billion and the euro area have already had a significant impact
under the existing asset-backed securities and covered on financial markets. In the euro area, much of this
bond purchase programs, to address the risks of persis- was achieved in the wake of ECB President Draghis
tently low inflation. This will result in a total program speech at Jackson Hole in August 2014a date widely
of about 1.1 trillion by September 2016 and will taken as the trigger for QE. Ten-year sovereign yields

International Monetary Fund | April 2015 15


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

in Germany, Italy, and Spain, among others, declined ment, through the combined Asset Quality Review
before the implementation of QE by as much as and Stress Tests, has credibly boosted the transparency
10-year U.S. Treasury bonds did during the first two of bank balance sheets, while the establishment of the
years of the Federal Reserves QE programs (Figure European Stability Mechanism, along with the Single
1.9, panel 1). Positive market impacts were reinforced Resolution Mechanism and Fund, has enhanced the
following the official announcement of QE in January capacity of the euro area economies to safeguard finan-
(Figure 1.9, panel 2). As of late March 2015, more cial stability. These actions have supported the founda-
than 30 percent (or 2.4 trillion) in short- and long- tions for recovery, helped reduce fragmentation, and
term euro area government bonds had negative yields boosted investor confidence in the euro area.
(Figure 1.9, panel 3). These improvements and associ-
ated positive ripple effects through credit markets have A strong portfolio rebalancing channel will be key to
helped significantly to reduce fragmentation, improve the transmission of QE
credit conditions, and raise demand for loans.4 More- A strong portfolio rebalancing channel is a key trans-
over, strong gains in equity markets in both Japan and mission channel for QE. Rebalancing could occur in
the euro area underscore progress through the portfo- three central ways. First, rebalancing lowers risk-free
lio rebalancing channel, laying the basis for positive rates, which translates into lower funding costs. Sec-
wealth effects. There has also been a positive impact on ond, rebalancing from sovereign bonds into more risky
inflation expectations in the euro area, as measured by assets should reduce lending spreads and thus credit
inflation swaps. In Japan, different measures of infla- costs. However, this is most likely to benefit large com-
tion expectations, which steadily rose until mid-2014, panies that have access to markets, with limited direct
have fallen recently and converged to about 1 percent. support for small and medium-sized enterprises. Third,
The ECBs QE program complements a broader set there could be portfolio outflows from the economies
of measures to address tail risks and safeguard mon- engaging in QE, primarily to the United States, but
etary transmission, for example, the lowering of policy also increasingly to emerging markets.
rates to historic levelsincluding negative territory, Institutional investors are key to the transmission of
in line with some other European countries (Figure QE to the private sector in the euro area. In particular,
1.9, panel 4).5 Although providing a credible signal of substantial intra-euro-area portfolio rebalancing within
the ECBs accommodative stance, prolonged negative credit markets would directly lower private funding costs
rates could cause disruptions to short-term funding and have similar beneficial knock-on effects for smaller
markets, particularly money market funds.6 Bank term entities. However, European life insurers, which hold
funding has been ensured through a combination of about 20 percent of EU government bonds, may have
expanded collateral eligibility, fixed-rate full allotment limited incentive to sell bond portfolios to the ECB,
facilities, and longer-term refinancing operations. The partly because of regulatory considerations, but also as a
announcement of the Outright Monetary Transac- result of their weak balance sheets (as discussed later in
tions program eliminated euro redenomination risks this chapter). Given significant duration mismatches, the
and lowered spreads on euro area government bonds. cash from a bond sale would need to be reinvested into
Progress toward banking union, including the Single similar-duration bonds, which have less attractive yields,
Supervisory Mechanism, has helped strengthen the putting further pressure on life insurers already weak cash
health of banks through enhanced and harmonized flow positions. Since rerisking by taking on lower-quality
regulation and supervision. The Comprehensive Assess- credit will further eat into their fragile capital buffers,
rebalancing will likely take place primarily in investment-
4See the ECBs Euro Area Bank Lending Survey for the fourth grade sovereign and corporate bonds, particularly in U.S.
quarter of 2014. bonds, given the combination of attractive yields, long
5See IMF 2014a for a more complete discussion of these policies.
duration, and low foreign currency hedging costs.
6If interest rates were to turn sharply negative and remain at those
In Japan, the government bond maturity extension
levels for a protracted period, including at the retail level, larger
distortions could arise with far-reaching financial stability implica- under QQE2 is expected to lead to more portfolio
tions. For example, savers could stop saving, bank deposits could be rebalancing at life insurers and pension funds. Life
turned into cash, new forms of cash management could emerge, and insurers and pension funds are now the largest hold-
borrowers could be encouraged to take on excessive leverage, with
long-term consequences for some asset markets, such as housing ers of Japanese government bonds and are in a better
markets. position to rerisk their balance sheets, including toward

16 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

higher-yielding securities. This partly follows the reform panels 1 and 2). Such heavy foreign rebalancing would
of the Government Pension Investment Fund (GPIF), weaken domestic currencies, underscoring the impact
which encourages higher allocations away from govern- of QE on the exchange rate, and could pose some
ment bonds, and will induce other public and private risks by adding to movements in global exchange rates.
pension funds to follow the GPIFs lead. Finally, the While these potential flows could also partly offset the
combination of low domestic yields and low foreign risks of outflows from emerging markets as the Federal
exchange hedging costs should boost the incentives of Reserve begins to exit, the potential for short-term
insurers and pensions funds to rebalance their portfolios volatility could increase.
abroad, particularly to the United States.
One way to gauge the potential portfolio flows is to Bank lending may take time to fully recover
simulate the effects of alternative policy outcomes on In previous episodes of QE, bank credit has taken time
portfolio choices through three stylized scenarios: to fully recover. Bank lending has accelerated only
Under a baseline scenario, central bank asset modestly in Japan and the United Kingdom since the
purchases under QE reduce risk-free rates and boost launch of their QE programs (Figure 1.11, panel 1).7
some asset prices, putting a floor under growth and Even in the United States, where bank credit is now
inflation and supporting inflation expectations. But growing quickly, it took at least a year after the launch
clogged bank balance sheets and continued private of its third QE program before lending started to pick
sector indebtedness limit a fuller transmission of QE up. Although the overall economic and policy environ-
to real activity. As a result, confidence and activity ment was different in these casesand QE should
are slower to recover, prompting investors to rebal- help credit conditions as it reduces bank funding costs,
ance a portion of their assets abroad. which should be reflected in lower lending ratespast
But there is a risk that a slower recovery will leave experience suggests that bank lending in the euro area
the economy vulnerable to adverse shocks or and Japan may pick up with a lag.
policy slippages, leading to a downside sce- Furthermore, the ability and willingness of banks to
nario. Here negative shocks leave QE (in its current supply more credit will depend on the business environ-
form) unable to put a floor under growth and infla- ment and regulatory conditions they are facing. Before
tion, resulting in further balance sheet weakness and the global financial crisis, banks were primarily concerned
drifting inflation expectations. The situation is exac- about meeting risk-weighted capital regulations. How-
erbated by a lack of progress on policies to repair ever banks now need to operate their businesses under a
private balance sheets, further eroding confidence multidimensional set of regulatory and economic targets
and prompting additional capital flight. that they need to meet simultaneously, including regular
This underscores the need for additional structural supervisory stress testing and the new Total Loss-Absorb-
measures to repair private balance sheets, a QE- ing Capacity requirement for global systemically impor-
plus scenario, complementing monetary policy tant banks (Figure 1.11, panel 2).8 Although differences
and helping boost growth and inflation. In this case, in national implementation are complicating this picture
investors would want to increase their relative domes-
tic exposures, resulting in fewer portfolio outflows. 7This, however, does not imply that there has been no impact on

And although this would limit exchange rate depre- bank lending from QE. For example, Saito and Hogen (2014) find that
a decrease in the interest rate risk at major Japanese banks under QQE1
ciation, it would reinforce positive domestic price has been associated with higher bank lending, after controlling for loan
effects as demand for higher-yielding assets increases. demand, interest rate spreads, and the nonperforming loan ratio.
8See, for example, EBA 2015 for a discussion of the implications

of regulatory measures on bank business models. The target ratios


The scenarios suggest that QE in the euro area and used in Figure 1.11, panel 2, are profitability (10 percent return on
Japan could lead to significant portfolio outflows. In equity target), loss-absorbing capital (estimated total loss-absorbing
the baseline scenario, additional gross outflows from capacity ratio at 18 percent of risk-weighted assets and 6 percent of
total assets), leverage (minimum 3 percent but requirements higher
the euro area economies amount to 1.2 trillion by the
in Switzerland, the United Kingdom, and the United States), asset
end of 2015, raising gross outflows from 50 percent to quality (nonperforming loans 10 percent of gross loans), stable fund-
55 percent of GDP. Similarly, insurance companies and ing (estimated net stable funding ratio of 100), and capital (Tier 1
pension funds in Japan could invest as much as 42 common capital of 7 percent, plus systemically important financial
institution buffers, or plus 0.5 percent for large domestic banks). The
million ($350 billion), or 8 percent of GDP, in foreign figure is based on a sample of more than 300 advanced economy
assets by the end of 2017 (Annex 1.2; Figure 1.10, banks.

International Monetary Fund | April 2015 17


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.10. Illustrative Baseline QE Portfolio Rebalancing Scenarios in Japan and the Euro Area

1. Portfolio Outows of Japanese Insurance Companies and 2. Euro Area Portfolio Outows
Pension Funds (Percentage of euro area GDP)
(Trillions of yen)
20 60
United States Forecast
Projection Other advanced economies
Public pensions
Insurance companies and private pensions Emerging markets
15 50

10 40

5 30

0 20

5 10

10 0
2005 07 09 11 13 15 17 2004 05 06 07 08 09 10 11 12 13 14 2015
Sources: Bank of Japan; and IMF staff projections. Source: Bloomberg L.P.; Coordinated Portfolio Investment Survey (CPIS);
Note: See Annex 1.2 for a description of the assumptions. European Central Bank; and IMF staff calculations.
QE = quantitative easing. Note: Foreign currency-denominated ows exclude central banks, banks and
government holdings, and are determined by European insurer asset allocations
as of 2013, while destinations are based on data for 2013 international portfolio
ows (CPIS). Data for 2014 are IMF estimates. Emerging markets (EM) comprise
Argentina, Brazil, Chile, China, Colombia, Hungary, India, Indonesia, Malaysia,
Mexico, Philippines, Poland, Russia, Saudi Arabia, Singapore, South Africa,
Thailand, and Turkey. Other Advanced Economies (OA) comprise Australia,
Canada, Denmark, Japan, New Zealand, Norway, Sweden, Switzerland, and the
United Kingdom.

for global banks, the introduction of the Single Super- government bond holdings. So banks may face limits on
visory Mechanism has helped strengthen bank balance the degree to which they can reallocate sovereign bond
sheets and further fostered the process of supervisory and portfolios toward riskier assets, because the average risk
regulatory harmonization across euro area countries, as weight would rise, eroding bank buffers.
discussed in Box 2.3. Even if banks have the capacity to expand their loan
The ECBs Comprehensive Assessment has credibly portfolios, there is a risk that they may reallocate their
boosted the transparency of bank balance sheets and portfolios toward more profitable strategies. Table 1.3
fostered significant improvements in capital. This has provides some stylistic examples of possible alternative
made capital and leverage less of a constraint for most investment choices. According to these estimates, banks
banks in both the euro area. Nonetheless, institutions may have incentives to invest in higher-yielding bonds,
may be reluctant to use current buffers to increase their such as U.S. and emerging market sovereign bonds.
lending, particularly given the challenges that histori-
cally low profitability are posing for business models, In the euro area, improving asset quality is
as discussed in the October 2014 Global Financial important to boost bank lending
Stability Report. One reason for this is that banks may be In the euro area, improving asset quality at some banks
reluctant to reduce capital ratios in the absence of clarity could further bolster bank credit. Asset quality contin-
on the amount of capitalization that will be required by ued to deteriorate in the euro area as a whole in 2014,
regulators over the medium term. A second reason is although at a slowing pace, with total nonperforming
that in many countries, the average risk weight of bank loans now standing at more than 900 billion (Figure
assets is low relative to the past, reflecting a high share of 1.12, panel 1). Furthermore, the stock of nonperforming

18 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.11. Bank Lending and Constraints

Bank lending growth has lagged in past QE episodes... ...and banks now face new constraints.
1. Bank Credit Growth to the Nonnancial Private Sector 2. Proportion of Banks Missing Target Ratios
(Percent) (Percent of sample assets)
6 United States
5 36 37 77 89 88 24 50 Protability
Euro area Loss-
4
Japan 19 87 52 17 41 absorbing
3 capital
2 United Kingdom 52 5 26 1 Leverage
1
Asset
0 3 36 3 quality
1 Stable
52 1 37 9
2 funding
1 0 1 2 3
Years since start of quantitative easing program 1 1 Capital
Sources: Bank of England; Haver Analytics; and IMF staff calculations. Specialist Global Regional Euro Other North Asia-
Note: Quantitative easing (QE) programs are United Kingdom (QE2, Oct. 2011); bank area European American Pacic
United States (QE3, Sep. 2012); Japan (QQE1, Apr. 2013); and euro area (QE,
Mar. 2015). QE = quantitative easing; QQE = quantitative and qualitative easing. Commercial banks
Sources: SNL Financial; and IMF staff calculations.
Note: See footnote 8 for a description of the targets.
3. Nonperforming Assets and Credit

Operational costs of Interest rates not cut


holding nonperforming loans as much to recoup prot
Initial
provisions and Lending
unexpected supply
provisions Prots and
Private Gross non- Provisions capital
sector performing Lower generation
debt loans Interest
income
Credit
Net non-
growth
performing Use of
loans scarce
balance
sheet
resources
Less willing to lend Lending
Less willing to borrow demand

Source: IMF staff.

loans in the euro area is unevenly distributed, with about suggesting less active bad debt management and more
two-thirds located in six euro area countries.9 In Cyprus, limited improvement in corporate indebtedness.
Greece, Ireland, Italy, Portugal, and Slovenia, a majority, if Nonperforming assets reduce banks willingness
not all, of the banks involved in the ECBs Asset Qual- and ability to supply credit (Figure 1.11, panel 3) in
ity Review were found to have nonperforming assets of three key ways. First, nonperforming assets are a drag
10 percent or more of total exposure (Figure 1.12, panel on profitability because they require provisioning and
2). These bad assets are large relative to the size of the generate less interest income than performing assets
economy (Figure 1.12, panel 3), even net of provisions. (Figure 1.12, panel 4).10 There are also operating costs
Euro area banks have lagged the United States and Japan to holding nonperforming assets on balance sheets
in the early 2000s in their write-offs of these bad assets,
10Banks with large nonperforming loan portfolios may also face
9The
stock of nonperforming loans in Cyprus, Greece, Ireland, Italy, higher funding costs, although banks may seek to offset this by
Portugal, and Spain in total amounts to more than 600 billion. charging a higher interest rate on new loans.

International Monetary Fund | April 2015 19


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Table 1.3. Reallocating Assets: Stylized Investment Choices


(Percent)
Sovereign Bond Corporate Loan SME Loan
DEU ITA/ESP JPN USA EM IG DEU ITA/ESP JPN DEU ITA/ESP
Investment Return 0.4 1.4 0.4 2.1 3.9 1.0 1.7 0.9 2.4 3.7
Foreign Exchange Hedge 0.3 0.3
Credit Risk 0.2 0.2 0.1 0.3 0.2 0.5 1.0
Operations 0.1 0.3 0.3 0.3 0.4 0.4
Pretax Return 0.4 1.2 0.4 1.9 3.3 0.6 1.1 0.4 1.5 2.3
Required Capital 3 3 3 3 5 6 6 8 10 10
Pretax Return on Required Capital 13 40 13 62 67 10 19 5 15 23
Sources: Bank of America Merrill Lynch; Bank of Japan; European Central Bank (ECB); and IMF staff estimates.
Note: Investment returns using current yields for 10-year sovereign bonds and an index of emerging market investment-grade sovereign bonds. Corporate loan rates
proxied using broad bond indices for large European firms, ECB (interest rate on all new loans), and Bank of Japan (average contracted interest rate on new loans).
SME loan rates are proxied using ECB data (interest rate on new loans under 1 million). The foreign exchange (FX) hedge has a one-year roll-over period; hedging
costs are currently comparable for European and Japanese banks. Credit risk is based on the probability of default for an investment-grade loan rated A/BBB+ and
for an SME loan rated BBB/BB, using sovereign credit risk as a floor. Operational costs are based on usual cost-to-income ratios for corporate and SME loans.
Emerging market credit risk assumed for a sovereign rated BBB. Capital requirements are the maximum of a leverage requirement of 3 percent and a Common
Equity Tier 1 target of 10 percent with risk-weighted assets of 50 percent for emerging market sovereign bonds, 60 percent for corporate loans, 100 percent for
SME loans, and 80 percent for loans to Japanese firms. DEU = Germany; EM IG = emerging market investment grade; ESP = Spain; ITA = Italy; JPN = Japan;
SME = small- and medium-sized enterprise; USA = United States.

(including administrative expenses, legal costs, and Policy actions are needed to support bank lending
maintenance of repossessed property). And even if capacity
banks appear adequately provisioned at a given point These observations suggest that policy actions are
in time, additional provisioning may be needed over needed to further help bank lending in the euro area
time if economic conditions do not improve. Second, and Japan. This can be illustrated through a simula-
nonperforming assetsnet of provisionsuse scarce tion, which is based on the assumption that necessary
resources on bank balance sheets. Net nonperforming actions are not taken. The simulation is estimated
assets need to be backed by capital. They are par- using more than 100 banks in the euro area and
ticularly costly for risk-weighted capital because net about 80 banks in Japan. The capacity of these banks
nonperforming loans on average have a significantly to supply credit is estimated for the period 201517.
higher risk weight than do performing loans. Third, The banks are assumed to preserve their capital buf-
banks with high levels of nonperforming loans on their fers through the simulation, so lending capacity is a
balance sheets may be less willing to lend to borrowers function of retained earnings, which here are based
with borderline credit quality. While many banks are on analysts forecasts. Banks also reallocate portfo-
chasing the same good-quality firmsoften in compe- lios by selling government bonds, in line with the
tition with capital marketsother weaker companies scenarios presented earlier in this chapter. However,
are finding it more difficult to obtain loans. the overall effect is limited, because this reallocation
As a result, banks with high levels of nonperform- raises the average risk weight of banks portfolios. The
ing assets may hamper the transmission of QE via results suggest that without corrective policy actions,
banks. Figure 1.12, panel 4 shows that banks with outlined later in this chapter, median bank lending
a higher ratio of nonperforming loans have tended capacity could be limited to a meager 1 to 3 percent
to lend less recently, even relative to average lend- on average a year, though some individual institutions
ing by banks in the same economy that have faced may be able to increase lending by more (Figure 1.12,
similar demand conditions. This negative relationship panel 5). For banks that have excess capital and are
between bank lending and nonperforming loans was willing to run down their capital buffers, bank lend-
also illustrated in the April 2014 Global Financial ing growth could be higher than suggested by these
Stability Report. simulations.

20 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.12. Bank Nonperforming Loans and Lending Conditions


Nonperforming loans remain at high levels... ...concentrated in a few economies.
1. Nonperforming Loan Stock 2. Nonperforming Assets: Distribution by Country
(Billions of euros) (Percent of sample assets)
Other 05 510 1025 Over 25
1,000 100
Cyprus, Greece, Ireland, and Portugal
900 Spain 90
800 Italy 80
700 70
600 60
500 50
400 40
300 30
200 20
100 10
0 0
2009 10 11 12 13 14 NL FR DE BE ES AT CY PT IT SI IE GR
Sources: National central banks; IMF Financial Soundness Indicators; and IMF Sources: European Central Bank; and IMF staff calculations.
staff estimates. Note: Based on a sample of 106 banks from 12 of the countries that took part in
Note: National denitions have been adjusted according to Barisitz (2013). Other the European Central Banks Asset Quality Review (AQR). Banks are sorted by their
comprises Austria, Belgium, France, Germany, and the Netherlands. nonperforming exposure (NPE) ratio. NPE ratio = AQR-adjusted NPE level as a
percentage of total credit exposure. Data labels use International Organization for
Standardization (ISO) codes.
Provisioning and write-offs have lagged... ...and so bank income and lending have been reduced.
3. Nonperforming Loans, Provisions, and Write-offs 4. Interest Income, Lending Growth, and Nonperforming Loans
Net NPLs Provision ratio (right scale) (Percentage points)
Provision Write-off ratio (right scale) Interest income to gross loans Lending growth
(Percent of GDP) (Percent of gross NPLs) (relative to average) (relative to average)
10 80 0.20 8
0.15 6
8
60 0.10 4

6 0.05 2
40 0.00 0
4 0.05 2
20 0.10 4
2
0.15 6
0 0 0.20 8
1 2 3 4 1 2 3 4
2010
11
12
13
14

2010
11
12
13
14

2000
01
02
03
04

Nonperforming loan quartiles Nonperforming loan quartiles


United States Euro area Japan
Sources: European Central Bank; Financial Services Agency; and IMF staff Sources: European Banking Authority; SNL Financial; and IMF staff calculations.
calculations. Note: Left chart shows annual interest income to gross loans, for over 100 euro area
Note: NPL = nonperforming loan; net NPL = gross NPL plus provisions; banks, relative to the yearly average for banks with the same nationality, calculated
provision ratio = provisions as a percentage of gross NPL; write-off ratio = over the period 200913. The right chart shows annualized lending growth relative
write-offs as a percentage of gross NPL. to average lending growth in the same economy, and uses European Banking
Authority data for a sample of more than 60 banks over the period 201013.
Outliers have been excluded, based on extreme values for lending growth,
nonperforming loans and interest margins.
(continued on next page)

Addressing the corporate debt overhang will help coverage ratios, a key indicator of borrower distress
support healthy credit demand (Figure 1.12, panel 6). Companies with high levels
Boosting credit demand will require tackling high of debt are less likely to demand more credit,
corporate indebtedness. In the euro area, there is potentially hampering growth in bank credit. High
a close correlation between countries with a high indebtedness is also likely to reduce the sensitivity
volume of nonperforming loans and those with of loan demand to a change in bank lending rates,
high corporate debt. This is illustrated by infor- reducing the effectiveness of a further compression
mation on the distribution of corporate interest in yields under QE.

International Monetary Fund | April 2015 21


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.12. Bank Nonperforming Loans and Lending Conditions (continued)


Policy actions are needed to support bank lending. Corporate debt-servicing capacity remains weak.
5. Simulated Bank Lending Capacity, 201517 6. Corporate Interest Coverage Ratios
(Percent of sample) (Percent of debt)
Below 1 1 to 2 2 to 3 Above 3
EBIT/interest expense
50 100
Euro area Japan
80
60
40 40
Median 20
0
30 2011 12 13 11 12 13 11 12 13 11 12 13 11 12 13
Germany France Italy Spain Portugal
EBITDA/interest expense
20 100
80
10 60
40
20
00 2 4 6 8 10 12 0 2 4 6 8 10 12 0
2011 12 13 11 12 13 11 12 13 11 12 13 11 12 13
Estimated growth in credit supply capacity (percent; annualized) Germany France Italy Spain Portugal
Source: IMF staff estimates. Sources: Amadeus database; national central banks; and IMF staff estimates.
Note: Shows the average annual bank lending capacity over the period 201517 Note: French data for 201213 are estimated using central bank data for a
for a sample of more than 100 euro area banks and around 80 Japanese banks. smaller number of rms. EBIT = earnings before interest and taxes. EBITDA =
earnings before interest, taxes, depreciation, and amortization.

In Japan, corporate leverage may also limit credit tion gaps. Moreover, many policies contain generous
demand for some smaller firms. Companies now have return guarantees, which are unsustainable in todays
significant cash holdings, amounting to 50 percent of low-interest-rate environment. According to the Euro-
GDP, up from 37 percent at the end of 2007.Firms pean Insurance and Occupational Pensions Authority
with large cash holdings are likely to demand less credit (EIOPA), more than half of European life insurers are
from banks. At the same time, firm-level data and sec- guaranteeing an investment return to policyholders
toral balance sheets show that some small and medium- that exceeds the yield on the local 10-year government
sized enterprises face the structural challenges of high bond, thereby incurring undesirable negative invest-
leverage and low profitability. Again, these indebted ment spreads (EIOPA, 2013).11
firms are likely to be less willing to take on more credit. Countries that suffer from both large duration
mismatches and negative investment spreads are
European life insurance: An unsustainable business particularly vulnerable to a prolonged low-interest-
model in a low-interest-rate environment rate environment. According to EIOPA, Germany
In the past three years, European life insurers equities and Sweden, which together accounted for about
have paid one of the most attractive dividends, outper- 20 percent of gross written premiums at the end of
forming on the back of waning euro area fragmenta- 2013, suffer from both duration mismatches of more
tion risk, high capital gains on bond holdings, and the than 10 years and negative investment spreads (Figure
release of excess capital due to lower claims inflation. 1.13, panel 1). In contrast, countries with positive
This trend, however, is likely to slow in response to ris- duration gaps (reflecting a higher share of saving- and
ing vulnerabilities, particularly in countries exhibiting unit-linked products), such as Ireland and the United
large duration mismatches. Kingdom, are less sensitive to the risks arising from
The current low-interest-rate environment, which low or falling interest rates. They may, however, face
QE will further exacerbate, poses severe challenges to
the EU life insurance industry. The industrys practice 11In Germany, for example, despite a recent reduction in the guar-

of writing long-term policies, sometimes of more than anteed policy rate on new products to 1.25 percent, the guaranteed
return on total policies is about 3.2 percent, whereas the 10-year bond
30 years, without assets of a correspondingly long yield is about 0.3 percent. For more information on the health and
duration has resulted in undesirable negative dura- challenges of German life insurers, see Elekdag and others 2014.

22 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.13. Life Insurance Industry Characteristics

1. Duration Mismatches and Negative Investment Spreads


(Baseline; percent)
2 SVK
GBR IRL
0
ESP PRT
Duration mismatch

BEL CZK
2 HUN ITA ROU GRC
POL DNK
4 BGR
FRA LTU
NLD FIN
6 EST
HRV CYP
8 SVN
MLT
10 AUT
SWE DEU LTV
12
2.0 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0
Internal rate of return mismatch

Source: European Insurance and Occupational Pensions Authority.


Note: AUT = Austria; BEL = Belgium; BGR = Bulgaria; CYP = Cyprus; CZK = Czech Republic; DEU = Germany; DNK = Denmark; ESP = Spain; EST = Estonia; FIN
= Finland; FRA = France; GBR = United Kingdom; GRC = Greece; HRV = Croatia; HUN = Hungary; IRL = Ireland; ITA = Italy; LTU = Lithuania; LTV = Latvia; MLT =
Malta; NLD = Netherlands; POL = Poland; PRT = Portugal; ROU = Romania; SVK = Slovak Republic; SVN = Slovenia; SWE = Sweden.

2. EIOPA Interest Rate Assumptions Relative to Current Euro Area Yield Curve
(Percent)
3.0 EIOPA baseline scenario
2.5
2.0 EIOPA Japanese-like scenario
1.5 Euro area market yields
(ve years out)
1.0
Euro area market yields
0.5
(current)
0.0
0.5
0 5 10 15 20 25 30
Years
Sources: Bloomberg, L.P.; European Insurance and Occupational Pensions Authority (EIOPA).

3. Share of European Asset Classes Held by Insurers


(Percent of outstanding)
45
40
35
30
25
20
15
10
5
0
Sovereign bonds Corporate bonds Financial bonds Nonnancial bonds Covered bonds Securitized credit European equities

Sources: Morgan Stanley; and IMF staff estimates.

International Monetary Fund | April 2015 23


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

other vulnerabilities, including high volatility in equity The absence of a policyholder protection scheme or set
markets. In the United States, life insurance companies of common minimum standards for the entire EU
also appear less sensitive to the risks associated with arrangements similar to those in Japan and the United
low interest rates, reflecting their product mix, which is Statesmagnifies the risk of market disruptions.
similar to that of U.K. insurers, and the more favorable The high and rising interconnectedness of the
U.S. economic outlook.12 insurance industry and the wider EU financial system
is another source of potential spillovers. The industry
A low interest rate scenario is materializing in Europe has a portfolio of 4.4 trillion in EU credit. Further-
The results of the 2014 stress tests conducted by more, insurers are traditionally closely linked to banks
EIOPA indicate the urgency and size of the insurance through liquidity swaps and bank bond holdings, a
industry problem. The stress tests show that 24 percent trend that could increase with the new Total Loss-
of insurers were not able to meet their 100 percent Absorbing Capacity requirements. A large mark-
Solvency Capital Ratio requirement under a Japanese- to-market shock could force life insurers into asset
like scenario.13 Although the industry was expected to reallocations and sales that could engulf the financial
have about 8 to 11 years before running into serious system (Figure 1.13, panel 3).
cash-flow pressures, even these results seem optimistic, Policies needed to maximize the effectiveness of QE
as interest rates are now significantly lower than in the
in the euro area
stress test scenarios (Figure 1.13, panel 2).
Solvency II adjustments (the Long-Term Guaran- QE provides a strong framework for addressing defla-
tee measures) help to mitigate the impact of stress tion risks, and some key transmission channels are
but may not be realistic under industry-wide stress.14 already beginning to work. But given the potential
Under the Japanese-like scenario, these Solvency II limits to bank credit growth, further steps to repair
adjustments eradicated the impact of the scenario on private balance sheets are needed for the full potential
insurers cash profiles by allowing the value of insur- benefits of QE to materialize:
ers assets to grow faster than that of their liabilities, First, regulators need to provide clarity about
which is counterintuitive in a prolonged low-interest- regulatory standardsand thus certainty for banks
rate environment.15 It seems reasonable for Solvency II adapting business modelsby promptly finalizing
adjustments to help an individual life insurer overcome the calibration of recent requirements, including the
temporary capital shortfalls, particularly in light of the leverage ratio, net stable funding ratio, and Total
long-term nature of its liabilities. But vulnerabilities Loss-Absorbing Capacity requirements.
become difficult to mitigate, even with regulatory Second, a number of actions are needed to compre-
adjustments, once an insurer hits negative cash flows or hensively tackle the burden of nonperforming loans.
the source of the vulnerabilities are industry wide and Supervisors must continue to provide strong incen-
likely to affect many insurers simultaneously. tives for banks to maintain adequate provisioning
levels and help reduce the current gap between bank
European life insurers are vulnerable to distress and market valuation of nonperforming loans. This
Midsize insurers in Europe face a high and rising risk includes encouraging banks to develop and use spe-
of distress. The failure of one or more midsize insur- cialized internal and external capacity for handling
ers could trigger an industry-wide loss of confidence if the stock of nonperforming assets, actively manage
the failure is believed to reflect a generalized problem. their provisions, and write off their nonperforming
assets (see Bergthaler et al, 2015).
Third, authorities should also ensure that legal
12Further analysis of U.S. insurers can be found in the forthcom-

ing 2015 U.S. Financial Sector Assessment Program.


frameworks for bankruptcy of firms and individuals
13Japanese-like scenario is used in EIOPA 2014a to test the resil- continue to be reviewed and reformed, where neces-
iency of the insurance sector by assuming a persistent low-interest- sary, and that institutional frameworks (judiciary
rate environment. See also EIOPA 2014b.
14The measures include both transitional arrangements and
and insolvency practitioners) and out-of-court pro-
permanent adjustments to eliminate the economic loss from negative cedurespossibly combined with corporate equity
investment spreads. financingare adequately resourced and supported
15For example, Solvency II requires insurers to recognize valuation
to deal with large volumes of distressed debt. Regu-
gains on the asset side fully and immediately, whereas losses on long-term
liabilities can be smoothed over a 16-year transitional period, adjusting latory measures should also be taken to encourage
for short-term credit spread volatility and other sources of volatility. the speedy disposal of problem loans by banks. In

24 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

addition, an active market for nonperforming loans extending the program to longer-maturity government
should be encouraged (for sample, Jassaud and Kang bonds, as necessary to achieve its 2 percent inflation
2015). target. A more forecast-oriented monetarypolicy com-
Fourth, the resilience of the financial system munication would increase the transparency of the
should be strengthened by diversifying the sources BOJs assessment of inflationprospects and also signal
of funding from banks to capital markets. A commitment to its inflation target, mainly through
deeper and broader capital market would improve thediscussion of envisaged policy changes if inflation
access to finance, particularly for smaller firms, is not on track.To further stimulate bank lending to
and make financial markets more efficient. In the the private sector, authorities should expand the special
euro area, this would require harmonization of lending facilities; jump-start the securitization market
company law, corporate governance, insolvency for bank loans to small and medium-sized enterprises
regimes, and taxation, in line with the latest and mortgages; and enhance risk capital provision,
Capital Markets Union proposal by the European including by encouraging more asset-based lending
Commission. and removing barriers to entry and exit of small and
medium-sized enterprises.
The challenges facing life insurers should also be
tackled promptly to ensure these institutions can play
an active role in the portfolio rebalancing channel. United States
Regulators need to reassess the viability of guarantee-
Despite the much-anticipated start of the process
based products and promptly bring minimum return
for monetary policy normalization in the United
guarantees offered to policyholders in line with any
States, long rates have been lower than expected as
secular trend in policy rates. At the same time, they
concerns over global growth and disinflation feed
must improve the sectors asset-liability matching and
back into U.S. markets. Plummeting crude oil prices
hedging capabilities. Prompt regulatory and supervi-
have raised concerns regarding the recent flurry of
sory actions are needed to mitigate damaging spillovers
high-yield debt issued by speculative-grade energy
from a failure of a medium-sized insurer. Introducing a
companies. Divergence between the expectations of
nationally harmonized policy holder protection scheme
financial market participants and those of policy-
would further increase the resilience of the industry
makers regarding the pace of U.S. monetary tighten-
by enhancing confidence. Partnerships combining the
ing reflects the challenge of normalizing monetary
credit risk expertise of banks with the balance sheet
policy in a world still addressing legacy problems
capacity of insurers could also help promote growth.
and trying to encourage economic risk taking.
Finally, regulators should continue to improve
transparency and public disclosure of life insur- U.S. recovery solidifies as economic risk taking takes hold
ers. Despite EU regulators significant efforts to
strengthen transparency, including through the publi- The fundamentals of the U.S. economy continue to
cation of comprehensive stress test results, it remains strengthen. The April 2015 World Economic Outlook
difficult to assess insurers true solvency positions. projects growth of 3.5 percent in 2015 amid low inter-
This situation could undermine public confidence est rates, dissipating fiscal headwinds, and lower energy
and exacerbate industry pressures if vulnerabilities prices. More people are returning to the workforce,
start materializing in smaller firms. and wage growth is widely expected to start picking
up. The World Economic Outlook projects three-year
The effectiveness of QQE in Japan depends on average growth at an annual rate of about 3 percent,
supporting policies the fastest annual pace since 2005.
Steadfast implementation of fiscal and structural Other indicators support the view that U.S. growth is
reforms is essential to boosting growth and making successfully making the transition from dependence on
QQE more effective. If these reforms are incomplete, asset appreciation and financial risk taking to an economy
efforts at pulling the economy out of deflation are led by economic risk taking. Capacity utilization is
less likely to succeed, hampering the effectiveness of returning to precrisis levels, and business fixed investment
QQE. The BOJ should consider strengthening the is rising, although at a slower pace than in previous cycles
portfolio rebalancing effects of its asset purchases by (Figure 1.14, panels 1 and 2). Growth in credit extended
increasing the share of private assets in purchases and to nonfinancial firms is on the rise, in contrast to growth

International Monetary Fund | April 2015 25


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.14. United States: Nonnancial Corporations

1. Credit Growth and Capital Expenditures 2. Capacity Utilization and Business Fixed Investment
(Percent) (Percent)
15 120 85 15

10
10 Historical
average since 100 80
1999 5

5
0
80 75
5
0

10
60 70
5 Credit growth to rms
Capital expenditures (percent Capacity utilization (left scale)
(year over year; four-quarter 15
of operating cash ow; right Nonresidential xed investment (year-over-year
moving average; percent; left
scale) growth; right scale)
scale)
10 40 65 20
2000 02 04 06 08 10 12 14 1990 94 98 2002 06 10 14
Sources: Federal Reserve; and IMF staff calculations.

in the euro area, where the trend is still negative. Funds review program. These developments are also indicative
raised through corporate debt issuance are increasingly of broader trends toward weaker underwriting standards.
devoted to capital expenditure rather than to equity buy- Relatively easy financing conditions and slower earnings
backs and other forms of financial engineering. The tepid growth could encourage higher leverage in future deals.
recovery of housing activity, however, remains a concern. Reflecting the search for yield, credit spreads remain
These developments are setting the stage for a nor- below historical averages (Figure 1.15, panel 1), and
malization of U.S. monetary policy. U.S. authorities are despite recent improvementdefault cushions are thin
preparing markets for a shift toward monetary policy tight- in lower-rated segments of high-yield corporate bonds
ening in 2015. Even though much anticipated, such an (Figure 1.15, panel 2).
exit remains challenging, as discussed in the next section. U.S. companies generally continue to add leverage,
as indicated by rising ratios of net debt to assets. How-
Financial risk taking continues at a strong pace in ever, measuring leverage through net debt to earnings
U.S. markets shows a widening disparity between large-capitalization
Alongside positive developments in economic funda- and small-capitalization firms (the latter with equity
mentals, the search for yield has continued in U.S. credit value between $100 million and $1 billion). The
markets. Signs of excesses in credit markets include the median small-cap firm has pushed leverage far higher
following: (1)underwriting standards continue to deteri- than the median large-cap firm, to levels above those
orate, with covenant-light loans now accounting for two- preceding the global financial crisis (Figure 1.15, panel
thirds of new issuance of leveraged loans; (2) issuance of 3). Smaller corporations are more vulnerable than the
other types of lower-standard loans, such as second-lien largest U.S. companies, which have the highest credit
loans, is at near-record rates; (3) there is an ongoing rise ratings among U.S. corporations and the easiest access
in leveraged buyouts and heightened activity in mergers to both the capital markets and banks. An examination
and acquisitions. Although the leveraged loan market is of the weak tail of corporations with the lowest debt
still a relatively small part of the U.S. credit market and repayment capacity, reveals a stark picture (Figure 1.15,
does not pose an immediate systemic threat, the sector panel 4). The weakest quartile of small-cap corporations
is growing rapidly, and weak underwriting standards are operating with relatively low interest-coverage ratios,
could pose problems down the road, as highlighted by leaving them more dependent on cash reserves and the
U.S. supervisors in their annual shared national credit continued ability to roll over debt to service interest.

26 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.15. U.S. Credit Spreads, Firm Leverage, and Interest Coverage
1. Investment-Grade and High-Yield Credit Spreads
(Z-score vis--vis the respective historical distribution) 2. U.S. B Rated Corporate Bonds: Breakeven Spreads
(Basis points)
6 1,400
U.S. investment grade Current spread
5 EU investment grade Breakeven spread in average default cycle 1,200
4 U.S. high yield
EU high yield 1,000
3
800
2
600
1
0 400

1 200
2 0
2001 03 05 07 09 11 13 15 1992 95 98 2001 04 07 10 13 15
Sources: Bank of America Merrill Lynch; and Bloomberg, L.P. Sources: Bank of America Merrill Lynch; Moodys; and IMF staff estimates.
Note: Z-scores relative to the respective historical distribution of
option-adjusted spreads.

3. Nonnancial Corporate Leverage 4. Nonnancial Interest Coverage Ratio


(Median) (Times; weak tail)
Debt to assets (all rms; left scale, percent)
30 Net debt to EBITDA (large-cap rms; right scale) 3.0 7
Net debt to EBITDA (small-cap rms; right scale) Small-cap rms
Large-cap rms 6
5
25 2.5
4
3
20 2.0
2
1
15 1.5 0
2003 04 05 06 07 08 09 10 11 12 13 2014 2003 04 05 06 07 08 09 10 11 12 13 2014
LTM LTM

Source: Standard & Poor's Capital IQ. Source: Standard & Poor's Capital IQ.
Note: Small-cap and large-cap rms are dened as having market capitaliza- Note: The weak tail is dened as the 25th percentile of the distribution of the
tions of $100 million to $1 billion, and greater than $5 billion, respectively. The interest coverage ratio within the sample. Small-cap and large-cap rms are
sample is a balanced panel of 1,695 rms. Standard & Poor's Capital IQ dened as having market capitalizations of $100 million to $1 billion, and
classies duty taxes related to exploration and production in the energy sector greater than $5 billion, respectively. The sample is a balanced panel of 1,695
as operating expenses. EBITDA = earnings before interest, taxes, depreciation, rms. S&P Capital IQ classies duty taxes related to exploration and production
and amortization; LTM = last 12 months. in the energy sector as operating expenses. LTM = last 12 months.

Leverage is being increasingly employed by equity the U.S. system increase the risk of minor shocks being
market participants. Although there are some recent propagated and amplified into sharp price corrections.
signs of stabilization, margin debt as a percentage of
market capitalization remains higher than it was during Declining oil prices could undermine credit quality
the late-1990s stock market bubble. The increasing in high-yield debt markets
use of margin debt is occurring in an environment of In the wake of the sharp drop in oil prices, market
declining liquidity. Average weekly trading volumes participants have grown concerned about exposed credit
continue to decline, and although the 52-week moving in the high-yield sector. Since oil prices started to decline
average of turnover has improved somewhat over recent in June 2014, the cumulative decline in total returns on
lows, it remains below its historical long-term average. energy-related issues in the Barclays High-Yield Index
Lower market liquidity and higher market leverage in peaked at 13 percent in January of this year, but a recov-

International Monetary Fund | April 2015 27


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.16. U.S. High-Yield Energy Markets

1. Cumulative Total Returns in High-Yield Index and 2. U.S. High-Yield and High-Yield Energy Spreads
High-Yield Energy (Percent)
(Percent)
10 20
High-yield indices 18
5 Composite
Energy 16
Exploration and production
Oil eld services 14
0
12
High-yield indices
5 10
Composite
Energy 8
Exploration and production
10
Oil eld services 6

4
15
2

20 Dec. 2004 0
Jul. 14
Mar. 14

Jun. 14
Dec. 2013
Jan. 14

Jan. 15
May. 14

Oct. 14

Dec. 14
Apr. 14

Aug. 14

Nov. 14
Sep. 14
Feb. 14

Feb. 15

Dec. 13
Dec. 11

Dec. 14
Dec. 07
Dec. 05

Dec. 09
Dec. 08

Dec. 12
Dec. 10
Dec. 06
Sources: Barclays; and IMF staff estimates.
Sources: Barclays; and IMF staff estimates.

ery in February on the back of rising oil prices limited the markets, energy-related bonds account for 56 percent
cumulative decline to 9 percent (Figure 1.16, panel 1). of the bonds trading at distressed levels, and virtu-
Accordingly, the divergence between the spreads of the ally allwere issued by firms engaged in extraction and
energy subcomponents of the Barclays High Yield Index production and oil field servicing (Table 1.4). A positive
and the broader index was in January at the widest it has point in this regard is that U.S. high-yield mutual funds
been in the past 10 years (Figure 1.16, panel 2). have relatively limited exposure to the energy sector, and
Oil-related issues comprise a significant portion of the accordingly they have only a limited ability to amplify
U.S. high-yield bond market. The share has tripled dur- volatility in any potential sell-off in the high-yield
ing the past 10 years, largely because of the U.S. shale energy sector. Also, thus far the contagion to the rest of
oil boom. Combining the high-grade and high-yield the high-yield bond market has been limited.

Table 1.4. Energys Impact on Two Barclays Corporate Credit Indices


U.S. High-Grade Index U.S. High-Yield Index
Number of Issues in Distress, January 30, 20151
Total Index 16 182
Energy 5 101
Exploration and Production 0 67
Servicers 5 33
Number of all Issues in Index 6,039 2,238
Index Amount (US$ trillion)
December 31, 2008 2.5 0.4
January 30, 2015 5.3 1.3
Sources: Barclays; and IMF staff calculations.
1Distressis defined as a bond trading below $80 per $100 par; this is a rule of thumb often used by market participants.

28 International Monetary Fund | April 2015


CHAPTER 1 EnhancIng polIcy tractIon and rEducIng rIsks

Figure 1.17. U.S. Interest Rates and Term Premiums

1. U.S. 10-Year Bond Yield and Term Premiums 2. Expected Terminal Federal Funds Rate
(Percent) (Percent)
2.7 0.6 4.25
Ten-year premium (right scale)
0.5
Ten-year nominal rate (left scale)
2.5 0.4
FOMC median projection
0.3 4.00
2.3 0.2
0.1
2.1 0.0 3.75
0.1
1.9 0.2 Primary dealers median
0.3 projection 3.50
1.7 0.4
0.5 Market-implied terminal
rate
1.5 0.6 3.25
Oct. 2014 Nov. 14 Dec. 14 Jan. 15 Feb. 15 Mar. 15 Jan. 2014 Apr. 14 Jul. 14 Oct. 14

Sources: Haver Analytics; and IMF staff estimates. Sources: Bloomberg, L.P.; Kim and Wright (K&W) (2005, updated); and IMF staff
estimates.
Note: The market-implied terminal rate is derived from the 10-year Treasury rate,
the 10-year term premium (Kim and Wright 2005), and the expected months to
liftoff in the federal funds rate. The pace of rate hikes is assumed to be 100 basis
points per year until the terminal rate is reached. FOMC = Federal Open Market
Committee.

Markets remain concerned that global indicates that, before this event, changes in the 10-year
disinflationary forces and downside risks may yet Treasury rate were more likely to precede (Granger
delay the U.S. recovery cause) changes in the 10-year German bund rate; after
Global developments are exercising strong influence on Jackson Hole, changes in bund yields were likely to
U.S. Treasury markets. The strengthening of the dollar precede (Granger cause) changes in Treasury yields
and lower yields in the euro area and Japan have made (Figure 1.18, panel 1).
U.S. Treasury bonds more attractive on a relative value Recent developments in global asset markets also
basis, because buyers can benefit from both the favor- reflect dissonance between financial market concerns
able yield differential and potential exchange rate gains. over global disinflationary pressures and the Federal
As a result, 10-year Treasury yields declined by 80 basis Reserves signaling of the path of U.S. monetary
points between October 2014 and the end of January policy. Both market-based and survey-based expecta-
2015, before rebounding by 50 basis points by mid- tions continue to point to mid- to late 2015 for the
March. A large part of this movement can be attributed first hike in the U.S. policy rate. But market-based
to a recompression of the term premium. Indeed, the expectations for the future path of policy rates remain
term premium on U.S. Treasuries briefly declined into notably below the forecasts of most of the participants
negative territory, pulling down U.S. long rates, even in the Federal Open Market Committees dot fore-
as the expected terminal federal funds rate remained casts (Figure 1.18, panel 2).16 These influences have
steady at about 3.25 to 3.50 percent (Figure 1.17), and persisted despite the continuing improvements in the
expected short-term rates remained stable.
Monetary developments in the euro area have had a 16Some market analysts forecasts for the first U.S. rate hike

particularly strong effect on U.S. interest rates. At the extend to early 2016, citing the absence of price pressure and an
Jackson Hole Conference in August 2014, the ECB expectation for a U.S. recovery slowed by a strong dollar and weak
foreign growth. Rates implied by futures contracts are also affected
president indicated a willingness to consider additional by risk premiums, and declines in those premiums can lower the
unconventional policy measures. Statistical analysis implied path of the policy rate.

International Monetary Fund | April 2015 29


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure1.18.Global Interest-Rate Developments


1. Granger Causality Tests for Precedence in 10-Year Bond Yields

January 1, 2010August 21, 2014 August 22, 2014March 15, 2015

ts ts
ke ke
ar ar

Un
gm gm

ite
Un
gin

dS
gin

it
ed
er er

tat
Em

S
Em

se
tat
es
Ja

Ja
pa

pa
n

n
y
an an
y
rm rm
Ge Ge

Sources:Bloomberg,L.P.;andIMFstaffestimates.
Note:The Jackson Hole Economic Policy Symposium was held on August 22, 2014. The arrows indicate the direction of the Granger Causality. The width of bands
indicates the signicance of the chi-squared statistic from the Granger causality test. The widest band represents signicance at the 1 percent level, the medium
band at the 5 percent level, and the narrowest at the 10 percent level.

2. U.S. Policy Expectations


(Percent)
5

FOMC participant
expectations 4

FOMC
median 2

Market implied
1

0
2015 2016 2017 Longer term

Sources:Bloomberg,L.P.;andIMFstaffestimates.
Note:Data is as of March 2015.Federal Open Market Committee (FOMC) meeting.

U.S. economic outlook and consistent signals from the States and the euro area, suggesting that markets are
committee on the likely trajectory for policy rates. taking a more benign view of inflation prospects. If this
Financial markets are effectively signaling a significant view is correct, it is possible that the Federal Reserve
risk that policy will not normalize as soon as the central may act more slowly than currently anticipated.
bank is forecasting, because disinflationary forces at On the other hand, as the Federal Reserve
work in the global economy will keep inflation con- approaches exit and rate hikes appear more imminent,
tained below target. Inflation swap markets are signaling Treasury yields could spike. This risk is not currently a
a lower level of expected inflation for both the United major focus for market participants. However, as was

30 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

seen in MayJune 2013, a sudden rise of 100 basis Two recent price disruptionsthe October 15, 2014,
points in the 10-year Treasury yield is quite conceiv- volatility in U.S. Treasuries and the January 15, 2015,
able, even in a generally disinflationary context and surge in the Swiss francinvolved an initial shock that
even when central banks work to communicate their was likely amplified by market makers withdrawal of
intentions in advance. Shifts of this magnitude can liquidity support. Many of the factors responsible for
generate negative shocks globally, especially in emerg- lower market liquidity also appear to be exacerbating
ing market economies. The anticipation of an immi- risk-on/risk-off market dynamics and increasing cross-
nent policy move could temporarily overwhelm global asset correlations during times of market stress. These
disinflation concerns and cause rapid decompression in phenomena suggest that low market liquidity may act
the term premium. Reduced structural liquidity could as a powerful amplifier of financial stability risks.
exacerbate the volatility of yield adjustments.
Rising market liquidity risks
Policies need to support economic risk taking, avert
financial excesses, and enhance financial resilience As discussed in the October 2014 Global Financial
Stability Report, capital markets are now more impor-
The impact of international market forces requires
tant providers of credit than they were in the past,
appropriately balanced policies, including strong
with a growing share of fixed-income instruments
macroprudential policies. In particular, regulators must
held by mutual funds. Inflows into mutual funds have
continue their efforts to understand the less closely
provided an illusion of liquidity in credit markets, but
regulated corners of the financial sector that could
changes in market structure may exacerbate illiquid-
cause problems for the banking system. Existing regu-
ity in times of stress.17 Banks have reduced their
latory frameworks may need to be reassessed to enable
market-making activities, and more investors are now
authorities to better identify and measure the activi-
following benchmarks. A combination of lower dealer
ties of nonbank entities. Policymakers should support
inventories, elevated asset valuations, flight-prone
further economic risk taking, such as tax reforms that
investors, and vulnerable liquidity structures have
could encourage firms to build capacity and increase
increased the sensitivity of key fixed-income markets to
employment.
increasing market and liquidity risks.
Given the risks and uncertainties surrounding the
normalization of U.S. monetary policy, central bank Economic and policy tensions leave global markets
officials must continue to follow a transparent and care- vulnerable to bouts of illiquidity that could prove
fully calibrated communications strategy to manage the systemic
policy-tightening process that is expected to commence
this year. The potential impact of increased volatility Asynchronous monetary policies and divergent eco-
and portfolio adjustments that could accompany the nomic prospects have led to a sharp increase in volatility
move toward policy rate normalization makes this task in foreign exchange markets. Global disinflationary pres-
especially crucial. The section titled When Market sures and accompanying policy responses have com-
Liquidity Vanishes examines some of the potential risks pressed longer-term U.S. bond yields. A sudden shift
from decreased market liquidity and changing patterns in market views that unwinds compressed premiums
of correlation in key financial markets. and sends yields higher could trigger a market liquid-
ity shock. Asset valuations remain elevated relative to
the past 10 years as monetary policies continue to exert
When Market Liquidity Vanishes downward pressure on spreads, but could widen on U.S.
As U.S. monetary policy normalizes, the temporary exit from monetary accommodation. This could reverse
boost to market liquidity provided by monetary accom- recent causality channels discussed elsewhere in this
modation will ebb, revealing a changed capital market chapter, sending shock waves through global markets.
landscape. Without the buoyant liquidity provided by Policy tensions led the central bank of Switzerland
the Federal Reserve, the liquidity-inhibiting impact of to unexpectedly abandon its support for a ceiling on
regulatory changes, industry consolidation, and other the value of the franc against the euro on January
secular factors will likely become more pronounced.
Markets could be increasingly susceptible to episodes in 17Financial stability risks related to mutual funds are also discussed

which liquidity suddenly vanishes and volatility spikes. in Chapter 3.

International Monetary Fund | April 2015 31


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.19. The October 15 Flash Rally in U.S. Treasuries

1. U.S. 10-Year Treasury Yields 2. U.S. 10-Year Treasury


(Percent) (Daily trading range in basis points)
2.8 60
Mar. 18, 2009 Aug. 9, 2011

2.6 50
Oct. 15, 2014
Oct. 8, 2008
40
2.4
30
2.2
20
2.0 October 15, 2014
10

1.8 0
Aug. 15 Sep. 2 Sep. 18 Oct. 3 Oct. 22 Nov. 6 Nov. 24 1996 98 2000 02 04 06 08 10 12 14
Source: Bloomberg, L.P. Source: Bloomberg, L.P.
Note: October 8, 2008: Federal Reserve bailed out American International Group;
March 18, 2009: Federal Reserve upsized quantitative easing; August 9, 2011:
Federal Reserve added mid-2013 language in the Federal Open Market
Committee statement.

3. Treasury Futures Contracts (10-Year) Available for Trading


(Number of contracts offered at the 10 best price limits)

8:30 a.m.: Usual drop in liquidity


Offered to buyers

40,000 10-year futures price 131.0


ahead of news release
30,000 (right scale)
130.5
20,000
130.0
10,000
129.5
0
129.0
10th best price
Offered to sellers

10,000
20,000 128.5

30,000 128.0
1st best price
40,000 127.5
8:00 8:10 8:20 8:30 8:40 8:50 9:00 9:10 9:20 9:30 9:40 9:50 10:00

Sources: CME DataMine; and IMF staff estimates.


Note: Liquidity is measured by the number of contracts offered to buyers and sellers at the top 10 best price limits.

15, 2015. The franc immediately surged by as much 1), and volatility spread to closely related asset classes
as 41 percent against the euro, and not surprisingly, (U.S. dollar swaps) and to equities (with a lag). To
some participants widened bid-ask spreads or refused put this event in perspective, the decline in yields was
to quote in the currency. Foreign exchange liquidity larger than that on September 15, 2008, when Lehman
overall collapsed and became less available than it was Brothers filed for bankruptcy. When compared with
during the 201112 euro crisis or the 2013 taper tan- recent trading, the massive intraday price change on
trum concerning prospective U.S. monetary policy. October 15 was an extremely rare occurrence (Figure
On October 15, 2014, U.S. Treasuries and related 1.19, panel 2).
markets experienced one of their largest intraday The rally in Treasuries on October 15 was report-
changes in yields in the past 25 years.18 Yields on edly initiated by a variety of poor data releases and
10-year bonds fell by 37 basis points from the previ- one-sided positioning, but was likely amplified by the
ous day before rebounding quickly (Figure 1.19, panel retreat of traditional market makers from their custom-
ary role of warehousing risk. As the number of Treasury
18See Bouveret and others, forthcoming, for a detailed analysis of futures contracts available for purchase or sale declined,
the events of October 15. individual trades had a larger effect on the market

32 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

price than they would normally have had (Figure 1.19, because of tighter requirements to minimize asset-
panel3). Although the rising price and vanishing liquid- liability mismatches.
ity allowed the flash dynamic to take off more than Inadequate market safeguardsExisting safeguards
an hour into the ongoing market reaction to the data can fail to limit abnormal price movements in mar-
releases, both were just as quickly reversed. The event kets dominated by automated trading. On October
remains under investigation by U.S. authorities, but a 15, 2014, circuit breakers were not triggered on
number of factors are likely to have contributed to it. futures markets because trades continued to take
place as the market moved through successive price
Why have market shocks become more amplified? levels without gaps. But because the number of con-
Market shocks are easily propagated when liquid- tracts available at each price level was small, prices
ity is low. As highlighted elsewhere in this chapter, rose rapidly with each successive execution, giving
technological change, regulation, and the shifting market participants no real opportunity to liquidate
composition of market participants have altered the significant positions at the market price.
microstructure of the Treasury market and fixed- Emergence of less-regulated nonbank market intermedi-
income markets more broadly. As a result, participants ariesAccess of leveraged retail investors to foreign
cannot always rely on dealers to provide sufficient currency brokers allowing bets against the Swiss franc
liquidity in volatile markets, making them more exacerbated the price surge. In many cases, heavily
vulnerable to liquidity shocks. Moreover, market leveraged positions involved little coordination or
safeguards may no longer be appropriately calibrated to oversight by authorities. Many retail investors were
changing market conditions. More specifically, either unaware of the risks or had explicit or implicit
Automation and the rise of high-frequency trading guarantees from their foreign exchange brokers that
Treasury bonds and Treasury futures trade almost they could not lose more than their deposits. How-
exclusively on electronic platforms, which allow ever, when the franc suddenly and sharply moved
algorithmic and high-frequency traders to capture an against their positions, their high degree of leverage
expanding market share. High-frequency trading is generated losses far greater than their account equity.
estimated to account for at least 50 percent of cash Two firms were driven into insolvency, and a retail
market volumes and 60 to 70 percent of futures trad- broker reported losses of nearly $225 million.
ing activity (Jiang, Lo, and Valente 2014; Tabb 2012; BenchmarkingMore market participants are using
and Chicago Mercantile Exchange 2010). Responding benchmarks by investing in indices or in underlying
to competition from these sources, even traditional baskets of securities.20 Several factors are driving this
market makers have increasingly adopted algorithmic trend, including restricted access to leverage from
trading strategies. Market participants report that prime brokers and demands from investors for tighter
liquidity provision has become more dependent on risk management and greater transparency. As more
programmed reaction functions and less on client- asset managers focus on benchmarks, assets not in the
based relationships. In a more anonymous, short- benchmark index suffer a decline in liquidity.
term, profit-oriented trading environment, fewer Use of derivatives and exchange-traded fundsThe
participants make their pools of liquidity available in increasing trading of index-based instruments
risky conditions to help stabilize the market. such as derivatives and exchange-traded funds may
Reduction in market making by traditional dealers amplify the effects of benchmarking in limiting
Banks claim that their ability to make markets and liquidity. When dealers use the cash market to hedge
therefore provide liquidity has diminished with the their exposure to a clients derivatives contract on an
tightening of regulation in recent years.19 Similarly, equity index, they need to replicate a simultaneous
pension funds and insurance companies are less able opposing order for each stock in the index.21 This
to play a countercyclical role in financial markets

20Mutual funds own a rising share of risky assets, particularly

in the less liquid credit markets, and hedge funds are increasingly
19To a degree this may be related to restrictions on proprietary behaving in a more benchmark-centric manner (see the October
trading and to more demanding capital requirements, which may 2014 Global Financial Stability Report).
have limited the capacity of banks to hold inventories and conduct 21Similar dynamics apply to broad-index exchange-traded funds.

repurchase agreement operations (see Powell 2015). Although buying a future does not directly lead to the purchase of the

International Monetary Fund | April 2015 33


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.20. Asset Comovements and Correlation Spillovers


Correlations among major asset classes have risen markedly since 2010. Cross-asset correlations increase with higher market volatility (VIX).
1. Correlation Levels 2. Cross-Asset Correlations
0.8 Precrisis Postcrisis Increasing cross-asset correlations 0.8
0.7

Median cross-asset correlations


Postcrisis
0.6 0.7

0.5 Crisis
0.6
0.4
Precrisis
0.3 0.5
0.2 These levels of the VIX are
associated with elevated market 0.4
0.1 uncertainty.
0.0 0.3
Median Asset (i) (ii) (iii) (iv) (v) (vi) 1 2 3 Top 10 Top 5
with: classes: MSCI US EMBI GBI-EM US Commodities percent percent
(i)(vi) EM Treasuries Global broad HY Quartiles of VIX probability distribution
loc cur

Declining liquidity is driving correlations. Greater use of derivatives is also driving correlations higher.
3. Correlation and Market Liquidity 4. Correlation and Derivative Usage
Median correlation with U.S. Treasuries

1.0 Increasing Precrisis Crisis Postcrisis Precrisis Crisis Postcrisis Increasing 1.0
correlation correlation

Median correlation with S&P 500


0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2 0.0
Declining liquidity
Greater use of derivatives
0.0 0.2
0 1 2 3 4 5 6 7 8 0 3 6 9 12 15
Treasury market liquidity Futures volume as a percentage of S&P 500 cash market

Sources: Bank of America Merrill Lynch; Bloomberg, L.P.; Federal Reserve; JPMorgan Chase and Co.; and IMF staff calculations.
Note: Precrisis period denotes January 1, 1997, to June 30, 2007; crisis period July 1, 2007, to December 31, 2009; and postcrisis period January 1, 2010, to
December 31, 2014. Cross-asset correlation is measured as the median of the absolute values of pair-wise correlations over a 60-day window between the daily
Sharpe ratios of the asset classes listed in panel 1. Market liquidity is measured as the ratio of returns on the U.S. Treasurywide index to the turnover of the U.S.
Treasury market. The higher the ratio the lower the liquidity, because large amounts cannot be traded without a signicant impact on prices. The median correlations in
panels 3 and 4 are of the U.S. Treasury 710-year index and the S&P 500 index against all six other asset classes as shown in panel 1. MSCI EM = MSCI Emerging
Markets Equity Index; U.S. Treasuries = 710-year U.S. Treasury Index; EMBI Global = JPMorgan Emerging Markets Bond Index Global; GBI-EM broad loc cur =
JPMorgan Government Bond Index-Emerging Markets in local currency; US HY = U.S. High-Yield Index; Commodities = Credit Suisse Index; VIX = Chicago Board
Options Exchange Market Volatility Index.

leads to further differentiation in liquidity between respective idiosyncratic fundamentals. Both the decline
securities included and excluded from indices. in market liquidity and the increasing use of deriva-
tives are associated with higher asset price correlations
Illiquidity events can spill over to other asset classes over the past five years (Figure 1.20, panels 3 and 4).22
and emerging markets This is particularly evident during periods of stress,
These structural shifts in markets may have also con- when flow liquidity reverses and volatility increases.23
tributed to higher asset price correlations. With lower
liquidity, less market making, and more benchmarking, 22The replication impact on the securities that make up an index

when derivatives are traded naturally pushes up intra-asset correlations.


asset prices are more likely to be driven by common
Increasing trading of derivatives also drives up cross-asset correlations.
shocks, particularly at higher frequencies, than by their For example, it is not uncommon for credit investors to hedge their
portfolios with liquid futures and options on equity indices.
constituents in the index, it will have an impact on the underlying secu- 23Flow liquidity, or the capacity to trade assets cheaply during

rities through the actions of index arbitrageurs such as hedge funds. normal market conditions, has been enhanced by the rise in flows

34 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

A rise in correlations during periods of stress is often and improve the liquidity and transparency of second-
seen as one of the main attributes of contagion (see, ary markets, specifically for longer-term debt markets.
for example, Pericoli and Sbracia 2004). Chapter 3 finds that the asset management industry
Correlations among risk-adjusted returns of major needs stronger oversight that combines better micro-
asset classes have increased markedly since 2010 prudential supervision of risks with the adoption of a
(Figure 1.20, panel 1).24 The correlation of the S&P macroprudential orientation. These findings suggest that
500 with U.S. high-yield indices has shown a steep securities regulators should shift to a more hands-on
increase, and the correlation with commodities has supervisory model, supported by global standards on
increased fourfold. The substantial rise in correlations supervision and better data and risk indicators. The roles
between asset markets in advanced and emerging and adequacy of existing risk management tools should
market economies points to an increased possibility of be reexamined, taking into account the industrys role in
contagion or spillovers in periods of stress. systemic risk and the diversity of its products.
Asset price comovement has become stronger during Policies are also needed to strengthen market
periods of high market volatility. Correlations normally structures, including in the more liquid fixed-income
increase during periods of market turbulence. How- markets such as government bond markets. Authorities
ever, over the past five years, correlations have been could consider encouraging market participants in gov-
rising to much higher levels, often to 0.7 or beyond, in ernment bond markets to provide liquidity in normal
periods of high volatility (Figure 1.20, panel 2). trading conditions, thereby forestalling the deteriora-
The increase in correlations during stress periods tion of trading liquidity. Drawing on examples from
suggests greater risks of contagion across asset classes other advanced markets, authorities could consider
or borders. It also points to the importance of liquid- either rewarding primary dealers with incentives and/
ity as an amplifier of other risk factors. Consequently, or obligating them to maintain their willingness to
policies that address the sources of low liquidity should trade passively (by providing quotes) or to participate
be seen as part of a comprehensive financial stability actively. Importantly, these approaches should not nec-
framework. essarily require dealers to maintain a market presence
during unusual bouts of extreme volatility. Reporting
What can policymakers do to address illiquidity and requirements could reinforce these approaches, typi-
stability spillovers? cally on an ex-post basis.
Policymakers should seek to address the liquidity Futures exchanges for U.S. Treasury markets could
mismatch in the asset management sector. As discussed consider introducing designated market makers.25
in the October 2014 Global Financial Stability Report, Unlike some equity markets, futures markets for
a major concern is the market liquidity risk arising Treasuries do not have designated market makers who
from the mismatch between the liquidity promised to provide liquidity. By providing fee rebates and other
mutual fund owners in good times and the cost of illi- incentives, exchanges could effectively charge market
quidity when meeting redemptions in times of stress, participants for the provision of risky market-making
particularly in the less liquid corporate and emerging services. Authorities could also consider best-practice
market bond markets. Policymakers should seek to guidelines for market makers.
address this mismatch by adopting policies that remove Market safeguards can help stop panics in periods
incentives of asset owners to run by aligning redemp- of heightened volatility. In the U.S. Treasury futures
tion terms of funds with the underlying liquidity in markets, current market safeguards should be recalibrated
the assets in which they are invested. They could also to prevent a market dislocation of the scale observed on
adopt policies that enhance the accuracy of net asset October 15, 2014, and periodically reviewed to ensure
values, increase liquidity cash buffers in mutual funds, that they are up to date and relevant. The authorities
could consider introducing similar market safeguards in
into mutual funds and exchange-traded funds. The effect may be
masking the negative impact of declining market making on other
the U.S. Treasury cash market. Adequate coordination of
measures of market liquidity, such as depth and breadth (see the such safeguards across cash and related derivatives markets
October 2014 Global Financial Stability Report. would help prevent liquidity arbitrage across platforms.
24The median correlation of the risk-adjusted returns between

the S&P 500 and the six major asset classes in the figure has almost
doubled from 0.44 in 19982007 to 0.70 in the past five years. 25For a discussion of how designated market makers with well-

Sharpe ratios are used to calculate risk-adjusted returns to control for designed obligations can support liquidity and price efficiency in
differing risk characteristics across asset classes. order-driven markets, see Bank of England 2012.

International Monetary Fund | April 2015 35


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.21. Wide Range in the Ination Outlook of Emerging Market Economies

Inationary pressures in emerging market economies are broad... but ination is expected to decelerate in most countries in 2015,
with a few exceptions.
1. Headline Ination Phase Curve 2. 2015 Forecast Headline Ination
(February 2015) (Year over year, percent)
15
Accelerating Decelerating Accelerating April 2014 Current Revision
ination ination ination 12
Brazil,
Colombia, Russia Neutral zone 9
Chile
Turkey Indonesia
6
Peru
Mexico 3
South Africa Philippines India
0

3
Hungary,

Poland
Singapore
China
India
Saudi Arabia
Hungary
Malaysia
Philippines
Indonesia
Bahrain
Mexico
South Africa
Hong Kong SAR
UAE
Thailand
Chile
Peru
Colombia
Turkey
Brazil
Nigeria
Russia
Poland, China
Thailand

Sources: National authorities; and IMF staff calculations. Source: IMF, World Economic Outlook database.
Note: Countries are placed on the phase curve according to where their February Note: UAE = United Arab Emirates.
2015 year-over-year headline ination print is relative to (1) central bank ination
target band (where available), (2) three-month trend, and (3) efcacy of monetary
policy (using ination targeting performance in previous two years). When ination
is in the central bank band, it is placed within the dotted lines, otherwise outside,
and the distance from the closest dotted line is determined by criterion 3, because
ination expectations are not considered. Future ination prints may not follow the
arc trajectory.

Real rates are expected to remain high in 2015, and higher than in the previous four years for many emerging markets.
3. Real Policy Rates in Major Emerging Markets
(Ex post, percent)
6 Latin America Real policy rates in 2015 are
EMEA expected to be higher than BRA
2015 forecast real policy rate

Asia previous 4 years


4
IDN
2 TUR ZAF
IND PHL POL
MEX COL CHL
0
HUN
THA
2
2 1 0 1 2 3 4
201114 average real policy rates
Sources: Bloomberg, L.P.; IMF, World Economic Outlook database; and IMF staff calculations.
Note: The 2015 forecast real policy rate is calculated using the end-2015 policy rate as implied by the market (using forward-rate agreements or interest rate swaps)
and the World Economic Outlook end-2015 headline ination forecast. BRA = Brazil; CHL = Chile; COL = Colombia; EMEA = Europe, Middle East, Africa; HUN = Hungary;
IDN = Indonesia; IND = India; MEX = Mexico; PHL = Philippines; POL = Poland; THA = Thailand; TUR = Turkey; ZAF = South Africa.

Risk management at trading firms should be on trading firms positions), or at the level of the clear-
reinforced, including from a macroprudential perspec- ing firms. Supervisors should also investigate whether
tive. Supervisors should provide coordinated guidance retail platforms are adequately capitalized to honor
to trading firms, allowing them to set consistent and guarantees on loss limits for leveraged retail investors
appropriate risk limits on individual retail investors, or under stressed conditions. Retail firms need to improve
at the level of the exchanges (circuit breakers and limits their ability to monitor the aggregate risk of their

36 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

clients in real time while, as discussed in Chapter 3, decelerate to their target bands by the end of 2015.27
regulators should enhance the microprudential supervi- As net commodity-importing economies, India, and
sion of risks from individual institutions that builds to a lesser extent Turkey, are expected to reduce their
on their own risk analysis and stress testing. Regulators external imbalances and have a chance to improve their
and monetary authorities should take the dynamics resilience by enabling necessary reforms.
of asset correlations and volatility into account when Easing inflation pressure provides a welcome
evaluating systemic risks in financial markets.26 increase in monetary policy space for countries in
which growth is expected to decelerate. Markets expect
real policy rates to decline relative to recent years in
Emerging Markets: Safeguarding the Financial economies with large inflation gaps, such as Hungary,
Sector against Global Headwinds Poland, and Thailand (Figure 1.21, panel 3), which in
Commodity price declines are exacerbating ongoing turn can help strengthen financial stability by reducing
corporate balance sheet strains in some emerging mar- the debt burden of domestic currency debt. Else-
ket economies, adding to headwinds from overcapac- where, central banks may have only limited ability, or
ity, real estate sector adjustments, and property price willingness, to significantly cut rates. For Brazil, India,
declines (particularly in China). This is despite the Indonesia, and Turkey, the expected increase in real
benefits of additional monetary policy space provided policy rates in 2015 relative to the previous four years
by lower commodity prices and lower inflationary may boost the cost of debt service in the private sector,
pressures. Elevated volatility and the rapid deprecia- where credit has grown strongly in recent years.
tion of local currencies for some economies jeopardize The following challenges confront some emerging
financial stability of firms that have borrowed heavily market economies to varying degrees:
in foreign currencies. These developments outweigh the Retrenchment of overinvested industries, real estate
financial stability benefits from improved competi- sector adjustments, and property price declines,
tiveness provided by depreciating currencies. Overall, particularly in China, which could spill over to
these shocks have increased financial stability risks in emerging markets more broadly.
emerging market economies, given the increased lever- Price declines in oil and other commodities, which
age in the public and private sectors, and authorities hurt commodity-exporting countries and related
need to enhance surveillance of vulnerable sectors. corporate sectors.
Ongoing dollar appreciation and the resulting
Inflation dynamics vary across emerging market upward revaluation of foreign currency liabilities,
economies, and some of those economies are gain- which creates balance sheet strains for indebted
ing monetary policy space to support growth emerging market firms and sovereigns.

Inflation dynamics in emerging market economies Disinflationary pressures in China may complicate
are diverse (Figure 1.21, panel 1). Most of South the transition to slower but safer growth, while
American economies and Russia continue to experi- real estate sector adjustments and overcapacity in
ence accelerating inflation pressure or above-target leveraged industries are key financial stability risks
inflation, while Hungary, Poland, and many Asian In addition to food and energy prices, Chinas disinfla-
economies have seen falling or low inflation (Fig- tion pressure may reflect more durable forces, includ-
ure 1.21, panel2). Some economies are benefiting ing debt-financed supply-demand imbalances that
substantially from the impact of lower oil prices and have built up since 2008. Overcapacity in some heavy
increased monetary policy space. India and South industries and excess supply in the real estate market are
Africa, for example, are expected to have inflation likely contributing to downward pressure on inflation.
Disinflationary pressures are keeping real interest rates
high (even when calculated using less volatile core infla-

26Other initiatives, such as the G20 Financial Stability Boards 27Hong Kong SAR and Singapore are categorized as advanced

recent proposal (issued jointly with the International Organization economies, but they are included in this section because as inter-
of Securities Commissions) on the supervision of global systemically national financial centers that cater primarily to emerging market
important financial institutions to cover traditional funds and their economies, their banking and corporate sectors are influenced by the
managers (rather than just the funds), also merit attention. forces analyzed here.

International Monetary Fund | April 2015 37


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.22. China: Real Estate and Interest Rate Developments


1. Effective Real Bank Lending Interest Rate 2. New Home Price Growth
(Contribution to deviation from average; percentage points) (Percent, year over year)

Spread over benchmark


3 Core ination 10 20
Benchmark Tier 1 cities
9 Tier 2 cities 15
2 Real lending rate (percent,
right scale) 8 Others
10
1 7
6 5
0 5 0
4
1 5
3
2 2 10
2006 08 10 12 14 2011 12 13 14
Sources: CEIC; and IMF staff calculations. Sources: CEIC; National Bureau of Statistics (NBS); and IMF staff calculations.
Note: Percentage point contribution to deviation of weighted average lending rate Note: New home price growth is the simple average of year-over-year change of
from the 200515 average. Weighted average interpolated from quarterly data and NBS-compiled property price indices for newly constructed residential buildings in
estimated from a linear model using benchmark rates and new loan relative 70 medium and large cities grouped by different tiers. Tier 1 cities include Beijing,
pricing before 2009. Guangzhou, Shanghai, Shenzhen, and Tianjin.

3. Lending to Real Estate Sector 4. Listed Firms ROA and Ination


(Trillions of renminbi, unless otherwise noted) (Percent)
30 6 25
Onshore bonds for real estate and
construction 20
25 5
Entrusted loans to developers 15
Trusts: Infrastructure 4 10
20 Trusts: Real estate
Bank loans: Mortgages 5
15 3
Bank loans: Developers 0
Percent of total social nancing 2 5
10
Percent of bank loans
10
5 1
15
0 0 20
2010 11 12 13 14 2008 09 10 11 12 13 14
Median ROA of all nonnancial sectors (four-quarter moving average)
Median ROA of overcapacity sectors (four-quarter moving average)
PPI ination for overcapacity sectors (year over year; right scale)
Sources: CEIC; China Trust Association; People's Bank of China; and IMF staff Sources: CEIC; WIND Information Co.; and IMF staff calculations.
calculations. Note: Overcapacity sectors include building materials, chemicals, and mining.
Note: Assumed 20 percent of entrusted loans are to the real estate sector. The PPI = producer price index; ROA = return on assets.
entrusted loans to developers and onshore bonds for real estate and construction
are captured by the total social nancing (TSF).

tion) and contributing to tighter real financial condi- (Figure 1.22, panel 2). Currently, levels of nonperform-
tions, notwithstanding slowing growth (Figure 1.22, ing property loans reported by banks remain subdued.
panel 1). If these trends intensify, they could engender a Credit exposures to real estate, excluding mortgages,
disinflationary feedback loop in which further declines stood at about 12 trillion yuan ($1.9 trillion, or 19
in inflation raise the real cost of debt service for highly percent of GDP) at the end of 2014 (Figure 1.22, panel
leveraged firms in weaker sectors, leading to potentially 3).28 Moreover, financial stress among real estate firms
abrupt and disorderly deleveraging, a further slowdown
in activity, and more downward pressure on prices. 28Assuming most trust real estate and infrastructure assets (often

Lower real estate prices are necessary in China for an related to property development) are in the form of loans. However,
the true total may be higher if lenders and borrowers found ways to
improved supply-demand balance, but they could lead overcome tighter restrictions placed in 2010 on lending for property
to higher-than-expected losses in the financial sector development, such as by classifying loans for other purposes.

38 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.23. Emerging Market Nonnancial Corporate Investment Continues to Shrink

Emerging market rms are expected to continue reducing capital ...and stripping out maintenance reveals reduced investment, particularly
spending across most sectors... among commodity rms.
1. Emerging Market Capital Expenditure Growth by Sector 2. Net Capital Expenditures to Total Debt
(Percent) (Percent)
70 25

60 201012
201214
50 201416
20

40

30 15

20

10 10

10 5

20 Energy and materials Total excluding energy and materials

30 0
Health care Utilities Industrials Consumer Energy, Tech, 2005 06 07 08 09 10 11 12 13 14
materials telecom LTM
Sources: S&P Capital IQ; and IMF staff calculations. Sources: S&P Capital IQ; and IMF staff calculations.
Note: Calculated on a balanced panel of 2,075 rms from 20 major emerging Note: Net capital expenditures = capital expenditures depreciation. Calculated
market economies. Estimates for 201416 from S&P Capital IQ. on a balanced panel of 1,274 rms from 20 major emerging market economies.
LTM = last 12 months.

could lead to direct cross-border spillovers, given gross Chinas case is instructive for some sectors of other
issuance of about $130 billion in external bonds since emerging market economies where excess capacity and
2010. An instance of such a spillover was a missed pay- overinvestment could create additional disinflationary
ment by the developer Kaisa in January 2015, which pressures. Emerging market firms, which have been
contributed to sharply curtailed issuance and wider reducing their capital investment since 2011 (see the
spreads across Asias high-yield bond market. Uncertain- April 2015 World Economic Outlook, Box 4.1, for a
ties related to the seniority of external creditors and their broader exposition), have more recently been cutting
access to borrower collateral could rise sharply. back across all sectors on the investment plans (Figure
Falling output prices are eroding the profitabil- 1.23, panel 1) that were funded by big debt increases.
ity of sectors with overcapacity and worsening their The share of net capital expenditures to total debt over
debt-service capacity (Figure 1.22, panel 4). These the past two years has declined, and is more pronounced
sectors, which include building materials, chemicals, among commodity firms, which also account for nearly
and mining, have also borrowed heavily since 2009. half of capital expenditures of nonfinancial firms (Figure
As with the property sector, falling output prices are 1.23, panel 2). As with China, these developments pose
welcome if they result in the exit of unprofitable firms the risk of a disinflationary feedback loop.
and a return to financially sustainable growth. Such an
adjustment, however, could mean potentially substan- Commodity price declines are exacerbating balance
tial losses for creditors. For banks, on-balance-sheet sheet strains in some emerging market economies
exposures to these sectors look manageable. But their In most emerging market economies, lower commodity
off-balance-sheet exposures, which some may have used prices are boosting consumption, helping to offset lost
to evade macroprudential edicts against lending to output from general trade shocks and providing greater
these sectors, may be much higher. As banks recognize monetary policy space. However, they may also give rise
these contingent liabilities, the losses could quickly to financial stability concerns. For others, the decline in
erode their seemingly ample capital buffers. commodity prices during the past nine months has led

International Monetary Fund | April 2015 39


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.24. Dependence of Emerging Market Sovereigns on Commodities, and Market Reaction

Emerging market economies that rely heavily on commodity exports... ...generally had the greatest growth revisions.
1. Net Exports of Commodities 2. 2015 Forecast Real GDP Growth Revision from October 2014
(Share of 2014 forecast GDP, percent) (Percentage points)
50 2
1
40
0
30 1
20 2
3
10 Net commodity exporter 4
Net commodity importer
0 5
10 6

Philippines
Mexico

Malaysia
Nigeria

Thailand

Bahrain
Brazil

India
Singapore
South Africa

Hungary
Indonesia
Saudi Arabia
Russia
Venezuela
Nigeria
Colombia
Malaysia
Chile
Argentina
Peru
Indonesia
Brazil
South Africa
Mexico
Poland
Thailand
Turkey
Philippines
India
China

Colombia

Chile
United Arab Emirates

China
Russia

Saudi Arabia

Turkey
Peru

Poland
Venezuela

Hong Kong SAR


Sources: IMF, World Economic Outlook database; UN Comtrade; and IMF staff Sources: IMF, World Economic Outlook database; UN Comtrade.
calculations. Note: See Figure 1.24, panel 1 for categorization of commodity exports.
Note: The UN Comtrade commodity net exports for 2013 comprise commodity
codes 0 through 4, using Standard International Trade Classication Revision 3.

3. Change in the CEMBI Index Spread 4. Change in the CEMBI Index Spread
(June 30, 2014, to March 18, 2015, by sector, basis points) (June 30, 2014, to March 18, 2015, by country, basis points)
250 796 400

200 300

150 200

100 100

50 0

0 100
Metals and mining
Infrastructure

CEMBI
Nigeria
Russia
Brazil
Colombia
South Africa
Mexico
Philippines
Poland
Turkey
Thailand
United Arab Emirates
India
Chile
China
Hungary
Malaysia
Bahrain
Saudi Arabia
Singapore
Argentina
Indonesia
Tech, media, and telecom

Industrial
Utilities

Diversied

Pulp and paper


Financial
Oil and gas

Consumer

Real estate
Transport
CEMBI

Source: JPMorgan Chase & Co.


Note: CEMBI = Corporate Emerging Markets Bond Index.

to sizable downward revisions of economic activity for ited the ability of those countries to react to the growth
some major commodity-exporting countries (Figure1.24, downturn (see the April 2015 Fiscal Monitor).
panels 1 and 2). Commodity price shocks have become Since 2007, energy firms have issued one-third of all
systemic for the oil and gas sector in Nigeria, Russia, and hard-currency nonfinancial emerging market corporate
Venezuela, and markets have reflected that fact (Figure bonds as they took advantage of accommodative finan-
1.24, panel 4). Lower revenue and higher public indebt- cial conditions to borrow heavily in international bond
edness in Nigeria and Venezuela, for example, have lim- and syndicated loan markets to expand their operations

40 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.25. Energy Corporate Sector Metrics

The balance sheet deterioration for the emerging market energy sector started even before the oil price decline...
1. Balance Sheet Metrics for the Energy Sector in Emerging Markets (Medians)
8 2.8 7
Return on assets Leverage
Interest coverage ratio
(percent) (net debt to EBITDA)
7 2.6 (EBIT to interest expense) 6

6 2.4
5
5 2.2
4
4 2.0
3
3 1.8

2 1.6 2
2003 04 05 06 07 08 09 10 11 12 13 14 2003 04 05 06 07 08 09 10 11 12 13 14 2003 04 05 06 07 08 09 10 11 12 13 14
LTM LTM LTM

Source: S&P Capital IQ.


...and is notable in Argentina, Brazil, Nigeria, and South Africa, among others.
2. Energy Sector Balance Sheet Metrics (Medians)
12 Return on assets Interest coverage ratio 14
2007 2014 LTM (EBIT to interest expense) 2007
10 (percent) 2014 LTM 12
8 10
6 8
4 6
2 4
0 2
2 0
4 2
Philippines
South Africa

Argentina
Malaysia
Nigeria
Brazil

Thailand

India

United Arab Emirates


Indonesia
China

Peru
Poland

Russia

South Africa
Poland
Brazil
Nigeria
Argentina
Philippines
China
Indonesia
Malaysia
India
United Arab Emirates
Peru
Thailand
Russia
Sources: S&P Capital IQ; and IMF staff calculations.
Note: S&P Capital IQ classies duty taxes related to exploration and production as operating expenses. EBIT = earnings before interest and taxes; EBITDA = earnings
before interest, taxes, depreciation, and amortization; LTM = last 12 months.

and finance investment. Given expectations of lower 2003 (Figure 1.25, panel 1). Price declines have cut into
energy prices, firms in the oil and gas and materi- the profitability of energy firms, particularly in China,
als sectors are significantly cutting back their capital Nigeria, and South Africa (Figure 1.25, panel 2). Strains
expenditure plans. Because these sectors account for, in the debt-repayment capacity of the oil and gas sector
on average, half of investment in the major emerging may become more evident in Argentina, Brazil, Nigeria,
market economies, this may affect long-term growth and South Africa, given their low interest-coverage ratios in
for other sectors as well (Figure1.23, panel2).29 2014 (on a last-12-month basis; Figure 1.25, panel 2).30
On average, the deterioration of balance sheets for many
oil and gas firms preceded the energy price decline of 2014. Dollar appreciation could test firms and countries
Profitability (for example, return on assets), leverage, and that have accumulated dollar debt
debt-servicing capacity are now at their worst levels since From October 2014 through February 2015 the U.S.
dollar appreciated by 14 percent in nominal terms and
29For emerging market energy firms with available data, capital
30In Brazil, Petrobrass corporate governance concerns have
expenditures in fiscal year 2015 will decline by 31 percent from the
previous year, and earnings before interest, taxes, depreciation, and resulted in credit rating downgrades and pushed its borrowing costs
amortization will decline by 20 percent. to their highest level in more than 10 years.

International Monetary Fund | April 2015 41


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.26. Large Increase in Emerging Market Debt

Indebtedness increased strongly across most major emerging markets.


1. Change in Private Sector and Government Indebtedness between 2007 and 2014
(As a share of GDP; percentage points)

Hong Kong SAR 8 14 93 98


China 17 58 6 81
Singapore 22 24 16 63
Malaysia 27 12 12 51
Thailand 28 7 9 44
Brazil 3 9 26 33 Change in private sector and government
indebtedness between 2007 and 2014
Turkey 10 18 3 32
Poland 14 11 3 28
Russia 17 6 5 28
Mexico 2 6 13 21
Hungary 4 6 13 16 Household
Nonnancial corporate
Indonesia 4 6 8 10 Government
South Africa 2 6 17 9
India 2 1 9 6

10 0 10 30 50 70 90 110
Sources: Bank for International Settlements; Morgan Stanley; national authorities; IMF, World Economic Outlook database; and IMF staff calculations.
Note: Data for Malaysia are from Bank Negara Malaysia, and the change in debt is between 2008 and 2014.

The private sector in many economies increased its foreign currency Most emerging market economies increased their foreign currency
debt since the crisis or kept it high. debt and exposure to foreign investors despite issuing relatively less
in foreign currencies.
2. Foreign Currency Debt of Nonnancial Firms and Households 3. Government Debt Breakdown in Foreign Currency and Nonresident
(Percent of GDP) Holdings of Local Currency
(Percent of total)
50 80
2007:Q4 FX 2007:Q4 NRLC
2007
2014:Q3 FX 2014:Q3 NRLC 70
40 2014
60
30 50
40
20 30
20
10
10
0 0
Argentina

Brazil
India

South Africa
Indonesia
Colombia

Thailand

Brazil

Colombia

Indonesia

Russia

Turkey

South Africa

Romania

Hungary

Mexico

Poland
Chile
Russia

Turkey

Poland

Sources: Bank for International Settlements; IMF, Financial Soundness Indicators; Sources: Bank for International Settlements; Haver Analytics; national authorities;
and IMF staff calculations. and IMF staff calculations.
Note: Foreign currency debt comprises domestic loans, international loans, and Note: FX is the share of foreign currency government debt; NRLC is the share of
international bonds. nonresident holdings of local currency government debt. The date for the left-hand
bar for Colombia is January 2010, for Romania is December 2010, for South Africa
is January 2011, and Russia is December 2011.

by 11 percent in real effective terms. This dramatic faster than GDP in all major emerging market econo-
movement in the exchange rate over a period of five mies and in the international financial centers of Hong
months has major implications for emerging market Kong SAR and Singapore, which lend to many emerg-
economies that have high debt levels denominated ing market economies in Asia. Most of the growth in
in foreign currencies. From 2007 to 2014, debt grew debt was in the nonfinancial private sector (firms and

42 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Figure 1.27. Firms in Countries with Large Currency Mexico, and Poland.32 Even though foreign currency
Selloffs Also Had High Equity Volatility exposure may not have increased for many emerg-
(Percent) ing market economies, the increased role of foreign
25
Latin America
investors in local bond markets creates an implicit
EMEA NGA debt-rollover risk, which can be loosely described as
Asia 20
original sin 2.0.
The dramatic moves in commodity prices and the

Change in equity volatility (percent)


COL 15
exchange rates of many emerging market economies
ZAF MEX BRA OLS regression over the past six months have already had a significant
line 10
impact on market valuations for emerging market cor-
PER MYS
HUN porations (Figure 1.24, panel 3). For some central and
THA 5
POL eastern European countries, such as Poland, the high
IND TUR share of foreign-currency-denominated or -linked debt
0
CHL ROU built up during the precrisis period also makes them
vulnerable to depreciation against other currencies, such
IDN 5
as the Swiss franc. Since the end of June 2014, financial
markets have reassessed equity valuations for firms in
10
2 0 2 4 6 8 10 12 14 16 Brazil, Colombia, Malaysia, Mexico, Nigeria, Peru, Rus-
Change in foreign currency volatility (percent) sia, and South Africa with the increase in the volatility
Sources: Bloomberg, L.P; and IMF staff calculations. of their currencies (Figure 1.27). For emerging markets
Note: Changes calculated over June 30, 2014March 9, 2015. Volatilities are three- more generally, higher volatility and loss of market con-
month realized. Currencies are quoted against the U.S. dollar, except in Hungary,
Poland, and Romania, where they are quoted against the euro. Volatility in Russia fidence can cause a sharp reduction in secondary market
(not shown) increased 44 points for the ruble and 39 points for equities. BRA = liquidity of emerging market assets and fast depreciation
Brazil; CHL = Chile; COL = Colombia; EMEA = Europe, Middle East, and Africa;
HUN = Hungary; IDN = Indonesia; IND = India; MEX = Mexico; MYS = Malaysia; of local currencies, similar to what has been observed in
NGA = Nigeria; OLS = ordinary least squares; PER = Peru; POL = Poland; ROU = Russia since the introduction of economic and financial
Romania; THA = Thailand; TUR = Turkey; ZAF = South Africa.
sanctions (Box 1.2).

Banks have large exposures to the corporate sector in


households; Figure 1.26, panel 1), and a significant countries with significant corporate debt at risk
portion is in foreign currencies, especially in Chile,
A significant share of debt in Argentina, Brazil, China,
Poland, and Turkey (Figure 1.26, panel 2), although
India, Nigeria, and Turkey is owed by firms with
in Chile foreign currency mismatches of corporate
relatively constrained repayment capacity in terms of
balance sheets appear limited, and households do not
interest-coverage ratios33 (Figure1.28, panel 1), and in
have debt in foreign currencies.31
Turkey a significant share of this debt is in foreign cur-
Rapid depreciation of the domestic currency can
rencies (Figure1.26, panel 2). The exposure of banks
lead foreign investors to abruptly reduce their holdings
to the nonfinancial corporate sector is particularly high
of local currency debt and thus create a debt-rollover
in some emerging market economies. In 11 of the
challenge to the public sector. Since 2007 the share
21 emerging market banking systems analyzed here,
of foreign currency and nonresident holdings of local
more than half of the bank loan books consist of loans
currency general government debt in total general gov-
to firms, rendering them more exposed to corporate
ernment debt has risen in a number of countries, such
weakness, particularly in Nigeria, Peru, Turkey, and
as Indonesia, Mexico, Poland, Romania, and South
Ukraine (Figure 1.28, panel 2). Although it is dif-
Africa, or remains elevated, such as in Hungary (Figure
ficult to match the precise exposure of banks to firms,
1.26, panel 3). This development is critical where the
the higher the overlap of these two metrics, the more
ability of the local investor base to absorb new debt
may be insufficient, such as in Hungary, Indonesia,

31The existence of foreign currency hedges, financial (via financial 32See the October 2012 Global Financial Stability Report, Chapter

derivatives) or natural (via offshore revenues), are significant offset- 1, for an analysis of the absorptive capacity of banks and asset man-
ting factors to foreign currency risks of emerging market firms. agers in emerging market economies.
Nevertheless, disclosures and data availability for such hedges are 33Defined as the ratio of earnings before interest, taxes, deprecia-

difficult to obtain or estimate. tion, and amortization to interest expenses.

International Monetary Fund | April 2015 43


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 1.28. Financial Stability of Emerging Market Banks

Nigeria, India, Turkey, and Brazil among others have a large share of Banks in Nigeria, Ukraine, Turkey, and Peru are highly exposed to
corporate debt-at-risk. nonnancial rms.
1. Corporate Debt-at-Risk 2. Bank Loans to Nonnancial Corporations
(Percent) (Share of total loans; percent)
50 100
2010 Change 201014 2014 LTM
40 80
30
60
20
10 40
0
20
10
20
Philippines 0
Mexico

South Africa
Argentina

Malaysia
Nigeria

Turkey
Brazil

Thailand

Nigeria
Ukraine
Turkey
Peru
China
Hong Kong SAR
Pakistan
Bulgaria
Russia
Argentina
Philippines
Lithuania
Brazil
Uruguay
Romania
Hungary
Croatia
Latvia
Poland
India
India

Indonesia

Colombia
United Arab Emirates
Hungary

Saudi Arabia
Chile
Peru
China

Poland

Russia

Sources: S&P Capital IQ; and IMF staff calculations. Sources: IMF, Financial Soundness Indicators; and IMF staff calculations.
Note: Debt-at-risk is the share of corporate debt held by the weak rms or those
with interest coverage ratios (EBITDA divided by interest expense) less than two. A
sample of more than 10,000 rms was used. EBITDA = earnings before interest,
taxes, depreciation, and amortization; LTM = last 12 months.

Banking buffers vary considerably among emerging market economies, and loan-to-deposit ratios have increased in places.

3. Loss-Absorbing Buffers 4. Banking System Loan-to-Customer-Deposit Ratio


(Share of risk-weighted assets; percent) (Percent)
20 250

2014 or latest 2010:Q4


15 200
Change since 2010 2014:Q3 or latest

10 150

5 100

0 50

5 0
South Africa
South Africa

Mexico

Philippines

Argentina
Nigeria
Turkey

Brazil
India
Singapore
Indonesia
Colombia
Chile
Philippines

Mexico

Romania
Argentina

Peru
Malaysia

Poland
Turkey

Russia
Brazil
India

Thailand

Indonesia
Hungary

Chile

China
Poland

Hong Kong SAR


Russia

Sources: IMF, Financial Soundness Indicators; and IMF staff calculations. Source: IMF, Financial Soundness Indicators.
Note: Loss-absorbing buffers = (Tier 1 capital + loan loss reserves Note: Does not include interbank loans. Dashed line corresponds to 100
NPL)/(risk weighted assets). Data are for 2014 or latest available (2013 for percent level.
China, Poland, and Thailand). NPL = nonperforming loan.

significant the risks of bank asset deterioration from that domestic banks still play the primary financing
weaknesses in the corporate sector. role in emerging market economies. An assessment of
The broader impact of a sudden deterioration in different measures of bank health is provided in Table
corporate health depends on the capacity of banks to 1.5. Bank balance sheets appear healthy in most emerg-
absorb losses and continue providing liquidity, given ing market economies, but some vulnerabilities are still

44 International Monetary Fund | April 2015


Table 1.5. Summary of Sovereign, Corporate, and Banking Indicators
Macroeconomy Commodity Exposure Corporate Sector
Deviation
CPI of CPI from Commodity- Oil-
(Percent Central Banks Commodity Oil Related Related Share of Private Nonfinancial
year- Inflation Target/ 2015 Exports/ Exports/ Share of Share of Share Debt at Nonfinancial Corporate
over-year, Middle of Band Policy Growth Total Total Corporate Corporate of Debt Risk after Sector FX Total Debt/
2015 (Percentage Rate Forecast Exports Exports Sector Sector at Risk Shock Debt/GDP GDP
forecast) points) (Percent) (Percent) (Percent)1 (Percent)1 (Percent)2 (Percent)2 (Percent)3 (Percent)3 (Percent) (Percent)
Latin Brazil 8.0 3.5 12.75 1.0 52 7 25 20 25.3 37.0 15.1 46.9
America Chile 2.9 0.1 3.00 2.7 47 1 26 6 10.0 22.4 39.4
Colombia 3.6 0.6 4.50 3.4 69 47 45 33 0.8 2.5 9.9
Mexico 3.1 0.1 3.00 3.0 20 12 39 30 17.9 19.4 20.9
Asia China 1.2 2.3 2.50 6.8 4 1 29 17 24.5 31.2 149.9
India 5.8 0.2 7.50 7.5 25 14 23 13 36.9 49.9 13.6 47.5
Indonesia 4.6 0.6 7.50 5.2 54 7 36 18 17.8 45.7 15.3 21.7
Malaysia 2.7 3.25 4.8 32 12 33 28 10.8 24.2 96.9
Philippines 2.4 0.6 4.00 6.7 15 2 9 5 3.2 31.5
Singapore 0.8 3.0 80.7


Thailand 2.1 0.4 1.75 3.7 20 5 47 31 13.7 17.5 53.7
Europe, Bahrain 1.5 0.50 2.7 82 68 37 2.1 2.1
Middle East, Hungary 1.7 1.3 1.95 2.7 13 2 85.2
Africa Nigeria 12.0 13.00 4.8 91 81 60 35 49.3 51.9
Poland 0.4 2.1 1.50 3.5 16 2 32 18 16.3 23.9 32.5 44.2
Russia 12.0 7.5 14.00 3.8 68 48 53 46 13.6 16.0 22.9 49.2
Saudi Arabia 2.0 2.00 3.0 83 81 50 5 6.4 6.4 13.0
South Africa 5.0 0.5 5.75 2.0 34 3 28 6 16.1 20.2 16.4 32.6
Turkey 7.0 2.0 7.50 3.1 13 3 12 7 27.3 31.0 29.6 50.8
United Arab Emirates 2.2 1.00 3.2 32 29 11 8 2.3 4.0
(continued)
CHAPTER 1 Enhancing policy traction and reducing risks

International Monetary Fund | April 2015 45


46
Table 1.5. Summary of Sovereign, Corporate, and Banking Indicators (continued)
External Vulnerabilities Banking Sector
Nonresident
Holdings of
Domestic
Current Reserves/ Government
Account Short-Term Debt (Percent Change in Sovereign
(Percent of External of all domestic Currency Weight Credit Default Nonperforming Loss-Absorbing
GDP, 2015 Financing government since June 30 in EMBIG Swap Spreads Loans Loan-to-Deposit Buffers
forecast) Requirements4 debt) (Percent)5 (Percent)5 (Basis points)5 (Percent)6 Ratio (Percent)6 (Percent)6
Latin Brazil 3.7 1.9 20.1 32.5 6.7 250 2.9 84.3 14.3

International Monetary Fund | April 2015


America Chile 1.2 1.2 12.7 2.6 89 2.2 163.7 10.0
Colombia 5.8 1.1 14.6 32.0 3.2 151 2.8 109.1
Mexico 2.2 1.2 38.5 17.8 12.9 115 2.9 108.9 14.7
Asia China 3.2 6.2 1.0 5.0 84 1.1 56.5 11.3
India 1.3 2.0 3.2 0.4 4.0 86.1 7.9
Indonesia 3.0 1.2 38.1 8.9 7.8 145 2.1 100.2 16.6
Malaysia 2.1 1.0 47.3 12.8 1.3 138 1.8 82.0 12.5
Philippines 5.5 7.8 1.0 4.6 85 2.4 73.4 13.1
Singapore 20.7 10.0 0.0 0.9
Thailand 4.4 2.3 18.6 0.4 0.0 98 2.2 109.4 13.5
Europe, Bahrain 2.1 7.9 0.0 0.0 278 4.6 46.0
Middle East, Hungary 4.8 2.1 34.1 22.0 2.8 115 16.6
Africa Nigeria 0.7 20.5 0.2 3.2 59.9
Poland 1.8 0.7 40.0 22.6 2.1 59 4.9 114.1 11.8
Russia 5.4 6.7 24.3 57.6 8.1 454 6.5 149.9 7.8
Saudi Arabia 1.0 18.6 0.0 0.0 73 1.3 78.1 16.8
South Africa 4.6 0.9 36.0 12.4 2.5 203 3.4 187.7 10.8
Turkey 4.2 0.6 21.8 21.4 7.4 218 2.7 124.0 13.0
United Arab Emirates 5.3 0.0 0.0 7.3 97.0
Sources: Bank for International Settlements; Bloomberg, L.P.; JPMorgan Chase & Co; S&P Capital IQ; UN Comtrade; IMF, World Economic Outlook database; IMF, Financial Soundness Indicators, and IMF staff calculations.
GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Note: CPI = consumer price index; EBITDA = earnings before interest, taxes, depreciation, and amortization; EMBIG = Emerging Markets Bond Index Global; FX = foreign currency; NPL = nonperforming loan.
1 2013 COMTRADE data. Commodity exports using codes 04 from Standard International Trade Classification, Revision 3. Oil exports using code 33.
2According to available firms from S&P Capital IQ, weighted by assets.
3Percentage of firms with interest coverage (EBITDA/interest expense) ratio below 2. The shock is composed of a 25 percent increase in borrowing costs, 20 percent appreciation of the U.S. dollar, and 25 percent reduction in earnings

of energy firms.
4Short-term external financing requirement is defined as short-term debt maturities plus current account deficit.
5Market data is as of March 6.
6As of 2014:Q3 or latest available data. Loss-absorbing buffers is (Tier 1 capital + loan loss reserves NPL)/(risk-weighted assets). 2014 data, except for China, Poland, and Thailand (2013).
CHAPTER 1 Enhancing policy traction and reducing risks

Box 1.2. Russias Financial Risks and Potential Spillovers


Russias economic outlook has deteriorated significantly
under the combined shocks of sanctions and the sharp Figure 1.2.2. Russian Banking System Key
drop in the price of oilinterest rates are higher, the Financial Soundness Indicators
ruble has depreciated, and the government has lost its (Percent)
investment-grade credit rating. These developments
threaten a further deterioration in asset quality and pos- 14.0 Capital to risk-weighted assets 20
Return on equity (right scale)
sible financial spillovers. 18
Foreign portfolio outflows amounted to $21 billion 13.5
in the first nine months of 2014 ($13 billion of which 16
was in equities), taking the stock of total foreign port- 14
folio investment down to $225 billion. In the same 13.0
period, Russians increased their portfolio investments 12
abroad by $10 billion, to $63 billion. 12.5 10
External debt is not insignificant (at $599 billion
as of December 2014). But the short-term repay- 8
ment burden$74 billion is due April-December 12.0
6
2015 (Figure 1.2.1), of which 61 percent is due to
the corporate sector and 36 percent to banks repre- 11.5 4
sents only one fifth of foreign exchange reserves ($352
2
billion as of the end of March). And the public and
private sectors hold significant assets abroad (includ- 11.0 0
Dec. 2012
Dec. 13
Jan. 14
Feb. 14
Mar. 14
Apr. 14
May 14
Jun. 14
Jul. 14
Aug. 14
Sep. 14
Oct. 14
Nov. 14
Dec. 14
ing $61 billion in portfolio assets, $184 billion in

The authors of this box are Allison Holland and Luigi


Ruggerone. Source: Central Bank of Russia.

Figure 1.2.1. Russian External Debt currency and deposits, and $32 billion in short-term
Amortization Schedule loans at the end of December) that can be liquidated
(Billions of U.S. dollars) as needed. Nevertheless, as a consequence of the
General government Banks Other sectors escalation of geopolitical tensions, lower oil prices,
30 and sanctions, Russias sovereign and corporate spreads
have risen sharply, reflecting the markets perception of
25 increased credit risk.
From a financial stability perspective, the Russian
banking sector deserves close attention. Solvency
20 risks in the sector appear contained overall, but some
pressure is evident: nonperforming loans increased
15 steadily through 2014 (to 6.7 percent as of the end of
December) and profitability declined (Figure 1.2.2).
Liquidity risk also appears relatively contained to
10 dateoverall deposits grew through 2014, with cen-
tral bank funding representing 12 percent of liabilities
5 as of the end of 2014; however, this may prove more
challenging in the future. With a loan-to-deposit
ratio of 150 percent, the sector is heavily dependent
0 on wholesale market financing, and rolling over in
2015:Q1 15:Q2 15:Q3 15:Q4 16:Q1 16:Q2
external markets the foreign financing that comes due
Source: Central Bank of Russia. in 2015 ($37 billion) will not be possible for the seven
sanctioned banks that account for about 75 percent

International Monetary Fund | April 2015 47


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 1.2. (continued)


of Russian bank assetsAlthough the Russian banking
sector weathered the crisis of 2009 (when conditions Figure 1.2.3. Foreign Bank Exposures
were arguably tougher), and official sector support can (Billions of U.S. dollars)
be expected to continue, a significant deterioration in
asset quality or earnings or a liquidity shock at a large 50
French banks
bank could signal a more systemic problem. 1.0%
Direct financial linkages between Russia and the rest
of the world are fairly limited, but the indirect connec- 40
tions with neighboring countries raise more serious
global financial stability concerns. Foreign bank expo- Italian banks

Exposure to Russia
U.S. banks 5.3% 30
sures to Russia have been reduced. But the stability of
0.4%
the European banking system could become signifi-
cantly stressed should geopolitical concerns boost
investors risk aversion, which would lead to a stronger Japanese banks 20
0.1%
dollar and higher rates. This could cause Russias inten-
sified difficulties to spill over to central and eastern U.K. banks
European countries, to which some large European 0.3% Austrian banks 10
29.4%
banking systems are highly exposed (Figure 1.2.3).

0
0 50 100 150 200 250 300
Exposure to central and eastern European economies

Sources: Bank for International Settlements (BIS); Bank of


Japan; European Central Bank; and IMF staff estimates.
Note: Size of bubbles is exposure (BIS Table 9D) to central
and eastern European economies as a percentage of total
assets.

present. Loss-absorbing buffers appear particularly low could yet arise via increasing nonperforming loans in
in Chile, Hungary, India, and Russia (between 5 and 10 places where firms have a high proportion of foreign
percent of risk-weighted assets; Figure 1.28, panel 3), currency debt.
and deterioration in loan quality could threaten capital
levels. Furthermore, in India, Russia, and Turkey loss- Policies to mitigate risks
absorbing buffers have deteriorated quite substantially in Emerging markets generally should aim to cushion the
recent years. System-wide Tier 1 ratios for most emerg- impact of global headwinds and disinflationary forces
ing market economies are above 10 percent. However, where possible, for example, by allowing exchange rate
the countries with the lowest ratios are China, India, adjustment if it does not jeopardize smooth market
and Russia, which account for about 70 percent of the functioning, or if the currency is already significantly
aggregate banking system assets in this sample of banks. undervalued, by boosting reserves, or by applying
Buffers are still fairly low in some commodity-sensitive policies to increase macroeconomic policy space and
economies (such as Russia), while some banking systems buffers. Furthermore, countries ought to safeguard the
are also sensitive to dollar funding and tighter liquidity resilience of the financial system through enhanced
conditions. This sensitivity could in turn put pressure on surveillance of vulnerable sectors.
banks funding channels, with many countries exhibit- In China, the overall priority must be to allow an
ing high levels of loan-to-deposit ratios, including Chile, orderly correction of excesses. This will require policies
Russia, South Africa, and Turkey (Figure 1.28, panel 4). to play a dual and finely balanced role. Policies should
Finally, although regulatory caps mean that banks direct contribute to a financial rebalancing, curtailing the
currency exposures are generally limited, vulnerabilities riskiest parts of shadow banking. Policies should also

48 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

facilitate corporate deleveraging and the transparent risks and regularly monitor corporate foreign currency
recognition of costs arising from the exit of nonviable exposures, including derivatives positions. The hedges
firms. Authorities should discourage the financing employed by corporations to limit their exposure risks
of nonviable borrowers, which will require tolerat- may be compromised when most needed, so regulators
ing more defaults, including in public bond markets. should assess them conservatively. These macropru-
Orderly deleveraging requires comprehensive policies dential and microprudential measures can be usefully
that allow a gradual slowdown in credit growth and, complemented by flexible exchange rates. Flexible
where necessary, provides mechanisms for orderly debt exchange rates can aid the adjustment to shocks and
restructuring. Given Chinas outsized level of gross facilitate an independent monetary response to credit
corporate debt and its importance to the global econ- booms. They can also discourage banks and corpora-
omy, managing this process smoothly will be critical tions from building up large foreign exchange expo-
in order to minimize the macroeconomic headwinds it sures in the first place. Renewed efforts by authorities
could create. globally to collect and provide better information on
Across emerging markets more generally, the large foreign currency corporate indebtedness and offsetting
portion of debt denominated in foreign currencies factors (such as hedges) is also desirable.
as well as in specific sectors, such as energy firms, To ensure properly functioning markets, authorities
means that micro- and macroprudential measures have need to adopt and enforce policies that protect against
an important role to play in limiting the risks from lapses of liquidity in local bond markets. This calls for
shocks, and authorities need to enhance supervision country authorities to potentially use cash balances
of these sectors. The relevant macroprudential tools when needed or to lower the supply of long-term debt
include higher risk weights (capital requirements) for to the market to help curtail bond spread increases.
corporate foreign currency exposures as well as caps on Policymakers can also adopt crisis management tools
the share of such exposures on banks balance sheets. that allow the smooth functioning of markets, by using
In the likely case of leakage, consideration should also bilateral and multilateral swap line agreements to help
be given to changes in the tax code that remove fiscal reduce excess volatility in currency markets and provide
incentives in favor of debt or that penalize foreign foreign currency funding in times of stress. Multilateral
currency debt (see also IMF 2014b). To avoid these resources, such as IMF facilities, could also provide
measures from becoming procyclical, they should be additional buffers. Overall, keeping emerging market
introduced cautiously and with sufficient phase-in economies resilient calls for authorities to maintain a
periods. At the microprudential level, regulators need strong focus on domestic vulnerabilities, as noted in
to conduct bank stress tests related to foreign currency previous Global Financial Stability Report issues.

The authors of this chapter are Peter Dattels and Matthew Jones (Team Leaders), Ali Al-Eyd, Serkan Arslanalp, Magally Bernal, Antoine
Bouveret, Peter Breuer, Yingyuan Chen, Martin ihk, Fabio Cortes, Reinout De Bock, Martin Edmonds, Jennifer Elliott, Michaela Erbenova,
Tryggvi Gudmundsson, Sanjay Hazarika, Geoffrey Heenan, Allison Holland, Eija Holttinen, Bradley Jones, David Jones, William Kerry, Daniel
Law, Andrea Maechler, Alejandro Lopez Mejia, Peter Lindner, Daniela Mendoza, Evan Papageorgiou, Vladimir Pillonca, Alvaro Piris Chavarri,
Jean Portier, Gabriel Presciuttini, Juan Rigat, Shaun Roache, Luigi Ruggerone, Luca Sanfilippo, Tsuyoshi Sasaki, Katharine Seal, Nobuyasu
Sugimoto, Narayan Suryakumar, Shamir Tanna, Chris Walker, and Jeffrey Williams.

International Monetary Fund | April 2015 49


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Annex 1.1. Progress on the Financial The reform agenda has taken steps to address the
Regulatory Reform Agenda too-big-to-fail problem. First, once finalized, inter-
national agreement on total loss-absorbing capacity
The main elements of the Financial Regulatory Reform
(TLAC) should support orderly resolution of global
Agendacapital, leverage, and liquidityhave been
systemically important banks (G-SIBs) without
substantially agreed to. This accomplishment achieves
recourse to public funds, by setting minimum stan-
a key postcrisis goal of strengthening the regulatory
dards on the amounts and characteristics of capital and
framework for banks.34 Progress on the implementa-
bail-inable debt that banks must issue. Second, 18
tion of the agenda, however, is uneven: several areas
G-SIBs recently signed the new International Swaps
require significant movement forward. Further, the
and Derivatives Association Resolution Stay Proto-
stakeholders need to devote more resources to moni-
col that suspends early termination and cross-default
toring implementation.
rights in over-the-counter (OTC) derivatives contracts
The last important element of the liquidity
during resolution. Nonetheless, resolution reforms
framework for banksthe net stable funding ratio
remain a work in progress. Further action is needed in
(NSFR)was finalized in late 2014. It requires banks
many jurisdictions to (1) make large, complex firms
to maintain a stable funding profile in relation to their
more resolvable and agree to living wills; (2) align
on- and off-balance sheet activities to address a vulner-
legal frameworks with international best practice in
ability that fueled the systemic stress in the global
resolution; (3) reduce legal and practical impediments
financial crisis. The NSFR complements the liquidity
to effective cross-border resolution; and (4) develop
coverage ratio (LCR), which targets short-term liquid-
policies for the recovery and resolution of key nonbank
ity risks and came into force on January 1, 2015. The
intermediaries such as central counterparties.
Basel Committee for Banking Supervision (BCBS)
Although the nonbank financial sector has seen
reports significant progress in meeting both of these
some progress from the reform agenda, movement for-
standards well ahead of implementation dates.
ward continues to lag in other areas. The International
The BCBS is now able to direct its attention to out-
Association of Insurance Supervisors has finalized
standing regulatory concerns, such as restoring the cred-
the Basic Capital Requirement for global systemi-
ibility of risk-weighted assets. Aligned with the Group
cally important insurers, and the insurance capital
of 20 (G20) objectives, the recently published propos-
and higher loss-absorption capacity requirements are
als on the standardized approach to credit risk would
under development. The Financial Stability Board
reduce reliance on external credit ratings. Once final-
(FSB) released a second public consultation draft on
ized, amendments to the calibration of the risk weights
methodologies to identify nonbank and noninsurer
should help enhance risk sensitivity and comparability
global systemically important financial institutions in
with the internal ratings-based (models-based) approach
March 2015. The FSB is continuing to finalize mini-
to credit risk. Importantly, the BCBS wishes to ensure
mum haircut requirements on securities lending and
the standardized approach is suitable for a wider range
repurchase agreements after the end of the consultation
of jurisdictions and banks, not just the main financial
period in December 2014.
centers and internationally active institutions.
The reform agenda has seen some progress on the
The BCBS has finalized revisions to the securiti-
application of new OTC derivatives rules across bor-
zation framework that enhance risk sensitivity and
ders, partly by deferring to home country regulatory
reduce the mechanistic reliance of capital requirements
regimes. In October 2014, the European Commission
on external ratings. Work is also progressing on criteria
deemed that central counterparty (CCP) regimes in
for identifying simple, transparent, and comparable
four jurisdictions are equivalent to European Union
securitizations, which should help support a sustain-
requirements. In December, the U.S. Commodity
able securitization market.
Futures Trading Commission extended the deadlines
for requiring certain foreign CCPs to register in the
Annex prepared by Katharine Seal, Michaela Erbenova, Alvaro United States until late 2015. However, agreement on
Piris, Nobuyasu Sugimoto, and Eija Holttinen. important decisions, in particular between the United
34This was the conclusion of the Financial Stability Board (FSB),

as reported to the Leaders of the Group of 20 (G20) at the Brisbane States and the European Union, remains elusive and
Summit. perpetuates regulatory uncertainty for industry players.

50 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Annex 1.2. External Portfolio Rebalancing pean Central Banks (ECBs) government bond holding
under Quantitative Easing in the Euro Area and statistics, nonbank investors hold about 1.8 trillion (70
Japan percent) of these negative yielding bonds, while banks
and national central banks (NCBs)hold the remaining
This annex considers portfolio rebalancing for the euro
600 billion. From an asset-liability management point of
area and Japan for three scenarios. This exercise is par-
view, there is little reason to expect institutional investors
tial and does not incorporate exchange rate effects.
to maintain a negative carry on these assets. Similarly, it
Euro area is assumed that the other domestic nonbank and foreign
investors will also forego holding negative yielding assets.
Under the baseline, euro area nonbank investors could
However, banks are assumed to either sell their EGBs to
allocate nearly 1.2 trillion abroad by the end of 2015
the ECB, or hold on to them for regulatory reasons.
(Table 1.2.1). This figure could rise or fall, depending on
The international allocation of assets by institutional
the attractiveness of euro-denominated assets. A nega-
investors is based on current domestic and foreign currency
tive shock in the absence of additional policy action to
investment allocations, implying that under the baseline
strengthen bank balance sheets (downside) could be
about one-third (630 billion) will be rebalanced into euro-
expected to increase outflows to nearly 1.3 trillion (an
denominated assets (namely high-yield corporate bonds)
additional 10 percent), whereas additional policy mea-
and the rest (1.2 trillion) will be rebalanced into foreign
sures (QE-Plus) could be expected to lower outflows to
assets. Based on data on international investment positions,
about 1.1 trillion (a symmetric decline of 10 percent).
about 420 billion of this could end up in U.S. dollar
Here, it is assumed that, at a minimum, the nonbank
assets, 480 billion in other advanced economies, and 130
investor portion of the 2.4 trillion in negative yielding
billion in emerging market economies.
European government bonds (EGBs) will be rebalanced
into alternative assets (Figure 1.2.1). Based on the Euro- Japan
Calculations indicate that Japanese financial institu-
Annex Table 1.2.1. Potential Portfolio Outflows by Euro tions could shed 165 trillion of Japanese government
Area Investors, 2015 bonds (JGBs) by the end of 2017, given the pace of
(Billions of euros)
the central banks purchases under the second round of
Baseline QE-Plus Downside quantitative and qualitative easing (QQE2) and addi-
Total 1,177 1,059 1,294 tional debt issuance by the government. In particular,
held by
Foreign 819 737 901 Japanese insurance companies and pension funds could
Pension and shed 55 trillion of government bonds, while banks
Insurance 233 210 256 could sell another 110 trillion by the end of 2017,
Firms 125 112 137
Destination
which would bring their sovereign exposure closer to
United States 421 379 463 international levels (Table 1.2.2).
Other Advanced This baseline scenario considers three types of
Economies 482 434 530
Emerging Markets 131 118 144 investorsdomestic banks, insurance companies, and
Other 142 128 157 pension fundswhich own nearly 80 percent of JGBs
Sources: Bloomberg, L.P.; European Central Bank; IMF, Consolidated Portfolio Invest- in private hands, and makes the following assumptions:
ment Survey (CPIS); and IMF staff estimates.
Domestic banks are assumed to reduce their JGB hold-
Note: The downside scenario assumes an additional 10 percent in portfolio outflows,
which is consistent with a return to the euros share of international reserves hold- ings to 5 percent of assets by 2017, in line with bank
ings (based on IMF Composition of Foreign Exchange Reserves data) at the onset of sovereign exposures in other Group of Seven (G7)
Economic and Monetary Union from the peak reached in 2009 (namely from 27 to 17
percent). For simplicity, a symmetric 10 percent is assumed for the QE-Plus scenario. economies (excluding Italy). Japan Post Bank reduces
As in Figure 1.10, foreign-currency-denominated flows are determined by European its domestic bond holdings to 35 percent of assets, in
insurer asset allocations as of 2013, while destinations are based on data for 2013
international portfolio flows (CPIS). Emerging markets comprise Argentina, Brazil, Chile, line with the Government Pension Investment Funds
China, Colombia, Hong Kong SAR, Hungary, India, Indonesia, Malaysia, Mexico, Philip- (GPIFs) new allocation to domestic bonds.
pines, Poland, Russia, Saudi Arabia, Singapore, South Africa, Thailand, and Turkey.
Other advanced economies comprise Australia, Canada, Denmark, Japan, New Zealand,
Insurance companies broadly follow the GPIF as
Norway, Sweden, Switzerland, and the United Kingdom. QE = quantitative easing. a benchmark and reduce their exposure to govern-
ment bonds to 35 percent of total assets, a drop of
The authors of this annex are Ali Al-Eyd and Serkan Arslanalp. 40 trillion.

International Monetary Fund | April 2015 51


gloBal FInancIal staBIlIty rEport: naVIgatIng MonEtary polIcy challEngEs and ManagIng rIsks

Annex Figure 1.2.1. Euro Area Negative-Yielding European Government Bonds and Baseline Portfolio Rebalancing

Expectations of quantitative easing drove some core yields negative. Rebalancing under the baseline scenario.
1. Sovereign Yields: Euro DebtWeighted Yield Curve 2. Potential Flows into Foreign-Exchange-Denominated Assets
(Percent) (Billions of euros)
4.0 2,500

3.5 Sold to national


Euro area range, post Jackson Hole 560 central banks under 2,000
3.0 Euro area, post Jackson Hole QE or held
Euro area, pre Jackson Hole
2.5
626 1,500
Maturities

2.0 Other
EUR assets
2,363
1.5 142 EM
Holders 131 1,000
1,803
1.0 United
Re- 421 States
0.5 1,177
balanced 500
FX
0.0 assets
482 Other
AE
0.5 0
1 2 3 4 5 6 7 10 15 30 EGBs < 0 Bank/Nonbank EUR/FX split Destination
Year Sources: Bloomberg, L.P.; and IMF staff calculations.
Sources:Bloomberg,L.P.; and IMF staff estimates. Note: AE = advanced economies; EGB = European government bond; EM =
emerging markets; EUR = euro area; FX = foreign exchange; QE = quantitative
easing. See Figure 1.10 panel 2 for country groupings.

Pension funds are assumed to follow the GPIFs As a result of this portfolio rebalancing, insurance
lead, reducing their domestic bond holdings to companies and pension funds could invest as much as
35 percent of assets through a reduction of JGB 42 trillion ($350 billion), or 8 percent of GDP, in for-
holdings. eign assets (Table 1.2.3). This scenario is in line with the
The outstanding stock of JGBs rises in line with the pace of their portfolio rebalancing abroad over the last
latest World Economic Outlook fiscal projection and year and the GPIFs new target allocation announced
the BOJ buys 80 trillion of JGBs every year, as in late 2014. If the insurance companies and pension
announced under QQE2. funds maintain present international allocation ratios,

Annex Table 1.2.2. Japan: A Potential Portfolio Rebalancing Scenario under


QQE2, 201517
(Trillions of yen)
End-2014 End-2017 Change
Bank of Japans JGB Holdings 207 447 240
Other Financial Institutions JGB Holdings 505 340 165
Pension Funds (public and private) 92 78 14
Insurance Companies 199 159 40
Domestic Banks (major and regional) 105 50 55
Japan Post Bank 110 53 57
Memo Items:
Outstanding Stock of JGBs 828 903 75
Sources: Bank of Japan (BOJ); Japan Post Bank; Ministry of Finance; and IMF staff projections.
Note: Pension fund holdings of domestic bonds decline to 35 percent of assets by a reduction in JGB holdings in line
with the GPIFs new target allocation. Similarly, insurance company holdings of JGBs and Japan Post Bank holdings
of domestic bonds decline to 35 percent of total assets. Domestic bank holdings of JGBs decline to 5 percent of total
assets (benchmark: other advanced economies). The BOJ buys 80 trillion yen of JGBs on a net basis every year, as
announced under QQE2. Outstanding stock of JGBs rises in line with World Economic Outlook fiscal projections. GPIF =
Government Pensions Investment Fund; JGB = Japanese government bond; QQE = quantitative and qualitative easing.

52 International Monetary Fund | April 2015


CHAPTER 1 Enhancing policy traction and reducing risks

Annex Table 1.2.3. Potential Portfolio Outflows by Bergthaler, Wolfgang, Kenneth Kang, Yan Liu, and Dermot
Japanese Institutional Investors, 201517 Monaghan. 2015. Tackling Small and Medium Sized
(Billions of U.S. dollars) Enterprise Problem Loans in Europe. Staff Discussion Note
15/04, International Monetary Fund, Washington.
QE-plus
(complete Bornhorst, Fabian, and Marta Ruiz Arranz. 2013. Indebtedness
Baseline policies) Downside and Deleveraging in the Euro Area. Country Report 13/232,
Insurance Companies 100 275 0 International Monetary Fund, Washington.
Private Pensions 25 58 0 Bouveret, Antoine, Peter Breuer, Yingyuan Chen, David A.
Public Pensions 225 225 225
Jones, and Tsuyoshi Sasaki. Forthcoming. Fragilities in U.S.
Total 350 559 225 Treasury Markets and Lessons from the Flash Rally of Octo-
Source: IMF staff projections. ber 15, 2014. IMF Working Paper, International Monetary
Note: All figures are expressed at end-2014 exchange rates. Under the baseline
Fund, Washington.
scenario, insurance companies and pension funds continue their portfolio rebal-
ancing abroad at the same pace as since 2012:Q3. Under the complete policies/ Cecchetti, Stephen G., Madhusudan S. Mohanty, and Fabrizio
QE-plus scenario, insurance and private pension funds accelerate their portfolio Zampolli. 2011. The Real Effects of Debt. Working Paper
rebalancing abroad at twice the pace as baseline. Under the downside scenario,
352, Bank for International Settlements, Basel.
they stop their portfolio rebalancing abroad. QE = quantitative easing.
Chen, Sally, Minsuk Kim, Marijn Otte, Kevin Wiseman, and
Aleksandra Zdzienicka. 2015. Private Sector Deleveraging
80 percent of the outflow would go into bonds of other and Growth Following Busts. Working Paper 15/35, Interna-
advanced economies, 14 percent into emerging market tional Monetary Fund, Washington.
bonds, and 6 percent into global equities. Chicago Mercantile Exchange. (2010). Algorithmic Trading and
This baseline scenario assumes a significant but Market Dynamics. Chicago Mercantile Exchange Group.
partial implementation of the other two arrows of Abe- July 15.
nomics (fiscal and structural reforms). If announced Dattels, Peter, Rebecca McCaughrin, Ken Miyajima, and Jaume
Puig. (2010). Can You Map Global Financial Stability?
policies are fully implemented and work to their fullest
IMF Working Paper No. 10/145 (Washington: International
extent across the three reform arrows (the QE-plus
Monetary Fund).
scenario, also referred to as the complete policies
Draghi, M. 2015. Speech at The ECB and its Watchers XVI
scenario), portfolio outflows could be as much as $550 Conference. Frankfurt, March 11.
billion, as insurance and private pension funds acceler- EIOPA, 2014a. EIOPA Insurance stress test 2014. EIOPA-
ate their portfolio rebalancing abroad (Table 1.2.3). BOS-14-203, November 28.
Alternatively, if the other two reform arrows are not EIOPA, 2014b. Low interest rate environment stock taking
effectively deployed and efforts at pulling the economy exercise 2014. EIOPA-BoS-14/103, 28 November 2014.
out of deflation are not successful (downside sce- Elekdag, Selim, Faezeh Raei, Jrme, Vandenbussche, and
nario), portfolio outflow could be less than anticipated, Vanessa Le Lesl. 2014. Selected Issues: Germany.
as private financial institutions continue to demand Country Report 14/217, International Monetary Fund,
JGBs as a hedge against deflation. This would imply Washington.
European Banking Authority (EBA). 2015. Overview of the
a partial return to the status quo before Abenomics
Potential Implications of Regulatory Measures for Banks
when home bias of Japanese institutional investors
Business Models. London.
was strong and portfolio outflows were limited. In this
European Central Bank (ECB). 2012. Corporate Indebtedness
case, portfolio outflows could be limited to $225 bil- in the Euro Area. Monthly Bulletin (February): 87103.
lion by end-2017. . 2015. Box 1: The Governing Councils expanded asset
purchase programe. Economic Bulletin, January 2015.
European Insurance and Occupational Pensions Authority
References (EIOPA). 2013. Financial Stability Report, Second Half-
Bank of England. 2012. The Role of Designated Market Makers in Year Report, Autumn.
the New Trading Landscape. Quarterly Bulletin Q4: 34353. Httl, Pia, and Guntram B. Wolff. 2014. What is Behind the
Barisitz, Stephan. 2013. Nonperforming Loans in Western Reduction of Private Sector Debt? Comparing Spain and the
EuropeA Selective Comparison of Countries and National UK. Bruegel (blog).
Definitions. Focus on European Economic Integration (1): 2847. International Monetary Fund (IMF). 2014a. Euro Area Article
Baum, Anja, Cristina Checherita, and Philipp Rother. 2013. IV Consultation14/198, Washington.
Debt and Growth: New Evidence from the Euro Area. . 2014b. Staff Guidance Note on Macroprudential
Journal of International Money and Finance 32: 80921. Policy. Washington.

International Monetary Fund | April 2015 53


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Jassaud, Nadge, and Kang, Kenneth. 2015. A Strategy for Devel- Reinhart, Carmen M., and Kenneth S. Rogoff. 2010. Growth
oping a Market for Nonperforming Loans in Italy. Working in a Time of Debt. American Economic Review 100 (2):
Paper 15/24, International Monetary Fund, Washington. 57378.
Kim, D. H., and J. H. Wright. 2005. An Arbitrage-Free Satio, Masashi and Yoshihiko Hogen. 2014. "Portfolio Rebal-
Three-Factor Term Structure Model and the Recent Behavior ancing Following the Bank of Japan's Government Bond
of Long-Term Yields and Distant-Horizon Forward Rates, Purchases: Empirical Analysis Using Data on Bank Loans
Finance and Economics Discussion Series (FEDS): 200533 and Investment Flows." BOJ Reports & Research Papers,
(Washington: Federal Reserve Board). June.
Kumar, M., and J. Woo. 2010. Public Debt and Growth. Work- Tabb, L. (2012). Written Testimony to the United States Senate
ing Paper 10/174, International Monetary Fund, Washington. Committee on Banking, Housing, and Urban Affairs, CEO,
Pericoli, Marcello, and Sbracia Massimo. 2004. A Primer on Finan- TABB Group. September 20.
cial Contagion. Journal of Economic Surveys 17 (4): 571608.

54 International Monetary Fund | April 2015


2
CHAPTER

INTERNATIONAL BANKING AFTER THE CRISIS:


INCREASINGLY LOCAL AND SAFER?

SUMMARY

T
wo developments stand out among the changes in international banking since the global financial
crisis. First, direct cross-border lending as a share of total banking assets has declined, mostly because of
the retrenchment of European banks. Second, the share of local lending by foreign bank affiliates has
remained steady. Global banks in particular have refocused their activities on some key markets, leaving
space for other banks to expand. As a result, intraregional financial linkages have deepened, especially in Asia.
Although the cutback in cross-border lending was triggered by the crisis, regulatory changes and weaknesses in
bank balance sheets have contributed significantly to the subsequent retrenchment. Better-capitalized banks were
more likely to maintain cross-border lending. Macroeconomic factors have also played a role.
The relative shift on the part of foreign banks away from cross-border lending and toward more local lending
through affiliates has a positive effect on the financial stability of host countries. Cross-border lending compounds
adverse domestic and global shocks. In contrast, foreign-owned subsidiaries, particularly those with better-capital-
ized parent banks, tend to behave less procyclically than domestic banks around domestic crises.
In principle, international banking has benefits that are not examined in this chapter. For example, global banks
contribute to the allocation of global savings across countries, with positive effects on investment and growth. The
reduction in cross-border lending may diminish some of those benefits.
Policymakers should therefore strive to maximize the benefits of international banking while mitigating risks.
The findings of this chapter lend support to recent financial reforms that strengthen the resilience of global banks.
They also emphasize the need for more international cooperation to deal with regional or global shocks.

Prepared by Frederic Lambert (team leader), Pragyan Deb, Johannes Ehrentraud, Brenda Gonzlez-Hermosillo, Hibiki Ichiue, Oksana
Khadarina, Win Monroe, Hiroko Oura, Martin Saldas, and Kai Yan, with contributions from John Bluedorn and Alexandra Peter, under the
overall guidance of Gaston Gelos and Dong He.

International Monetary Fund | April 2015 55


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Introduction countries can restrict the ability of parent banks to


withdraw liquidity from their subsidiaries.1 The overall
International banking has changed since the global financial stability effect of the observed patterns of
financial crisis. Two developments stand out. Interna- changes in global banking is therefore unclear without
tional banks, especially European ones, have reduced further examination.
their cross-border lending, that is, their direct lending This chapter provides a comprehensive picture of
to non-affiliated entities in other countries. At the recent changes in international banking, analyzes what
same time, loans extended locally by banks affiliates is driving those changes, and investigates the potential
abroad have remained steady. Other changes include a consequences for financial stability. The results should
retrenchment of international banks from certain mar- not be interpreted as providing a full cost-benefit
ket segments, the emergence of new actors to fill the analysis of the changing patterns of banking globaliza-
resulting gaps, and some regionalization where global tion. In particular, the role of international banks in
banks are replaced by ones with a more regional focus. the global allocation of savings and in contributing to
The drivers of these changes have been both internal financial deepening is not explored.2
and external to the banking sector. The sharp and The analysis finds that, jointly with bank balance
prolonged process of deleveraging of banks and house- sheet weaknesses, regulatory changes can explain a sub-
holds since 2008 has had a strong effect on credit sup- stantial portion of the decline in the ratio of cross-bor-
ply and demand. Large U.S. and European banks have der claims to GDP between the precrisis and postcrisis
been cleaning up their balance sheets and selling legacy periods. The results are based on data collected in a
assets while trying to reduce their reliance on less survey conducted specifically for this chapter. Macro-
stable funding sources, such as short-term wholesale economic factors, including monetary policy factors,
funding. At the same time, banks have been pressed have also played a role.
by supervisors to shore up capital, while abstaining The financial stability implications of the rela-
from reducing domestic credit supply. Different eco- tive shift away from cross-border lending and toward
nomic conditions across countries and recent financial more local lending by branches and subsidiaries may
reforms, such as those aiming at restricting certain be positive from the perspective of host countries. A
types of operations by banks, as well as new capital comparison reveals that cross-border banking flows
and liquidity standards, have also affected banks global have historically been much more volatile and sensi-
operations and their organizational structure. tive than portfolio flows to global financial conditions.
The reduction in cross-border banking flows can in Consequently, a reduction in their relative importance
principle have opposite effects on financial stability. is likely to reduce the global transmission of volatility
The retrenchment in cross-border lending may reduce and contagion. The analysis also finds that cross-border
risk sharing and diversification for banking groups lending is associated with a strong transmission of
(Allen and others 2011), because investing or lending global shocks to domestic banking systems, and does
abroad allows banks to reduce their exposure to domes- not help dampen local shocks. By contrast, confirming
tic shocks (Schoenmaker and Wagner 2011). From the existing findings in the literature, the chapter finds that
perspective of recipient countries, cross-border lending local lending by foreign subsidiaries is more resilient in
may also lower the volatility of domestic credit because the face of domestic shocks.
foreign banks, which are less exposed to domestic The strengthening of regional linkages, particu-
shocks, are more able to withstand local stress. Then larly in Asia, implies a heightened exposure to shocks
again, cross-border flows are also likely to contribute emanating from within the region. It also means that
to the transmission of foreign shocks and may thus shocks originating outside the region can propagate
increase volatility (Bruno and Shin, forthcoming; faster within the region once they hit a countrys
IMF 2014c). For example, deleveraging by interna-
1The shift toward local funding may also enhance the effectiveness
tional banks can reduce funding sources for banks in
of monetary policy by tightening the link between domestic interest
host countries. These banks in turn may be forced to rates and credit supply (Forbes 2014).
contract lending even in the absence of domestic credit 2For example, foreign bank presence is also often associated with

problems. Moreover, cross-border lending is often seen greater efficiency and competition in host countries banking sec-
tors (Claessens and Laeven 2004; Cull and Martnez Pera 2010).
as less stable than local lending through local subsidiar- Enhanced competition in turn may also affect financial stability;
ies and branches (Schnabl 2012), partly because host these issues are not explored here.

56 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

banks. This prospect may call for a strengthening local operations whereas internationally operating
of regional safety nets to address idiosyncratic and Japanese banks continue to conduct mostly cross-bor-
regional shocks. der operations. Differences in business models can be
Financial reforms that contribute to strengthen- related to differences in funding models. Multinational
ing the soundness of parent banks can help limit the banks tend to rely less on wholesale funding and were
transmission of negative foreign shocks by affiliates of thus less affected by disruptions in the wholesale fund-
foreign banks. Increased cooperation among national ing market during the crisis.
regulators and supervisorsnot only in matters of One question is whether the precrisis level of cross-
cross-border resolution, but also on the implemen- border claims reflected an anomalythat is, the out-
tation of Basel standards and on accounting stan- come of a temporary, unsustainable boom. Although
dardsis key to reconciling banking globalization with this question extends beyond the scope of the chapter,
financial stability. it is worth noting that international claims (which
include cross-border claims and local claims of foreign
bank affiliates in foreign currencysee Figure 2.2)
What Has Changed? grew steadily between 2002 and 2007, with the growth
From Cross-Border Banking to Multinational Banking rate picking up only somewhat in 2007 (Figure 2.1,
Cross-border bank lending has declined since the panel 1). This at least indicates that the observed levels
global financial crisis, while international banks have in 200708 were part of a longer-term trend (which
shifted their international business models toward may well have been unsustainable).
more local operations. Cross-border claims as a share The reduction in cross-border lending and lending
of total banking assets of host countries have not through affiliates is mainly due to euro area banks;
recovered to their precrisis level (Figure 2.1, panel 1).3 banks from other areas have only partially offset that
Local loans extended by affiliates of foreign banks did reduction (Figure 2.1, panel 3). Foreign claims of
fall slightly in 2007 and 2008 but have since stabilized. European banks dropped sharply in the wake of the
Their share in total foreign claims (the sum of cross- global crisis and have continued to decline since then.
border claims plus loans extended through affiliates The drop in claims from euro area banks has been
abroad) has thus grown from less than 43 percent to general across all regions of the world. Claims vis--vis
about 49 percent. Most of those loans are in local non-euro-area countries have dropped more than intra-
currency; their share rose mildly after the crisis, most euro-area claims. U.S. and U.K. banks also retrenched
likely because of foreign currency funding pressures in 2008, but their foreign claims have partially recov-
(McGuire and von Peter 2009), and has not returned ered. Foreign claims from other areas, particularly from
to its precrisis level even after the pressures abated Japan, have grown quickly.
(Figure 2.1, panel 2). Foreign claims on emerging market and develop-
The shift from cross-border banking to multina- ing countries dropped in all regions in 2008 and have
tional banking with more local and likely locally exhibited different recovery patterns. Claims on the Asia
funded operations is more pronounced in some bank- and Pacific region have nearly doubled since their 2008
ing systems than in others. McCauley, McGuire, and trough (Figure 2.1, panel 4). Those on Latin America
von Peter (2012) show that global French and Spanish and the Caribbean have also exceeded their precrisis peak,
banks in particular have increased the share of their although growth has slowed since 2011. Meanwhile,
outstanding claims on emerging and developing Europe
are still hovering slightly below their precrisis levels.
3Strictly speaking, banking claims include not only loans but

also deposits with other banks and holdings of securities and partici-
Overall, international banks have somewhat reduced
pations. Following the Bank for International Settlements terminol- the number of branches and subsidiaries they hold
ogy, foreign banking claims are defined as the sum of cross-border abroad. Based on a sample of 64 countries, including
claims (for example, a direct loan of a bank in a given country to
both advanced and emerging market economies, the
a firm in another country) and local claims of affiliates of foreign
banks in local or foreign currency (for example, a loan from a branch total number of affiliates of foreign banks shrank by
or subsidiary of a foreign bank in a given country to a firm in that about 5 percent between 2008 and 2013. The drop
same country). International claims include cross-border claims essentially comes from a reduction in the number of
and only the part of local claims denominated in foreign currency.
See Figure 2.2. This chapter considers claims reported on a consoli- subsidiaries, especially in the European Union, while
dated basis; that is, intragroup positions are netted out. the total number of branches has risen marginally

International Monetary Fund | April 2015 57


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 2.1. Developments in Foreign Banking Claims

1. Cross-Border and Local Claims Relative to 2. Share of Local Currency Claims in Total Local Claims
Total Banking Assets of Recipient Countries (Percent)
(Percent)
18 87
16
14 86
12
10
85
8
6 International claims (cross-border plus local claims in foreign currency)
4 Cross-border claims 84
2 Local claims
0 83
2002 03 04 05 06 07 08 09 10 11 12 13 2005 06 07 08 09 10 11 12 13

3. Foreign Claims by Home Country of Banks 4. Foreign Claims on Emerging Market and Developing Economies
(Trillions of U.S. dollars) (Trillions of U.S. dollars)
18 2.5
Euro area Other Europe United States Other Europe Latin America and the Caribbean Asia and Pacic
16
14 2.0
12
1.5
10
8
1.0
6
4 0.5
2
0 0.0
2005 06 07 08 09 10 11 12 13 2005 06 07 08 09 10 11 12 13
Sources: Bank for International Settlements (BIS), Consolidated Banking Statistics; IMF, International Financial Statistics database; and IMF staff calculations.
Note: Claims include deposits and balances placed with other banks, loans and advances to banks and nonbanks, and holdings of securities and participations. Foreign
claims are the sum of cross-border claims and local claims of afliates of foreign banks. International claims include cross-border claims and local claims in foreign
currency. In panel 1, the ratios are calculated by dividing claims of all BIS reporting countries by total bank assets for all host countries with available data for each
period. In panel 2, the share of local claims in local currency is estimated by dividing local claims in local currency on an immediate risk basis by the total local claims on
an ultimate risk basis, after correcting for the difference in foreign claims in the two bases. Panels 1, 3, and 4 are based on ultimate risk basis data. The data in panels 1,
2, and 3 are adjusted for statistical breaks following Cerutti (2013). The observation period ends in 2013:Q3. In panel 3, Other consists of Australia, Canada, Chile, India,
Japan, Singapore, Taiwan Province of China, and Turkey.

Figure 2.2. Types of Claims in Bank for International


Settlements Consolidated Statistics

Local claims of foreign banks afliates


A B C
Cross-border claims Local claims of foreign Local claims of foreign
banks afliates in banks afliates in local
foreign currency currency

International claims (A + B)

Foreign claims (A + B + C)

Source: Cerutti, Claessens, and McGuire 2012.


Note: Cross-border claims and total local claims of foreign banks afliates are
reported on an ultimate risk basis (that is, allocated to the country in which the
nal risk lies) whereas international claims are compiled on an immediate risk
basis (allocated to the country of residence of the immediate counterparty).
Cross-border claims do not include intragroup positions. See Annex 2.1.

58 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Figure 2.3. Number of Branches and Subsidiaries of Foreign Banks in 2008 and 2013, by Region

1. Number of Branches 2. Number of Subsidiaries


1,200 1,200

1,000 2008 2013 2008 2013 1,000

800 800

600 600

400 400

200 200

0 0
Africa and Middle Asia and Pacic Europe Latin and North Africa and Middle Asia and Pacic Europe Latin and North
East America East America

Sources: National authorities; and IMF staff estimates.


Note: Africa and Middle East = Bahrain, Botswana, Israel, Jordan, Kuwait, Nigeria, Oman, Qatar, Saudi Arabia, South Africa, United Arab Emirates; Asia and Pacic =
Australia, China, Hong Kong SAR, India, Indonesia, Japan, Republic of Korea, Malaysia, New Zealand, Philippines, Singapore, Thailand; Europe = Austria, Belgium,
Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands,
Poland, Portugal, Romania, Russia, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine, the United Kingdom; Latin and North America = Argentina,
Brazil, Canada, Chile, Colombia, Costa Rica, Mexico, Paraguay, Peru, and the United States. See Fiechter and others 2011.

(Figure 2.3). There is no evidence of increased subsid- A Trend toward Regionalization?


iarization at the expense of branches.4 Since 2008, only
The reduction in the exposures of euro area banks
7 of the 64 sample countries experienced an increase in
to some regions has left a gap that local banks have,
the number of foreign subsidiaries and a simultaneous
at least partially, filled. In Asia in particular, the
decline in the number of foreign branches.
retrenchment of euro area banks has been accom-
The decline in the number of foreign affiliates
panied by increased regionalization. According to
partially reflects the refocusing of global banks inter-
data from the Bank for International Settlements
national operations on core markets and businesses.
(BIS), foreign banking claims of euro area banks in
Having strengthened their balance sheets and reduced
the emerging and developing Asia and Pacific region
risk exposures to meet risk-based requirements, global
have declined since 2008 and have not recovered to
banks are reallocating capital to core businesses and
their precrisis level, despite the regions high growth
markets, shrinking capital markets activities, rebalanc-
(Figure 2.4, panel 1). This decline has been more
ing their business models away from capital-intensive
than offset by the expansion of banks from Asian
activities to more fee-based businesses, and refocusing
countries, particularly Japan. The increase in claims
their geographical presence on fast-growing markets
of other European countries, which likely reflect
(Claessens and van Horen 2014) or on markets in
those of British banks with a very large Asian pres-
which they have a competitive edge (see Chapter 1 of
ence, such as HSBC and Standard Chartered, was
the October 2014 Global Financial Stability Report).
remarkable in 2009 and 2010, but growth has since
slowed. Claims of Chinese banks are not reported to
4Operating in the form of a subsidiary versus a branch has legal
the BIS, but anecdotal evidence suggests a significant
implications. Subsidiaries are entities legally independent from the
increase.
parent bank and have to fulfill regulatory requirements, including An analysis of Asian banks geographical alloca-
capital and liquidity ratios, on a stand-alone basis in the host coun- tion of assets shows an increased concentration in
try. In addition to consolidated supervision by the home supervisor,
the region. The share of regional assets more than
subsidiaries are regulated and supervised by the authorities in the
host country. In contrast, branches are an integral part of the par- doubled between the precrisis and postcrisis periods,
ent company and are typically subject to more limited supervision rising from about 10 percent to close to 20 percent
by host supervisors (Fiechter and others 2011; IMF 2013b). Host of total assets, whereas the share of domestic assets
country authorities generally prefer the subsidiary model, and some
countries are implementing measures that require foreign banks to declined from 84 percent to 73 percent (Figure 2.4,
operate as subsidiaries under certain conditions. panel 2). These changes reflect in particular the recent

International Monetary Fund | April 2015 59


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 2.4. Banking Regionalization in Asia

1. Foreign Claims on Emerging Market and Developing Asia 2. Geographic Breakdown of Assets of Asian Banks
and Pacic by Home Country of Banks (Percent)
(Billions of U.S. dollars)
700 90
Euro area Other Europe Asia Other Before crisis After crisis 80
600

Average asset share


70
500
60
400 50
300 40
30
200
20
100 10
0 0
2005 06 07 08 09 10 11 12 13 Domestic Asia Rest of the world
Sources: Bank for International Settlements, Consolidated Banking Statistics; and Sources: Datastream/Worldscope; and IMF staff calculations.
IMF staff calculations. Note: Average geographic breakdown of Asian banks assets as a percentage of
Note: Foreign claims are the sum of cross-border claims and local claims of their total assets before and after the global nancial crisis (from 2002 to 2007
afliates of foreign banks. Asia consists of Australia, India, and Japan. Other and from 2008 to 2013, respectively).
consists of Canada, Chile, Turkey, and the United States. The panel is based on
ultimate risk basis data. China does not report its banking claims to the Bank for
International Settlements.

internationalization and regionalization of Chinese different markets.6 Figure 2.6 shows the networks
banks (see Box 2.1 for a comparison of the inter- in 19982007 and 201014 using data from both
nationalization strategies of Chinese and Japanese advanced and emerging market and developing
banks). economies. Each colored square represents a bilateral
Other regions of the world do not show a com- correlation between two banks stock returns after
parable degree of regionalization. In Latin America, removing the effect of strong common factors (for
the retrenchment of European banks was short-lived instance, a shock to the whole banking industry). Sig-
and has been accompanied by an increase in lend- nificant correlations tend to be clustered by countries
ing by U.S., Canadian, and Latin American banks. and regions, which underscores the importance of
Colombian banks, for instance, have aggressively local factors such as common balance sheet or market
expanded in Central America.5 In emerging Europe, exposures, common accounting practices, or techno-
the share of European banks in total foreign claims logical linkages. More than 90 percent of the signifi-
declined slightly, reflecting both the deleveraging that cant correlations in both periods are between banks
took place in the region in the aftermath of the crisis within the same region. Although most banks are not
and the effect of the Vienna Initiative in preventing a directly connected to one another, the combination
sudden and massive reduction in cross-border financ- of strong linkages within countries and regions and
ing (Figure 2.5, panel 2). In Africa, the rapid regional the presence of a few cross-regional links (via so-
expansion of pan-African banks in recent years has called hub banks) may allow for rapid transmission of
contributed to increasing cross-country linkages across shocks across regions.
that continent (Box 2.2).
Correlation networks based on banks stock returns
illustrate patterns in financial interconnections across 6The correlation networks used in this section are derived from spa-

tial-econometric techniques described in Saldas and Craig (forthcom-


ing) and Bailey, Holly, and Pesaran (forthcoming), applied to banks
daily stock returns. These networks are obtained by applying first spa-
5Colombia does not report international banking statistics to the tial dependence methods to detect and filter the effects of strong com-
BIS; the regional expansion of Colombian banks is therefore not mon factors and then a thresholding procedure to select the significant
reflected in Figure 2.5, panel 1. bilateral correlations.

60 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Intraregional linkages increased in the postcrisis Figure 2.5. Trends in Latin America and Europe
period in Europe, Middle East, and Africa (EMEA)
countries and especially in Asia. Cross-regional 1. Foreign Claims on Emerging Market and Developing Latin America
linkages, represented by colored squares outside the and the Caribbean, by Home Country of Banks
(Billions of U.S. dollars)
diagonal blocks in Figure 2.6, were more frequent dur- 700
Euro area Other Europe America Other
ing the precrisis period (19982007). EMEA banks in
600
particular exhibited many linkages with banks in Asia
500
and the Americas, which contributed to the propaga-
tion of the crisis across regions. The regionalization of 400
banking linkages since 2010 partially reflects increased 300
correlations within countries, illustrated by a larger 200
concentration of colored squares within each diagonal
100
block, especially in Asia, but also actual growth in the
0
share of regional cross-country interconnections after 2005 06 07 08 09 10 11 12 13
the crisis. Sources: Bank for International Settlements, Consolidated Banking Statistics; and
IMF staff calculations.
Note: Based on ultimate risk basis data. America consists of Canada, Chile, and
Changes in Corporate Borrowing the United States. Other consists of Australia, India, Japan, and Turkey.

The decline in cross-border lending by banks has been 2. Foreign Claims on Emerging Market and Developing Europe, by
Home Country
accompanied by a surge in international nonfinancial (Billions of U.S. dollars)
corporate bond issuances (Figure 2.7, panel 1). This 1,400 Euro area Other Europe Other
surge has been driven to a large extent by the rapid 1,200
increase in bond issuances from emerging markets (see 1,000
Chapter1 of the October 2014 Global Financial Sta-
800
bility Report). Faced with bank credit constraints, firms,
600
especially large ones, may have turned to capital mar-
kets to obtain financing. The low level of interest rates 400
has also encouraged risk taking by private investors and 200
fueled the demand for higher-risk debt securities. One 0
question is to what extent the reduction in cross-bor- 2005 06 07 08 09 10 11 12 13

der banking and the expansion in direct capital market Sources: Bank for International Settlements, Consolidated Banking Statistics;
and IMF staff calculations.
borrowing by nonfinancial firms may have affected Note: Based on ultimate risk basis data. Other consists of Australia, Canada,
their borrowing costs. Chile, India, Japan, Turkey, and the United States.
All else equal, a less globalized banking system may 3. Geographic Breakdown of Assets of European Banks
imply greater heterogeneity of bank funding costs for (Percent)
90
firms across countries. The decline in cross-border 80
lending may limit arbitrage opportunities for firms 70 Before crisis After crisis
and reduce competitive pressures for domestic banks 60
when capital markets are shallow. It also makes lend- 50
ing interest rates more dependent on the condition of 40
the domestic banking sector. The cross-country diver- 30
gence of bank lending rates was one of the features of 20
the euro area crisis and the main sign of the frag- 10
mentation of euro area financial markets (see Box 2.4 0
Domestic Europe Rest of the world
and Chapter 1 of the October 2013 Global Financial Sources: Datastream/Worldscope; and IMF staff calculations.
Stability Report). Higher dispersion of corporate Note: Average geographic breakdown of European banks assets as a percentage
borrowing costs at the global level would potentially of their total assets before and after the global nancial crisis (from 2002 to
2007 and from 2008 to 2013, respectively).
have adverse consequences for private investment in
some countries because firms with profitable invest-
ment opportunities may struggle to obtain funding

International Monetary Fund | April 2015 61


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 2.6. Precrisis and Postcrisis Geographic Correlation Networks from Banks Stock Returns

Correlation network Correlation network


19982007 201014
EMEA EMEA

Asia Asia

Americas Americas

EMEA Asia Americas EMEA Asia Americas

Medium negative correlation


Weak negative correlation
Weak positive correlation
Medium positive correlation
Strong positive correlation

Sources: Bloomberg, L.P.; and IMF staff estimates.


Note: The networks are constructed from daily stock returns of 506 banks located in 62 countries. Each colored square represent a bilateral correlation between two
banks after removing the effect of strong common factors. The matrix is symmetric, which allows for identifying clusters by square areas. The banks are grouped into
nine sub-regions and three regions (Europe, Middle East, and Africa; Asia; and Americas), then sorted by country (alphabetically) and size (market capitalization)
within each region. The nine sub-regions are advanced European economies, emerging and developing Europe, Commonwealth of Independent States, advanced
Asian economies, emerging and developing Asia, advanced American economies, Latin America and the Caribbean, Middle East, North Africa, Afghanistan, and
Pakistan, and sub-Saharan Africa. The sub-regions follow the country classication in the World Economic Outlook. EMEA = Europe, Middle East, and Africa.

or face higher borrowing costs as a result of lower debt crisis in Europe in 2011, it has recently declined.8
banking competition. Panel 2 in Figure 2.7 illustrates However, in euro area countries, the dispersion of cor-
the changes in the dispersion of manufacturing firms porate borrowing costs did rise after 2008 compared
borrowing costs since 1990, after accounting for firm with the precrisis period.
and country characteristics.7
There is no clear evidence of increased dispersion of
corporate borrowing costs following the global finan- Summary
cial crisis. Corporate borrowing costs have converged Cross-border lending is the dimension of global
across countries since 1990, in line with the rise of banking that has shrunk most sharply since the global
financial globalization. The recent changes in interna- financial crisis. Local claims of affiliates of foreign
tional banking patterns described in this chapter do banks have remained more resilient despite an overall
not seem to have reversed this trend. Although the reduction in the number of foreign subsidiaries and
cross-country dispersion of corporate funding costs branches. Euro area banks retrenched the most. Where
seemingly rose slightly after 2008 and again after the they were replaced by other, more regionally focused

8Because the borrowing cost measure is backward looking (it


7This
dispersion is interpreted as a sign of financial frictions that represents the average interest cost on outstanding debt and not the
distort the allocation of resources among firms (Gilchrist, Sim, and cost on newly obtained loans), the estimation does not capture the
Zakrajek 2013). most recent changes in borrowing costs.

62 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Figure 2.7. Changes in Corporate Borrowing banks, international banking linkages have become
more regional. Yet these developments do not seem to
1. Net Issues of International Debt Securities by Nonnancial have led to a larger dispersion of corporate borrowing
Corporations costs.
(Billions of U.S. dollars)
140

120 The Drivers of the Changes in International


Banking
100
Changes in Regulations on Banks International
80 Operations
This section examines the drivers of the previ-
60
ously described changes in international banking.
40
The analysis builds on the results of a confidential
survey about the regulations applicable to banks
20 international operations in both home and host
countries (see Annex Table 2.1.1 for a list of the
0 survey questions). Answers were collected from
bank supervisors in 40 countries that are among
20
2005 06 07 08 09 10 11 12 13 14 the top recipients of international banking claims
Sources: Bank for International Settlements, Debt Securities Statistics; and IMF according to BIS data.9
staff estimates. The survey results show that many countries tight-
ened regulations on banks international operations
2. Cross-Country Dispersion of Corporate Borrowing Costs or strengthened their supervision between 2006 and
(Percent)
2014, while a more limited number loosened them
3.2 (Figure 2.8). The supervisory authorities in many
3.0 countries are now more likely than before to limit
2.8
banks activitiesfor instance, by imposing ring-fenc-
ing measures in a discretionary way. Many resolution
2.6
authorities obtained more powers over local branches
2.4 of foreign banks. Some countries amended banking
2.2 secrecy laws to enhance information sharing about
2.0
banks operations and balance sheets with foreign
supervisors. In contrast, a few countries have loosened
1.8
regulations regarding foreign banking presence (for
1.6 example, conditions for a foreign banks acquisition
1.4 of a domestic bank) and activity (for example, cross-
border lending and borrowing).
1.2
The proportion of countries that tightened their
1.0 regulations on banks international operations is
1991 93 95 97 99 01 03 05 07 09 11 13
higher in advanced economies than in emerging
Sources: Datastream/Worldscope; IMF staff estimates.
Note: The gure plots the standard deviation of median manufacturing rms market economies (Figure 2.9). There is, however,
borrowing costs across countries, after accounting for rm and macroeconomic little evidence that countries that experienced
characteristics, including country risk. The two spikes in the gure correspond to
the years following the Asian crisis in 1998, and the Argentine crisis in 2001.
9Survey respondents were Argentina, Australia, Austria, Belgium,
Corporate borrowing costs are computed from listed rms balance sheet and
income statement data as the ratio of rms interest expenses to total debt. Brazil, Canada, Chile, China, Croatia, Denmark, Finland, France,
Germany, Greece, Hong Kong SAR, Hungary, Indonesia, Ireland,
Italy, Japan, Luxembourg, Malaysia, Mexico, the Netherlands,
Norway, the Philippines, Portugal, Romania, Russia, Saudi Arabia,
Singapore, the Slovak Republic, South Africa, Spain, Sweden,
Switzerland, Thailand, Turkey, the United Kingdom, and the United
States.

International Monetary Fund | April 2015 63


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 2.1. The International Expansion of Chinese and Japanese Banks


This box compares the international expansion strategies of Although Chinese banks expanded rapidly after the
Chinese and Japanese banks and discusses some implica- financial crisis, their global business is still limited in
tions for financial stability. scale and much smaller than that of Japanese banks,
which were among the worlds biggest creditors
Banks headquartered in China and Japan expanded before the Japanese banking crisis of the late 1990s.
rapidly after the global financial crisis. Strong balance The internationalization of Chinese banks remains
sheets, growth opportunities outside the domestic primarily driven by a follow-your-customer strategy.
economy, and the retrenchment of euro area and U.S. In contrast, limited domestic growth prospects and
banks from Asia have been common factors behind new business opportunities abroad for Japanese banks,
their international expansions. However, their growth particularly following the retrenchment of European
also differs in several important ways, including scales, banks, added incentives for them to expand abroad
business lines, and funding patterns. (Lam 2013). The degree of internationalization also
The scale of international expansion varies greatly among the four largest Chinese banks.
The proportion of both international assets and inter-
Japanese banks and, to a more limited extent, Chinese national loans has exceeded 20 percent for the Bank of
banks, have increased their assets and loans overseas China, but is still less than 5 percent for the Agricul-
as a share of total assets and loans, respectively (Figure tural Bank of China.
2.1.1). From 2009 to 2013, the average ratio of over-
seas loans to total loans for the three largest Japanese Business models and expansion strategies
banks rose from 15 percent to about 26 percent.1 The
same numbers for the four largest Chinese banks were Both Chinese and Japanese banks generate major
6.1 percent and 9.2 percent. portions of their revenues abroad from net interest
income (Figure 2.1.2). For Chinese banks, corpo-
This box was prepared by Kai Yan.
rate loans amount to more than 80 percent of the
1The data set includes the four largest banks in China (Indus- total loan portfolio, with most of them coming from
trial and Commercial Bank of China, China Construction Bank, Chinese customers foreign subsidiaries. For Japanese
Bank of China, and Agricultural Bank of China), and the three banks, which showed resilience during the global
largest banks in Japan (Mitsubishi, Mizuho, and Sumitomo). financial crisis and which benefit from strong capital
Mizuho does not report assets and liabilities based on geographic
segments. The average for Japanese banks in Figure 2.1.1, panel
buffers, longer-term project finance and syndicated
1, and Figure 2.1.3, panel 2, is thus computed using data for the lending have also played a major role in their overseas
two remaining banks. expansion.

Figure 2.1.1. The Internationalization of Chinese and Japanese Banks

1. Share of Foreign Assets 2. Share of Foreign Lending


0.5 0.5
Ratio of overseas assets to total assets

Ratio of overseas loans to total loans

0.4 0.4
Japanese banks
0.3 0.3

0.2 Japanese banks 0.2

0.1 Chinese banks Chinese banks 0.1

0.0 0.0
2009 10 11 12 13 2009 10 11 12 13

Sources: Banks annual reports; and IMF staff estimates.

64 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Box 2.1 (continued)


Funding pattern vulnerabilities
Figure 2.1.2. Ratio of Foreign Interest Income The risks of foreign expansion for banks can come
to Foreign Revenue for Chinese and Japanese from both the asset and liability sides. Such risks can
Banks stem from the concentration of exposure to certain
0.8 countries and certain industries, or from dependence
Japanese banks
0.7
on unstable funding sources. This section focuses on

Ratio of foreign net interest


income to foreign revenue
funding vulnerabilities.
0.6
For Japanese banks, the overseas loan-to-deposit
0.5
Chinese banks ratio is about 1.3, with little variation across banks
0.4
(Figure 2.1.3). Chinese banks average overseas loan-
0.3 to-deposit ratio increased from about 1.5 to more than
0.2 2 during the past five years. The rise was primarily
0.1 driven by the growth of the ratio for the Agricultural
0.0 Bank of China, the least globalized of the four largest
2009 10 11 12 13
Chinese banks. At the opposite end, Bank of China,
Sources: Banks annual reports; and IMF staff estimates.
which is the most international of the four, has a loan-
to-deposit ratio of less than 1. The inverse correlation
between Chinese banks foreign loan-to-deposit ratios
Japanese banks have also expanded assertively in and the degree of international activity suggests that
non-lending activities. Overseas business strategies the least globalized banks embarked on aggressive
differ across banks, however. Mizuho Bank, which strategies to expand overseas.
experienced 240 percent growth in foreign non- Another indicator of vulnerability is the ratio of
interest income in the past three years, emphasizes its total overseas liabilities to overseas deposits, which
syndicated loan business as one of the main sources measures banks dependence on funding sources other
of fee income. The revenue generated by Mitsubishis than local deposits for their operations abroad. The
three business lines (foreign exchange, corporate and overseas total liabilities-to-deposits ratio for Chinese
investment banking, and fees and commissions), grew banks has been rising steadily since 2009, indicating a
by 33 percent during the past three years. growing reliance on nontraditional funding. By con-
Similarly, expansion strategies differ for the leading trast, the same ratio has been declining for Japanese
banks in the two countries. Chinese banks tend to banks.
expand their global presence through organic growth Both Chinese and Japanese banks have loan-to-
by opening foreign offices and branches. The increase deposit ratios consistently larger than 1. This shows
in their business coverage mainly occurred in their that despite the increase in deposits collected abroad,
subsidiaries in Hong Kong SAR.2 In contrast, Japanese banks still fall short of funding for their total external
banks have completed major mergers and acquisitions loans and have to rely on external wholesale funding
to expand globally. The three Japanese megabanks to fill the gap. This growing reliance on wholesale
combined spent more than 1 trillion yen acquir- funding could raise potential vulnerabilities from cur-
ing foreign companies between 2012 and 2014. The rency and liability mismatches.
acquisition targets range from banks to asset manage-
ment companies. Future prospects
Growth opportunities still abound for both Chi-
nese and Japanese banks, as their domestic clients
increase their outward expansion. Japanese banks
2For example, of the 623 overseas affiliates of the Bank of
can build on their already well-established market
China, almost all of those outside of mainland China and Hong shares in project finance and syndicated loans to
Kong SAR are overseas branches and offices. Besides the tradi-
take advantage of a rise in infrastructure invest-
tional deposit, loan, and payment business conducted by those
branches, all the other banking business abroad is conducted by ment in Asia, whereas Chinese banks will benefit
Bank of China International, which is in Hong Kong SAR. from the further liberalization of financial markets

International Monetary Fund | April 2015 65


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 2.1 (continued)

Figure 2.1.3. Funding Vulnerabilities for Chinese and Japanese Banks


1. Ratio of Foreign Loans to Foreign Deposits 2. Ratio of Total Overseas Liabilities to
Overseas Deposits
4 4

Ratio of total liabilities to deposits


Chinese banks
Ratio of loans to deposits

3 3

Chinese banks Japanese banks


2 2

1 Japanese banks 1

0 0
2009 10 11 12 13 2009 10 11 12 13
Sources: Banks annual reports; and IMF staff estimates.

in China combined with the internationalization of frameworks, the difficulty of raising local deposits,
the renminbi. and the need to rely on external funding. In addition,
Both groups of banks face challenges, however. Chinese banks relatively simple business model and
Constraints to their global expansion include cross- heavy reliance on domestic customers may also weigh
country differences in regulatory and supervisory on their ability to expand.

higher banking stress, such as some countries in the of lending because their effects may differ across types
euro area, consistently tightened more than other of exposures.
countries.
Changes in regulations targeting banks interna-
tional operations, as well as more general regulatory Econometric Evidence
changes (such as those on bank capital requirements),
According to the econometric analysis, regulatory
can affect foreign banking claims in at least three ways.
changes can explain a sizable fraction of the decline
First and most simply, tighter regulations may reduce
in cross-border claims on recipient countries. The
foreign bank lending just because bank activities in
analysis relates changes in cross-border lending and
general are curtailed. Second, regulatory arbitrage may
in lending by foreign affiliates to changes in regula-
induce a countervailing effect: banks in countries that
tions on international banking operations in both
tighten banking regulations may increase their claims
home and host countries. It uses the results of the
on countries that are less regulated (Houston, Lin, and
above-mentioned survey, as well as changes in capital
Ma 2012; Ongena, Popov, and Udell 2013; Bremus
regulations and official supervisory power (Barth,
and Fratzscher 2014).10 Third, regulatory changes may
Caprio, and Levine 2013), an indicator of the health
bring about a substitution effect between various types
of the banking sector in home countries, and other
macroeconomic variables, including GDP growth
10The literature finds some evidence of regulatory arbitrage across
and changes in exchange rates and real policy interest
countries, and Chapter 2 of the October 2014 Global Financial
Stability Report shows the presence of regulatory arbitrage between
rates. The growth rate of international claims before
banks and the nonbank financial sector. 2007 is used to control for the precrisis boom (see

66 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Figure 2.8. Share of Countries that Changed Regulations on International Banking Operations between 2006 and 2014
1. Home Countries 2. Host Countries
(Percent) (Percent)
35 Tightened Tightened 80
30 Loosened Loosened 70

25 60
50
20
40
15
30
10
20
5 10
0 0
Supervisory discretion

Supervisory discretion
Any regulation

Any regulation
Activity

Information

Activity

Information
Depositor insurance

Other

Other
Resolution
Presence

Presence
Source: IMF staff calculations.
Note: Shares are calculated from the results of a survey about regulations on banks' international operations. The any regulation bar represents the share of countries
that changed any regulation related to banks international operations during 200614. The other bars correspond to the shares of countries that have changed any
corresponding type of regulation.

Annex 2.1 for details). The results show that roughly regulatory changes on local claims is not statistically
half of the drop in cross-border claims (as a percent- significant.
age of GDP) since the precrisis period (200507) The effect of regulatory changes in host countries
can be attributed to regulatory changes. Figure 2.10 depends on the type of regulation (Figure 2.10, panel
examines the sensitivity of the various types of claims 2). Countries that tightened their regulations on banks
to each explanatory variable and the contributions international operations received lower volumes of
of the various factors to the observed changes in the cross-border loans. Changes in capital requirements do
claims-to-GDP ratio. not seem to affect total foreign, cross-border, or local
Tighter regulations on banks international opera- claims. However, tighter capital regulations are posi-
tions or capital regulations in home countries are tively associated with changes in foreign claims on the
associated with a reduction in lending from those public sector, which may be explained by a portfolio
countries (Figure 2.10, panel 1). This effect is intui- shift to safer assets to satisfy more stringent capital
tive, given that both impose limitations on banks requirements (see Annex 2.1).
operations abroad and imply indirect restrictions Higher precrisis bank-capital-to-total-assets ratios
through, for example, higher risk weights on foreign in the home country (a proxy for the health of the
assets.11 There is some indication that home coun- home country banking system) are associated with
tries with more powerful supervisors tend to experi- higher postcrisis growth in foreign claims (Figure 2.10,
ence stronger growth in foreign claims, possibly panel 3).13 The initial sharp drop in claims (up to
as a result of regulatory arbitrage.12 The effect of 2009) may to a large extent be due to this factor, along
with possible expectations of a tightening of regula-
tory standards. In particular, precrisis capitalization
11Figuet, Humblot, and Lahet (2015) estimate that the Basel III
levels of European banks were on average substantially
regulatory reforms could lead to a drop of 20 percent in cross-border
claim inflows to emerging markets. lower than in other countries. This result suggests
12Whereas the literature emphasizes the role of regulatory arbi-

trage, by which banks facing stronger supervisory power at home 13This result is consistent with previous studies on syndicated

may increase foreign claims on countries with less supervisory power, lending that also find that banks with strong balance sheets were
the use of consolidated supervision by home supervisors weakens this better able to maintain lending during the crisis (Kapan and Minoiu
argument. 2013).

International Monetary Fund | April 2015 67


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 2.9. Share of Countries that Tightened Regulations on International Banking Operations between 2006 and 2014, by
Region
(Percent)
1. Home Countries 2. Host Countries
50 100
90
40 80
70
30 60
50
20 40
30
10 20
10
0 0
AE excluding Selected Other euro area Emerging AE excluding Selected Other euro area Emerging
euro area euro area economies market euro area euro area economies market
economies economies economies economies

Source: IMF staff calculations.


Note: AE = advanced economies. Selected euro area economies are those with high borrowing spreads during the 201011 sovereign debt crisis and comprise Greece,
Ireland, Italy, Portugal, and Spain. Other euro area economies comprise Austria, Belgium, Finland, France, Germany, Luxembourg, and the Netherlands. Countries that
tightened regulations are dened as countries with a positive index of changes in regulations on banks international operations (see Annex 2.1).

that although tighter capital requirements in home ables) provide additional evidence of the role played
countries may initially curtail international banking by regulatory changes (see Annex 2.1). In particular,
operations, they can contribute to stabilizing banking the contribution of regulatory changes remains sig-
flows later on once banks have built capital buffers. nificant even when euro area countries are excluded
Countries with higher precrisis growth rates of foreign from the sample or when the euro area is treated as a
claims experienced a larger subsequent contraction in single country.
these claims, as foreign banks deleveraged to strengthen Accommodative monetary policies in the wake of the
their balance sheets. Greater physical distance between crisis may have slowed the decline in international bank-
home and host countries is associated with lower ing activities while also supporting a shift to portfolio
growth, particularly for local claims. investment. After the global financial crisis, short-term
The overall effect of regulatory changes on foreign interest rates effectively hit the zero lower bound in
banking claims is comparable to that of nonregula- many economies, and central banks engaged in uncon-
tory factors (Figure 2.10, panel 4). Among regulatory ventional monetary policies aimed at stimulating their
changes, those directly targeted at the international economies. Those policies helped reduce uncertainty
operations of banks have a larger effect than more and market volatility, lowered banks funding costs, and
general banking regulatory or supervisory changes. bolstered their balance sheets, with a potentially positive
All these results still need to be considered with cau- effect on foreign banking claims. The results indeed
tion. It is possible that the correlation between regu- suggest that international banking activities would likely
lations and foreign claims does not reflect a causal have contracted more without such accommodative
relationship, but may rather be driven by other policies, confirming previous findings in the literature
factors. For instance, the vulnerabilities revealed dur- (Bremus and Fratzscher 2014; IMF 2014c).14
ing the crisis may have caused both bank deleverag-
ing and regulatory reforms in the postcrisis period. 14Empirically examining the effect of unconventional monetary
This concern is alleviated by adding many control policies on capital flows is challenging, in part because long-term
variables, including banks precrisis capital-to-assets interest rates are endogenous to capital flows (Bernanke 2005).
ratios and the precrisis growth rate of international Estimates computed after incorporating long-term interest rates in
the regression model broadly confirm the robustness of the results on
claims, to the regression. Moreover, extensive robust- the effect of regulatory changes while pointing to a significant effect
ness checks (among others, with instrumental vari- of monetary easing (see Annex 2.1).

68 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Figure 2.10. Effects of Regulations and Other Factors on International Banking Linkages
(Percent)

1. Response to Home Country Regulatory Changes 2. Response to Host Country Regulatory Changes
20 20
Total foreign claims
15 Cross-border claims 15
10 Local claims 10
5 5
0 0
5 Total foreign claims 5
10 Cross-border claims 10
Local claims
15 15
20 20
International operations Capital regulations Supervisory power International operations Capital regulations Supervisory power

3. Response to Nonregulatory Variables 4. Contributions of Regulatory Changes to Growth in


Claims-to-GDP Ratio
Total foreign claims Regulations on international operations of banks
Cross-border claims General banking regulations such as capital requirements
Local claims Other
60 Claims to GDP growth 20
50
10
40
30 0
20 10
10
0 20
10 30
20
40
30
40 50
GDP (host) Bank-capital-to- Distance Foreign Cross border Local
total-assets ratio
Types of claims
Source: IMF staff estimates.
Note: Panels 1, 2, and 3 show the effects of a one standard deviation increase in each variable on the growth rate of different types of claims. These are calculated by
multiplying the estimated coefcient of the regression and the cross-sectional standard deviation of the corresponding independent variable. Nonshaded bars
correspond to coefcients that are not signicant at the 10 percent level. Panel 4 decomposes factors contributing to the growth of the claims-to-GDP ratio from
200507 to 201113 averaged across the observations of the regression. The factor contribution is calculated by multiplying the estimated coefcient by the
average of the independent variable. Country samples vary depending on the type of claims. Regulations on international operations of banks is the sum of the
contributions of international operations regulatory changes in home and host countries. General banking regulations such as capital requirements is the sum of
the contributions of the other regulation variables. See Annex 2.1 for details.

Effects on Financial Stability panel 1). Yet there is no clear evidence of substitu-
tion between the various types of flows at the country
Cross-Border Lending and the Volatility of Capital level.
Flows All else equal, the reduction in cross-border banking
Cross-border banking flows dropped more sharply flows can be expected to reduce the sensitivities of total
and more durably than other capital flows in reac- capital inflows to global financial shocks. A compari-
tion to the global financial crisis. Both cross-border son of the sensitivity of different types of flows to the
banking flows and portfolio flows declined strongly Chicago Board Options Exchange Market Volatility
in 2008, but portfolio flows recovered much more Index (VIX) shows that cross-border banking claims
quickly and have remained positive on average since are more sensitive to global conditions than are local
early 2009. By contrast, cross-border banking flows claims, whose sensitivity to global shocks is close to
have been slightly negative since 2009 (Figure 2.11, that of portfolio flows (Figure 2.11, panel 2). This

International Monetary Fund | April 2015 69


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 2.2. The Expansion of Pan-African Banks: Opportunities and Challenges

This box describes the recent expansion of pan-African bank to follow their clients, and the declining role of more
groups (cross-border banks headquartered in Africa), the traditional players such as European banks.
benefits these groups offer, and the financial stability risks The growth of pan-African banks offers a number
they entail. of opportunities and benefits. Anecdotal evidence sug-
gests that the expansion of these banks has improved
The face of African finance is changing rapidly with competition and given rise to economies of scale,
the strong expansion of pan-African banks across the especially in host countries with small local markets.
continent in recent years. Reflecting a number of con- Pan-African banks are driving innovation, offering
verging push and pull factors and aided by improved opportunities to enhance financial inclusion, and
political and macroeconomic stability and robust in some cases contributing to lowering borrowing
economic growth, the number of operations of the costs. For example, in the East African Community,
seven largest groups has more than doubled since the Kenyan banks have introduced innovative business
mid-2000s (Figure 2.2.1). Specific factors contribut- models such as agency banking into neighboring
ing to this expansion include increasing trade linkages countries. Similarly, Moroccan banks focus on small
between African countries, which have induced banks and medium enterprise development is being exported
to francophone West Africa, while Nigerian banks

This box was prepared by Alexandra Peter.

Figure 2.2.2. Major Pan-African Banks:


Figure 2.2.1. Major Pan-African Banks: Systemic Importance by Country, 2013
Cross-Border Expansion, 200214
(Number of subsidiaries in sub-Saharan Africa;percent)
40

2002 2006 2010 2014 35

30

25
No presence
20 Systemically important
presence
15 Non-systemically
important presence
10

5
Sources: Banks annual reports; Bankscope; IMF Monetary
0
Attijariwafa Bank of Groupe United Standard Ecobank Oragroup and Financial Statistics; and IMF staff calculations.
(Morocco) Africa/ Banque Bank for Bank (Togo) (Togo) Note: The countries highlighted in red are those where
BMCE Centrale Africa (South any of the seven largest pan-African banks has a
(Morocco) Populaire (Nigeria) Africa) systemically important presence dened as a deposit
(Morocco) share of more than 10 percent of the banking systems
deposits. This includes the home countries of the
Source: Bank websites and annual reports. pan-African banks.

70 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Box 2.2 (continued)


are expanding their branch networks across their host expansion, the largest pan-African banks have become
countries, including in rural areas. African banks have systemically important in many of their host countries,
also become lead arrangers for syndicated loans, filling raising concerns about spillover risks (Figure 2.2.2).
the gap left by European banks (IMF 2014a). From Most groups conduct their foreign operations through
a home country perspective, the geographical expan- subsidiaries, which rely on local deposits for funding,
sion of pan-African banks increases diversification and somewhat mitigating potential contagion. However,
provides further growth and profit opportunities for with limited information about intragroup exposures
banks. and interconnections within pan-African banks and
However, as these groups have developed in reach cross-border cooperation between supervisors just
and complexity, significant supervision gaps, governance emerging, undetected risks could be mounting. In addi-
issues, and questions about cross-border resolution have tion, pan-African groups have become more complex,
emerged that could pose risks to national and regional encompassing nonbank activities that could give rise to
financial stability if unaddressed. With their rapid additional contagion channels (IMF, forthcoming b).

result confirms previous evidence that net bank flows in response to shocks. Bank credit is one of the main
have consistently been the most volatile type of capital channels of transmission of financial shocks to the
flow (see Chapter 4 of the April 2011 World Eco- real economy and plays a crucial role in the ability of
nomic Outlook). It suggests that the observed changes economic agents to withstand negative shocks.
in international banking may yield a reduction in International banking linkages for each country are
contagion, but potentially may also reduce flows that measured in three ways. The first measure is the ratio
help countries dampen external and domestic shocks.15 of cross-border claims to the total assets of the banking
These issues are examined next. sector in recipient countries. This measure excludes
local lending by foreign branches and subsidiaries in
both foreign and domestic currencies (and, given the
International Banking Linkages, Adverse Shocks, and consolidated nature of the data, also excludes intra-
Credit group lending). The second measure uses international
The analysis now turns to the role that foreign banks claims (the sum of cross-border claims and local claims
can play in mitigating or amplifying the effect of in foreign currency) relative to banking sector assets
adverse local and foreign shocks.16 This question is in recipient countries. Because local claims in foreign
tackled from both a macroeconomic (country-level) currency are more likely to be funded by external
and microeconomic (bank-level) perspective. The borrowing, this measure may better capture the overall
analysis focuses on the effect of international bank- dependence of a country on foreign bank lending. The
ing linkages on the changes in domestic credit growth third measure uses the ratio of foreign subsidiaries and
branches local claims in local currency to total bank-
15Recent changes, such as the growing issuances of nonfinancial ing assets.
corporate bonds or changes in the mix of global portfolio investors
(see Chapter 2 of the April 2014 Global Financial Stability Report) Measuring linkages through cross-border and
might, however, affect the sensitivity of portfolio flows to future international claims
shocks.
16Many studies have looked at the role of international banking Host countries with higher cross-border or inter-
linkages in the transmission of shocks to host countries (for example, national claims tend to be more exposed to global
Cetorelli and Goldberg 2011), while ignoring the role those linkages
shocks.17 In times of global stress, credit growth
may play in smoothing the effect of domestic shocks. The analysis
in this chapter considers both effects, thereby providing a more drops more in these countries (Figure 2.12, panel 1).
comprehensive assessment of the stabilizing role of foreign banks. This finding can be related to the literature pointing
For other effects of banking globalization, in particular the role of to the financial stability risks associated with bank
foreign bank participation in financial development in developing
countries, see Goldberg (2009) and Detragiache, Tressel, and Gupta
(2008). 17Global stress (shocks) is measured by the VIX.

International Monetary Fund | April 2015 71


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 2.11. Changes in Capital Flows

1. Cross-Border Banking Flows and Portfolio Inows 2. Sensitivity of Cross-Border Flows to Global Shocks
(Billions of U.S. dollars) (Percent)
3,000 Cross-border banking ows 5
Gross portfolio inows
2,000 4
1,000
3
0
2
1,000
1
2,000

3,000 0
Cross-border Local claims Total Debt ows Equity ows
2005:Q2
05:Q4
06:Q2
06:Q4
07:Q2
07:Q4
08:Q2
08:Q4
09:Q2
09:Q4
10:Q2
10:Q4
11:Q2
11:Q4
12:Q2
12:Q4
13:Q2
13:Q4
claims
Banking claims Portfolio ows
Sources: Bank for International Settlements (BIS); IMF, International Financial Statistics; and IMF staff calculations.
Note: Cross-border banking ows are computed as changes in cross-border banking claims from the BIS Consolidated Banking Statistics on an ultimate risk basis.
These data are not compiled on a residency basis and therefore are not fully consistent with the ows reported in the balance of payments. For panel 2, all ows are
normalized by the average of their absolute values over the sample period. The bars in panel 2 represent the changes in ows following a one-unit increase in the VIX
(Chicago Board Options Exchange Market Volatility Index).

wholesale funding (see Berkmen and others 2012). cross-border lending does not dampen the impact of
In fact, a substantial portion of precrisis cross-border domestic shocks.
lending by major banks was financed by tapping By contrast, countries that are home to banks with
wholesale markets. Cross-border lending itself may large foreign assets experience some stabilizing benefits.
also reflect cross-border wholesale funding between Domestic credit is less affected during times of global
non-affiliated banks.18 stress in countries that are home to banks with large
Similarly, host countries do not enjoy a diversifica- international operations (Figure 2.12, panel 2). This
tion benefit when they are hit by domestic shocks. outcome may be related to the fact that banks in these
All else equal, cross-border lending by international countries have more leeway to adjust their operations
banks may be expected to be more resilient around worldwide and support the domestic entitiesa form
domestic shocks. For example, the balance sheets of of home bias in which international banks are more
global banks will be less affected by economic stress inclined to maintain credit at home during times of
in any given host country. This should enable these global stress, potentially at the expense of their foreign
banks to curtail lending less than their local peers operations (Giannetti and Laeven 2012). No such
do. However, the opposite seems true. In the face of result is observed, however, for domestic shocks. One
higher domestic banking stress, countries with more possible reason is that international banks, in the face
international banking linkages in the form of cross- of troubles at home, would rather maintain or expand
border or international claims tend to see a larger, not their more profitable overseas operations than support
smaller, contraction in lending.19 This suggests that domestic credit. The underlying assumption is that a
global shock affects global banks activities in a similar
18Other than during stress periods, cross-border and interna- way both at home and abroad, while a domestic shock
tional claims are associated with higher domestic credit growth in hurts the profitability of domestic operations relative to
host countries. This may reflect the role that cross-border lending foreign ones.
can play as a complement to domestic lending in relaxing credit
constraints, and in contributing to financial deepening in countries
with small domestic banking sectors. It may, however, also suggest
a contribution of cross-border lending to unsustainable local credit size of the domestic banks). The average expected default frequency
booms (see next section). of all listed domestic firms, which represents a broader measure
19Domestic stress (shocks) is measured by the average expected of domestic stress, is used as a robustness check; the main results
default frequency of the domestic banking sector (weighted by the remain unchanged.

72 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Figure 2.12. Effect of International Banking Linkages on These results do not depend on the severity of
Domestic Credit Growth domestic or foreign shocks. The analysis finds little
(Percent) evidence that the stabilizing role of global banks may
1. Host Country be either impeded or enhanced during extreme shocks
2.5 or crises.
Cross-border claims
2.0 International claims
1.5 (cross-border plus local Measuring linkages through local currency lending
in foreign currency)
1.0 by foreign banks
0.5 Linkages in the form of higher local currency lend-
0.0 ing by foreign subsidiaries or branches do not amplify
0.5 domestic shocks (Figure 2.12, panel 3). Cross-border
1.0 and international claims do not capture the local
1.5 activities of foreign branches and subsidiaries well.
IBL x IBL x One reason is that local claims are mostly denomi-
Real GDP Global Domestic
IBL Global Domestic
growth shock shock nated in local currency and are therefore more likely
shock shock
2. Home Country to be funded by local deposits. Another reason is that
2.5
on a consolidated basis, cross-border claims cannot
2.0 Cross-border claims
International claims (cross-border plus local in
account for intragroup funding flows, although these
1.5 foreign currency) are known to play a stabilizing role during periods of
1.0 heightened risk (Reinhardt and Riddiough 2014; De
0.5 Haas and van Lelyveld 2010). Intragroup funding can
0.0 help support local lending by foreign banks affiliates.
0.5 In fact, countries with a high share of local lending
1.0 in local currency by foreign banks do not experience
1.5 stronger credit contractions when they are hit by
IBL x IBL x domestic shocks.
Real GDP Global Domestic
IBL Global Domestic
growth shock shock
shock shock
A more in-depth look at subsidiaries lending
3. Host Country
1.5 An examination of the behavior of individual banks
Local claims in local currency
0.8 suggests that lending by foreign-owned subsidiaries is
0.0 in fact more stable during domestic crises. The micro-
0.8 level analysis uses balance sheet data for a large number
1.5 of domestic and foreign-owned banks (see Annex 2.2
2.3 for details). The regression model compares the growth
3.0 rate of loans by foreign-owned subsidiaries in a given
3.8 country with that of domestic banks in periods of
4.5 stress.20 The growth rate of lending by foreign-owned
IBL x IBL x subsidiaries is higher than that of lending by domestic
Real GDP Global Domestic
IBL Global Domestic
growth shock shock banks during domestic banking crises, but lower dur-
shock shock
ing global crises (Figure 2.13). These results emphasize
Source: IMF staff estimates. the beneficial role played by local lending of foreign-
Note: IBL = international banking linkages. The bars show the effects of a one
standard deviation increase in each variable on domestic credit growth. These owned subsidiaries during domestic crises. This finding
are calculated by multiplying the estimated coefcient of the regression and the is consistent with the literature reporting that lending
standard deviation of the corresponding independent variable. Nonshaded bars
correspond to coefcients that are not signicant at the 10 percent level. The by subsidiaries is more stable than direct cross-border
estimation period spans 200213, depending on data availability. lending (Peek and Rosengren 2000; De Haas and van

20Branches of foreign banks are excluded from the analysis

because of the lack of balance sheet data. Using regulatory data,


Hoggarth, Hooley, and Korniyenko (2013) provide an interesting
analysis of the behavior of foreign bank branches in the United
Kingdom.

International Monetary Fund | April 2015 73


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Figure 2.13. Lending Growth by Domestic and Foreign- owned (Figure 2.14, panel 3) and further under-
Owned Banks during Crises scores the importance of banks liability structures for
(Percent) financial stability (see Chapter 3 of the October 2013
10 Global Financial Stability Report).

0
International Banking Linkages and the Incidence of
Crises
5
If certain forms of international banking linkages
10 can aggravate the effect of domestic shocks, do they
15 also increase the incidence of crises more generally?
The previous section found that cross-border banking
20
Global crisis Foreign ownership Domestic crisis x linkages tend to facilitate the transmission of global
Foreign ownership shocks and aggravate the effect of domestic ones on
Domestic crisis Global crisis x
Foreign ownership host countries but are also associated with higher
domestic credit growth on average. Given that rapid
Source: IMF staff estimations.
Note: The bars represent the values of the estimated coefcients of the
credit growth is considered a powerful indicator of
independent variables in a regression of lending growth at the bank level systemic risk buildup, this section directly investi-
comparing lending by domestic and foreign-owned banks (see Annex 2.2). The gates the effect of international banking linkages on
estimation period spans 19982013, depending on data availability.
the probability of a banking crisis (see Annex 2.3 for
more details).
On average, a higher degree of international
banking linkages does not seem to be significantly
Lelyveld 2006; McCauley, McGuire, and von Peter correlated with the probability of domestic banking
2012; Schnabl 2012). crises (Table 2.1). This result is not surprising, since
Foreign subsidiaries with better-capitalized parent the existing literature does not provide a definitive
banks and parent banks with more stable funding answer. Although Demirg-Kunt, Levine, and Min
sources tend to react less procyclically. Higher capi- (1998) find that foreign bank presence tends to lower
talization of the parent bank is associated with higher the probability that a country will experience a bank-
lending growth by its subsidiaries during stress periods ing crisis, more recent work by Minoiu and others
(Figure 2.14, panel 1; and Annex Table 2.2.3). High (forthcoming) suggests a positive relationship between
dependence of parent banks on nondeposit fund- a countrys banking interconnectedness and the prob-
ing sources is destabilizing during both domestic ability of a banking crisis.
and global crises (Figure 2.14, panel 2). The results
highlight the role played by parents dependence on Policy Implications
nondeposit funding sources in increasing contagion,
As evidenced by the regulatory survey results, the
an intuitive and well-known result in the literature
challenges of the recent financial crisis prompted a
(Cetorelli and Goldberg 2012; Porter and Serra
number of countries to take crisis-resolution mea-
2011).21
sures and impose new requirements on banks. The
A high reliance of subsidiaries on domestic deposits
response was global, with the Group of 20 playing
for their funding is also found to help stabilize lending
a major role in setting up the agenda for financial
during both domestic and global stress. This result
reforms (Vials and others 2010). National regula-
holds for all banks, whether domestically or foreign
tory reforms followed, although they were not always
well coordinated across countries. Structural banking
21A comparison of the credit growth of foreign banks with that reforms aiming to reduce interconnectedness between
of domestic banks in Central, Eastern, and Southeastern Europe intermediaries may have intentionally introduced
showed that the tightening in parent banks funding conditions
explained most of the difference in the credit slowdown in 200811
some degree of fragmentation to the market, includ-
(IMF 2013a). ing across borders (FSB 2014). Measures frequently

74 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Figure 2.14. Effect of Parent and Subsidiary used include the separation of specific activities
Characteristics on Subsidiary Lending Growth into different legal entities, restrictions on business
(Percent) models, heightened regulatory requirements on a
1. Sensitivity of Lending Growth by Banks to Parent Banks subconsolidated basis, and requirements to operate
10 Equity Ratio
as subsidiaries instead of branches. These regulatory
changes clearly had an effect on the patterns of inter-
national banking.
5
With regard to financial stability, the findings of
the empirical analysis in this chapter lend support to a
multinational banking model rather than a cross-
0
border one (see Table2.2). In contrast to international
banks, which are mainly engaged in cross-border
5
transactions out of their home countries, multinational
Parent Domestic Domestic crisis banks operate locally through subsidiaries or branches
equity crisis Parent bank Foreign Parent bank
ratio equity ratio equity ratio (McCauley, McGuire, and von Peter 2012). All else
equal, the shift to more local as opposed to cross-
2. Sensitivity of Lending Growth by Banks to Parent Banks
10 Dependence on External Funding border operations results in a decline in the sensitivity
of capital flows to global shocks and yields a reduction
in contagion. Foreign banks operating locally rather
5 than through cross-border transactions tend to contract
credit much less following domestic shocks in host
countries. More local claims may also translate into
0 higher effectiveness of macroprudential policies given
that local measures are less likely to be circumvented
(Vials and Nier 2014; IMF 2014d).
5 Governments can enhance the resilience to financial
Parent Domestic Domestic crisis
dependence on crisis Foreign Parent shocks. A higher reliance of affiliates on local funding
external funding Parent dependence dependence on
on external funding external funding
sources increases their resilience to global shocks. At
the parent level, higher capitalization levels and more
3. Sensitivity of Lending Growth by Banks to Banks stable funding sources positively contribute to finan-
10 Reliance on Local Deposits
cial stability in host countries. The results therefore
support recent financial reforms aimed at strengthen-
ing banks capital and liquidity buffers, especially the
5
buffers of global systemically important banks. The
results also call for the close monitoring of cross-border
0 and foreign currency lending, given that both tend to
compound domestic and global shocks.22
However, limiting cross-border lending across the
5
Global crisis Global crisis
board may jeopardize other benefits and create new
Bank local funding ratio
Bank local Foreign Bank risks, most of them not examined here. The analy-
funding ratio local funding ratio sis finds a positive effect of cross-border lending on
Source: IMF staff estimations.
Note: The bars represent the values of the estimated coefcients of the domestic credit growth in host countries in normal
independent variables multiplied by the standard deviation of the parent or times. Moreover, home countries benefit from hav-
subsidiary characteristic of interest in a regression of lending growth at the bank
level, comparing lending by domestic and foreign-owned banks (see Annex 2.2). ing cross-border banking claims during times of
Nonshaded bars correspond to coefcients that are not statistically signicant at global stress. However, the chapter does not consider
the 10 percent level. The marginal effect of each parent and subsidiary
characteristic on lending growth by foreign subsidiaries during crises corresponds
to the sum of the coefcients on the interaction terms Crisis x Characteristic and 22Lower dependence of banks on external funding, along with
Crisis x Foreign x Characteristic. The estimation period spans 19982013.
stronger supervision, was shown to also reduce the fiscal costs of
banking crises (IMF, forthcoming a).

International Monetary Fund | April 2015 75


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Table 2.1. Effects of International Banking Linkages on the Incidence of Crises


International Banking Linkages Measured with
Cross-Border Claims International Claims Local Claims
Real GDP Growth (year-over-year change, lagged) 0.03 0.05* 0.05*
Credit Growth (lagged) 0.08*** 0.06*** 0.06***
Foreign-Exchange-Reserves-to-GDP Ratio (lagged) 2.59 1.96 1.81
Foreign-Debt-to-GDP Ratio (lagged) 0.39** 0.48*** 0.43***
Current-Account-Balance-to-GDP Ratio (lagged) 0.14*** 0.16*** 0.15***
International Banking Linkages (lagged) 0.16 0.19 0.14

Observations 1,324 1,840 1,792


Number of Countries 46 46 45
Chi-squared 41.8 47.5 46.5
Source: IMF staff estimates.
Note: IBL = international banking linkages. Banking crises are defined as in Laeven and Valencia (2013). The estimates are derived from a random effects panel
probit model. The estimation period spans 200213, depending on data availability. *p < 0.10; **p < 0.05; ***p < 0.01.

the positive role cross-border flows can play in the In particular, stronger intraregional banking linkages
allocation of global savings across countries, and the call for enhanced regional cooperation. Regionalization
resulting benefits for investment and growth. Some may increase vulnerability to regional crises. Dealing
of these benefits would likely be lost if divergences in with such crises requires agreement on the resolution
the implementation of reforms agreed to at the global of regional banks and the availability of adequate fis-
level and the ensuing regulatory fragmentation were cal backstops at the regional level. Box 2.3 provides a
to lead to a further retrenchment of global banks.23 description of the progress made in this regard with
In addition, the changes in the provision of cross- the European banking union.
border credit could raise new financial stability risks. International forums have an important role to play
As international issuances of corporate bonds con- in the advancement of regulatory standards and in
tinue to increase and bank direct cross-border lending ensuring their consistent application across countries
declines, the locus of risks is shifting away from banks (see Box 2.4 for a discussion of areas that warrant
to nonbanks. Such a shift may complicate surveillance attention by financial regulators). Progress along these
of the global financial system (see Chapter 3 of this dimensions would reduce the scope for regulatory arbi-
Global Financial Stability Report and Chapter 2 of the trage between countries as well as between regulated
October 2014 Global Financial Stability Report). banks and the shadow banking system.
One policy challenge would therefore be to make
the global financial system safer for cross-border lend-
ing. Doing so requires a more harmonized institutional Conclusion
and regulatory framework, with more cooperation and The reduction in cross-border lending and the move
coordination among national regulators and supervi- toward more local and locally funded operations,
sors. The analysis highlights the destabilizing effects of partly fostered by regulatory reforms, should positively
cross-border lending during shock episodes; therefore, affect financial stability in host countries. The analysis
the efforts should first focus on reducing the risks in in this chapter provides evidence that cross-border
times of crisis. In that regard, mutually compatible banking tends to aggravate adverse domestic and global
resolution frameworks could provide a global safety shocks in host countries. In contrast, local lending by
net, preventing the ad hoc imposition of ring-fencing foreign banks is less sensitive to global shocks than are
measures. cross-border lending and portfolio inflows in general.
Moreover, lending by foreign-owned subsidiaries, espe-
23Furthermore, the chapter does not consider the particular case of
cially when their parents are well capitalized and less
banking unions, within which the distinction between cross-border
and local claims is less relevant because of full regulatory and super- dependent on nondeposit funding sources, can help
visory integration and the existence of common safety nets. stabilize credit growth in the face of adverse domestic

76 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Table 2.2. Main Findings of the Analysis of the Effects of International Banking Linkages on Domestic Credit Growth
Effect on Domestic Credit Growth by Banks during Periods of
Measure of International Banking Linkages Adverse Domestic Shocks Adverse Global Shocks
Cross-Border Claims Amplifies the effect of the shock Amplifies the effect of the shock
Local Lending through Branches and
Subsidiaries Dampens the effect of the shock Amplifies the effect of the shock
Effect on Lending Growth by Foreign Subsidiaries during Periods of
Parent and Subsidiary Characteristics Domestic Crises Global Crises
Higher Parent Capitalization and Lower Parent
Dependence on Nondeposit Funding Dampens the effect of the crisis Dampens the effect of the crisis
Higher Reliance of Subsidiaries on Local
Deposits Dampens the effect of the crisis Dampens the effect of the crisis
Source: IMF staff.

shocks. Countries that are home to banks with large Overall, the findings lend support to recent regula-
foreign assets still enjoy some risk diversification ben- tory reforms strengthening the resilience of global banks
efits from their international exposures. while calling for further progress on the consistent
However, the chapter does not look into the other implementation of regulatory standards and cross-border
benefits usually associated with cross-border banking resolution. Given the trade-offs, an important policy
flows. Although the decline in cross-border lending challenge is to make the global financial system safer for
may reduce the international transmission of shocks, cross-border lending. Only with sufficient international
it may dampen benefits in other domains, such as cooperation on the regulation and supervision of global
financial deepening, the efficient allocation of global banks can the full benefits of banking globalization be
savings, and the diversification of financing sources. realized with no increased risk to financial stability.

International Monetary Fund | April 2015 77


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 2.3. Banking Union in Europe


This box describes the banking union in Europe as a policy be replaced by a shared and common framework.1
response to financial fragmentation in the euro area. The banking union goes a step further than European
Unionwide initiatives to harmonize banking practice
The global financial crisis and its aftermath led to frag- across countries, by establishing centralized mecha-
mentation of euro area financial markets along national nisms for these functions.2
borders, peaking in the summer of 2012. Bank borrow- Like many European institutions, the euro area
ing and lending costs became highly correlated with sov- mechanisms are layered on top of existing national
ereign yields and both diverged markedly across countries institutions. Under the Single Supervisory Mecha-
(Figure 2.3.1; Goyal and others 2013). Local banks relied nism, which began operation in November 2014,
on their sovereigns as backstops in times of stress, linking the European Central Bank (ECB) is the overarch-
the financial health of the sovereign and the banking ing supervisory authority, directly supervising 120
sector: when banking sector conditions deteriorated, the significant bankswhich together make up almost 85
sovereigns fiscal space to backstop shrank, and vice versa. percent of total euro area bank assetsand overseeing
Moreover, in a currency union, individual member states the supervision of the other 3,500 or so less significant
cannot use interest or exchange rates to support banks in banks in the euro area by their respective national
response to local macroeconomic conditions. competent authorities. Moreover, the ECB can take
To short-circuit bank-sovereign linkages and safe- over the direct supervision of any less significant bank
guard the functioning of the currency union and single if it deems it necessary to ensure the integrity of euro
market, policymakers formulated a plan for a banking area supervision or if the bank becomes systemically
union in the euro area, in which nationally distinct important.
banking supervision and resolution frameworks would Similarly, under the Single Resolution Mechanism,
the newly established, stand-alone Single Resolution
Board oversees the resolution of banks by national res-
olution authorities and directly handles the resolution
of large and cross-border banks. Following European
Unionwide practice, resolution may involve a bail-in
Figure 2.3.1. Sovereign Bond and Corporate of up to 8 percent of bank liabilities. Importantly, as
Lending Rates in the Euro Area of January 2016, the board will also have access to a
(Percent; GDP weighted) common, industry-funded backstop called the Single
8 Resolution Fund to facilitate resolution if needed. The
eventual size of the industry backstop is planned to
7
be 55 billion by 2024 (about 1 percent of covered
6 deposits in the euro area). Together, these tools should
help minimize recourse to taxpayer-financed bail-
5
outs. Moreover, as of December 2014, the European
4
This box was prepared by John Bluedorn.
3 1Plans for banking union began in earnest shortly after the

European Central Banks August 2012 announcement of the


Sovereign selected 2 Outright Monetary Transactions instrument that contained and
Corporate selected
Sovereign other 1 alleviated the turmoil in euro area financial markets.
2The key European Union initiatives include the Single
Corporate other
0 Rulebook, to establish a common bank capital definition and
2007 08 09 10 11 12 13 14 15 implement Basel III prudential requirements (adopted in June
2013; phased in by 2019); the Bank Recovery and Resolution
Sources: Bloomberg, L.P.; and Haver Analytics.
Directive, establishing common practices for bank resolution
Note: Sovereign rates are the yields on ve-year bonds.
Corporate lending rates are for bank loans longer than ve years. at the national level, which minimizes taxpayer support for
The rates for Belgium and Portugal reect all maturities. banks, partly through the bail-in of bank creditors in resolu-
Selected countries are those which experienced high tion (adopted April 2014; in full force January 2016); and the
borrowing spreads during the 201011 debt sovereign crisis Deposit Guarantee Scheme Directive, harmonizing rules for
and comprise Ireland, Italy, Portugal, and Spain. Other national deposit guarantee schemes and ensuring their upfront
countries are Belgium, France, Germany, and the Netherlands. funding and uniform functioning (adopted April 2014; phased
in by 2025).

78 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Box 2.3 (continued)


Stability Mechanism may directly recapitalize banks euro area banks; enforce a high, common supervisory
under restructuring, acting as a kind of common fiscal standard; enable the cross-border flow of bank liquid-
backstop to the banking union. However, the hurdles ity; and ensure common and consistent treatment of
for its use are very high (for example, bail-in must be investors and depositors in cases of bank distress. This
exhausted), and the funding available is capped at 60 centralization should help foster the single market and
billion, which could be rapidly depleted in a systemic reduce fragmentation. However, a number of the prac-
crisis. ticalities and modalities still need to be worked out for
By centralizing and sharing bank supervision and the new institutions. Moreover, without an effective
resolution, the banking union will eliminate the common fiscal backstop, the risk that bank-sovereign
distinction between home and host supervisors for linkages could reemerge in a systemic crisis remains.

Box 2.4. Global Banks: Regulatory and Supervisory Areas in Need of Attention
This box highlights areas that warrant further attention impediments to cross-border cooperation among
from policymakers to make regulation and supervision of supervisory authorities, thus enabling them to share
globally active banks more effective. information effectively.3
Establish a dedicated framework for reforms with
Cooperation and coordination a cross-border reach: The unilateral adoption of
A pragmatic approach is needed to tackle the challenges measures without international agreement can
global banking poses to national policymakers. Mutually encourage other countries to take similar unilateral
shared objectives as well as a stronger cooperation and measures, leading to a spiral of regulatory fragmenta-
coordination process among regulators and supervisors tion. Financial stability might be compromised if
are paramount. national approaches, introduced in the absence of
Build trust through strengthened cooperation and coor- an international standard, confront global banks
dination: The international response to the financial with competing or contradictory requirements. In
crisis has markedly improved the regulatory frame- the long term, countries should consider moving
work. However, more attention could be devoted to toward an international system for mutual consulta-
strengthening supervision (Vials and others 2010). tion of reform proposals with considerable cross-
Building and maintaining trust among supervisors border reach. While retaining national autonomy for
is essential for effective cooperation among more safeguarding financial stability, such a process could
integrated countries, especially during times of ensure broader application of substituted compliance
crisis. Confidence-building measures include the with foreign regulatory regimes and internalize the
signing of memoranda of understanding (MoUs) or effects of extraterritorial measures.4
active participation in regional networks.1,2 In gen-
eral, policymakers should strive to remove any legal Consistency
The details of the implementation and application of
reforms deserve more attention. Inconsistent implemen-
This box was prepared by Johannes Ehrentraud.
1MoUs establish a set of details for cooperation and informa- tation of international standards across countries may
tion exchange with other supervisory authorities. Although such
agreements failed to facilitate cooperation during the global 3In some countries, banking secrecy laws prevent authorities

financial crisis, their format could be revamped to include from sharing information with others if their counterpartys
specific timelines and escalation procedures (IMF 2014a). For legal system provides the option of sharing the data with tax
systemically important institutions, the Financial Stability Board authorities.
(FSB) recommends setting up crisis management groups and 4Substituted compliance describes the circumstances in which

institution-specific cross-border cooperation arrangements (FSB authorities permit legal subjects to use compliance with regula-
2014). tions in another jurisdiction as a substitute for compliance with
2Examples include the Group of Banking Supervisors from local regulations. Deferring to the regulatory regimes of other
Central and Eastern Europe or the Association of Supervisors of countries often involves the determination of equivalence of the
Banks of the Americas. other countries regulatory regimes.

International Monetary Fund | April 2015 79


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 2.4 (continued)


cause global banks to book their transactions in jurisdic- be required to hold different loan loss reserves for a
tions with light-touch regulation or more preferential given level of loan portfolio riskiness.
accounting rules.
Basel framework: In 2012, the Basel Committee Resolution and organizational banking structures
on Banking Supervision established a Regulatory Effective cross-border resolution regimes would allow
Consistency Assessment Program to facilitate consis- for more flexibility in the choice of legal structures for
tency in the adoption and implementation of Basel banking groups.
standards.5 Current challenges to ensuring a level Advancing cross-border bank resolution: The Key
playing field include different phase-in requirements Attributes, which are the international standard for
and transitional adjustments in banks regulatory resolution regimes for financial institutions, are to be
capital calculations, and excessive variability in the implemented in Financial Stability Board member
calculation of risk-weighted assets in banks using jurisdictions by end-2015. They provide resolution
an internal-ratings-based approach. In Europe, authorities with comprehensive resolution pow-
the Capital Requirements Regulation and Capital ers. However, a number of considerable challenges
Requirements Directive include a large number remain. In some cases, there may be significant asym-
of options allowing for national discretion in the metry of power in interactions between home and
application of certain regulatory rules (Lauten- smaller host countries where the operations are not
schlger 2014). Further efforts are thus required to material to the institutions overall health. Moreover,
ensure that national discretion does not undermine national interests may still trump incentives for coop-
the consistency of agreed-upon reforms.6 erative cross-border strategies. More work is needed
Accounting: Although commissioned by the Group of on proposals for total loss-absorbing capacity, greater
20 countries in 2009, convergence efforts by the Inter- harmonization of creditor hierarchies, and depositor
national Accounting Standards Board and the U.S. preference between countries (IMF 2014a).7
Financial Accounting Standards Board have not yet Legal banking structures: Given a cooperative inter-
produced a single set of global standards. For banks, national environment, banking groups that find it
one key area of divergence is the standards for credit more useful to be organized either as branches or as
loss provisioning. Diverging accounting approaches subsidiaries can be consistent with financial stability
are costly for compliance and hamper comparability in outcomes.8 In some situations, however, imposing
loan loss estimates. They also create an uneven playing subsidiarization might seem preferable from a financial
field because banks in different parts of the world will stability perspective but has efficiency costs for banks
that would otherwise prefer to organize themselves
through a branch structure in light of their business
5Main elements of the Regulatory Consistency Assessment model. Harmonizing cross-border resolution regimes
Program are the implementation and monitoring of Basel stan- and burden-sharing agreements, along with effective
dards and consistency assessments carried out on a jurisdictional cooperation and information sharing in crisis times,
and thematic basis.
6In 2010, the FSB established a framework for encouraging may change authorities current preference for certain
stronger adherence to international standards. The three main structures with regard to financial stability.
elements are (1) FSB members commitment to implement stan-
dards and publish evidence of their adherence, (2) periodic peer
reviews for FSB and non-FSB members, and (3) a toolbox with 7In November 2014, the FSB issued a consultation paper on a

positive and negative measures, including identification of non- common international standard on total loss-absorbing capacity
cooperative jurisdictions (FSB 2010a, 2010b). This framework for global systemic banks.
could be strengthened. 8See Fiechter and others (2011) for an exhaustive discussion.

80 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Annex 2.1. Regression Analysis of the Drivers of each for home and host countries, as shown in Annex
the Decline in Foreign Banking Claims24 Table 2.1.1. Each country-category pair is assigned a
value of 1, 0, or 1 when the number of answers report-
This annex describes the data and the regression model
ing a tightening of regulations is greater than, equal to,
used to examine the drivers of the decline in foreign
or smaller than, respectively, the number of answers
claims and provides more detailed results. Annex Table
reporting a loosening. The final index is calculated as a
2.1.1 lists the questions used to construct the models
simple average of the scores for the six categories.
regulatory index. Annex Table 2.1.2 provides a sum-
mary of data definitions and sources, and Annex Table
2.1.3 gives the coefficient estimates.25 Regression Model
The regression model takes the following form:
Data on Foreign Banking Claims and the Regulatory Dclaimsij = a + b homei + g hostj
Index
+ d bilateralij + eij ,
The dependent variable is the growth rate of foreign
banking claims from a home country to a host coun- in which Dclaimsij denotes the growth rate of claims from
try. In addition to total foreign claims, subcategories home country i to host country j. The terms homei and
by type of claim and counterparty sector are also used. hostj are vectors of variables specific to home and host
The data come from the BIS Consolidated Banking countries, respectively. Each of these vectors includes three
Statistics on an ultimate risk basis.26 Statistical breaks indices of regulatory changes (one based on the survey
are adjusted following Cerutti (2013). Quarterly claims results and two based on World Bank data on capital
over the period 2005:Q22013:Q3 are annualized and requirements and supervisory power; see Annex Table
averaged over the precrisis (200507) and postcrisis 2.1.3), the change in the exchange rate against the U.S.
periods (201113). The growth rate is computed by dollar,28 the GDP growth rate, and the real policy interest
dividing the change in claims between the two periods rate. In addition, homei includes an indicator of banking
by the average level in the two periods.27 sector health in the precrisis period. The term bilateralij
The main explanatory variables of interest are the is a vector of bilateral variables, comprising the log of the
indices of changes in regulations on banks international physical distance between the home and host countries,
operations in home and host countries, based on the a common language dummy, two variables capturing the
results of a survey conducted for the purpose of this importance of the claims from the home country in the
chapter. Survey questions are classified into six categories host country and of the claims in a given host country
from the home country perspective in the precrisis period,
24The author of this annex is Hibiki Ichiue.
25For
and the growth rate of bilateral international claims in
more details, see Ichiue and Lambert (forthcoming).
26The BIS Consolidated Banking Statistics record the consolidated the precrisis period.29 The coefficients a, b, g, and d are
positions of reporting banks worldwide offices, excluding interoffice parameters or vectors of parameters, and eij is the residual.
positions. They comprise two subsets, compiled on different bases: The results reported in the text are broadly robust to
an immediate risk basis and an ultimate risk basis. The immediate
risk basis data allocate banking claims to the country of residence
the following specification changes: First, the indices on
of the immediate counterparty; the ultimate risk basis data allocate changes in capital requirements and supervisory power
claims to the country in which the final risk lies. The immediate (computed from World Bank data) are excluded, which
risk basis data offer better coverage of time series and countries.
is an important robustness check given that the indi-
In addition, they distinguish between international claims (sum of
cross-border claims and local claims in foreign currency) and local ces are not available for some BIS reporting countries,
claims in local currency, whereas the ultimate risk basis data provide including Japan and the United Kingdom. Second, real
a breakdown between cross-border claims and total local claims (sum long-term interest rates in home and host countries are
of local claims in both foreign and local currencies). The immediate
risk basis data, however, do not reflect risk transfers and have limita- used instead of real policy interest rates to control for
tions in capturing banks bilateral risk exposures. These issues are unconventional monetary policy effects. Third, euro
irrelevant when immediate risk basis data are aggregated by country
of origin. The analysis described in this annex uses bilateral claims 28The BIS Consolidated Banking Statistics are reported in U.S.

and thus relies on ultimate risk basis data. dollars by converting claims in other currencies. Changes in claims
27The literature often uses log differences to calculate growth rates. from one period to another may then only reflect valuation effects
However, such a method naturally discards data when claims are zero following exchange rate fluctuations with the actual underlying posi-
at the start or end of the period and cannot capture home countries tion remaining unchanged (Cerutti 2013).
entry into or exit from host countries, which may actually result 29Precrisis values of the variables of bank health and bilateral

from changes in regulations or other factors. importance are used to mitigate endogeneity concerns.

International Monetary Fund | April 2015 81


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

area countries are either excluded from the sample or of instruments is justified by the possibility of regulation
aggregated and treated as a single country. Fourth, the contagion as discussed in Demirg-Kunt and Detragia-
International Country Risk Guide country risk rating is che (2002) and Houston, Lin, and Ma (2012).
added to the variables for host countries. Fifth, home Annex Table 2.1.3 reports the detailed results for
countries sovereign rating index or a banking crisis different types of banking claims. The model is also
dummy is added to the regression. Finally, the indices estimated using the difference between the growth rates
for the changes in regulations in home countries are of different types of claims as the dependent variable.
instrumented by the capital regulation index and super- Significant nonzero coefficients confirm that two differ-
visory power index from the World Bank in 2003 and ent types of claims have different sensitivities to some of
2006, to deal with possible endogeneity bias. The choice the explanatory variables. These results are not reported.

Annex Table 2.1.1. Survey on the Regulation of Banks International Operations


Category Questions
Home Country Regulations
Presence Are domestic banks prohibited from acquiring foreign banks?
Do domestic banks need their domestic supervisors approval to acquire a foreign bank?
Are domestic banks prohibited from establishing branches overseas?
Do domestic banks need their domestic supervisors approval to establish a branch overseas?
Are domestic banks prohibited from establishing subsidiaries overseas?
Do domestic banks need their domestic supervisors approval to establish a subsidiary overseas?
Are the requirements to obtain permission to establish a branch stricter than those applicable to subsidiaries?
Activity Are domestic banks prohibited from making cross-border loans?
Are domestic banks prohibited from purchasing foreign securities?
Are there restrictions on the type of activities (for example, corporate and retail lending, residential mortgage,
trade finance, long-term infrastructure finance, investment banking) that domestic banks can conduct overseas
that do not apply to domestic operations?
Are there additional regulatory requirements for domestic banks operating outside their home country beyond
what would be required for similar operations conducted domestically?
Depositor Insurance Are foreign depositors covered by deposit insurance?
Information Do banking secrecy laws in your country limit your ability to share information about banks operations and
balance sheets with foreign supervisors?
Supervisory Discretion Can the supervisor limit the range of activities a consolidated group may conduct and/or the locations in which
activities can be conducted (including the closing of foreign offices) in specific circumstances (as per Basel
Core Principle 12.6)?
Other Did the authorities introduce other structural measures (such as Volcker reform, Vickers proposals, and others)
that could weigh on the decision of some banks to expand internationally?
Host Country Regulations
Presence Is foreign ownership of domestically incorporated banks prohibited?
Do foreign banks need the host country supervisors authorization to acquire a domestic bank?
What is the maximum percentage of foreign ownership of a domestic bank legally allowed?
Are foreign banks prohibited from operating in the form of branches?
Are the requirements for establishing a branch stricter for foreign banks than for domestic banks?
Are there additional and/or different regulatory requirements for foreign-owned banks versus domestic banks?
Activity Are there restrictions on the type of activities (for example, corporate and retail lending, residential mortgage,
trade finance, long-term infrastructure finance, investment banking) that foreign banks can conduct
domestically and that do not apply to domestic banks?
Are there restrictions on domestic currency cross-border borrowing by banks?
Are there restrictions on foreign currency cross-border borrowing by banks?
Are banks required to fund part or all of their domestic operations with local deposits?
Are there restrictions on the share of funding a domestically incorporated bank can obtain from a foreign parent?
Are there restrictions on lending by domestically incorporated banks to a foreign parent?
Supervisory Discretion Can the supervisory authorities impose ring-fencing measures in a discretionary way?
Information Do banking secrecy laws in your country limit your ability to share information about banks operations and
balance sheets with foreign supervisors?
Resolution Does the resolution authority have resolution powers over local branches of foreign firms and the capacity to use
its powers either to support a resolution carried out by a foreign home authority or, in exceptional cases, to
take measures on its own initiative (as per Key Attribute 7.3)?
Other Did the authorities introduce other structural measures (such as Volcker reform, Vickers proposals, and others)
that could weigh on the decision of some banks to retrench from your country?
Source: IMF staff.

82 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Annex Table 2.1.2. Definition of the Variables


Variable Description Source
Claims The dependent variable is the growth rate of bilateral claims from the BIS
precrisis period (200507) to the postcrisis period (201113), which
is calculated from the change in average claims between the pre- and
postcrisis periods. The precrisis growth rate of bilateral international
claims, computed between 200204 and 200507, is used as a control
variable.
International Operations An index constructed from answers to survey questions about regulation IMF
Regulatory Index changes for 200614. See the text of this annex for more detail.
Capital Regulatory Index Difference between Barth, Caprio, and Levine (2013) indexes in 2006 and Barth, Caprio, and Levine
2011. (2013)
Official Supervisory Power Index Difference between Barth, Caprio, and Levine (2013) indexes in 2006 and Barth, Caprio, and Levine
2011. (2013)
Exchange Rate Change in the exchange rate against the U.S. dollar between 200507 and IMF, IFS
201113.
GDP Growth rate from 200507 to 201113. IMF, WEO
Real Policy Interest Rate Change in the policy rate (or an alternative interest rate if not available) minus Central banks, Consensus
the one-year-ahead expected inflation rate between 200507 and 201113. Forecasts
Bank-Capital-to-Total-Assets Average of the ratio in 2005, 2006, and 2007. World Bank
Ratio
Distance Log distance between two cities, mostly capitals, in home and host countries. http://privatewww.essex
The distance to Hong Kong SAR is proxied by the distance to Taiwan .ac.uk/~ksg/data-5.html
Province of China.
Common Language Dummy The variable is equal to 1 when the home and host countries use a common Rose (2004)
language and zero otherwise.
Importance of Host in the Claims Ratio of bilateral claims from a home country to a host country to total BIS
from Home claims from the home country to all host countries, averaged over 2005,
2006, and 2007.
Importance of Home in the Ratio of bilateral claims from a home country to a host country to total BIS
Claims on Host claims from all home countries to the host country, averaged over 2005,
2006, and 2007.
Source: IMF staff.
Note: BIS = Bank for International Settlements; IFS = International Financial Statistics; WEO = World Economic Outlook.

International Monetary Fund | April 2015 83


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Annex Table 2.1.3. Results of Country-Level Regression for the Drivers of the Changes in Foreign Banking Claims
By Instrument By Sector
Foreign Claims Cross Border Local Nonbank Banks Public
Regulatory Index (changes)
International Operations (home) 179.60*** 136.95* 131.74 184.27** 20.88 249.65
International Operations (host) 41.62** 42.73** 9.26 28.17 42.23* 6.91
Capital (home) 7.09*** 2.02 3.96 6.67*** 0.43 4.14
Capital (host) 0.66 0.97 1.50 2.52 2.01 7.47***
Supervisory Power (home) 3.88*** 3.89*** 1.73 2.23* 1.24 10.17***
Supervisory Power (host) 1.08 1.96 3.51 2.10 0.93 0.02
Exchange Rates (percent appreciation against US$)
Home 2.89*** 3.01*** 7.23*** 0.01 4.21*** 10.26**
Host 0.07 0.20 1.28** 0.07 0.25 0.18
GDP (percent change)
Home 0.39 0.44 7.07*** 0.15 8.87*** 1.82***
Host 0.88*** 0.93*** 1.22*** 1.24*** 0.12 0.65***
Real Policy Interest Rate (percentage point changes)
Home 1.54 0.68 55.21*** 7.60* 61.62** 7.77
Host 5.00*** 6.27*** 2.58 8.45*** 5.71** 1.27
Bank-Capital-to-Total-Assets Ratio (percent in
200507)
Home 10.50*** 12.07*** 18.52*** 13.82*** 12.51** 8.82***
Bilateral Geographic and Cultural Variables
Distance (log, km) 11.72*** 10.78** 33.19*** 14.08*** 1.76 9.26*
Common Language Dummy 3.50 3.73 3.66 13.60 15.01 1.40
Bilateral Share (percent in 200507)
Host Countrys Share of Claims from Home 0.77 0.82 1.31 1.00 2.65** 1.11*
Home Countrys Share of Claims on Host 0.86** 0.10 0.63 0.17 1.40** 0.05
Bilateral Lagged Claims (percent changes from
200204 to 200507)
International Claims 0.18*** 0.17** 0.36** 0.19** 0.27** 0.03
Number of Observations 518 433 328 424 352 417
R2 0.27 0.27 0.22 0.28 0.26 0.19
Source: IMF staff estimates.
Note: km = kilometer; Whites (1980) robust standard errors are used. *p < 0.10; **p < 0.05; ***p < 0.01.

84 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Annex 2.2. Analysis of the Role of International global financial crisis (200809) and a dummy for
Banking Linkages in Mitigating or Amplifying domestic banking crises (Laeven and Valencia 2013).
Shocks30 The econometric specification is as follows:
This annex summarizes the analysis of the role played Dbankcrediti,t = ai + b1 Xi,t1 + b2 domestic shocki,t
by global banks in mitigating or amplifying domestic
+ b3 global shockt + b4 IBLi,t1
and global shocks. The analysis uses panel data tech-
niques on country-level and bank-level data to estimate + g1 IBLi,t1 domestic shocki,t
the impact of international banking linkages on credit
growth. + g2 IBLi,t1 global shocki,t

Country-level analysis + b5 domestic crisisi,t

International banking linkages are measured in three + b6 global crisist + g3 IBLi,t1


ways, by (1) the ratio of cross-border claims to the
domestic shocki,t domestic crisist
total assets of the banking sector, (2) the ratio of
international claims to total banking assets, and (3) + g4 IBLi,t1 foreign shocki,t
the ratio of foreign subsidiaries and branches local
claims in local currency to total banking assets. The global crisist + ei,t ,
second measure includes foreign currency domestic in which Dbankcrediti,t is the quarterly growth in bank
claims of foreign bank affiliates whereas the first claims to the private sector available from the IMF
one focuses exclusively on cross-border claims.31 International Financial Statistics; ai and Xi,t1 capture
All of these variables are available from the BIS country-level effects with country fixed effects and the
and adjusted for statistical breaks following Cerutti real GDP growth rate; domestic shocki,t and global shockt
(2013). Other measures, such as the ratio of foreign are measured by the EDF of the banking sector and
claims to the nonfinancial sector to total domestic the VIX, respectively; and IBLi,t1 is the measure of
credit to the nonfinancial sector, are used for robust- international banking linkages. The main coefficients of
ness checks. interest are the coefficients that capture the interaction
Global (foreign) stress is measured by the VIX. between the level of international banking linkages and
Results are similar when an alternative measure is used the sensitivity of credit to domestic and foreign shocks.
(such as average credit default swap (CDS) prices of The baseline model is supplemented by the inclusion of
the global systemically important banks identified by dummies for domestic and global crises (domestic crisisi,t
the Financial Stability Board). Domestic stress is mea- and global crisist) and their interactions.
sured by the average expected default frequency (EDF) Annex Table 2.2.1 summarizes the results from the
of the domestic banking sector (weighted by the size panel regressions. Driscoll-Kraay standard errors are
of the domestic banks). The EDF is used instead of used to account for the potential heteroskedasticity
CDS prices because the former has much better data and autocorrelation of standard errors. The results are
coverageCDS data are only available for the largest robust to adding one lag of the dependent variable on
banks. Since the EDF can be contaminated by global the right-hand side to account for the persistence of
stress, a measure of domestic stress purged of the effect credit growth or the possibility of boom-bust cycles,
of global stress (residual of a regression of the EDF on and to including additional country-level control
the VIX) is used as a robustness check. The average variables. They also hold for subsamples of advanced
EDF for all listed firms, a broader measure of domestic economies and emerging markets and when the Euro-
shock, is also considered. The results are unchanged. pean countries are excluded from the sample. Finally,
Alternative specifications include a dummy for the the results are robust to the exclusion of Vienna Initia-
tive countries.
The above analysis is from the perspective of coun-
30The authors of this annex are Pragyan Deb and Kai Yan. tries that are host to foreign banks. Annex Table 2.2.2
31To be precise, the first measure is not exactly a subset of the sec- summarizes the results of the panel regressions from the
ond measure because cross-border claims are reported on an ultimate
risk basis whereas international claims are compiled on an immediate perspective of the home country of international banks.
risk basis. See Annex 2.1. For this specification, international banking linkages are

International Monetary Fund | April 2015 85


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

measured by the ratio of nondomestic claims of banks lished subsidiaries operating for only a few months in
domiciled in the country to the total domestic banking their year of incorporation and represent fewer than 3
sector assets of the country. International banking link- percent of the total number of observations.
ages are measured in two ways: (1) ratio of cross-border The econometric specification is the following:
claims to domestic banking assets and (2) ratio of inter-
national claims (including both cross-border claims and Dloani,j,k,t = a Xi,t1 + r foreigni + b bankcrisisk,t
local claims of affiliates in foreign currency) to domestic
+ q bankcrisisk,t foreigni + d bankcrisisk,t
banking assets. Local claims in local currency are less rel-
evant from a home country perspective and are therefore Xi,t1 + g bankcrisisk,t Xi,t1 foreigni
not considered in this analysis.
+ controlsj,k,t + ei,j,k,t ,

in which foreigni is a dummy variable equal to 1 if


Bank-Level Analysis of the Stabilization Role of Foreign the bank is owned by a foreign bank. The variable
Banks bankcrisisk,t is now a dummy variable equal to 1 if the
The analysis uses balance sheet data for a panel of host country of the bank is having a banking crisis. In
banks during the period 19982013. The data set con- some specifications, bankcrisisk,t is replaced by a global
tains 25,568 domestic- or foreign-owned subsidiaries financial crisis dummy, which equals 1 during the global
over 15 years, though the number of active banks for financial crisis (200809). The term Xi,t1 still denotes
which balance sheet data are available is much smaller the bank-level characteristics of interest. We subtract the
and varies from year to year. mean of Xi,t1 from Xi,t1 to facilitate the interpretation of
The data set is constructed in two steps. First, the results. The two-way interaction terms can therefore
subsidiary banks are matched with their parent banks be interpreted as the marginal impact of being in the
using ownership data from 2007 to 2013 from Bank- treatment group (when the dummy is equal to 1) when
scopes ownership database, which is extended back to the banks characteristics are that of an average bank.
1998 (Porter and Serra 2011). The data set includes The coefficients r, q, and g are the focus of the anal-
commercial banks, savings banks, cooperative banks, ysis. A statistically significant r suggests that the lend-
and bank holding companies. Adjustments are made ing behavior of foreign-owned subsidiaries differs on
to correct for missing or incorrectly identified parents, average from that of domestic banks. The coefficient q
when possible. Independent banks or banks with no measures the stabilization role played by foreign-owned
parent are considered to be their own parent. Second, subsidiaries during banking crises. The coefficient g
bank parents and subsidiaries financial statement data measures the way in which different characteristics of
since 1998 are obtained from Bankscope. Balance sheet the parent bank or subsidiaries affect foreign sub-
data are annual, as of year-end, and on a consolidated sidiaries credit growth during crises. A negative and
basis. Unconsolidated balance sheet data are used to significant g suggests that foreign-owned subsidiaries
control for subsidiaries characteristics. Country-level of a banking group with certain characteristics are less
data are the same as used in the macro-level analysis. likely to support credit growth during financial crises.
Observations that show an annual growth rate of The model is estimated with a standard fixed effects
loans of more than 100 percent are dropped. These panel estimation method, with Driscoll-Kraay standard
observations are likely to correspond to newly estab- errors. Annex Table 2.2.3. reports the detailed results.

86 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Annex Table 2.2.1. Credit Growth Panel Regressions from the Perspective of Host Countries of Foreign Banks
International Banking Linkages Measured with
Cross-Border Claims International Claims Local Claims
Real GDP Growth (year-over-year change, lagged) 0.26 0.31 0.35** 0.36** 0.34** 0.35**
Domestic Shock (average EDF) 2.43* 2.38 2.29* 1.19 2.81** 1.6
Global (foreign) Shock (VIX) 12.99** 17.19*** 11.19** 13.35** 12.00** 14.03**
International Banking Linkages (lagged) 2.36*** 2.10*** 1.29** 1.25* 1.47 1.72
IBL Domestic Shock 4.43*** 4.97*** 3.37*** 3.44*** 0.77 0.51
IBL Global Shock 2.26** 0.15 2.34*** 1.48 0.76* 1.14*
Domestic Crisis 1.06 2.27* 2.35*
IBL Domestic Shock Domestic Crisis 12.22* 2.28 2.2
Global Crisis (200809) 1.98 1.78 1.77
IBL Foreign Shock Global Crisis 1.96 0.83 0.24
Number of Observations 1,486 1,486 2,174 2,174 2,135 2,135
Number of Countries 49 49 49 49 49 49
R2 0.12 0.13 0.09 0.10 0.09 0.11
Source: IMF staff estimates.
Note: EDF = expected default frequency; IBL = international banking linkages; VIX = Chicago Board Options Exchange S&P 500 Volatility Index. The dependent variable is
the quarterly growth in bank claims to the private sector. Country fixed effects are included, but not reported. Driscoll-Kraay standard errors are used to take into account
potentially heteroscedastic and autocorrelated standard errors. *p < 0.10; **p < 0.05; ***p < 0.01.

Annex Table 2.2.2. Credit Growth Panel Regressions from the Perspective of Home
Countries of Foreign Banks
International Banking Linkages Measured with
Cross-Border Claims International Claims
Real GDP Growth (year-over-year change, lagged) 0.25 0.17
Domestic Shock (average EDF) 2.64 2.8
Global (foreign) Shock (VIX) 13.99** 15.69**
International Banking Linkages (lagged) 2.86 2.11
IBL Domestic Shock 4.48 0.05
IBL Global Shock 19.49 25.39*
Number of Observations 749 1,250
Number of Countries 23 27
R2 0.12 0.09
Source: IMF staff estimates.
Note: EDF = expected default frequency; IBL = international banking linkages; VIX = Chicago Board Options Exchange S&P
500 Implied Volatility index. The dependent variable is the quarterly growth in bank claims to the private sector. Country fixed
effects are included, but not reported. Driscoll-Kraay standard errors are used to take into account potentially heteroscedastic
and autocorrelated standard errors. *p < 0.10; **p < 0.05.

International Monetary Fund | April 2015 87


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Annex Table 2.2.3. Bank-Level Evidence on Foreign Bank Stabilization Role during Crises
(1) (2) (3) (4) (5) (6) (7)
Loan Growth
Host Country GDP Growth 0.52** 0.41*** 0.43*** 0.58*** 0.35*** 0.34*** 0.54***
Domestic (host country) Banking Crisis 16.46*** 20.02*** 19.97*** 21.12***
Global Crisis 6.92** 10.35*** 11.42*** 13.76***
Foreign Ownership Dummy 4.35*** 0.82 2.36 0.14 6.69* 4.41 3.89**
Domestic Crisis Foreign 7.05*** 3.06 4.00** 2.23*
Global Crisis Foreign 8.59*** 7.22*** 5.54** 4.85***
Parent Equity Ratio 62.95*** 66.45***
Foreign Parent Equity Ratio 38.08 73.72
Domestic Crisis Parent Equity Ratio 30.21*
Domestic Crisis Foreign Parent Equity Ratio 126.05*
Global Crisis Parent Equity Ratio 38.80
Global Crisis Foreign Parent Equity Ratio 143.25***
Parent Dependence on Ext. Funding 0.36 0.61
Foreign Parent Dependence on Ext. Funding 2.02** 3.42***
Domestic Crisis Parent Dependence on Ext. 1.46**
Funding
Domestic Crisis Foreign Parent Dependence 2.61
on Ext. Funding
Global Crisis Parent Dependence on Ext. 3.48***
Funding
Global Crisis Foreign Parent Dependence on 3.07**
Ext. Funding
Subsidiary Local Funding Ratio 21.77*** 11.18
Foreign Subsidiary Local Funding Ratio 8.00** 8.87
Domestic Crisis Subsidiary Local Funding Ratio 16.29***
Domestic Crisis Foreign Subsidiary Local 0.18
Funding Ratio
Global Crisis Subsidiary Local Funding Ratio 26.08***
Global Crisis Foreign Subsidiary Local 9.41
Funding Ratio
Constant 18.35*** 15.84*** 15.90*** 16.28*** 16.12*** 16.50*** 18.37***
Observations 13,167 7,557 7,437 11,022 7,557 7,437 11,022
Number of Banks 2,031 1,491 1,471 1,751 1,491 1,471 1,751
R2 0.14 0.10 0.10 0.13 0.09 0.09 0.14
Source: IMF staff calculations.
Note: Ext. = external. The dependent variable is the annual growth rate of loans by banks. *p < 0.10; **p < 0.05; ***p < 0.01.

88 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

Annex 2.3. Analysis of the Effect of exchange reserves, foreign debt, and the current account
International Banking Linkages on the balance. These variables are obtained from the IMFs
Probability of a Banking Crisis32 International Financial Statistics database. IBLi,t1
measures the level of international banking linkages
This annex summarizes the analysis of the effect of
in country i. The term global shockt captures global
banking linkages on the incidence of banking crises
(foreign) stress measured by the VIX. The Greek letters
using a discrete response model (probit). International
a, , g, and d are parameters or vectors of parameters of
banking linkages are measured as in Annex 2.2.
the explanatory variables and their interactions, and ei,t
The dependent variable, host country banking crisis,
is the residual.
is defined as in Laeven and Valencia (2013). Following
Annex Table 2.3.1 shows the detailed results from
the literature, the crisis variable takes the value 1 in the
the probit regressions. Similar results are obtained using
first year of a crisis, is set to missing for the subsequent
a logistic (or logit) regression model. Although these
two years (as banks are impaired in the aftermath of a
regressions include country-level control variables,
banking crisis), and is zero in the noncrisis years.33 The
they do not include country fixed effects. Whereas the
sample period covers the period 200213 (200513
inclusion of fixed effects biases the results of the probit
when international banking linkages are measured with
regressions but not those of the logit regressions, the
cross-border claims). The probit model takes the fol-
logit specification with fixed effects ignores all countries
lowing form:
that did not have a crisis during the sample period, leav-
P(hostcrisisi,t|X) = F(a Xi,t1 + b IBLi,t1 ing a relatively small and potentially non-representative
sample of countries. Including or substituting the
+ global shockt + d IBLi,t1 measure of global stress with a dummy for the global
financial crisis does not change the results.
global shockt + ei,t), The results are robust to the use of additional
in which Xi,t1 denotes the set of variables used in the explanatory variables such as financial depth (mea-
benchmark specification. Drawing on the crisis predic- sured by credit-to-GDP ratio and a more inclusive
tion literature, Xi,t1 controls for credit growth in the measure developed by IMF [forthcoming c]), govern-
run-up to the crisis, real GDP growth rate, foreign ment primary deficit, inflation, real effective exchange
rate misalignment, and other country-level controls
32The author of this annex is Pragyan Deb. for governance and supervisory powers. In addition,
33Minoiu and others (forthcoming) and Gourinchas and Obstfeld alternate definitions of crises, derived from episodes of
(2012) drop four years of observations after the crisis. This chapter
uses only two years to account for quarterly frequency of the data slowdown in GDP growth rates and domestic credit,
and the shorter period under consideration. yielded similar results.

Annex Table 2.3.1. Detailed Probit Regression Results


International Banking Linkages Measured with
Cross-Border Claims International Claims Local Claims
Real GDP Growth (year-over-year change, lagged) 0.03 0.03 0.05* 0.01 0.05* 0.01
Credit Growth (lagged) 0.08*** 0.05*** 0.06*** 0.04** 0.06*** 0.04**
Foreign-Exchange-Reserves-to-GDP Ratio (lagged) 2.59 4.89* 1.96 3.02 1.81 2.94
Foreign-Debt-to-GDP Ratio (lagged) 0.39** 0.36* 0.48*** 0.48*** 0.43*** 0.42**
Current-Account-Balance-to-GDP Ratio (lagged) 0.14*** 0.15*** 0.16*** 0.17*** 0.15*** 0.17***
International Banking Linkages (lagged) 0.16 0.29 0.19 0.31 0.14 0.69
Global (foreign) Shock (VIX) 7.26*** 6.36*** 5.78***
IBL Global Shock 0.86 0.82 6.57
Observations 1,324 1,284 1,840 1,800 1,792 1,753
Number of Countries 46 46 46 46 45 45
Chi-squared 41.78 44.60 47.51 59.72 46.51 62.54
Source: IMF staff estimates.
Note: IBL = international banking linkages; VIX = Chicago Board Options Exchange S&P 500 Implied Volatility Index. Banking crises are defined as in Laeven and
Valencia (2013). The estimates are derived from a random effects panel probit model. *p < 0.10; **p < 0.05; ***p < 0.01.

International Monetary Fund | April 2015 89


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

References Know about the Drivers and Consequences of This Phenom-


enon? Policy Research Working Paper 5398, World Bank,
Allen, Franklin, Thorsten Beck, Elena Carletti, Philip Lane, Dirk Washington.
Schoenmaker, and Wolf Wagner. 2011. Cross-Border Banking De Haas, Ralph, and Iman van Lelyveld. 2006. Foreign Banks and
in Europe: Implications for Financial Stability and Macroeco- Credit Stability in Central and Eastern Europe. A Panel Data
nomic Policies. London: Center for Economic and Policy Analysis. Journal of Banking and Finance 30 (7): 1927952.
Research. . 2010. Internal Capital Markets and Lending by Mul-
Bailey, Natalia, Sean Holly, and Hashem Pesaran. Forthcom- tinational Bank Subsidiaries. Journal of Financial Intermedia-
ing. A Two Stage Approach to Spatiotemporal Analysis with tion 19 (1): 125.
Strong and Weak Cross-Sectional Dependence. Journal of Demirg-Kunt, Asli, and Enrica Detragiache. 2002. Does Deposit
Applied Econometrics. Insurance Increase Banking System Stability? An Empirical
Barth, James R., Gerard Caprio, and Ross Levine. 2013. Bank Investigation. Journal of Monetary Economics 49 (7): 1373406.
Regulation and Supervision in 180 Countries from 1999 Demirg-Kunt, Asli, Roos Levine, and Hong-Ghi Min. 1998.
to 2011. Working Paper No. 18733, National Bureau of Opening to Foreign Banks: Issues of Stability, Efficiency, and
Economic Research, Cambridge, Massachusetts. Growth. In The Implications of Globalization of World Finan-
Berkmen, Pelin, Gaston Gelos, Robert Rennhack, and James cial Markets, edited by Seongtae Lee. Seoul: Bank of Korea.
Walsh. 2012. The Global Financial Crisis: Explaining Detragiache, Enrica, Thierry Tressel, and Poonam Gupta. 2008.
Cross-Country Differences in the Output Impact. Journal of Foreign Banks in Poor Countries: Theory and Evidence.
International Money and Finance 31 (1): 4259. Journal of Finance 63 (5): 212360.
Bernanke, Ben S. 2005. The Global Saving Glut and the U.S. Fiechter, Jonathan, nci tker-Robe, Anna Ilyina, Michael Hsu,
Current Account Deficit. Remarks at the Sandridge Lecture, Andr Santos, and Jay Surti. 2011. Subsidiaries or Branches:
Virginia Association of Economists, Richmond, Virginia, Does One Size Fit All? Staff Discussion Note No. 11/04,
March 10. International Monetary Fund, Washington.
Bremus, Franziska, and Marcel Fratzscher. 2014. Drivers of Figuet, Jean-Marc, Thomas Humblot, and Delphine Lahet.
Structural Change in Cross-Border Banking since the Global 2015. Cross-Border Banking Claims on Emerging Coun-
Financial Crisis. Discussion Papers of DIW Berlin 1411, tries: The Basel III Banking Reforms in a Push and Pull
German Institute for Economic Research, Berlin. Framework. Journal of International Financial Markets,
Bruno, Valentina, and Hyun Song Shin. Forthcoming. Cross- Institutions, and Money 34: 294310.
Border Banking and Global Liquidity. Review of Economic Financial Stability Board (FSB). 2010a. FSB Framework for
Studies. Strengthening Adherence to International Standards. Finan-
Cerutti, Eugenio. 2013. Banks Foreign Credit Exposures and cial Stability Board, Basel.
Borrowers Rollover Risks: Measurement, Evolution, and . 2010b. Promoting Global Adherence to International
Determinants. Working Paper No. 13/9, International Mon- Cooperation and Information Exchange Standards. Financial
etary Fund, Washington. Stability Board, Basel.
Cerutti, Eugenio, Stijn Claessens, and Patrick McGuire. (2012). . 2014. Structural Banking Reforms. Financial Stability
Systemic Risks in Global Banking: What Can Available Data Board, Basel.
Tell Us and What More Data Are Needed? BIS Working Forbes, Kristin. 2014. Financial Deglobalization? Capital
Papers No. 376, Monetary and Economic Department. Bank Flows, Banks, and the Beatles. Speech at Queen Mary Uni-
for International Settlements, Basel, Switzerland. versity, London, November 18.
Cetorelli, Nicolas, and Linda S. Goldberg. 2011. Global Banks Giannetti, Mariassunta, and Luc Laeven. 2012. The Flight Home
and International Shock Transmission: Evidence from the Effect: Evidence from the Syndicated Loan Market during
Crisis. IMF Economic Review 59 (1): 4176. Financial Crises. Journal of Financial Economics 104 (1): 2343.
. 2012. Banking, Globalization and Monetary Transmis- Gilchrist, Simon, Jae W. Sim, and Egon Zakrajek. 2013.
sion. Journal of Finance 67 (5): 181143. Misallocation and Financial Market Frictions: Some Direct
Claessens, Stijn, and Luc Laeven. 2004. What Drives Bank Evidence from the Dispersion in Borrowing Costs. Review of
Competition? Some International Evidence. Journal of Economic Dynamics 16 (1): 15976.
Money, Credit, and Banking 36 (3): 56383. Goldberg, Linda. 2009. Understanding Banking Sector Global-
Claessens, Stijn, and Neeltje van Horen. 2014. The Impact ization. IMF Staff Papers 56 (1): 17197.
of the Global Financial Crisis on Banking Globalization. Gourinchas, Pierre-Olivier, and Maurice Obstfeld. 2012. Stories
Working Paper No. 14/197, International Monetary Fund, of the Twentieth Century for the Twenty-First. American
Washington. Economic Journal: Macroeconomics 4 (1): 22665.
Cull, Robert, and Mara Soledad Martnez Pera. 2010. Foreign Goyal, Rishi, Petya Koeva Brooks, Mahmood Pradhan, Thierry
Bank Participation in Developing Countries. What Do We Tressel, Giovanni DellAriccia, Ross Leckow, Ceyla Pazarba-

90 International Monetary Fund | April 2015


CHAPTER 2 INTERNATIONAL BANKING AFTER THE CRISIS: INCREASINGLY LOCAL AND SAFER?

sioglu, and an IMF Staff Team. 2013. A Banking Union to-Day Supervision. Speech at the Euro Finance Week,
for the Euro Area. IMF Staff Discussion Note No. 13/01, Frankfurt, November.
International Monetary Fund, Washington. McCauley, Robert, Patrick McGuire, and Gtz von Peter. 2012.
Hoggarth, Glenn, John Hooley, and Yevgeniya Korniyenko. After the Global Financial Crisis: From International to
2013. Which Way Do Foreign Branches Sway? Evidence Multinational Banking? Journal of Economics and Business 64
from the Recent UK Domestic Credit Cycle. Financial (1): 723.
Stability Paper 22, Bank of England, London. McGuire, Patrick, and Gtz von Peter. 2009. The US Dollar
Houston, Joel, Chen Lin, and Yue Ma. 2012. Regulatory Shortage in Global Banking and the International Policy
Arbitrage and International Bank Flows. Journal of Finance Response. BIS Working Paper 291, Bank for International
67 (5): 184595. Settlements, Basel.
Ichiue, Hibiki, and Frederic Lambert. Forthcoming. The Minoiu, Camelia, Chanhyun Kang, V. S. Subrahmanian, and
Decline in Foreign Banking Claims: An Analysis with New Anamaria Berea. Forthcoming. Does Financial Connected-
Regulatory Survey Data. IMF Working Paper, International ness Predict Crises? Quantitative Finance.
Monetary Fund, Washington. Ongena, Steven, Alexander Popov, and Gregory F. Udell. 2013.
International Monetary Fund (IMF). 2013a. Central, Eastern, When the Cats Away the Mice Will Play: Does Regula-
and Southeastern EuropeRegional Economic Issues, Washing- tion at Home Affect Bank Risk-Taking Abroad? Journal of
ton, April. Financial Economics 108 (3): 72750.
. 2013b. Key Aspects of Macroprudential Policy. IMF Peek, Joe, and Eric S. Rosengren. 2000. Implications of the
Policy Paper, International Monetary Fund, Washington, Globalization of the Banking Sector: The Latin American
June. Experience. Federal Reserve Bank of Boston Conference Series
. 2014a. Cross-Border Bank Resolution: Recent Devel- 44 (June): 14585.
opments. IMF Policy Paper, Washington, June 2. http:// Porter, Nathan, and Cesar Serra. 2011. Evidence from Cross-
www.imf.org/external/np/pp/eng/2014/060214.pdf. Border Banking Groups. In Mapping Cross-Border Finan-
. 2014b. The Financing of Infrastructure in Sub-Saharan cial Linkages: A Supporting Case for Global Financial Safety
Africa: A Changing Landscape. Chapter 3, Regional Economic Nets. IMF Strategy, Policy, and Review Department (June
Outlook: Sub-Saharan Africa, Washington, October. 1), International Monetary Fund, Washington.
. 2014c. Global LiquidityIssues for Surveillance. Reinhardt, Dennis, and Steven J. Riddiough. 2014. The Two
IMF Policy Paper, International Monetary Fund, Wash- Faces of Cross-Border Banking Flows: An Investigation into
ington, March 11. http://www.imf.org/external/np/pp/ the Links between Global Risk, Arms-Length Funding, and
eng/2014/031114.pdf. Internal Capital Markets. Working Paper 498, Bank of
. 2014d. Staff Guidance Note on Macroprudential England, London.
Policy. IMF Policy Paper, International Monetary Fund, Rose, Andrew K. 2004. Do We Really Know That the WTO
Washington, December. Increases Trade? American Economic Review 94 (1): 98114.
. Forthcoming a. From Banking to Sovereign Stress Saldas, Martin, and Ben Craig. Forthcoming. Spatial Depen-
Implications for Public Debt. IMF Policy Paper, Interna- dence and Data-Driven Networks of International Banks.
tional Monetary Fund, Washington. Federal Reserve Bank of Cleveland Working Paper.
. Forthcoming b. Pan-African Banks: Opportunities and Schnabl, Philipp. 2012. The International Transmission of
Challenges for Cross-Border Oversight. African Department Bank Liquidity Shocks: Evidence from an Emerging Market.
Paper, International Monetary Fund, Washington. Journal of Finance 67 (3): 897932.
. Forthcoming c. Rethinking Financial Deepening: Schoenmaker, Dirk, and Wolf Wagner. 2011. The Impact of
Stability and Growth in Emerging Markets. Staff Discussion Cross-Border Banking on Financial Stability. Tinbergen
Note, International Monetary Fund, Washington. Institute Discussion Paper 11-054/DSF18.
Kapan, Tmer, and Camelia Minoiu. 2013. Balance Sheet Vials, Jos, and Erlend Nier. 2014. Collective Action Problems in
Strength and Bank Lending during the Global Financial Macroprudential Policy and the Need for International Coordi-
Crisis. Working Paper No. 13/102, International Monetary nation. Financial Stability Review 18, Banque de France, Paris.
Fund, Washington. Vials, Jos, Jonathan Fiechter, Aditya Narain, Jennifer Elliott,
Laeven, Luc, and Fabian Valencia. 2013. Systemic Banking Ian Tower, Pierluigi Bologna, and Michael Hsu. 2010. The
Crises Database. IMF Economic Review 61 (2): 22570. Making of Good Supervision: Learning to Say No. IMF
Lam, Raphael. 2013. Cross-Border Activity of Japanese Banks. Staff Position Note No. 10/08, International Monetary Fund,
Working Paper No. No. 13/235, International Monetary Washington.
Fund, Washington. White, Halbert. 1980. A Heteroskedasticity-Consistent Covari-
Lautenschlger, Sabine. 2014. Start of the Single Supervisory ance Matrix Estimator and a Direct Test for Heteroskedastic-
Mechanism: From the Comprehensive Assessment to Day- ity. Econometrica 48 (4): 81738.

International Monetary Fund | April 2015 91


3
CHAPTER

THE ASSET MANAGEMENT INDUSTRY


AND FINANCIAL STABILITY

SUMMARY

F
inancial intermediation through asset management firms has many benefits. It helps investors diversify
their assets more easily and can provide financing to the real economy as a spare tire even when banks
are distressed. The industry also has various advantages over banks from a financial stability point of view.
Nonetheless, concerns about potential financial stability risks posed by the asset management industry
have increased recently as a result of that sectors growth and of structural changes in financial systems. Bond funds
have grown significantly, funds have been investing in less liquid assets, and the volume of investment products
offered to the general public in advanced economies has expanded substantially. Risks from some segments of the
industryleveraged hedge funds and money market fundsare already widely recognized.
However, opinions are divided about the nature and magnitude of any associated risks from less leveraged,
plain-vanilla investment products such as mutual funds and exchange-traded funds. This chapter examines sys-
temic risks related to these products conceptually and empirically.
In principle, even these plain-vanilla funds can pose financial stability risks. The delegation of day-to-day
portfolio management introduces incentive problems between end investors and portfolio managers, which can
encourage destabilizing behavior and amplify shocks. Easy redemption options and the presence of a first-mover
advantage can create risks of a run, and the resulting price dynamics can spread to other parts of the financial
system through funding markets and balance sheet and collateral channels.
The empirical analysis finds evidence for many of these risk-creating mechanisms, although their importance
varies across asset markets. Mutual fund investments appear to affect asset price dynamics, at least in less liquid
markets. Various factors, such as certain fund share pricing rules, create a first-mover advantage, particularly for
funds with high liquidity mismatches. Furthermore, incentive problems matter: herding among portfolio managers
is prevalent and increasing.
The chapter does not aim to provide a final verdict on the overall systemic importance of the potential risks or
to answer the question of whether some asset management companies should be designated as systemically impor-
tant. However, the analysis shows that larger funds and funds managed by larger asset management companies do
not necessarily contribute more to systemic risk: the investment focus appears to be relatively more important for
their contribution to systemic risk.
Oversight of the industry should be strengthened, with better microprudential supervision of risks and through
the adoption of a macroprudential orientation. Securities regulators should shift to a more hands-on supervisory
model, supported by global standards on supervision and better data and risk indicators. The roles and adequacy
of existing risk management tools, including liquidity requirements, fees, and fund share pricing rules, should be
reexamined, taking into account the industrys role in systemic risk and the diversity of its products.

Prepared by Hiroko Oura (team leader), Nicols Arregui, Jonathan Beauchamp, Rina Bhattacharya, Antoine Bouveret, Cristina Cuervo,
Pragyan Deb, Jennifer Elliott, Hibiki Ichiue, Bradley Jones, Yoon Kim, Joe Maloney, Win Monroe, Martin Saldias, and Nico Valckx, with
contributions by Viral Acharya (consultant), and data management assistance from Min-Jer Lee, under the overall guidance of Gaston Gelos
and Dong He.

International Monetary Fund | April 2015 93


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Introduction industry now intermediates assets amounting to $76


trillion (100 percent of world GDP and 40 percent of
In recent years, credit intermediation has been shifting
global financial assets; Figure 3.1).
from the banking to the nonbank sector, including the
The larger role of the asset management industry
asset management industry.1 Tighter regulations on
in intermediation has many benefits. It helps inves-
banks, rising compliance costs, and continued bank
tors diversify their assets more easily and can pro-
balance sheet deleveraging following the global finan-
vide financing to the real economy as a spare tire
cial crisis have contributed to this shift. In advanced
even when banks are distressed. The industry also
economies, the asset management industry has been
has advantages over banks from a financial stability
playing an increasingly important role in the financial
point of view. Banks are predominantly financed with
system, especially through increased credit intermedia-
short-term debt, exposing them to both solvency and
tion by bond funds.2 For emerging markets, portfolio
liquidity risks. In contrast, most investment funds
flowsmany of which are channeled through funds
issue shares, and end investors bear all investment risk
have shown steady growth since the crisis. Globally, the
(see Figure 3.2, and see Annex 3.1 for a primer on the
industry). High leverage is mostly limited to hedge
1In this chapter, the definition of the asset management indus-
funds and private equity funds, which represent a small
try includes various investment vehicles (such as mutual funds,
exchange-traded funds, money market funds, private equity funds,
share of the industry.3 Therefore, solvency risk is low in
and hedge funds) and their management companies (see Annex 3.1).
Pension funds and insurance companies are excluded, as are other 3However, these funds can still be a source of systemic risk, as

types of nonbank financial institutions. shown during the Long-Term Capital Management episode in 1998.
2See October 2014 Global Financial Stability Report. Mutual funds and exchange-traded funds do incur portfolio leverage

Figure 3.1. Financial Intermediation by the Asset Management Industry Worldwide

The asset management industry intermediates substantial amounts of The growth of investment funds has been particularly pronounced
money in the nancial system. among advanced economies during the past decade.
1. World Top 500 Asset Managers Assets under Management1 2. Size of Investment Funds in Selected Advanced Economies

Trillions of U.S. dollars (right scale) AUM, trillions of U.S. dollars (right scale)
Percent of world GDP AUM, percent of sample economies GDP
Percent of global nancial assets excluding loans

140 90 100 30

80 90
120
25
80
70
100 70
60 20
60
80 50
50 15
60 40
40
30 10
40 30
20
20
5
20
10 10

0 0 0 0
2000 01 02 03 04 05 06 07 08 09 10 11 12 13 2001 02 03 04 05 06 07 08 09 10 11 12
Sources: Bloomberg, L.P.; McKinsey (2013); Pensions and Investments and Towers Sources: Organisation for Economic Co-operation and Development; and IMF
Watson (2014); IMF, World Economic Outlook database; and IMF staff estimates. World Economic Outlook database.
1
The change of asset under management is determined both by valuation Note: AUM = assets under management. Economies comprise Canada, Germany,
changes of underlying assets as well as net inows to funds. Ireland, Japan, Luxembourg, United Kingdom, and United States. Investment funds
include mutual funds, money market funds, and exchange-traded funds.

94 International Monetary Fund | April 2015


CHAPTER 3 THE ASSET MANAGEMENT INDUSTRY AND FINANCIAL STABILITY

Figure 3.2. Products Offered by Asset Managers and Their Recent Growth

Plain-vanilla products and privately offered separate account services Open-end funds, exchange-traded funds, and private equity funds
dominate the markets as measured by assets under management. have shown strong growth since the global nancial crisis.
1. Asset Managers Intermediation by Investment Vehicles 2. Recent Growth of Selected Investment Vehicles
(Percent of $79 trillion total assets under management, end-2013) (Assets under management in trillions of U.S. dollars)

Private Hedge funds Other 30 2007


equity 3% alternatives
Exchange- 2008
5% 2%
traded funds 25 2009
4% 2010
20
Money market 2011
funds 15 2012
8% 2013
10
Closed-end Separate
mutual funds accounts 5
1% 36%
Open-end 0
mutual funds Open-end Money market Exchange- Private equity Hedge funds
41% mutual funds funds traded funds

Sources: BarclayHedge; European Fund and Asset Management Association; Sources: BarclayHedge; European Fund and Asset Management Association;
ETFGI; Organisation for Economic Co-operation and Development; Pensions and Organisation for Economic Co-operation and Development; Preqin; and IMF
Investments and Towers Watson (2014); Preqin; and IMF staff estimates. staff calculations.

most cases (see October 2014 Global Financial Stabil- and money market funds are already well recognized.
ity Report). Intermediation through funds also brings However, the importance of plain-vanilla products
funding cost benefits and fewer restrictions for firms is less well understood (Feroli and others 2014). At
compared with bank financingit does, however, also the individual fund level, plain-vanilla funds face
expose firms to more volatile funding conditions, so liquidity risk: the shares of open-end mutual funds
the advantages have to be weighed against the risks. and exchange-traded funds are usually redeemable or
Nevertheless, the growth of the industry has given tradable daily, whereas assets can be much less liquid.
rise to concerns about potential risks.4 By now, the However, the extent to which such risks at the level
assets under management of top asset management of an individual institution can translate into systemic
companies (AMCs) are as large as those of the largest risk is subject to ongoing research and debate.
banks, and they show similar levels of concentration.5 Potential systemic risks from less leveraged segments
For emerging markets, the behavior of fund flows has of the industry are likely to stem from price externali-
for some time been a key financial stability concern, as ties in financial markets and their macro-financial
extensively discussed in the April 2014 Global Finan- consequences. Systemically important effects may arise
cial Stability Report. Similarly, risks from hedge funds if features of the industry tend to amplify shocks or
increase the likelihood of destabilizing price dynam-
through derivatives and securities lending, about which only limited ics in certain asset markets compared with a situation
information is disclosed. However, most publicly offered products
have regulatory leverage caps that are generally much lower than
in which investors invest directly in securities. These
those for banks (see Table 3.1). effects can have broader economic implications. For
4A report by the Office of Financial Research (2013) summariz-
example, if intermediation through funds raises the
ing potential systemic risks emanating from the industry spurred an
probability of fire sales of bonds that are held by key
active discussion among academics, supervisors, and the industry.
A large number of qualitative analyses on this topic (CEPS-ECMI players in the financial sector or that are used as col-
2012; Elliott 2014; Haldane 2014) are available, but comprehensive, lateral, then the risk of destabilizing knock-on effects
data-based evidence is still limited. on other institutions rises, with potentially important
5In this chapter, the term AMC does not include asset manage-

ment companies set up to handle distressed assets in the context of macro-financial consequences. Similarly, if funds
bank restructuring and resolution. exacerbate the volatility of capital flows in and out of

International Monetary Fund | April 2015 95


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

emerging markets or increase the likelihood of conta- many market-making activities, possibly contributing to
gion, significant consequences will be endured by the a reduction in market liquidity (October 2014 Global
recipient economies.6 Financial Stability Report). Consequently, large-scale
Some key features of collective investment vehicles trading by funds could potentially have a larger effect on
may give rise to such destabilizing dynamics compared markets than in the past. Moreover, the role of fixed-
with a situation without intermediaries. Conceptually, income funds has expanded considerablyand price
it is important to distinguish clearly between the types disruptions in fixed-income markets have potentially
of risks that result from the presence of intermediaries larger consequences than large price swings in equity
and those that are merely a reflection of the behavior markets. The volume of products offered to the general
of end investors and would occur in the absence of public in advanced economies has grown considerably.7
intermediaries (Elliott 2014). Two main risk channels Finally, the prolonged period of low interest rates in
that are important in this context, even for unlever- advanced economies has resulted in a search for yield,
aged funds, are (1) incentive problems related to the which has led funds to invest in less liquid assets, and
delegation of portfolio management decisions by end is likely to have exacerbated the risks described above
investors to funds, which, among other things, may (October 2014 Global Financial Stability Report).
lead to herding, and (2) a first-mover advantage for These considerations have sparked a policy discus-
end investors (that is, incentives not to be the last in sion about intensifying oversight across advanced and
the queue if others are redeeming from a fund), which emerging economies. In 2014, the Financial Stability
may result in fire-sale dynamics. These issues are dis- Board (FSB) and International Organization of Securities
cussed in detail in this chapter. Commissions (IOSCO) proposed assessment methodolo-
In recent years, the importance of such risks is gies to identify investment funds that might be global
likely to have risen in advanced economies because of systemically important financial institutions (G-SIFIs)
structural changes in their financial systems. Not only and as such would be regulated differently from the oth-
has the relative importance of the asset management ers (FSB and IOSCO 2014). This proposal was revised
industry grown, but banks have also retrenched from in March 2015, and includes approaches for identifying
both investment funds and asset managers as G-SIFIs
6Other risks include operational risks and risks related to securi- (FSB and IOSCO 2015). Market regulators in major
ties lending, which are not discussed in detail in this chapter. See jurisdictions (Figure 3.3), such as the U.S. Securities and
Cetorelli (2014).
Exchange Commission (SEC), are considering revising
their approach to the oversight of asset managers and
Figure 3.3. Key Domiciles of Mutual Funds the products they offer, including through stress testing
(Mutual funds by domicile, percent of total assets under management, requirements. This is a paradigm shift. Until recently,
end-2014) securities regulators have mainly focused on investor pro-
tection, with limited attention to financial stability risks.
The mutual fund industry is dominated by U.S. and European funds.
Among emerging market economies, Brazil has the largest fund sector. This chapter aims to shed more light on the empiri-
cal relevance of these issues, thereby contributing to
the understanding of the systemic risk implications of
Other
Brazil the asset management industry. This task is challeng-
developed China
9% 3% 2% Other emerging
markets 4% ing given that the risks of concern have not yet or only
Japan
3% partially materialized in advanced economies; inference,
Luxembourg 10%
therefore, often has to be indirect. So far, the literature
Developed Ireland 5% has only examined partial aspects of these problems in
Europe 31% France 5% individual markets. This chapter provides an account of
United Kingdom 4%
key risk profiles of the largest segments in the industry
Other developed
Europe 7% and an in-depth, original, data-based analysis of some of
United States
7Retail investors are often seen to be less sophisticated and
49%
informed than institutional investors, and more prone to chase
Sources: European Fund and Asset Management Association; and IMF staff returns (Frazzini and Lamont 2008). This possibly exacerbates the
calculations. incentive problems mentioned earlier.

96 International Monetary Fund | April 2015


CHAPTER 3 THE ASSET MANAGEMENT INDUSTRY AND FINANCIAL STABILITY

the main issues featured in the public discussion, backed Moreover, macroprudential oversight frameworks
by interviews with asset managers and supervisors. The should be established to address financial stability
key questions are the following: risks stemming from the industry. These stability risks
What are the potential sources of financial stability originate in price externalities that can be missed by
risks from the asset management industry, particu- microprudential regulators and asset managers.
larly from the less leveraged, plain-vanilla segments? The roles and adequacy of existing risk management
What is the empirical evidence on the various spe- tools, including liquidity requirements, fees, and
cific risk channels? fund share pricing rules, should be reexamined, tak-
What existing internal risk management and over- ing into account the industrys role in systemic risk
sight tools can be used to mitigate financial stability and the diversity of its products.
risks? What needs to be done to better monitor and
mitigate these risks? The chapter first lays out conceptual issues related
to the nature of potential financial stability risks from
The detailed empirical analysis finds evidence for the industry. Next, various empirical exercises are
many mechanisms through which funds can create and conducted to identify different behavioral patterns
amplify risks, although their importance varies across of mutual fund investors and their financial stability
asset markets: implications. The chapter then examines the industrys
Mutual fund investments appear to affect asset oversight framework and makes recommendations for
price dynamics, at least in less liquid markets. The reducing financial stability risks.
impact, however, does not seem to have risen over
time. Assets that are held in a concentrated manner
by funds perform worse during periods of stress. Financial Stability Risks of Plain-Vanilla Funds:
Various factors create run risk, including certain Conceptual Issues
fund share pricing rules. To some extent, however, Plain-vanilla mutual funds and ETFsthe largest
risks are mitigated by funds liquidity management. segment of the industrydo not suffer much from
The evidence points to the importance of incen- the known vulnerabilities of hedge funds and money
tive problems between end investors and portfolio market funds. Reforms are already underway to address
managers. Herding among U.S. mutual funds has risks related to hedge funds (which can incur high
been rising across asset markets, particularly among leverage and engage in complex strategies with few dis-
retail-oriented funds (whose end investors are more closure requirements) and money market funds (some
fickle and for whom assessing the skills of portfolio of which offer redemptions at a constant nominal
managers is more difficult). The patterns of fund value per fund share, making their liabilities similar to
inflows by end investors also encourage poorly per- deposits and vulnerable to runs). In general, these spe-
forming portfolio managers to take excessive risks. cific risks apply less to typical mutual funds and ETFs
However, larger funds and funds belonging to (Table 3.1 and Annex 3.1).
larger AMCs do not necessarily contribute more to
systemic risk. The investment focus appears to be
relatively more important than size when gauging Risk Transmission Channels
systemic risk. Intermediation through plain-vanilla funds is, however,
not risk free (Figure 3.4):8
Overall, the evidence calls for strengthening the
microprudential supervision of risks and adopting
macroprudential oversight of the industry: 8Apart from Table 3.1 and Annex 3.1, this chapter does not cover

separate accounts in detail because of data limitations. However,


Currently, most securities regulators focus on investor
SIFMA (2014) indicates that these accounts mainly invest in simple
protection and do not intensively supervise risks of securities portfolios with little leverage. For pension fund and insur-
individual institutions with the help of risk indica- ance company investors, separate accounts are bound by overall
tors or stress tests. This practice needs to be changed, investment restrictions set by their respective regulators. Redemption
risks appear to be limited as well because institutional investors tend
supported by global standards on microprudential to internalize the cost of their sales, and large redemptions can be
supervision and more comprehensive data. settled in kind.

International Monetary Fund | April 2015 97


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Table 3.1. Summary Characteristics and Risk Profiles of Major Investment Vehicles
Vehicle 2013 AUM Publicly Collective Typical Typical Solvency Leverage Portfolio Main Disclosure
(trillions of Offered Investment Redemption Settlement Risk through Leverage2 Investor Gap3
U.S. dollars) Schemes and Trading Method Borrowing1,2 (Derivatives) Clientele
Practice
Open-End 25 Yes Yes End of day Cash Low Possible Yes with cap Retail, Low
Mutual Fund with cap institutional
Closed-End 0.5 Yes Yes N.A. Cash Low Some yes Yes with cap Retail, Low
Mutual Fund (primary) with cap institutional
Intraday
(secondary)
Money 4.8 Yes Yes End of day Cash Low Possible Yes with cap Retail, Low
Market Fund with cap institutional
Exchange- 2.3 Yes Yes Infrequent In kind Low Possible Yes with cap Retail,
Traded Fund (primary) (primary) with cap institutional
Intraday Cash
(secondary) (secondary) Low
Synthetic 0.14 Cash Low Possible High Institutional
ETF with cap derivative
use
Private 3.5 No Yes N.A. Cash High5 Some yes, No Institutional Medium
Equity Fund (closed-end no cap information
with long-
term finite
life)
Hedge Fund 2.2 No Yes Quarterly Cash High5 High no cap High no cap Institutional Medium
+ lock-up
period +
90 days
advance
notice
Separate 227 No No No Cash or in Low No No Institutional High
Account6 information kind information8 information8
Sources: BarclayHedge; Deutsche Bank (2014); ETFGI; EFAMA (2014); ICI (2014a, 2014c); McKinsey (2013); Metrick and Yasuda (2011); Morningstar (2012); OFR (2013); Preqin;
PriceWaterhouseCoopers (2013); and IMF staff estimates.
Note: AUM = assets under management; ETF = exchange-traded fund; N.A. = not applicable.
1Borrowing includes issuing debt or taking bank loans.
2No cap means no regulatory cap, and with cap means there are regulatory caps on the leverage. For public funds in the United States, leverage is capped at 33 percent of assets

including portfolio leverage. European Undertakings for Collective Investment in Transferable Securities (UCITS) funds can borrow up to 10 percent of assets, but only temporary bor-
rowing is allowed and it should not be used for investment.
3Disclosure in this column is about securities, borrowing through loans, and cash holdings information. Across all products, there is very little information about derivatives and

securities financing transactions (repurchase agreements and securities lending transactions), their counterparties, and collateral.
4The figure covers European-listed synthetic exchange-traded funds. Synthetic products are mainly seen in Europe and to a lesser extent in Asia. See Annex Table 3.1.1 for a descrip-

tion of synthetic products.


5In addition to taking leverage, these types of funds risk their own capital and balance sheets when investing given that they comingle client investors money with their own money for investment.
6This is different from separate account used among insurance companies. See Annex Table 3.1.1 for description.
7The figure is based on the U.S. data reported in OFR (2013) and the European data reported in EFAMA (2014).
8Investment strategy should be in line with the mandate set by clients and their regulatory requirements (such as insurance and pension fund regulations).

The delegation of investment decisions introduces incentives is to evaluate funds relative to their peers
incentive problems between end investors and port- and relative to benchmarks. This form of evaluation,
folio managers that can induce destabilizing behavior in turn, can lead to a variety of trading dynamics
and amplify shocks. Investors delegate day-to-day with potentially systemic implications, such as herd-
portfolio management to portfolio managers. Inves- ing or excessive risk taking (Box 3.1).10,11
tors cannot directly observe managers daily actions
10Similarly, the same type of informational issues can make it dif-
or their skills, and therefore provide incentives to
ficult for investors to distinguish between problems at the fund level
managers to act in investors interests (Rajan 2005).9 versus problems at the AMC level, possibly leading to brand name
A common (and imperfect) way of establishing effects, in which operational and reputational concerns about one
fund spill over to others in the same fund family.
9Legally, asset managers have a duty to act as fiduciaries on behalf 11Separate issues arise from passive, index-linked investing.

of their clients. Increasing investment of this form has been argued to distort asset

98 International Monetary Fund | April 2015


CHAPTER 3 THE ASSET MANAGEMENT INDUSTRY AND FINANCIAL STABILITY

Figure 3.4. Unleveraged Open-End Funds and Systemic Risk

Information gap between managers and investors


Incentive Benchmark-based evaluation Macronancial
problems Excessive risk taking consequences
of Herding
managers Brand name effects (spillovers of redemption
within fund family)

Price
externalities
re sales,
contagion,
volatility

First-mover advantage
Liquidity mismatch
Run risk Managers sell liquid assets rst
Some fund share pricing rules impose cost of
liquidity risk unfairly on second movers

Source: IMF staff.

Easy redemption options can create run risks due A large proportion of funds issue easily redeem-
to a first-mover advantage.12 Investors can have an able shares, and liquidity mismatches have been rising
incentive to exit faster than the others even without (Figures 3.5 and 3.6). Open-end funds are exposed
constant net asset value (NAV) or guaranteed returns to redemption risk because investors have the ability
if the liquidation value of fund shares declines as to redeem their shares (usually on a daily basis) while
investors wait longer to exit. This decline in value funds have increasingly been investing in relatively
could happen for various reasons. First, asset man- illiquid securities such as high-yield corporate bonds
agers may use cash buffers and sell relatively more and emerging market assets.
liquid assets first in the face of large redemptions. Large-scale sales by funds may exert significant
Second, certain funds have fund share pricing rules downward asset price pressures, which could affect
that pass the costs of selling assetspossibly at fire- the entire market and trigger adverse feedback loops.
sale priceson to the remaining investors (Box 3.2). The effects on asset prices could have broader macro-
Such effects are intensified when funds are investing financial consequences, affecting the balance sheets
in relatively less liquid assets, and thereby create large of other actors in financial markets; reducing collat-
mismatches between the market liquidity of assets eral values; and reducing credit financing for banks,
and liquidity offered to end investors (October 2014 firms, and sovereigns. The effects could also be spread
Global Financial Stability Report).13 unevenly across jurisdictions. For instance, the main
impact of trades by funds domiciled in advanced
prices and risk-return tradeoffs (Wurgler 2010 and Box 3.1). This
chapter does not explore these issues.
economies could be felt in emerging markets (see
12The incentive to redeem quickly is often referred to as strategic April 2014 Global Financial Stability Report for
complementarity, and is similar to the mechanism behind bank details).
runs (as in Diamond and Dybvig [1983]). More generally, problems
Although these potential risks and propagation
related to the delegation of investment decisions or first-mover
advantage are also present in other forms of financial intermediation, channels are recognized as theoretical possibilities,
albeit to different degrees. For instance, pension funds and insurance there is disagreement about their importance in prac-
companies face much lower redemption risks. tice. Advanced economies have experienced few cases
13A related issue concerns the pricing of infrequently traded

securities. The October 2014 Global Financial Stability Report dis- in which asset management activities outside of hedge
cusses some of the issues related to the so-called matrix pricing. funds and money market funds triggered or amplified

International Monetary Fund | April 2015 99


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 3.1. Possible Incentive Problems Created by Delegated Management


The delegation of investment decisions introduces ContagionBy contrast, if fund managers become
incentive problems between end investors and fund more risk averse in response to past losses, and if
managers, which can induce destabilizing behavior they are evaluated against their peers or bench-
and amplify shocks. As discussed in the primer on the marks, they may be induced to retrench to the
asset management industry (Annex 3.1), end investors benchmark in response to losses. This behavior, in
delegate day-to-day control of portfolios to managers. turn, can induce the transmission of shocks across
Investors cannot directly observe managers abilities, nor assets and result in momentum trading (Broner,
do they see every single trade and portfolio position. Gelos, and Reinhart 2006). See Calvo and Mendoza
Investors, therefore, provide incentives to asset managers (2000), Chakravorti and Lall (2003), and Ilyina
to act in investors interests (Rajan 2005). A common (2006) for other types of models linking bench-
way of providing incentives is to evaluate asset managers mark-based compensation to contagion.
relative to their peers and to benchmarks. This evalu- Herding, return chasing, and incentives to runEvalu-
ation can take direct or indirect forms: (1) managers ation relative to average performance tends to induce
compensation can be linked to relative performance risk-averse portfolio managers to mimic the behavior
(Ma, Tang, and Gomez 2013) or (2) investors inject of peers (Scharfstein and Stein 1990; Arora and
money into funds that perform well relative to their Ou-Yang 2001; Maug and Naik 2011). Incentives
benchmarks. The effect of the latter is similar to the to herd are reinforced because end investors can exit
effect of the former if compensation increases with funds quickly, and mutual fund managers cannot
assets under management (AUM). These incentive afford to wait until their peers private information is
problems, in turn, can lead to a variety of dynamics revealed and incorporated fully in asset prices (Froot,
with potentially systemic implications (Stracca 2006). OConnell, and Seasholes 2001). Vayanos (2004)
More specifically, they can lead to the following: shows that when fund managers lose AUM because
Excessive risk takingIf a funds AUM grow more of poor performance, flights to quality may occur.
with good performance than shrink with poor Feroli and others (2014) construct a model in which
performance, incentives are created to incur more performance evaluation relative to benchmarks cre-
risk when the fund is falling behind (Chevalier ates incentives for fund managers to join sell-offs
and Ellison 1997; Ferreira and others 2012; see the during downturns and chase yield during upturns.
example in Table 3.1.1). Similar incentives exist in a Buffa, Vayanos, and Woolley (2014) discuss theoreti-
tournament setting, in which funds are evaluated cally how such benchmark-centric assessments can
based on their interim performance (say, in the contribute to the buildup of bubbles.
middle of the year) compared with peers (Basak, Churning and noise tradingDelegated portfolio
Pavlova, and Shapiro 2008).1 management may induce managers to churn (engage

Table 3.1.1. An Illustrative Example of Asset Managers Incentives for Risk Taking
Because investors reward winners more than they punish poor performers, it pays to take risks.

Additional Fee Income


Net Inflows to Fund (1 percent of assets
Outcome: Change in Net (millions of under management, in
Options Likelihood (percent) Asset Value U.S. dollars) millions of U.S. dollars)
Benchmark Portfolio 100 Same as benchmark 0 0
50 10% in excess of 100 1
benchmark
Gamble
50 10% below benchmark 20 0.2
Expected outcome Same as benchmark 40 0.4

1This is also known as the risk-shifting problem. More generally, risk shifting arises when earnings for managers are convex based

on their compensation. Limited liability also contributes to the convexity of manager earnings. See Ross (2004) for a qualification of the
payoff convexity argument. See also Massa and Patgiri (2009).

100 International Monetary Fund | April 2015


CHAPTER 3 THE ASSET MANAGEMENT INDUSTRY AND FINANCIAL STABILITY

Box 3.1 (continued)


in noise trading) to signal their talent and superior contracts, the depth of the market may be reduced
knowledge, given that it is difficult to identify talent (Igan and Pinheiro 2012). Basak and Pavlova
and effort (Allen and Gorton 1993; Dow and Gor- (2014) develop a general-equilibrium asset price
ton 1997; Dasgupta and Prat 2006). model that incorporates incentives for institutional
Market depth and volatilityPerformance evalu- investors to do well relative to their index. The
ation relative to a benchmark may lead to higher induced investment patterns create excess correla-
price volatility of securities that are included in tions among stocks belonging to an index. It also
the benchmark. Since information acquisition may increases the volatility of index stocks and of the
be hindered by these relative-performance-based overall market.

Box 3.2. Fund Share Pricing Rules and First-Mover Advantage


Certain forms of fund share pricing can give rise to Inflexible net asset value (NAV) pricing can gener-
a first-mover advantage for investors to run. The key ate a first-mover advantage for an open-end mutual
factor is how investment losses and trading costs are fund (Table 3.2.1). In the United States, funds
distributed between buy-and-hold and redeeming fund issuing redeemable securities are required to sell,
shareholders. If these are borne by the fund and there- redeem, or repurchase such securities based on the
fore by the buy-and-hold shareholders, investors can NAV of the security next computed after receipt
recover more cash by redeeming early. of the order. Transaction coststrading fees, market

Table 3.2.1. Comparison of Fund Pricing Rules


(Millions of U.S. dollars)

UCITS UCITS-AIF U.S. Open-End Mutual Fund


Transactions Swing Pricing (Full) Dual Pricing (1940 Act)
Beginning NAV 100 100 100
Net Flows 15 15 15
Purchases +5 +5 +5
Redemptions 20 20 20
Total Costs of Selling Assets 0.015 0.015 0.015
(0.1 percent, including bid-ask
spread)
Transaction Costs Incurred 0.0051 0 0
by Investors Purchasing
Fund Shares
Transaction Costs Incurred 0.020 0.015 0
by Investors Redeeming
Fund Shares
Transaction Costs Incurred by 0 0 0.0152
Fund and Remaining Investors
Ending NAV 85.000 85.000 84.985
Memo Estimated transaction costs borne by trading investors Actual transaction costs
borne by fund
Source: BlackRock (2014b).
Note: AIF = Alternative Investment Fund (European directive governing products including hedge funds and private equity funds); NAV = net
asset value (mutual fund share price, per share); UCITS= Undertaking of Collective Investment in Transferable Securities (European Union direc-
tive governing publicly offered investment funds). In the United States, investment companies (as defined) are regulated primarily under the U.S.
Investment Company Act of 1940.
1Because fund NAV has swung to the bid price because of net redemptions, purchasing investors benefit to the extent that they purchase units

that are cheaper than preswung NAV. This benefit is offset by the costs paid by redeeming clients.
2In certain circumstances, portfolio managers may choose to use cash buffers or borrow funds (or both) to meet redemptions without incurring

transaction costs.

International Monetary Fund | April 2015 101


GLOBAL FINANCIAL STABILITY REPORT: NAVIGATING MONETARY POLICY CHALLENGES AND MANAGING RISKS

Box 3.2 (continued)


impact, and spread costsare borne by the funds. broker-dealerstrade in between. Only autho-
This reduces a funds NAV, possibly by a substantial rized participants trade with ETFs in the primary
amount if market liquidity dries up. The European market, and trades are usually settled in kind.
framework, in contrast, allows for pricing rules such Intraday liquidity to end investors is offered in the
as swing- or dual-pricing rules, as described in Table secondary market by authorized participants.1 The
3.2.1, that adequately impose transaction costs on key difference between ETFs and mutual funds in
redeeming shareholders instead of the fund. This the context of first-mover advantage is that ETFs
helps reduce remaining shareholders incentive to are not required to pay cash back to investors at
run. NAV.2 Authorized participants trade ETF shares
The share pricing practice of exchange-traded with clients or on stock exchanges at the ETF share
funds (ETFs) is different from that of open-end price determined in the secondary market. There-
mutual funds. As shown in Figure 3.2.1 and Annex fore, depending on market conditions, an ETFs
3.1, ETFs do not directly transact with end inves- share price could be higher or lower than the ETFs
tors. Authorized participantstypically major indicative NAV.

Figure3.2.1.Structure of Exchange-Traded Funds

Primary Market Secondary Market

Hold shares, arbitrage trading

Shares Shares Investors,


Authorized
ETF stock
participant
Securities Cash exchange

Physical Liquidity
NAV represents ETF share premium or
basket of
market value of price discount paid
securities by investors
ETFs assets

NAV may not be equal to ETF share price, depending on arbitraging capacity of APs

Source:IMF staff.
Note:AP=authorized participant; ETF=exchange-traded fund;NAV=net asset value.

1Although there is a widespread perception that ETFs face higher redemption risks because they offer intraday liquidity to share-

holders, intraday liquidity (offered in the secondary market) is not the same as intraday redemption (offered in the primary market).
Primary market activities, which result in fund flows, are much less frequent than secondary market trading (ICI 2014c; BlackRock
2014a).
2In the United States, ETFs operate with the Securities and Exchange Commissions special exemption from the 1940 Act

requirement that open-end funds repay redeeming shareholders at the next NAV calculated after an order is submitted (ICI
2014b).

102 International Monetary Fund | April 2015


CHAPTER 3 THE ASSET MANAGEMENT INDUSTRY AND FINANCIAL STABILITY

Box 3.2 (continued)


Redeeming shareholders need to pay for the
cost of market liquidity risk by accepting an ETF Figure 3.2.2. Difference between NAV and
share price below NAV if market liquidity dries up. ETF Share Price
(Percent of NAV, all countries, equity funds)
Authorized participants are usually arbitrageurs,
and if they see a major gap between NAV and ETF The ETF share can be traded in the secondary
share prices, they trade in the direction to close the market at a discount to NAV when markets are
gap. If investors find it easier to sell ETF shares under generalized stress.
1.0
relative to the underlying assets, this will tend to ETF share price > NAV
result in a discount to NAV. The discount can be Premium for ETF share
accentuated when funding conditions reduce autho-
0.5
rized participants arbitrage capacity (Figure 3.2.2).
The cost of fire sales of ETF shares is borne by the
trading shareholders, not by the ETF or buy-and-
0.0
hold shareholders, reducing buy-and-hold share-
holders incentive to run.
Economically, these flexible fund share pricing rules
0.5
are similar to countercyclical redemption and purchase
fees that reflect market liquidity cost and are added
to NAV. If a U.S. 1940 Act fund imposes purchase
1.0
and redemption fees that are retained by the fund3
ETF share price < NAV
and reflect the bid-ask spreads for transactions (or Discount for ETF share
ETF NAV and share price gap), the outcome would
1.5
be similar to that of funds with flexible share pricing
Jan. 2000
Jul. 01
Jan. 03
Jul. 04
Jan. 06
Jul. 07
Jan. 09
Jul. 10
Jan. 12
Jul. 13
rules. At the same time, such fees also help ensure
equality between buy-and-hold investors and trading
investors. Sources: Bloomberg, L.P.; and IMF staff calculations.
Note: ETF = exchange-traded fund; NAV = net asset
value.

3Current U.S. rules do allow for the introduction of fees that are added to funds NAV, which can then be distributed to remaining

shareholders.

systemic distress.14 The realization of brand risk and markets, in particular emerging markets.15 Moreover,
redemptions from PIMCO funds in September 2014 recent structural shifts in many markets following
did not result in major disruptive market movements the global financial crisis require a fresh review of the
because, overall, bond funds continued to receive net evidence.
inflows. However, the academic literature has docu- Against this backdrop, this chapter empirically
mented contagion and amplification effects for some explores the precise channels through which mutual
funds and ETFs can affect financial stability. The aim
14There have been some cases of nonmoney market mutual

fund distress in emerging markets. For example,