Académique Documents
Professionnel Documents
Culture Documents
APR
17
Getting the Policy Mix Right
17
APR
Getting the Policy Mix Right
IMF
I N T E R N A T I O N A L M O N E T A R Y F U N D
2017 International Monetary Fund
Cataloging-in-Publication Data
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 4, 2017. The views expressed in this publication are those of the IMF staff and do not
necessarily represent the views of the IMFs Executive Directors or their national authorities.
Recommended citation: International Monetary Fund. 2017. Global Financial Stability Report: Getting
the Policy Mix Right. Washington, DC, April.
Tables
Table 1.1. Emerging Market Economies: External and Trade Vulnerabilities 19
Table 1.2. Asset Quality and Capital Indicators for a Sample of Emerging Market Banks 20
Table 1.3. Details of the Sample Used in the European Bank Analysis 31
Table 1.4. Structural Factors Affecting Bank Revenues and Costs 33
Table 1.5. Asset Quality Position and Recent Progress 35
Table 1.6. Selected IMF Policy Recommendations 36
Table 2.1. Classification of Bank Business Models 59
Table 2.2. Bank Profitability and Equity Values in Periods of Normal and Prolonged
Low Interest Rates 60
Table 2.3. Capital Charges for Risky Investments by Insurers 65
Table 2.3.1. Risk-Return Calibration and Portfolio Allocations 74
Annex Table 2.2.1. Data Sources 79
Table 3.1. Determinants of the Sensitivity of Domestic Financial Conditions to
Global Financial Shocks 95
Annex Table 3.1.1. Country Coverage 101
Annex Table 3.1.2. Data Sources 102
Annex Table 3.3.1. Domestic Financial Conditions Drivers 104
Figures
Figure 1.1. Global Financial Stability Map: Risks and Conditions 2
Figure 1.2. Global Financial Stability Map: Assessment of Risks and Conditions 3
Figure 1.3. Reflation and Market Optimism 4
Figure 1.4. Assessments of U.S. Equity Valuations 5
Figure 1.5. United States: Policies under Discussion and Financial Stability Risks 6
Figure 1.6. United States: Business Confidence and Economic Risk Taking 7
Figure 1.7. Policy Stimulus and Corporate Balance Sheets 8
Figure 1.8. United States: Corporate Internal Funds and External Sources of Finance 10
Figure 1.9. Corporate Leverage and the Credit Cycle 11
Figure 1.10. Debt Service, Interest Coverage Ratios, and Vulnerability to Higher Interest Rates 12
Figure 1.11. United States: A Retrospective on Economic versus Financial Risk Taking 14
Figure 1.12. Emerging Market Economies: Asset Prices and Fundamentals 15
Figure 1.13. Transmission of External Risks to Emerging Market Economies 16
Figure 1.14. Capital Flows to Emerging Market Economies 17
Figure 1.15. Emerging Market Corporate Debt under Rising Risk Premiums and Protectionism 18
Figure 1.16. Emerging Market Bank Capital and Asset Quality 21
Figure 1.17. Underprovisioning in the Weak Tail of Banks 22
Figure 1.18. Emerging Market Economy Challenges 24
Figure 1.19. China: Credit and Bank Balance Sheets 25
Figure 1.20. China: Capital Flows and Foreign Exchange Reserves 26
Figure 1.21. Recent Turmoil in Chinese Financial Markets 26
Figure 1.22. Banking Sector Market Valuations and Return Performance 28
Figure 1.23. European Bank Profitability, 2016 30
Figure 1.24. European Banking System Actions to Reduce Costs 34
Figure 1.25. Global Systematically Important Bank Business Model Challenges 37
Figure 1.26. Bank and Sovereign Nexus 38
Figure 1.1.1 Weighted Average of Cross-Currency Swap Bases in Selected Advanced Economies 40
Figure 1.1.2 Foreign-Currency Maturity Matches 41
Figure 1.3.1. Measures of Financial Linkage between the United Kingdom and the European Union 45
Figure 2.1. Interest Rates, Term Spreads, and Volatility in Advanced Economies 51
Figure 2.2. Term Premiums in Economies with Normal versus Low Natural Rates 54
Figure 2.3. Banking under Low Natural Rates: Theoretical Predictions 56
Figure 2.4. Japan: Evolution of Bank Net Interest Margins in Normal and Low-for-Long Settings 57
Figure 2.5. Japan: Banks Adaptation to the Deposit Rate Lower Bound Period 58
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, 201617 or JanuaryJune) to indicate the years or months covered,
including the beginning and ending years or months;
/ between years or months (for example, 2016/17) to indicate a fiscal or financial year.
Billion means a thousand million.
Trillion means a thousand billion.
Basis points refers to hundredths of 1 percentage point (for example, 25 basis points are equivalent to of
1percentage 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.
This version of the Global Financial Stability Report (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
Copyright and Usage website, www.imf.org/external/terms.htm.
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 analysis in this report has been coordinated by the Monetary and Capital Markets (MCM) Department under
the general direction of Tobias Adrian, 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 com-
ments and suggestions from the senior staff in the MCM Department.
Individual contributors to the report are Ali Al-Eyd, Adrian Alter, Sergei Antoshin, Nicolas Arregui, Luis Brando-
Marques, John Caparusso, Jorge Chan-Lau, Qianying Chen, Sally Chen, Yingyuan Chen, Fabio Cortes,
DimitrisDrakopoulos, Martin Edmonds, Selim Elekdag, Jennifer Elliott, Caio Ferreira, Rohit Goel, Lucyna Gornicka,
Thomas Harjes, Sanjay Hazarika, Geoffrey Heenan, Dyna Heng, Paul Hiebert, Eija Holttinen, Gee Hee Hong,
Henry Hoyle, Viacheslav Ilin, Nigel Jenkinson, Andy Jobst, David Jones, Mitsuru Katagiri, Will Kerry,
Oksana Khadarina, John Kiff, Robin Koepke, Romain Lafarguette, Frederic Lambert, Tak Yan Daniel Law,
Yang Li, Sherheryar Malik, Rebecca McCaughrin, Naoko Miake, Win Monroe, Vladimir Pillonca, Thomas Piontek,
Luca Sanfilippo, Dulani Seneviratne, Juan Sol, Ilan Solot, Nobuyasu Sugimoto, Jay Surti, Narayan Suryakumar,
Laura Valderrama, Francis Vitek, Jeffrey Williams, Dmitry Yakovlev, and Kai Yan. Magally Bernal, Lilit Makaryan,
Breanne Rajkumar, and Annerose Wambui Waithaka were responsible for word processing.
Gemma Diaz from the Communications Department led the editorial team and managed the reports produc-
tion with support from Michael Harrup, Linda Kean, and Joe Procopio, and editorial assistance from Sherrie Brown,
Christine Ebrahimzadeh, Susan Graham, Linda Long, Alastair McIndoe, Lucy Scott Morales, Nancy Morrison,
Annerose Wambui Waithaka, Katy Whipple, Eric Van Zant, AGS, and Vector.
This particular issue of the GFSR draws in part on a series of discussions with banks, securities firms, asset
management companies, hedge funds, standard setters, financial consultants, pension funds, central banks, national
treasuries, and academic researchers.
This GFSR reflects information available as of March 31, 2017. The report benefited from comments and sug-
gestions from staff in other IMF departments, as well as from Executive Directors following their discussion of the
GFSR on April 4, 2017. 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.
spillovers from advanced economies. But overall finan- include banking systems with assets that are large rela-
cial stability risks remain elevated because global politi- tive to the economy, with a long weak tail of banks, or
cal and policy uncertainties are opening new channels with too many banks with a regional focus or a narrow
for negative spillovers. A sudden reversal of market mandate. These features can result in limited lending
sentiment or a global shift toward inward-looking pro- opportunities or a high number of branches relative to
tectionist policies could reignite capital outflows and the assets in the banking system, adding to costs and
hurt growth prospects, testing the resilience of these reducing operational efficiencies. Although measures
economies. are being taken to address profitability concerns, more
Countries with strong international financial and progress needs to be made in reducing overbanking in
trade links in particular could be challenged by tighter the countries with the biggest challenges.
global financial conditions or adverse trade measures. System-wide headwinds are a problem not only
These risks could exacerbate existing vulnerabilities within countries but can also affect the profitability of
in the corporate sector and could increase the debt large, systemically important banks in Europe. These
at risk of the weakest firms by $130$230 billion. A institutions find it difficult to keep up with their global
sharp turn away from the current supportive external competitors, and in some cases this may be partly due
environment could reinforce risks in countries whose to profitability problems in their home countries. Until
weakest banks are challenged to maintain asset quality these structural impediments are addressed, a simple
and adequately provision for bad loans after long credit restructuring of their business models is unlikely to
booms. yield sufficient profitability. Left unresolved, a combi-
China faces mounting risks to financial stability as nation of weak profits, lack of access to private capital,
credit continues to rise rapidly. Chinas bank assets are and large bad debt burdens impedes recovery and
now more than triple its GDP, and other nonbank could reignite systemic risks.
financial institutions also have heightened credit expo-
sure. Many financial institutions continue to be overly
dependent on wholesale financing, with sizable asset- It Is Crucial to Get the Policy Mix Right
liability mismatches and elevated liquidity and credit Securing and building on improvements in stabil-
risks. Recent turbulence in money markets illustrates ity and market expectations will require concerted and
the vulnerabilities that remain in Chinas increasingly careful efforts by policymakers at the national and
large, opaque, and interconnected system. global levels. Policymakers should adjust the policy mix
to deliver a stronger path for long-term and inclu-
sive growth while avoiding politically expedient but
European Banking Systems Must Address ultimately counterproductive inward-looking policies.
Structural Challenges In the United States, policymakers should vigilantly
Considerable progress has been made in the monitor increased leverage and deteriorating credit
European banking sector over the past few years, and quality. Regulators should preemptively address exces-
optimism about a cyclical upturn in advanced econo- sive financial risk taking. Prudential and supervisory
mies has helped boost European banks equity prices. actions should be taken if policy stimulus leads to an
However, as assessed in the October 2016 GFSR, a increase in debt-financed investment and rising corpo-
cyclical recovery will likely be insufficient on its own rate vulnerabilities. Tax reforms that reduce incentives
to restore the profitability of persistently weak banks. for debt financing could help attenuate risks of a fur-
Although many banks face profitability challenges, this ther buildup in leverage, and possibly even encourage
is particularly true for domestic banks, which are most firms to lower existing tax-advantaged leverage.
exposed to their home economies: almost three-quar- In Europe, further actions should be taken to
ters of these banks had weak returns in 2016 (defined address bank profitability and legacy challenges. Banks
as return on equity of less than 8 percent). This report have the primary responsibility for developing sustain-
examines the system-wide structural features that are able earnings by tackling business model problems
compounding profitability challenges. One structural through consolidation, branch rationalization, and
challenge is overbanking, which varies by nature investment in technology to increase medium-term
and degree from country to country. Some examples efficiency. Encouragingly, supervisors are increasingly
emphasizing the examination of bank business models consider the impact and unintended consequences of
in their supervisory frameworks. To determine weak reforms, such a review should not unravel the broad
links in banking systems with significant asset quality improvements achieved in buttressing the resilience of
challenges, consideration could be given to targeted the global financial system.
asset quality reviews for banks that have not undergone This report also includes two thematic chapters
such an exercise. Regulators should then take action to analyzing the long-term implications of low growth
resolve unviable institutions to remove excess capacity. and low interest rates for financial intermediation, and
Authorities should also focus on removing system-wide the ability of country authorities to influence domestic
impediments to profitability, including addressing financial conditions in a financially integrated world.
nonperforming loans and developing frameworks that
accelerate recovery.
Emerging market economies should address A Long Period of Low Growth and Low
domestic vulnerabilities to enhance their resilience to Interest Rates Would Challenge Financial
external shocks. They should seek to preserve financial Intermediation
stability by taking further steps to strengthen supervi- Advanced economies have experienced a pro-
sion and bank governance while maintaining a robust longed episode of low interest rates and low growth
macroprudential toolkit. Bank regulators should closely since the global financial crisis. From a longer-term
monitor vulnerabilities in countries with wide net perspective, real interest rates have been on a steady
foreign-currency positions or foreign-currency matu- decline over the past three decades. Despite recent
rity gaps. Policymakers should focus on strengthening signs of an increase in longer-term yields, particularly
the health of corporates and the banking system by in the United States, Japans experience suggests that
proactively monitoring and reducing vulnerabilities an imminent and permanent exit from low rates is
and improving restructuring mechanisms. In China, not necessarily guaranteed, especially in view of the
although the authorities have recognized the urgent prevalence of slow-moving structural factors, such
need to deleverage the financial system and have as demographic aging in many advanced economies.
undertaken substantive corrective measures, supervi- Chapter 2 analyzes the potential long-term impact of
sory attention should concentrate on banks emerg- a scenario of sustained low growth and low real and
ing risks, especially fast asset growth among smaller nominal rates for the business models of banks, insur-
banks, increasing reliance on wholesale funding, and ers, and pension funds and for the products offered
risks from interconnections between shadow products by the financial sector. It finds that yield curves would
and interbank markets. But staving off further bouts likely flatten, lowering bank earningsparticularly of
of market instabilityand ultimately, macro instabil- smaller, deposit-funded, and less diversified institu-
itywill require measures to address the policy tension tionsand presenting long-lasting challenges for life
between maintaining a high level of growth and the insurers and defined-benefit pension funds. If bank
need for deleveraging. deposit rates cannot drop (significantly) below zero,
The postcrisis reform agenda has strengthened over- bank profits would be squeezed even further. Smaller,
sight of the financial system, raised capital and liquid- deposit-funded, and less diversified banks would be
ity buffers of individual institutions, and improved hurt most. As banks reach for yield, new financial
cooperation among regulators. Caution is needed when stability challenges would arise in their home and host
considering any future regulatory rollback. While regu- markets.
lation is never costless, neither is its removal; weaken- More generally, a low-for-long interest rate envi-
ing regulatory standards comes at the cost of higher ronment, driven by population aging, rising longevity,
financial stability risks. Decisions to opt out of mutu- and stagnation in productivity, could fundamentally
ally established regulations in an uncoordinated or uni- change the nature of financial intermediation. For
lateral manner could result in financial fragmentation example, credit demand would likely be lower in this
and could threaten to reignite a race to the bottom in scenario, whereas household demand for transaction
regulatory standards. Completing the regulatory reform services would likely rise. Consequently, bank busi-
agenda is vital to ensure that weaknesses are addressed ness models in advanced economies may evolve toward
and to reduce uncertainty. Although there is scope to fees-based and utility banking services. Demographic
changes would also increase demand for health and indices that make it possible to compare a large set of
long-term-care insurance, and low asset returns would advanced and emerging market economies. It finds that
accelerate the transition to defined-contribution private global financial conditions account for 20 to 40 percent
pension plans. Demand would weaken for guaranteed- of the variation in countries domestic financial condi-
return, long-term savings products offered by insurers, tions, with notable differences among economies. The
and it would strengthen for passive index funds offered importance of this global factor does not, however, seem
by asset managers. Policies could help ease the adjust- to have increased much over the past two decades.
ment to such an environment. In general, prudential Despite the significant role of global financial
frameworks would need to provide incentives to ensure shocks, countries seem to be able to influence their
longer-term stability instead of falling prey to demands own financial conditions to achieve domestic objec-
for deregulation to ease short-term pain. tivesspecifically, through monetary policy. But
because domestic financial conditions react strongly
and rapidly to global financial shocks, countries may
Policymakers Challenged to Effectively Steer find it difficult to implement timely policy responses.
Domestic Financial Conditions amid Increased Emerging market economies, which are more sensi-
International Financial Integration tive to global financial conditions, should prepare for
Chapter 3 shows that countries can retain influence tighter external financial conditions. Governments can
over their domestic financial conditions in a globally promote domestic financial deepening to enhance resil-
integrated financial system. Although greater financial ience to global financial shocks. In particular, develop-
integration can complicate the management of domes- ing a local investor base, as well as fostering greater
tic financial conditions, it need not result in a loss equity- and bond-market depth and liquidity, can help
of control. The chapter develops financial conditions dampen the impact of such shocks.
The following remarks were made by the Chair at the conclusion of the Executive Boards discussion of the
Fiscal Monitor, Global Financial Stability Report, and World Economic Outlook on April 4, 2017.
E
xecutive Directors broadly shared the assess- of financial regulation, which could spur excessive
ment of global economic prospects and risks. risk taking; and a potential rise in protectionist and
They welcomed the positive developments inward-looking policies.
since the second half of2016: global eco- Against this backdrop, Directors emphasized
nomic activity has accelerated, headline inflation has the need for comprehensive, consistent, and well-
generally risen following a rebound in commodity communicated policy actions to achieve strong, sus-
prices, and financial market sentiment has strength- tainable, and balanced growth; enhance resilience; and
ened. Global growth is expected to pick up further ensure that the benefits of economic integration and
in201718, reflecting a stronger-than-expected technological progress are shared more widely. Policy
recovery in many advanced economies and projected priorities vary across individual economies depend-
higher growth in many emerging market and develop- ing on cyclical positions, structural challenges, and
ing economies, including from improved conditions vulnerabilities facing them. Multilateral cooperation is
in several commodity exporters. However, growth as essential as ever to complement national efforts as
momentum is still modest and downside risks continue well as tackle common challenges, including preserv-
to dominate, with heightened policy uncertainty and ing a rules-based, open trading system; ensuring a level
persistent structural headwinds. Directors underscored playing field in international taxation; and strengthen-
the importance of using all policy tools at the national ing the global financial safety net. Multilateral efforts
level and strengthening multilateral cooperative efforts are also needed to address the withdrawal of corre-
to sustain a stronger recovery, ward off downside risks, spondent banking relationships and the refugee crisis.
safeguard hard-won gains in global integration and Both deficit and surplus countries should implement
financial stability, and promote inclusion. appropriate policies to reduce persistent global excess
Directors noted that the balance of risks remain imbalances.
tilted to the downside, especially over the medium Directors agreed that a common challenge across
term. In advanced economies, while the ongoing advanced economies is to boost potential output,
cyclical recovery is encouraging, output remains below through fiscal and structural reforms that target
potential and unemployment above precrisis levels in country-specific priorities, including to upgrade public
many countries. Population aging, low labor pro- infrastructure where needed; improve labor force
ductivity growth, and crisis legacies are weighing on participation and skills; eliminate product market
growth potential. In emerging market and developing distortions; and reform corporate income taxation to
economies, medium-term prospects are closely linked promote private investment, research and development,
to developments in commodity markets, global finan- and resource reallocation to productive areas. Resist-
cial conditions, the ongoing economic transition in ing a retreat from global economic integration must
China, and progress in resolving domestic imbalances also be part of the agenda to secure strong, sustainable
and structural challenges in some economies. global growth.
Directors observed that elevated political and policy Directors saw a need to tackle the adverse side
uncertainties in many parts of the world pose diffi- effects of technological change and trade integra-
cult challenges to the economic outlook and financial tion with appropriate policies. In this context, they
stability. They cited, among other things, faster-than- noted the staffs finding that technological progress
expected normalization of interest rates; a rollback appears to be the main factor explaining the decline
in labor income share in advanced economies, while macroeconomic adjustment. For many countries,
trade integrationwhich has contributed to signifi- priorities include proactively monitoring vulnerabili-
cant improvements in living standards and poverty ties and addressing weaknesses in the corporate and
reduction around the worldseems to be the domi- banking sectors, improving corporate governance, and
nant driver in emerging market economies. Directors reducing infrastructure bottlenecks and barriers to
stressed that the design of inclusive fiscal policies, entry. These should be complemented by measures to
such as transfer and tax instruments, should strike the enhance resilience, such as developing a local inves-
right balance between promoting redistribution and tor base, fostering depth and liquidity in the equity
maintaining incentives to invest and work. They also and bond markets, and upgrading the tax system to
emphasized the importance of improving education, promote efficient use of resources.
training, health services, social insurance, and pension Directors stressed that solidifying improvements
systems. In some cases, active labor market policies in financial stability and market expectations requires
could be an effective tool in the short term. concerted efforts across countries. In the United States,
Directors agreed that strengthening the recovery where tax reform and financial deregulation could have
remains a priority in many economies, requiring sup- a significant impact on the financial and corporate
port from both monetary and fiscal policies, combined sectors globally, authorities should be vigilant to the
with growth-enhancing structural reforms. Where core increase in leverage and deterioration in credit quality
inflation is persistently low and/or the risk of deflation and should take preemptive measures against exces-
remains tangible, unconventional monetary policies sive risk taking. In Europe, where important progress
remain appropriate to support economic activity and has been achieved, further efforts are still needed to
lift inflation expectations, while their potential negative adjust bank business models, facilitate the disposal of
consequences on financial stability should be closely nonperforming loans, and remove structural impedi-
monitored. Fiscal policy can play an important role, ments to bank profitability. In China, where major
particularly when monetary policy has become less reforms to the financial system are taking place, special
effective. Directors agreed that, as a general principle, attention should be paid to the rapid growth in assets
fiscal policy should be countercyclical, be growth among smaller banks, the increasing reliance on whole-
friendly, and promote inclusion, anchored in a credible sale funding, and the close interconnections between
medium-term framework that ensures debt sustainabil- shadow products and interbank markets. At the global
ity. Depending on country-specific circumstances in level, completing the regulatory reform agenda remains
terms of economic slack, fiscal space, and debt levels, important, and a rollback of regulatory standards
policy choices range from discretionary fiscal sup- should be resisted.
port to budget recomposition and rebuilding of fiscal Directors observed that commodity-exporting low-
buffers. income developing countries have faced a difficult
Directors concurred that, while emerging market adjustment process since the commodity cycle turned
and developing economies can retain influence over in2014. In light of rising debt and weaker external
their domestic financial conditions, many could face positions in several of these economies, Directors
elevated risks that arise from external negative spill- called for intensified policy efforts to mobilize rev-
overs, including a sudden reversal of market sentiment enue, improve tax administration, enhance spending
and sharp volatility in capital flows and exchange rates. efficiency, and contain the buildup of debt. For many
Directors urged policymakers in these countries to be diversified countries, the priorities are to build fiscal
prepared for less favorable external conditions. Specifi- buffers while growth remains relatively strong and to
cally, it will be critical to maintain sound policies and achieve a better balance between meeting social and
strong frameworks, including exchange rate flexibility developmental needs and securing debt sustainability.
and a robust macroprudential toolkit, while capital A common challenge across all low-income developing
flow management measures may be used temporarily countries is to maintain progress toward attaining their
as warranted, though not as a substitute for warranted sustainable development goals.
Financial Stability Overview economic and financial stagnation have led to a shift
Financial stability has improved since the October 2016 in consensus and market-implied expectations toward
Global Financial Stability Report (GFSR). Growth is higher growth, inflation, and long-term interest rates.
gaining momentum and reducing macroeconomic risks, Reflation expectations have taken hold across advanced
rekindling hopes for reflation. Rising equity prices and economies (Figure1.3).
steeper yield curves have mitigated some of the negative Against this stronger economic backdrop, risk
side effects of low interest rates for banks and insurance appetite has increased, as reflected in more buoyant
companies. Emerging market risks remain elevated but investor confidence (Figure1.2, panel5). Market
unchanged, as recovering commodity prices and modest and liquidity risks have eased from elevated levels
deleveraging in some corporate sectors are offset by higher as risk premiums have fallen and volatility remains
external financing risks and rising financial vulner- subdued. These trends in market indicators have been
abilities in China. Despite these improvements, there a global phenomenon, starting last September and
are new downside risks and uncertainties around the accelerating following the U.S. elections (Figure1.3,
policy outlook. A key risk is that U.S. policy imbal- panel3). Expectations for policy stimulus have also
ances could lead to tighter-than-expected financial contributed to a stronger dollar and higher nominal
conditions and a rise in volatility and risk aversion. A and real U.S. Treasury security yields, spilling over to
global shift toward protectionism could adversely affect other advanced economy bond markets. Steeper yield
trade and global growth. Thus, anchoring stability will curves have helped banks enhance profitability, while
depend heavily on policy choices at the national and tighter corporate bond spreads, low rates, and ample
global levelsit is crucial to get the policy mix right. market access have reduced refinancing risks, lead-
ing to a reduction in credit risks. Although emerging
market economies have continued to enhance their
resilience, higher inflation volatility in some countries
Financial Stability Is Advancing
and rising financial vulnerabilities in China have left
Better-than-expected incoming data and gathering emerging market risks unchanged.
growth momentum, as outlined in the April 2017 Looking ahead, U.S. policy proposals under
World Economic Outlook (WEO), have reduced near- discussion aim to increase business confidence and
term macroeconomic risks (Figures 1.1 and 1.2). Hopes investment, and the nonfinancial corporate sector is
for reflation have risen, as monetary and financial well positioned to benefit. But rising corporate leverage
conditions remain highly accommodative, and antic- may challenge the capacity of some firms to expand
ipated U.S. fiscal measures and other reforms are investment without increasing stability risks. Growing
expected to bolster growth. Reduced concerns about signs of stretched valuations and the outperformance
of certain sectors exposed to potential fiscal stimulus
Prepared by staff from the Monetary and Capital Markets Depart-
ment (in consultation with other departments): Peter Dattels (Deputy measures raise the risk that valuations may reflect
Director), Matthew Jones (Division Chief ), Paul Hiebert (Advisor), overestimations of the potential benefits from policy
Ali Al-Eyd (Deputy Division Chief ), Will Kerry (Deputy Division initiatives and underestimations of downside risks (Fig-
Chief ), Sergei Antoshin, Magally Bernal, John Caparusso, Sally
Chen, Yingyuan Chen, Fabio Cortes, Dimitris Drakopoulos, Martin
ure1.4). Policies should aim to enhance the effective-
Edmonds, Jesse Eiseman, Jennifer Elliott, Rohit Goel, Thomas ness of proposed measures while safeguarding against
Harjes, Sanjay Hazarika, Geoffrey Heenan, Dyna Heng, Henry the excesses of financial risk and market stability. These
Hoyle, Nigel Jenkinson, David Jones, Robin Koepke, Tak Yan Daniel
trade-offs and policies are examined in the section
Law, Yang Li, Lilit Makaryan, Rebecca McCaughrin, Aditya Narain,
Vladimir Pillonca, Thomas Piontek, Luca Sanfilippo, Juan Sol, Ilan Is the U.S. Corporate Sector Ready to Accelerate
Solot, Narayan Suryakumar, Francis Vitek, and Jeffrey Williams. ExpansionSafely?
European bank equity prices have risen on opti- and raising credit and funding risks for banks as the
mism about a cyclical upturn in the economy and external environment deteriorates, which would expose
some further steps toward resolving weak banks. vulnerabilities (Box1.1). A shift toward protectionist
However, a cyclical recovery is unlikely to be sufficient policies in advanced economies could adversely affect
to restore the profitability of persistently weak banks, global trade and growth, capital flows, and market
and more needs to be done to improve resilience. The sentiment, resulting in adverse spillovers to emerging
system-wide structural impedimentscharacterized by markets. Many emerging market economies would face
operational inefficiencies, weak business models, inef- rising vulnerabilities in their weakest banks as a result
ficient allocation of credit, excess capacity, and a large of asset quality and provisioning challenges following
legacy of bad debtpose challenges, particularly for long credit booms that facilitated rising corporate
domestically oriented banks. Large international banks sector leverage. Emerging market resilience is assessed
are also affected by these system-wide challenges, and against this increasingly uncertain global policy mix in
unless these impediments are removed, business model the section Emerging Market Economies Face Trying
restructuring alone is likely to be insufficient. More Times in Global Markets.
systematic and comprehensive policies are needed to
address these profitability and legacy challenges and
to reduce financial stability risks, as discussed in the Getting the Policy Mix Right
section European Banking Systems: Addressing Struc- Given these challenges, securing and building on
tural Challenges. improvements in financial stability and validating
optimistic market expectations will require concerted
and careful efforts by policymakers at the national and
Policy Uncertainty Is a Key Downside Risk global levels. Policymakers need to adjust the policy
Despite these improvements in financial stability, mix to deliver a stronger path for long-term and inclu-
elevated political and policy uncertainty pose signif- sive growth while avoiding politically expedient but
icant challenges. In the United States, policies could ultimately counterproductive inward-looking policies.
increase fiscal imbalances and raise global risk premi- Furthermore, the potential for a broad rollback of
ums (see the April 2017 WEO and Fiscal Monitor). financial regulationsor a loss of global cooperation
Such an outcome could generate negative spillovers could undermine hard-won gains in financial stability
to emerging markets, reigniting capital outflows (Box1.2).
Figure 1.2. Global Financial Stability Map: Assessment of Risks and Conditions
(Notch changes since the October 2016 Global Financial Stability Report)
1. Macroeconomic risks have declined, driven by improving 2. Emerging market risks remain elevated, as higher ination volatility
economic activity and lower ination risks. offsets improvements in the corporate sector and external nancing.
2 1
0 Unchanged
0
1
Lower risk
2
Lower risk
3 1
Overall Sovereign Ination or Economic Economic Overall Fundamentals Volatility Corporate External
(7) credit deation risks activity uncertainty (10) (4) (2) sector nancing
(1) (1) (4) (1) (2) (2)
3. Credit risks have declined amid improvement in banks and the 4. Monetary and nancial conditions are unchanged, as tighter
corporate sector. monetary policies are offset by easier nancial conditions.
1 1
Higher risk
Easier
0
Unchanged
0
2 1
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)
5. Risk appetite has strengthened as a result of improved condence 6. Market and liquidity risks have moderated from an elevated
and gains in risk assets. level against the backdrop of better liquidity conditions.
4 0
3
Higher risk appetite
2
1
1
0
Lower risk 2
1
2 Lower risk appetite
3 3
Overall Asset Relative asset Emerging Uncertainty Overall Liquidity and Volatility Market Asset
(4) allocations returns markets (1) (11) funding (2) positioning valuations
(1) (1) (1) (5) (2) (2)
Market expectations for the U.S. economy and monetary policy ... and hopes are rising for reation across advanced economies ...
normalization have improved ...
1. Consensus Forecasts for End-2017 U.S. 10-Year 2. Ten-Year Ination Compensation
Treasury Yield (Cumulative breakeven yield change in basis points)
(Probability density)
Germany Japan United Kingdom United States
2.0 120
100
After U.S. election
(Mar. 31, 2017) 80
1.5
Before U.S. election 60
(Nov. 6, 2016) 40
1.0 20
0
20
0.5
40
60
0.0 80
0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 Jan. Apr. Jul. Oct. Jan. Apr.
Percent 2016 16 16 16 17 17
... generating market optimism and a compression in volatility across a number of global markets.
3. Financial Market Risk Dashboard
September 30, 2016 March 31, 2017 Range between September 30, 2016, and March 31, 2017
German Bund (10-year)
JGB (10-year)
EMBI global (yield)
GBI-EM (yield)
S&P 500
S&P 500 nancials
Euro Stoxx 50
Euro STOXX nancials
Japan equity
MSCI EM
China H-shares
Mexico equity
EU investment grade
U.S. investment grade
EM investment grade
VIX
V2X
China H-shares
Mexico equity
U.S. 1y10y swaptions
EU 1y10y swaptions
Japan 1y10y swaptions
MOVE
EURUSD 6-month
USDJPY 6-month
USDCNH 6-month
USDMXN 6-month
Commodity index
U.S. Treasuries (10-year)
EU high-yield
U.S. high-yield
EM high-yield
Rates Equity prices Credit spreads Equity volatility Rate volatility FX volatility
Despite greater policy uncertainty, implied volatility has declined to ... while U.S. equity valuations have become increasingly overvalued.
multiyear lows ...
1. Policy Uncertainty and Implied Equity Volatility 2. S&P 500 Index and Price-to-Earnings Ratio
350 Global economic policy uncertainty (index, left scale) 50 50 Shiller price-to-earnings ratio (multiples, left scale) 2,500
VIX (percentage points, right scale) S&P 500 (index, right scale)
300 40 40 2,000
250
30 30 1,500
200
20 20 1,000
150
Some sectors have beneted disproportionately from the interplay of ... while expectationsnot actual earningsare driving valuations in
potential policies ... sectors that would benet from stimulus.
3. Performance of U.S. Equity Industry Subsectors 4. Valuation of U.S. Equities Exposed to Policy Shifts
(Percent change since U.S. election) (Percent change since U.S. election)
35
Price
Deregulation (+28%) Banks (+38%) Twelve-month-forward P/E 30
Oil exploration (+13%) Infrastructure Twelve-month-forward EPS
spending
(+26%) 25
Consumer Autos and
trucking
Internet services nance (+12%) 20
(+13%) (+20%)
Health care (+18%) Engineering (+21%)
Consumable fuels Materials (+13%)
15
(+11%) Equipment (+48%)
Pharmaceuticals
Tech Capital
(+10%) 10
hardware goods
(+32%) (+11%)
Corporate tax Retail (+9%) 5
cuts (+13%) or Logistics (+12%)
repatriation Adverse trade
Semiconductors (+19%)
(+14%) policies (+21%) 0
Auto parts (+9%)
Deregulation Infrastructure Adverse Repatriation Corporate S&P 500
spending trade tax cuts
policies
Sources: Bloomberg L.P.; Haver Analytics; JPMorgan Chase & Co.; and IMF staff estimates.
Note: In panels 3 and 4, Corporate tax cuts = companies with high effective tax rates and domestic revenue exposures; Repatriation = companies with the largest
total cash balances held by foreign subsidiaries; Adverse trade policies = trade-linked importers, outsourcers, and logistics rms; Infrastructure spending = rms that
generate a signicant portion of revenue from civil construction activities and revenue from within the United States; Deregulation = companies in sectors likely to
experience regulatory relief, such as oil and gas, banks, consumer nance, and autos and trucking. EPS = earnings per share; P/E = price-to-earnings ratio; S&P =
Standard and Poors.
Figure 1.5. United States: Policies under Discussion and Financial Stability Risks
Key elements of policy stimulus proposals: Financial stability risks in the corporate sector:
Corporate sector taxation Excessive nancial risk taking
Potential reduction in corporate tax rate
Interest deductibility/expensing investment Corporate leverage peaking
Incentives for repatriation Credit cycle maturing
Other (infrastructure spending, deregulation, trade, and other Heightened vulnerability to default risk
policies)
Equity markets
Financial risk Economic (cost of equity nance Internal
taking = risk taking = and issuance) funds
Acquisition of Capital Debt markets (income,
nancial assets, spending (cost of borrowing operating
M&A, and share (including and leverage limits) cash, and
buybacks/ R&D) buffers)
Banks
dividends
(cost and availability
of loans)
Is the U.S. Corporate Sector Ready to Accelerate assets have rallied, and financial market sentiment has
ExpansionSafely? improved in anticipation of the stimulative elements
U.S. policies under discussion aim to increase economic of the policies being discussed. Such reforms could
growth. Healthy corporate balance sheets will be essential lead to a direct boost to the cash flow of firms and an
to facilitate the necessary increase in economic risk taking. indirect boost as a result of more favorable financial
Although the corporate sector has considerable balance market sentiment.
sheet capacity to support an expansion, overall corporate The U.S. corporate sector will be a central conduit
leverage is elevated, leaving some segments vulnerable to for such policies to gain traction and stimulate
higher financing costs. The sectors responsible for the most economic activity (Figure1.5). Tax policy reforms,
capital spending in recent years, such as energy, real estate, in particular, harbor the potential to boost economic
and utilities, may be challenged to expand investment risk takingin the form of corporate capital spend-
without resorting to further debt financing. Policies ingin two key ways. First, a cut in the statutory
should maximize the economic effectiveness of proposed tax rate for corporations would directly boost corpo-
measures while safeguarding against excesses of financial rate internal funds. The cash flow boost from such a
risk taking that could undermine financial stability. tax cut could be amplified by policies to encourage
the repatriation of foreign earnings. Second, elimi-
nating interest deductibility of debt and immediate
U.S. Policies under Discussion and Economic Risk Taking expensing capital expenditure could reduce the
Policies under discussion by the new U.S. admin- debt bias inherent in corporate financing decisions,
istration in the areas of tax reform and deregulation putting equity finance on a more equal footing with
could significantly boost economic growth. Risk debt financing.
Taken together, tax policy reforms under discussion Figure 1.6. United States: Business Confidence and Economic
could lay the groundwork for the corporate sector to Risk Taking
support higher economic growth. Surveys capturing
Business optimism has spiked ...
business sentiment have jumped to highs not seen in
more than a decade (Figure1.6, panel1), suggesting 1. Small Business Optimism Index
(Feb. 2017 = 100)
an expected rise in corporate capital spending. This
105
could help close a gap in corporate capital spending
relative to higher historical growth by almost 2per- 100
centage points of assets, or some $750billion a year 95
(Figure1.6, panel2).
90
85
Is the Corporate Sector Well Poised to Expand Economic Recessions
Risk Taking? 80
Utilities
Energy
Real estate
Telecom-
munications
Materials
Consumer
discretionary
Industrials
Information
technology
Consumer
staples
Total
Utilities
Energy
Real estate
Telecom-
munications
Materials
Consumer
discretionary
Industrials
Information
technology
Consumer
staples
Heatlh care
Cash constrained Cash abundant
Cash windfalls from repatriation would likely accrue to cash Three cash constrained sectors account for almost half of capital
abundant sectors. expenditure.
3. Unrepatriated Income, by Sector 4. Capital Expenditures by S&P 500 Firms, by Sector
(Total in U.S. dollars, sectoral shares in percent of total) (Share of total assets, 201216 average)
Utilities Energy
2% 8% Consumer Health care
Health care
1% Real estate staples 4%
28% 1% Telecommunications 13% Utilities
7%
3% Materials Information
8% Consumer technology 12%
discretionary
Cash and
investments: 3% Industrials Energy
~$1.3 trillion Industrials 11% 27%
Consumer
staples 9% Consumer
38% Information discretionary 11%
technology 4% Real estate
Materials 4% 7%
$2.2 trillion Telecom-
munications
Debt has been used to finance both economic and financial risk taking.
5. Cash Flow Decomposition for S&P 500 Firms, by Sector
(201216 average)
Total Telecom- Consumer Information Consumer Health
S&P 500 Utilities Energy Real estate munications Materials discretionary Industrials technology staples care
(1,896) (106) (210) (32) (99) (61) (256) (179) (495) (166) (289)
Sources of financing
303 29 20 6 67 25 60
27 7 58
179
(billions of U.S. dollars)
25 34
11 26 92 72 185 48 91
940 877 36 44 33 155 92 289
177 28 109 183
85
Sources: Bloomberg L.P.; S&P 500 company reports; Securities and Exchange Commission; and IMF staff estimates.
Note: See Figure 1.5 for more on the concepts underlying charts in panel 5. Capex = capital expenditures; S&P = Standard and Poors.
While positive effects of tax stimulus on cash flow coincide with higher corporate debt (Figure1.8,
could be considerable, they would be insufficient panel6)particularly if additional share buybacks are
for firms in a number of cash-constrained sectors to financed through debt.
finance increased capital spending. These sectors There has been a stronger reliance on debt financing
energy, utilities, and real estateare particularly as the credit cycle entered a mature phase. Corporate
important as they have contributed to nearly half of credit fundamentals have started to weaken (Fig-
overall capital spending among S&P 500 firms over ure1.9, panel1), creating conditions that have histor-
the past few years (Figure1.7, panel4). The cash ically preceded a credit cycle downturn (Figure1.9,
flow boost from a cut to the statutory tax rate may be panel2). Asset qualitymeasured, for example, by
insufficient to spur the nearly $140billion needed to the share of deals with weaker covenantshas dete-
boost capital expenditure to the level prevailing before riorated. At the same time, a rising share of rating
2000 (Figure1.7, panel1). Adding in changes to tax downgrades suggests rising credit risks in a number of
treatments of interest expense and capital expenditures, industries, including energy and related firms in the
along with repatriation, would attenuatebut likely context of oil price adjustments and also in capital
not eliminatefinancing needs for these sectors. goods and health care.
Perhaps more important, cash flow from tax reforms Also consistent with this late stage in the credit
may accrue mainly to sectors that have engaged in cycle, corporate sector leverage has risen to elevated
substantial financial risk taking. Such risk taking is levels. Median net debt across S&P 500 firms
associated with intermittent large destabilizing swings which collectively account for about one-third of
in the financial system over the past few decades (Fig- the $36trillion economy-wide corporate sector
ure1.11). It has averaged $940billion a year over the balance sheetis close to a historic high of more
past three years for S&P 500 firms, or more than half than 1times earnings (Figure1.9, panel3). A
of free corporate cash flow (Figure1.7, panel5). At the look beyond the S&P 500, at a broader set of nearly
sectoral level, such spending has been strongest in the 4,000firms accounting for about half of the econ-
health care and information technology sectorswhere omy-wide corporate sector balance sheet, suggests
purchases of financial assets, mergers and acquisitions, a similar rise in leverage across almost all sectors
and net payouts have been capturing more than half of to levels exceeding those prevailing just before the
free resources since 2012amounting collectively to global financial crisis (Figure1.9, panel4). Leverage
nearly $500billion a year. is uneven, though: the upward drift is limited by low
debt in cash-rich sectors such as information technol-
ogy, but debt is very high in the energy, real estate,
Where Are the U.S. Corporate Sectors Vulnerabilities? and utilities sectors, ranging between four and six
The health of the corporate sector will be central times earnings.
not only to the economic effectiveness of fiscal policy
reforms but also for financial stability (Figure1.5).
While U.S. corporate sector balance sheets are strong High Leverage Combined with Tighter Borrowing
in aggregate, cash flow has tapered recently as corpo- Conditions Could Affect Financial Stability
rate profits have come off peaks (Figure1.8, panels1 As leverage has risen, so too has the proportion of
and 2). income devoted to debt servicing, notwithstanding
The corporate sector has tended to favor debt low benchmark borrowing costs (Figure1.10, panel1).
financing, with $7.8trillion in debt and other lia- Although the absolute level of debt servicing as a pro-
bilities added since 2010 (Figure1.8, panel3). Bank portion of income is low relative to what it was during
lending to the corporate sector has continued to the global financial crisis, the 4percentage point rise
recover and could well rise further in response to more has brought it to its highest level since 2010, which
favorable market valuations (Figure1.8, panel4). In leaves firms vulnerable to tighter borrowing conditions.
contrast, equity finance has traditionally been out- The average interest coverage ratioa measure of the
stripped by share buybacks and has recently leveled ability for current earnings to cover interest expenses
off (Figure1.8, panel5). A drop in the cost of equity has fallen sharply over the past two years. Earnings
capital may stimulate equity financing, but it could have dropped to less than six times interest expense,
Figure 1.8. United States: Corporate Internal Funds and External Sources of Finance
Corporate cash holdings are tapering ... ... as prots recede from a high level.
1. Corporate Cash Holdings on Balance Sheet 2. Corporate Prots
(Percent of GDP)
Cash (trillions of U.S. dollars, left scale)
Cash (percent of assets, right scale)
1.8 5.0 13
1.6 Recessions Recessions
4.5 12
1.4
11
1.2 4.0
1.0 10
3.5
0.8 9
0.6 3.0
8
0.4
2.5 7
0.2
0.0 2.0 6
1980 84 88 92 96 2000 04 08 12 16 1960 70 80 90 2000 10
Net equity nancing has been falling the past four decades, as debt A sharp improvement in bank equity valuations may portend stronger
nance has continued to rise. willingness to lend.
3. Corporate Liabilities and Net Equity Issuance 4. Bank Equities and Corporate Lending
(Percent of assets)
S&P 500 bank stocks
12 25 (price index, right scale) 140
10 Increase in debt and other 20 Lending to nonnancial corporate 120
liabilities 15 rms (year-over-year
8 $7.8 trillion percent change, left scale)
10 100
6 5 80
4 0
5 60
2
10 40
0
15
2 20
20
Negative net equity issuance (gross issuance minus share buybacks) $3.0 trillion
4 25 0
1980 84 88 92 96 2000 04 08 12 16 2000 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16
Gross equity issuance has abated, despite favorable valuations ... ... while a lower cost of equity capital could boost business investment
(and, eventually, debt).
5. Corporate Sector Gross Equity Issuance 6. Illustrative Impacts of Improving Equity Sentiment
(Billions of U.S. dollars, unless otherwise stated) (Percent deviation from baseline)
300 30 6 2
Energy Real estate
Utilities Information technology 5
Health care Other 25 4 1
200 S&P 500 P/E (multiple,
3
right scale)
20 2 0
1
100
15 0 1
1
0 10 2 2
2000 02 04 06 08 10 12 14 16 2017 18 19 20 21 22 2017 18 19 20 21 22
Business investment Nonnancial corporate debt
Sources: Bloomberg L.P.; Dealogic; Federal Reserve; Morgan Stanley Capital International; Standard and Poors (S&P); Vitek 2017; and IMF staff estimates.
Note: In panel 4, the series for lending is lagged 12 months. Bank balance sheet improvements are modeled using an equivalent decrease in the share of bank capital
devoted to lending, phased in over ve years. Equity risk premium compression in panel 6 consists of a 10 percent increase in the real equity price, phased in over
one year. Shaded areas represent economic recessions. P/E = price-to-earnings ratio.
Deteriorating balance sheet fundamentals and credit conditions ... ... signal a late stage of expansion in the credit cycle.
1. Asset Valuations, Balance Sheet Fundamentals, and Credit Conditions 2. Stages of a Stylized Credit Cycle
(Unweighted average in percentile rank, normalized to zero)
3 Valuation August 2007March 2009 March 2009December 2009
Credit conditions Overvaluation Falling credit growth Companies try to shore up
2
Fundamentals Falling prots, rising leverage, balance sheets
weakened cash ow Weak credit growth
1 Credit undervaluation Falling prots and leverage
0 Downturn Repair Credit undervaluation
1
Increased leverage, declining ICRs July 2004August 2007 Expansion Recovery December 2009
Excessive risk taking August 2010
2 Increased risk taking
Recessions Rapid credit growth Improving momentum
Restrictive credit conditions Rising prots, rising leverage, Accelerating credit
3 large-scale M&A and capex Rising prots, falling leverage
Jan. Oct. Jul. Apr. Jan. Oct. Jul. Apr. Jan. Oct.
Credit overvaluation Valuations close to fair value
2000 01 03 05 07 08 10 12 14 15 and rising
Median corporate leverage among big rms has grown steadily and is Eight out of ten sectors witness an increase in leverage across a broad
close to a historical peak. set of rms.
3. Net Leverage of S&P 500 Companies 4. Net Leverage by Sector
(Ratio of net debt to EBITDA) (Ratio of net debt to EBITDA)
Median (all rms) Median (excluding energy) Mean (all rms)
2.5 8
7
2.0 200406 2016 6
Recessions
5
1.5 4
3
1.0 2
1
0.5 0
1
0.0 2
Overall
Energy
Real estate
Utilities
Consumer
discretionary
Consumer staples
Industrials
Materials
Telecommunications
Information
technology
1980 85 90 95 2000 05 10 15
Sources: Bloomberg L.P.; National Bureau of Economic Research; S&P Capital IQ; Thomson Reuters Datastream; and IMF staff estimates.
Note: 2016 estimates refer to the rst three quarters of the year, wherever full-year estimates are not available. Panel 1, Valuation = distress ratio, deviation in
high-yield bond spreads from fair value. Fundamentals = capital expenditures, interest coverage, leverage, liquidity, prot margins. Credit conditions = bank credit,
lending conditions, net bond issuance. Above zero represents an improvement in credit fundamentals (for example, high valuations, supportive credit conditions,
rising prots, ample liquidity). Below zero represents a deterioration (for example, excessive risk taking, reduced access to credit, high leverage, diminishing prots,
falling valuations) in fundamentals, credit conditions, and valuation. Shaded areas indicate economic recessions. Capex = capital expenditure; EBITDA = earnings
before interest, taxes, depreciation, and amortization; ICR = interest coverage ratio; M&A = mergers and acquisitions; S&P = Standard & Poors.
Figure 1.10. Debt Service, Interest Coverage Ratios, and Vulnerability to Higher Interest Rates
The debt service burden for the corporate sector as a whole has risen Interest coverage ratios have undergone a corresponding fall at the
strikingly despite low rates. rm level, particularly for smaller companies.
1. Corporate Debt Service and Interest Rates 2. Evolution of the Distribution of ICRs across Firms by Size
(Ratio of EBIT to interest payments)
10 48 8
Recessions Recessions
9 46 7
8 6
44
7 5
Debt service ratio 42
6 (percent of income, 4
right scale) 40 Average
5 Smallest 25 percent by assets 3
Prime lending
rate (percent, 38 Smallest 5 percent by assets
4 2
left scale) ICR = 2.0x
3 36 1
1999
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
1996
97
98
99
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
Market pricing of corporate risk has decoupled from the decline in Higher nancing costs could signicantly weaken rms interest
interest coverage ratios. coverage ratios ...
3. High Yield Option-Adjusted Corporate Spread and Average 4. Average Interest Coverage Ratio
Interest Coverage Ratios across Firms (Ratio of EBIT to interest payments)
2,500 High-yield spreads 4.0 7.5
(basis points, Recessions
Recessions left scale) 4.5 7.0
2,000
Mean ICR 5.0 6.5
(ratio, right scale,
1,500 5.5 6.0
inverted)
1996
98
2000
02
04
06
08
10
12
14
16
Scenario
... resulting in a growing set of rms at risk of default. The share of challenged rms has risen in the energy, real estate,
and utilities sectors.
5. Percentage of Challenged Firms 6. Evolution of Challenged Firms, by Sector
(Percent of total assets) (Share of total rms with ICR < 2)
Energy Consumer discretionary Information technology
Weak (ICR < 1) Vulnerable (1 ICR < 2) Real estate Industrials Others
Utilities Materials
30 100
22.1
Recessions ($3.9 trillion) 80
20.1
20
60
40
10
20
0 0
1996
97
98
99
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
1996
97
98
99
2000
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
Scenario
Sources: Bank for International Settlements; Bloomberg L.P.; S&P Capital IQ; and IMF staff estimates.
Note: 2016 calculations reect the rst three quarters of the year, wherever full year estimates are not available. Shaded areas indicate economic recessions. EBIT =
earnings before interest and taxes; ICR = interest coverage ratio.
close to the weakest multiple since the onset of the risk taking, abetted by a broad rollback of regulations.
global financial crisis (Figure1.10, panel2). Histori- Similarly, a tax holiday for offshore unremitted profits
cally, deterioration of the interest coverage ratio cor- in 2004, amid financial deregulation that started in the
responds with eventual widening in credit spreads for 1990s, was followed by a surge in financial risk taking.
risky corporate debt (Figure1.10, panel3). Declines In general, increased financial risk taking is associated
in the interest coverage ratio have been concentrated with pronounced leverage cycles that gradually build
mostly in smaller firms, which may have less access to up and end abruptly in recessions, as for example in
capital market financing than their larger counterparts. both 2001 and 2008.
Under the adverse scenario in Scenario Box1.1 of Policymakers must balance the economic benefits
the WEO, an unproductive fiscal expansion could of policy stimulus and tax reform against broader
lead to a sharp rise in borrowing costs. Such a sharp policy considerations and guard against financial
rise in interest rates amid tepid earnings growth could stability risks. Authorities need to be vigilant to the
further compromise the ability of firms to service increase in leverage and deteriorating credit qual-
their debt (Figure1.10, panel4).3 Under this sce- ity. Tax measures now under discussion that reduce
nario, the combined assets of challenged firms could incentives for debt financing could help attenuate
reach almost $4trillion. The number of firms with risks of a further buildup in leverage and may even
very low interest coverage ratiosa common signal of encourage firms to unwind existing tax-advantaged
distressis already high: currently, firms accounting debt. Existing leverage and a deterioration in interest
for 10percent of corporate assets appear unable to coverage ratios may, nonetheless, still represent a risk.
meet interest expenses out of current earnings (Fig- Tighter financial conditions could lead to distress for
ure1.10, panel5). This figure doubles to 20percent the weak tail of firms, with losses borne by banks,
of corporate assets when considering firms that have life insurers, mutual funds, pension funds, and over-
slightly higher earnings cover for interest payments, seas institutions.
and rises to 22percent under the assumed interest To mitigate the financial stability risks, regulators
rate rise. should preemptively address any areas in which risk
The stark rise in the number of challenged firms has taking appears excessive. Additional financial pruden-
been mostly concentrated in the energy sector, partly tial and supervisory action could be deployed should
as a result of oil price volatility over the past few years. policy stimulus lead to an increase in debt-financed
But the proportion of challenged firms has broadened investment and a rise in medium-term corporate
across such other industries as real estate and utilities. vulnerabilities, acknowledging lags and limits to
Together, these three industries currently account for scope.4 The Comprehensive Capital Analysis and
about half of firms struggling to meet debt service Review stress-testing exercise is already being used to
obligations and higher borrowing costs (Figure1.10, identify where risks may have a meaningful impact on
panel6). the balance sheets of systemic banks. A case can also
be made for using stress testing to assess the risks to
nonbank financial intermediary balance sheets from
Policies Should Be Carefully Calibrated and Attuned to severe losses in nonfinancial corporate debt, taking
Stability Risks into account likely associated liquidity strains and
Historical experience suggests that financial risk correlated risks in related sectors (such as commercial
taking in the form of asset acquisition, mergers and real estate).
acquisitions, and net payouts often follows tax policy More generally, policymakers should resist efforts
changes (Figure1.11). Tax cuts in the United States to weaken bank regulatory requirements that reduce
in the 1980s coincided with an increase in financial resilience (Box1.2). Although there is room to
fine-tune existing regulations, policymakers should
3Calculations capture partial sensitivity of the interest coverage
guard against wholesale dilution or backtracking on
ratio to an interest rate shock, based on a scenario with tighter
financial conditions, assumed to pass through into higher effective the important progress made in strengthening the
interest rates based on an assumed loan maturity of five years. The
number of firms considered in the analysis ranges from 1,800to 4For instance, after bank regulators instituted leverage caps in
4,000, depending on the availability of historical information from 2013, growth in leveraged lending eased, but more aggressive risk
S&P Capital IQ data. taking was evident in capital-market-based financing.
Figure 1.11. United States: A Retrospective on Economic versus Financial Risk Taking
Past corporate tax initiatives have been associated with a limited increase in economic risk taking.
Balance Sheet Evolution of Nonnancial Corporate Sector
(Percent of assets)
10
5
Operating cash
0
Capital expenditure Economic risk taking
Uses of nancing
5
Financial risk taking,
including net payouts
10
Net payouts (dividends
and share buybacks) Purchases of nancial assets and M&A
15
1980 84 88 92 96 2000 04 08 12 16
resilience of the financial system, particularly at a uncertain global policy mix, policymakers should con-
time when balance sheet fundamentals are deterio- tinue to address corporate and bank vulnerabilities.
rating for U.S. companies. The successful comple-
tion of the global regulatory reform agenda is vital Emerging Market Economies: Resilience Tested
to ensuring that the global financial system is safe Faster Growth in Advanced Economies and Ongoing
and resilient and can continue to promote economic Adjustment in Emerging Market Economies Support
activity and growth. Resilience
The world economy is gaining speed, boosting the
Emerging Market Economies Face Trying Times appetite for risk, reinforcing the recovery in commod-
in Global Markets ity prices, and supporting the rebound in emerging
Emerging market economies have continued to enhance market economy asset prices. U.S. market interest rates
their resilience. Their macroeconomic outlook has have risen notably amid the improving outlook and
improved due to stronger growth and lower corporate expectations of fiscal stimulus and monetary tightening
leverage, alongside prospects for positive growth spill- in the United States. The recent episode of rising rates
overs from advanced economies. But overall financial has been marked by a combination of higher real yields
stability risks remain elevated because political and and increased inflation compensation, portending
policy uncertainty in advanced economies opens chan- stronger U.S. growthin contrast to some previous
nels for negative spillovers. A sudden repricing of risk periods of rising U.S. interest rates, such as during the
or a rise in protectionism could trigger capital outflows 2013 taper tantrum. During that period, rising U.S.
and hurt demand. This would exacerbate existing interest rates hit emerging market economies hard,
vulnerabilities in corporate sectors and raise risks in the particularly those with weak macroeconomic funda-
weakest banking systems. To ensure resilience against an mentals (Figure1.12, panels1 and 2).
Emerging market assets were hurt during the taper tantrum in May This time is different: a brighter U.S. outlook and reation support the
2013 as higher U.S. real yields did not signal higher U.S. growth. assets of emerging market economies.
1. U.S. Rates and Emerging Market Spreads 2. U.S. Rates and Emerging Market Spreads
(Cumulative basis point change; May 22, 2013 = 0) (Cumulative basis point change; Nov. 7, 2016 = 0)
120 U.S. 10-year breakeven ination rate U.S. 10-year breakeven ination rate 120
100 U.S. 10-year real yield U.S. 10-year real yield 100
80 EM sovereign spread EM sovereign spread 80
60 EM corporate spread EM corporate spread 60
40 40
20 20
0 0
20 20
40 40
60 60
10 8 6 4 2 0 +2 +4 +6 +8 +10 +12 +14 10 8 6 4 2 0 +2 +4 +6 +8 +10 +12 +14
Weeks Weeks
Emerging market external balances have improved since the taper Emerging market corporate leverage has moderated but still remains
tantrum, reinforcing positive nancial market sentiment. elevated, especially in Latin America.
3. Current Account and Foreign Reserves Adequacy, 4. Emerging Market Economy Corporate Leverage, 200716
Change 2012 to 2016 (Debt to equity, percent)
230 Both improved 100
China Asia excluding China
210 One improved Europe, Middle East, Latin America 90
Both worsened and Africa
(percent of ARA metric)
190 80
Reserve adequacy
Russia
170 Brazil 70
150
India Colombia 60
130 Hungary
Poland Malaysia
50
110 Turkey Chile Mexico
90 40
Indonesia
South Africa
70 30
8 6 4 2 0 2 4 6 2007 08 09 10 11 12 13 14 15 16
Current account balance (percent of GDP)
Sources: Bloomberg L.P.; Haver Analytics; IMF, World Economic Outlook database; JPMorgan Chase & Co.; S&P Capital IQ; and IMF staff calculations.
Note: ARA = Assessing Reserve Adequacy; EM = emerging market.
Since the taper tantrum in 2013, many emerging Political and Policy Uncertainty in Advanced
market economies have reduced external imbalances Economies Opens Channels for Negative Spillovers
and strengthened policy buffers (Figure1.12, panel3). What would happen if current market optimism
Furthermore, credit booms have begun to wane. At the suddenly turned to pessimism because of concerns
same time, corporate leverage has started to decline, but that U.S. policies could deliver a less benign path for
remains elevated (Figure1.12, panel4). These develop- growth and debt than expected? Financial markets
ments have enhanced the resilience of emerging market would deliver faster normalization of the U.S. term
economies, while their overall growth is projected to rise premium, leading to higher worldwide term premiums
from 4.1percent in 2016 to 4.5percent in 2017. This (see Scenario Box1.1 in the April 2017 WEO). As a
increase is driven mainly by gains in commodity export- result, emerging market economies could face rising
ers, while a number of countries still face more challeng- risk premiums, increased asset price volatility, capital
ing growth prospects (see the April 2017 WEO). outflow pressures, a stronger U.S. dollar, and balance
Figure 1.13. Transmission of External Risks to Emerging sustained shift away from emerging market economies,
Market Economies undermining a vital source of external financing (Fig-
ure1.14, panels1 and 2).
Rising global risk premiums Rising protectionism Such an outcome could also amplify asset price vola-
tility induced by retail investors. Until recently, capital
Unproductive scal flow reversals were driven mainly by herd behavior on
expansion leads to Lower global trade
the part of retail investors, while continued buying
faster rise in U.S. yields
by institutional investors helped offset some of the
Higher global risk Lower output growth and
premiums commodity prices downward pressure on emerging market economy asset
prices (Figure1.14, panel3). However, inflows from
External liabilities Negative demand
become more effects and lower institutional investors have declined in recent quarters.
burdensome external revenues The period following the U.S. election in November
2016 marked the first notable retrenchment by these
Emerging Market Bank and
Financial Stability Risks investors since the global financial crisis (though flows
rebounded in early 2017). Moreover, disruptions could
Lower debt- Corporate leverage
servicing capacity, and FX mismatches
stem from portfolio reallocations by large, opportunis-
higher NPLs rise tic investment funds. For example, multisector bond
funds have sizable holdings in many emerging markets,
and a sharp unwinding of their positions could severely
Source: IMF staff. affect funding and liquidity conditions in some emerg-
Note: FX = foreign exchange; NPLs = nonperforming loans.
ing market economies (Figure1.14, panel4).5
In a scenario of rising global risk premiums, the weak
tail of emerging market economy firms would increase
sheet stresses (Figure1.13). Countries that are more to over 16percent of total nonfinancial corporate debt,
sensitive to external financial conditionsincluding which is an increase of $135 billion (Figure1.15).6,7
from large external financing needs, high corporate This would exceed the 15percent peak in 2015 when
foreign-currency indebtedness, or a large foreign pres- the collapse in commodity prices hit corporate balance
ence in local bond marketswould be most at risk, as sheets. Brazil, China, and India experience the greatest
would frontier market economy borrowers. impact in this scenario given their sensitivities to changes
And what would happen if there were a shift toward in earnings and corporate interest rates. A sustained
protectionism in a number of countries? Emerging reversal of capital inflows would put pressure on coun-
market economies with high trade openness would face tries with high external financing requirements and/or
rising risk premiums amid declining global trade and low reserve adequacy (Table1.1).
commodity prices (Figure1.13). In turn, corporate
earnings would suffer, especially for firms dependent 5For example, in 2016 the multisector bond funds of a single asset
on exports, placing strains on companies with high manager reduced their combined emerging market bond exposure
leverage and banking systems with weaker asset quality. by $15billion. Almost $11billion of that total was concentrated in
a single country. This represented an estimated 13percent of that
countrys total sovereign local and hard currency bonds.
Rising Global Risk Premiums 6The weak tail is defined as the proportion of all nonfinancial
If increases in U.S. interest rates push up global risk corporate debt that is issued by firms with interest coverage ratios
less than 1; the interest coverage ratio is earnings before interest, tax,
premiums and interest rates across emerging market
depreciation, and amortization (EBITDA) divided by interest expense.
economies, borrowing costs would increase for coun- 7Corporate EBITDA is adjusted using the expected changes in
tries with external weaknesses or significant foreign a countrys GDP output from the IMFs G20 model (G20MOD).
exchange exposures. Emerging market currencies would Earnings changes are calculated using the historical relationship
between a sectors earnings and the growth in the economy. Earnings
come under pressure as capital flows reverse, limiting of commodity-related firms are adjusted based on the models expected
space for monetary policy to ease and keeping long- change in commodity prices in the given shocks. Interest expenses are
term interest rates high. Such an environment would adjusted by the change in the corporate interest rate output from the
model. Interest expenses are also adjusted using the expected change
reduce firms debt-servicing capacity and could prompt in the exchange rate, based on the proportion of a given countrys
institutional investors to undertake a more forceful and nonfinancial corporate debt that is denominated in foreign currency.
Capital ows to emerging market economies have been subdued in Retail investors represent a small source of nancing but are a large
recent years. source of volatility.
1. Nonresident Capital Flows to Emerging Market Economies 2. Emerging Market Quarterly Portfolio Flows by Investor
(Billions of U.S. dollars) Type, 201016
450 Other (Billions of U.S. dollars) 70
400 Portfolio debt Source of 45-degree
350 Portfolio equity line 60
volatility
300 Foreign direct investment
Standard deviation
50
250
Total
200
Total retail ows institutional 40
150
ows 30
100
50 Retail equity Institutional debt
20
0
50 Retail debt Source of
nancing 10
100 Institutional equity
150 0
2012 13 14 15 16 10 0 10 20 30 40 50 60 70
Mean
Most capital ow reversals are driven by retail investors. Individual fund families often own large portions of emerging market
bonds in selected markets.
3. Emerging Market Portfolio Flows by Investor Type 4. Concentration of Foreign Emerging Market Bond Holdings
(Billions of U.S. dollars, three-month moving average) of Largest Fund Family Owner
Retail ows Institutional ows (Percent of total bonds outstanding)
50 8
40
30 6
20
10
4
0
10
20 2
Global nancial Taper Renminbi U.S.
30 crisis tantrum shock election
40 0
2008 09 10 11 12 13 14 15 16 17
Ukraine
Ghana
Uruguay
Colombia
Mexico
Indonesia
Serbia
Malaysia
Equities of manufacturing exporters with high U.S. trade exposure ... while currencies of manufacturing exporters have underperformed
have not performed as well as other emerging market equities ... those of commodity exporters.
5. Emerging Market Equity Returns versus Gross 6. Emerging Market Exchange Rates
Manufacturing Exports (Median and 25th75th percentiles, Index 100 = Nov. 1, 2016)
20 POL 108
Equity market return since Nov. 8, 2016
SAU
100
5
ARE BRA INDPER JOR VNM
RUS THA 98
UKR CHN MYS 96
ZAF
0 CHL
IDN MEX 94
COL
5 92
2 1 0 1 2 3 4
Nov. 02
Nov. 09
Nov. 16
Nov. 23
Nov. 30
Dec. 07
Dec. 14
Dec. 21
Dec. 28
Jan. 04
Jan. 11
Jan. 18
Jan. 25
Feb. 01
Feb. 08
Feb. 15
Feb. 22
Mar. 01
Mar. 08
Sources: Bloomberg L.P.; EPFR Global; Institute of International Finance (IIF); UN Comtrade; and IMF staff calculations.
Note: Panels 2 and 3 show proxies for portfolio ows by institutional and retail investors, which are estimated using IIF portfolio ows data and EPFR data on ows
into investment funds dedicated to emerging markets. IIF data capture ows by all types of investors, but fund ows are predominantly driven by retail investors. In
panel 2, standard deviations are calculated as the average eight-quarter rolling standard deviation over the 201016 period. The country sample for IIF data
encompasses 25 emerging market economies (used in panels 1, 2, and 3), while the country sample for EPFR data encompasses about 85 emerging and developing
economies (used in panels 2 and 3). Other differences between the two datasets are discussed in Koepke and Mohammed 2014. In panel 5, data labels in the gure
use International Organization for Standardization (ISO) country codes. In panel 6, commodity exporting countries include Brazil, Chile, Colombia, Indonesia,
Kazakhstan, Malaysia, Peru, Russia, and South Africa. Manufacturing exporting countries include Bulgaria, China, Hungary, India, Malaysia, Mexico, Poland, Taiwan
Province of China, Thailand, and Vietnam. EM = emerging market; FDI = foreign direct investment.
Figure 1.15. Emerging Market Corporate Debt under Rising Rising Protectionism
Risk Premiums and Protectionism
If protectionist pressures increase and start to affect
Emerging market economies would see an increase in the size of the global trade, emerging market economies closely
weak tail of their corporate sectors. integrated into global trade and capital markets will
1. Corporate Debt with Interest Coverage Ratio < 1 face lower external revenues and rising risk premiums.8
Current Debt at Risk Change under Scenarios The combination of declining global trade and growth
(percent) (percentage points)
would increase corporate vulnerability, especially for
India those with high leverage and large foreign exchange
mismatches. The resulting higher corporate risk
Indonesia
premiums and borrowing costs will increase financial
China stability risks in these economies.
Both direct and indirect transmission channels
Turkey
would come into play in such an environment,
Brazil including through disruptions to principal trading
partners. For example, manufacturing exports account
South Africa
for some 25percent of Mexicos GDP, and 80 percent
Russia Rising global of all its goods exports are bound for the United States
risk premiums (Table1.1). Some emerging market economies in
Mexico Protectionism
Asia (for example, Malaysia, Thailand, and Vietnam)
Saudi Arabia have high manufacturing exports as a share of GDP.9
Similarly, a decline in Chinese exports would not only
25 20 15 10 5 0 0 2 4 6 8
weaken Chinas growth and add to domestic vulnera-
The weak tail of corporate debt rises signicantly in a scenario of rising bilities: it would also weigh on demand for imported
global risk premiums and rising protectionism. intermediate and capital goods.
2. Emerging Market Corporate Debt with Interest Coverage Ratio < 1
This would further affect exporters in Asia, as well as
(Percent of total nonnancial corporate debt) commodity exporters. The broader negative repercus-
Rising protectionism sions for emerging market economies underscore the
20
Utilities
Rising global risk premiums potential for rising domestic vulnerability in China to
18 Other drive higher global risk premiums.
Industrials
16 Emerging market economy asset prices reflect
Commodities
some of these trade exposure risks. Equities in
14
countries with substantial manufacturing exports to
12 the United States (Mexico, Vietnam), or that form a
part of major supply chains (Chile, Malaysia), have
10
underperformed other emerging markets (Fig-
8 ure1.14, panel5). Commodity exporters currencies
6 have notably outperformed those of manufacturing
exporters in recent months (Figure1.14, panel6).
4
This performance likely reflects the boost from rising
2 commodity prices, but it may also indicate less mar-
ket concerns that protectionism would affect trade in
0
2010 11 12 13 14 15 16 17 Scenarios commodities.
(est)
India 1.1 1.5 10 30 156 13 1.5 3.3 10 3 4 2 42 0.9 2.3 0.0 3.7
Indonesia 1.5 2.4 39 126 131 41 1.9 0.8 7 0 39 7 35 0.9 2.0 1.7 3.8
Malaysia 0.5 1.4 52 9 115 37 1.8 3.4 40 6 34 3 128 2.0 4.3 4.1 5.6
Mexico 0.6 0.9 39 55 113 5 2.5 1.0 13 5 34 3 78 10.7 6.5 1.5 0.0
Peru 0.7 1.4 8 62 311 5 1.9 0.9 8 0 33 23 44 0.6 0.1 9.0 6.0
Philippines 1.1 2.0 4 43 227 29 0.1 2.9 6 2 8 6 58 0.5 3.4 1.1 3.2
Poland 0.9 6 126 8 1.7 2.0 21 11 34 2 101 0.2 2.9 3.6 0.2
Russia 1.1 1.4 58 91 238 74 3.3 0.0 7 4 23 1 46 0.1 0.9 4.1 16.3
Saudi Arabia 590 114 1.5 20.9 7 25 62 0.7 0.1 15.8 20.3
South Africa 0.2 1.6 22 34 74 5 3.4 1.7 18 4 35 3 61 0.1 2.6 2.3 1.7
Thailand 1.5 10 220 5 9.7 10.1 7 8 14 3 124 3.0 4.6 3.8 0.3
Turkey 0.8 0.2 9 43 90 9 4.7 0.8 31 9 18 6 47 0.5 2.7 1.4 1.4
United Arab Emirates 1.3 43 3.5 16.3 169 4.3 3.1 14.1 10.6
Vietnam 1.0 19 70 7 4.1 1.9 4 3 182 13.1 10.0 5.7 0.7
Median 0.9 1.4 20 47 136 4 1.3 0.8 9 2 22 2 58 0.2 2.6 1.4 1.7
Sources: Sovereign investor base estimates by Arslanalp and Tsuda (2014, updated); Bloomberg L.P.; IMF, World Economic Outlook database; JPMorgan Chase & Co.; UN Comtrade; and IMF staff calculations.
Note: The UN Comtrade data on net exports comprise commodity codes 0 through 4 for commodities and codes 6 through 8 for manufacturing, using Revision 3 or 4. For details on the reserve adequacy metric (ARA), see IMF
2015d. E = estimate.
19
GLOBAL FINANCIAL STABILITY REPORT: Getting the Policy Mix Right
Table1.2. Asset Quality and Capital Indicators for a Sample of Emerging Market Banks
(Data based on bank-reported financial statements; 2016 or latest available)
Banks with Provision Share of Banks with Tier 1
NPL and NPLs and Provision Needs Needs Ratio below 10 Percent
In a scenario of rising protectionism, the size of the ing market economies and spill over to the banking
weak tail of firms would increase to 17percent of total system. This underscores the importance of ensur-
nonfinancial corporate debt, an increase of $235 bil- ing the health of emerging market banking systems
lion, which is somewhat higher than under the case of through swift and transparent recognition of nonper-
rising global risk premiums (Figure1.15, panel1). The forming assets and by strengthening capital buffers.
greatest deterioration in corporate balance sheets would On the positive side, bank capital ratios have been
occur in China, India, and South Africa. Commodity rising steadily over the past several years, with a sample
sectors especially would come under pressure because of about 300emerging market banks showing aggre-
metal and oil prices would fall as a result of the sharp gate Tier 1 capital ratios now at comfortable levels
decline in global growth. (Table1.2; Figure1.16, panel1).10 Shrinking risk
weightings have been a contributing factor, particularly
Are Emerging Market Banks Capital Buffers
Sufficient to Absorb Increased Corporate Stress? 10Banking sector data in the remainder of this section are based on
Stronger external headwinds from tighter global a 294-bank sample covering banks from 14countries with $32tril-
lion in assets. Bank-level data are used instead of official Financial
financial conditions or increased trade protectionism Soundness Indicator (FSI) data because they offer better granularity
could worsen corporate vulnerabilities in some emerg- and allow for cross-sectional analysis.
Protability is low at some weak banks ... ... and asset quality is deteriorating in many countries.
3. Bottom Quartile of Banks by Protability: Return on Assets 4. Nonperforming and Problem Loans as a Ratio of Gross Loans
(Percent) (Ratio)
China: NPL ratio including problem loans
EM excluding China: NPL ratio including problem loans
2.0 10
China: NPL ratio
1.0 EM excluding China: NPL ratio
8
0.0
1.0 6
2.0 4
3.0 Range for the bottom quartile
2
4.0 Weighted average for the
bottom quartile
5.0 0
Saudi Arabia
Colombia
South Africa
Mexico
United Arab
Emirates
Turkey
Thailand
Indonesia
China
Malaysia
Brazil
Poland
EM excluding
China
India
Russia
2010 11 12 13 14 15 16
Sources: Bank for International Settlements; Bloomberg L.P.; Morgan Stanley Capital International; SNL Financial; and IMF staff calculations.
Note: EM = emerging market; NPL = nonperforming loan.
in Brazil, but banks in most markets have also actively compared with that in the United States and Europe
reduced leverage. Lenders outside China have increased heavy credit losses continue to erode profits at many
capital by 20percent since the end of 2014, compared banks, notably in Russia and India (Figure1.16,
with 15percent growth in assets over the same period, panel3). Furthermore, nonperforming and problem
reflecting a combination of public recapitalization and loans have climbed in many countries, reflecting
banks efforts in response to increased regulatory and various challenges: economic weakness (Brazil, Russia),
market scrutiny. Nonetheless, asset quality concerns continued corporate leverage growth (China), and sec-
have not been fully addressed after several years of tor-specific downturns (India) (Figure1.16, panel4).
rapid growth in lending. Bank equity valuations are Banks have raised provisioning levels in response,
relatively weak in China and Turkey, where credit has but not quickly enough to keep pace with bad loan
grown rapidly relative to GDP (Figure1.16, panel2). formation (Figure1.17, panel1). As a result, the weak
Although the profitability of banks in emerging tail of banks with poor loss coverage (nonperforming
market economies is generally strongin particular and problem loans as a proportion of bank buffers)
Russia
India
South Africa
Colombia
Brazil
Poland
Turkey
Indonesia
Mexico
Thailand
China
United Arab
Emirates
Saudi Arabia
Malaysia
EM excluding
China
Provision needs exceed annual prots in 30 percent of emerging If provisions were deducted from equity, weak banks would jump up to
market banks outside China. 35 percent of assets.
3. Number of Years to Absorb Additional Provisions through 4. Percentage of Assets with Tier 1 Ratio below 10 Percent
Earnings, by Share of Assets (Percent)
(Percent) Current
Based on Preprovision
After raising provisions
Based on Net Income Prots
India 100
Russia 90
EM excluding China
South Africa 80
Brazil 70
Poland
Indonesia 60
Mexico 50
Thailand
Turkey 40
China 30
United Arab Emirates
Malaysia 20
Colombia 10
Saudi Arabia
0
0 25 50 75 100 0 25 50 75 100
Colombia
India
South Africa
Russia
China
EM excluding China
Brazil
Turkey
Poland
Indonesia
Mexico
Thailand
United Arab Emirates
Saudi Arabia
Malaysia
has swelled in emerging market economies (excluding capital by tapping the equity market given generally
China) to about 40percent of sample assets (Fig- favorable valuations.
ure1.17, panel2).
More Forceful Policy Action Is Needed to Ensure
Further Deterioration in Asset Quality Would Erode Resilience of Emerging Market Economies
Capital Levels for Several Banks
Emerging market economies have become more
Restoring provisioning coverage among the weakest resilient, benefiting from a recovery in global com-
banks is important to ensure the banking system has modity prices and still-supportive external conditions.
resilience to withstand further asset quality deteriora- However, the preceding analysis highlights that these
tion. In an illustrative exercise to assess the potential economies face challenges along several channels (Fig-
extent of underprovisioning of weaker banks, banks ure1.18). Those reliant on trade openness (Hungary,
provision coverage ratios are raised to at least 50per- Malaysia, Thailand, Vietnam, United Arab Emirates)
cent of nonperforming and problem loans, or to or with large external financing needs (Malaysia,
their countrys average provision-to-loan ratio.11 This Poland) or low reserve adequacy (South Africa, Viet-
exercise generates some $120billion (5percent of nam), or a combination (Turkey), would be challenged
capital) in additional provisions, which would have to by tighter global financial conditions and unfavorable
be fulfilled through retained earnings, existing capital, trade developments. Others, challenged in the cor-
or new equity. More profitable banking systems such porate sector (China, India, Indonesia, Turkey) or
as those in Colombia and Indonesia would be well banking sector (China, India, Russia), could face more
positioned to absorb such costs; however, for about broad-based risks.
30percent of emerging market bank assets (outside Risks of an abrupt tightening in financial conditions
of China), additional provisions would exceed average and increased protectionism pose new challenges for
annual net income (Figure1.17, panel3).12 In more policymakers. Therefore, policymakers should continue
than a third of the banking systems in India and Rus- to address corporate and bank vulnerabilities to ensure
sia, provisioning needs would amount to at least three resilience against an increasingly uncertain global
years of net income, unless profits recover from cyclical environment.
lows. To account for cyclical weaknesses in some coun- Restoring the health of corporate balance sheets:
tries, which may reduce net income, provision needs Authorities should prioritize improving corporate
can be compared with preprovision profits.13 Based on debt-restructuring mechanisms, including for-
this approach, some banks in India and Russia would mal insolvency frameworks and out-of-court debt
still require more than one year of earnings to boost restructuring. Policymakers should develop an
provisioning. If the provisioning needs were fulfilled in-depth understanding of both the sources and
with equity, the share of banks with Tier 1 capital composition of credit extended to nonfinancial
ratios below 10percent, excluding China, would jump firms and proactively monitor corporate vulnerabil-
from about 20percent to 35percent of total assets ity. Authorities should continue to monitor firms
(Figure1.17, panel4). Many large banks could raise foreign exchange exposure, and the extent to which
foreign-currency debt is hedged, either naturally
(through foreign exchange income) or through
11Problem loans are those reported by banks and are valid leading financial instruments. Moreover, authorities should
indicators of nonperforming loans. Problem loans are usually not stand ready to provide additional foreign exchange
defined by supervisors, but certain categories, such as restructured
loans, receive increasing supervisory attention. Differences in cover-
hedging tools to help firms absorb sharp currency
age ratios may be driven by differences in reliance on collateral, so a movements without causing financial distress (as
coverage ratio of less than 50percent of nonperforming and problem undertaken in Brazil and Mexico in recent years).
loans does not necessarily imply underprovisioning.
12Three-year average profits are used for the calculation, reflecting
Strengthening the health of the banking system: Bank
the current cyclical position of a country. supervisors in countries whose banks are charac-
13For example, retained earnings may be reduced by higher terized by weak balance sheets or have expanded
provisions because of asset quality deterioration or more aggres- rapidly should carry out comprehensive asset quality
sive provisioning. The use of preprovision profits as a comparator
assumes that banks do not need to set aside additional provisions for assessments to gauge the extent of unrecognized
new nonperforming loans. credit losses. These assessments should be followed
Figure 1.18. Emerging Market Economy Challenges credit. Banks continue to play a major role in the pro-
vision of credittotal assets of Chinas banks are now
External nancing vulnerabilities Banking system weaknesses
more than triple the size of its GDPwith the fastest
expansion from city commercial, joint-stock, and other
Poland Russia
South Africa smaller banks (Figure1.19, panels3 and 4). At the
same time, other nonbank financial institutions have
Malaysia
China raised their credit exposure and leverage with the help
Vietnam
Hungary India of short-term wholesale funding, raising counterparty
Mexico Indonesia
concerns, while the issuance of corporate bonds surged
Thailand
UAE throughout 2016.
A large credit overhang has built up (the Bank for
International Settlements calculates that the credit gap
Turkey now stands at about 25percent), and there is evidence
that credit booms of this size are often dangerous (Fig-
ure1.19, panel2).14 The likelihood of a financial crisis
Trade linkages Corporate fragilities rises the longer a boom lasts and the larger it grows,
especially if exchange rate flexibility is very limited (see
Source: IMF staff. IMF 2012).
Note: Elevated risks for each category are dened as follows: Trade linkages = Capital account pressures remain significant, with
countries that rank in the top quartile among all countries for either trade openness
or exports to the United States and the United Kingdom. Corporate fragilities = outflows picking up again in the second half of 2016,
countries whose percentage of debt issued by rms with an interest coverage ratio although they moderated substantially in the first
below 1 rank in the top quartile of countries in Table 1.1. External nancing
vulnerabilities = countries with reserves as a percent of ARA metric below 100 two months of 2017. The Peoples Bank of China
percent in 2016 or projected external nancing requirements above 15 percent of has continued foreign exchange interventions to
GDP in 2017, as shown in Table 1.1. Banking sectors = countries with Tier 1
capital ratios below 11.5, using Financial Soundness Indicators data as shown in
maintain broad exchange rate stability (Figure1.20,
Table 1.2. ARA = Assessing Reserve Adequacy; UAE = United Arab Emirates. panels1 and 2). Foreign asset purchases by Chinese
residents account for most of the recent outflows,
and Chinese firms have increased their investments in
foreign companies abroad since late 2015. But foreign
by concrete steps to cover the losses andwhere direct investment by overseas firms in China has also
applicableensuing capital needs. Capital needs declined markedly over the past few quarters. Narrow-
should be tackled promptly while global financial ing interest rate differentials and market expectations
conditions are favorable. This should be achieved of bilateral depreciation versus the U.S. dollar have
preferably through private channels, including added to capital outflow pressures.
equity issuance and bail-ins. Public support should The Chinese authorities have continued to adjust
be used as a last resort, when issues are systemic and policies to address rising vulnerabilities from rapid
fiscal space is sufficient. In addition, bank regulators credit growth. In late 2016 they tightened monetary
should monitor limits on foreign exchange open conditions. But the market turbulence that followed
positions and assess the offsetting effect of foreign illustrates the risks that remain in Chinas increasingly
exchange hedging. large, opaque, and interconnected financial system.
Tighter liquidity conditions in interbank and repo
China: Rising Risks and Financial Vulnerabilities markets pushed up repo rates (Figure1.21, panel1),
While credit booms are waning in many emerging causing losses for financial institutions investing in
markets, credit continues to grow at a rapid pace in bond market vehicles (Figure1.21, panel2). This
China (Figure1.19, panel1). Despite stabilization of caused leveraged investors to sell bonds, pushing up
the near-term growth outlook, policy efforts to contain bond yields sharply (Figure1.21, panel3).
leverage and financial risks remain constrained by the
authorities long-term growth objective: doubling the
average income and size of Chinas economy by 2020. 14A comprehensive discussion of Chinas credit boom and debt
Achieving this requires ever increasing amounts of problem is provided by Maliszewski and others (2016).
0 100 50
2009 10 11 12 13 14 15 16 1980 83 86 89 92 95 98 2001 04 07 10 13 16
Chinese banks are now among the largest in the world, also relative ... and smaller city commercial and joint-stock banks are still growing
to the size of the economy ... rapidly.
3. Bank Total Assets to GDP 4. Total Asset Growth
(Percent) (Year over year, percent)
Large commercial banks City commercial banks
350 30
Joint-stock commercial banks Rural nancial institutions
300 25% 25
250
20
200 17%
15
150 17%
10
100
11%
50 5
0 0
2014 15 16 IND USA RUS ZAF BRA JPN DEU Jan. Jul. Jan. Jul. Jan. Jul.
CHN 2016 2014 14 15 15 16 16
Sources: Bank for International Settlements; CEIC; Haver Analytics; IMF, World Economic Outlook database; and IMF staff calculations.
Note: Data labels use International Organization for Standardization (ISO) country codes. In panel 1, credit is total social nancing stock (mainly private sector credit).
Falling bond prices, rising global interest rates, and ties, which calmed the markets and helped reduce
surging repo rates combined to cause distress in the yields in bond markets.
informal repo market called the entrusted bond
market. This led to increased counterparty concerns This episode highlights a number of pressure points
in this largely unregulated market characterized by that remain in the financial system:
weak documentation standards, and segments of the Many financial institutions continue to be overly depen-
repo market started to freeze up in mid-December dent on wholesale financing with sizable asset-liability
(Figure1.21, panel1). mismatches. As emphasized in the October 2016
To avoid systemic stress, the Peoples Bank of China GFSR, the very short-term nature of Chinas repo
instructed several large, state-owned banks to pro- funding implies that borrowers must roll over their lia-
vide broad-based liquidity support through so-called bilities on average almost daily, whereas funded credit
X-repurchase agreements (whose counterparties are products have much longer maturities. This maturity
anonymous), in some cases to institutions that do mismatch makes borrowers highly vulnerable to a sud-
not have access to the central banks lending facili- den liquidity crunch, as evidenced in December.
Figure 1.20. China: Capital Flows and Foreign Exchange Figure 1.21. Recent Turmoil in Chinese Financial Markets
Reserves
Tighter liquidity conditions pushed up repo rates, which surged in
Foreign asset purchases by Chinese residents have driven the recent December for riskier institutions.
pressure on capital outows ... 1. Repo Rates
(Seven day, percent)
1. Resident and Nonresident Capital Flows
(Billions of U.S. dollars) U.S. election U.S. rate hike Money market 12
300 Weighted average turmoil
repo rate (seven-day 10
Residents asset purchases (includes errors and omissions) maturity)
Nonresident inows (residents liabilities) 8
Maximum rate
200 6
2
100
0
Sep. Oct. 16 Nov. 16 Dec. 16 Jan. 17
2016
Liquidity and credit risks are sizable, amid increased efforts continue to adapt business models. More recently,
reliance on bond issuance and elevated redemption bank equity prices have gained as a result of inves-
needs. Low interest rates, a relaxation of bond issu- tor optimism about a cyclical upturn in the economy.
ance requirements, and expectations of a stronger However, a cyclical recovery alone is unlikely to fully
U.S. dollar triggered a surge in issuance beginning restore the profitability of persistently weak banks,
early in 2015. China now accounts for more than and more needs to be done to improve resilience. A
two-thirds of total emerging market bond issuance number of system-wide structural features are com-
and a third of U.S. dollar issuance, and maturities pounding profitability challenges for domestic banks
are shortening. and may be affecting some international institutions.
Investor composition has grown increasingly complex. One structural challenge is overbanking, the features
Banks continue to be the largest bond holders, but of which vary from country to country. Although
wealth management products and securities firms measures are being taken to address concerns, coun-
also have significant exposure and, in some cases, are tries with the biggest challenges need to make more
highly leveraged to boost returns. Moreover, leverage progress. Until these structural impediments have been
is often established through informal markets with fully addressed, business model restructuring alone may
limited documentation and transparency. not yield sufficient profitability. Left unresolved, the
combination of weak banks, lack of access to pri-
The difficult task of deleveraging the system is vate capital, and large bad debt burdens impedes the
thus as crucial and urgent as ever. This is increasingly scope for recovery and could reignite systemic risks.
recognized by the authorities, who have started a
host of new regulatory initiatives to close loopholes
for regulatory arbitrage, rein in leverage, and increase Sustainable Profitability Remains Elusive for Many
transparency of nonbank financial institutions and Banks
wealth management products. As discussed in previous There has been substantial progress in the Euro-
GFSR reports, proactive recognition of losses, com- pean banking sector. Bank capital ratios have been
bined with restructuring of overly indebted but viable raised, and banks have recently recapitalized in Italy
firms, is needed. Supervisory attention should con- and Portugal. Banks now make less use of short-
centrate on banks emerging risks, especially fast asset term wholesale funding. Regulations continue to be
growth among the small unlisted local banks, increas- strengthened and supervision has been enhanced.
ing reliance on wholesale funding, risks packaged into Steps are being taken to address the burden of
shadow products, and possible contagion through the nonperforming loans. Efforts continue to be made
interbank market. to adapt business models, and there has been some
But staving off further bouts of market instability consolidation within the banking sector in a number
and ultimately, macro instabilityrequires addressing of countries.
the policy tension between maintaining a high level of At the same time, the long-awaited cyclical recov-
growth and the need for deleveraging. To the extent ery is gathering momentum. European bank equity
that credit growth remains excessive, the underpricing prices have increased, rising by about 40percent
of credit risks remains an endemic characteristic of on average since mid-2016 (Figure1.22, panel1).
the financial system, and the search for yield remains Bank profits should be helped by the steepening in
a driving motivation, leverage will continue to build, yield curves, which has relieved some of the building
and financial risks will continue to grow. pressures on bank net interest margins in a low rate
environment (see Chapter2). Earnings should also
be buoyed by the strengthening economic outlook
European Banking Systems: Addressing as provisions fall and lending grows. Despite this
Structural Challenges improvement, market valuations (price-to-book
Considerable progress has been made in the Euro- ratios) continue to reflect concerns about the ability
pean banking sector over the past few years. Banks of European banks to generate sustainable profits
have higher levels of capital, regulations have been (Figure1.22, panel2). Indeed, in a large sample of
strengthened, supervision has been enhanced, and European banks, the 2016 return on equity was weak
Bank equity prices have increased ... ... but European equity valuations remain low.
1. European Bank Equity Prices and the Slope of the Yield 2. Bank Price-to-Book Ratios
Curve (Multiple)
A signicant proportion of banks have weak prots ... ... and analysts do not expect this to change quickly ...
3. European Banks, by Return on Equity Thresholds over Time 4. Selected European Bank Return on Equity
(Percent of sample, by assets) (Percent)
Weak (ROE < 8%) Healthy (ROE 10%)
Challenged (8% ROE < 10%)
100 25
14 14 Median
16 15
25th percentile Analysts
80 12 17 forecasts 20
40th percentile
36 32 60th percentile
60 75th percentile 15
Percentiles
75th
40 74 ROE 810% 60th 10
69 50th
49 53 40th
20 25th 5
0 0
2013 14 15 16 2005 06 07 08 09 10 11 12 13 14 15 16 17E 18E 19E
NPL ratio
Loan loss Bad debt
NPL coverage
provisions overhang
NPL disposals
(less than 8percent) for about half of the banks, by ments also play an important role? To answer this
assets (Figure1.22, panel3).15 questionwhich has important policy implications
Although cyclical support for bank profits is we divide banks in our sample of $35trillion by
welcome, it is likely to be insufficient to resolve the assets of 172large European banks into three groups:
profitability challenge that many banks and banking global, Europe focused, and domestic (Table1.3 pro-
systems face. The October 2016 GFSR concluded vides further information on the grouping of banks
that even after a cyclical recovery in profits, a group and on the sample used in this analysis). Although
of structurally weak banks, representing about the challenge of bank profitability is widespread,
$8.5trillion in assets (or about one-third of bank domestic banks (banks with more than 70percent
assets), would be stuck with a return on equity less of revenues or assets in their home market) as a
than 8percent. This finding is corroborated by mar- group struggled especially with profitability in 2016.
ket analysts who do not expect the economic upturn Overall, three-quarters of domestic banks in our
to increase bank profits significantly and predict sample had a weak return on equity, compared with
that the asset-weighted average return on equity for about 65percent of sample global banks and just
about 80European banks will remain below 8per- 15percent of Europe-focused banks in our sample
cent until 2019, and the majority will have a return (Figure1.23, panel1).
on equity below that level over the next three years In the euro area, significant strides have been made
(Figure1.23, panel4). to forge a full-fledged banking union. However,
Persistently weak profitability is a systemic sta- differences in national supervisory practices, legal
bility concern. Low profits can prevent banks from frameworks, impediments to cross-border mergers and
organically building cushions against unexpected acquisitions, and the role of government policies and
losses and thereby make them more vulnerable to institutions in influencing credit distribution mean
adverse shocks. Sustained returns below the cost that the country-level system operating environment
of equity can also inhibit banks access to private and features can influence profitability. Domes-
capital, because investors are generally more willing tic banks, in particular, face more limited scope to
to recapitalize banks if their profitability will sustain improve profitability by shifting their exposures across
valuations above book value and so avoid future markets, and hence the profitability of these banks
dilution. At the same time, banks facing profit- more clearly reflects the structural features of their
ability pressures may look to drive up returns by home systems. Therefore, the discussion of structural
taking greater risks, for example by seeking higher features in this section focuses on the performance of
yields, lending to less creditworthy borrowers at domestic banks.
higher spreads, or increasing the maturity mismatch There is great variability in domestic banks profit-
between loans and funding. Weak returns also limit ability across countriesmeasured on either a return
banks ability to expand balance sheets and lend on equity or return on assets basis (Figure1.23,
without depleting their capital base, and therefore panel2). Although sample domestic banks in Italy and
place a drag on recovery. Portugal suffered losses overall in 2016, and the Ger-
man, Spanish, and U.K. domestic banks in our sample
were barely profitable, sample domestic institutions
System-wide Operating Environments Are Compounding
in Ireland, Norway, and Sweden were able to generate
Challenges to Bank Profitability
much higher returns in the same year.
Does weak profitability result from poor business This variability in profits suggests that it is not
models only, or do system-wide operating environ- necessarily the domestic bank business model as
a whole that is the problem, but that conditions
15Much of the analysis in this section is based on 2016 profit data.
and system-wide features in each country can also
If annual 2016 data have not yet been published, available figures
have been annualized. In the few cases where no 2016 numbers have limit profitability. Cyclical economic conditions
been reported, 2015 profits have been used. An 8percent return on including interest rates, the slope of the yield curve,
equity benchmark is used because, as discussed in the October 2016 asset quality, and credit growthcould drive some
GFSR, investor surveys suggest that banks cost of equity is at least
8percent (though some investors indicated that the cost of equity is of this variability. But there are also a number of
above 10percent). system-wide structural impediments that could
Return on equity varies by type of bank ... ... and protability varies across countries.
1. Sample European Banks by ROE Thresholds 2. Sample Domestic Bank ROE and ROA
(Percent of sample, by assets)
Percent of Healthy (ROE 10%) Challenged
sample Weak (ROE < 8%) (8% ROE < 10%)
assets 100 40 29 31
100 16
15 8 SWE
18 19 14
Revenue
0 CHE l cos
r o t DNK BEL equa 1.5
1 Hig h p
NLD revenues
rot rot (
DEU ing p 1.0
2 Mid p n op e r a t
pro
t ovisio
3 Low e r o prepr 0.5
Z
4 0.0
0.0 0.5 1.0 1.5 2.0
Ireland
Sweden
Norway
Switzerland
France
Netherlands
Austria
Belgium
Germany
United
Kingdom
Spain
Italy
Portugal
Denmark
Operating costs
Protability also varies across groups of institutions ... ... and within countries.
5. Sample Domestic Banks by ROE Thresholds and Institution Type 6. Sample Domestic Bank ROA within Countries
(Percent of sample, by assets)
Healthy (ROE 10%) Challenged (8% ROE < 10%)
Weak (ROE < 8%)
100 6 IRE 2.0
30 6
Variation in ROA within a country
ITA 1.7
80
(standard deviation)
7 1.3
60
PRT GBR 1.0
40 88
BEL 0.7
63 ESP
AUT SWE
DEU
20 0.3
NLD DNK
CHE NOR
0 FRA 0.0
Commercial banks Savings, cooperative, 1.0 0.8 0.6 0.4 0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
development, policy, and Return on assets (percent)
state-owned banks
Type of bank: The analysis classifies banks into three main types: domestic, Europe focused, and global. Bank classification: The classification of institutions into commercial banks and others (Landesbanken,
The distinction is based on the geographic location of bank revenues, or assets if revenues are not cooperative banks, state-owned banks, and policy banks) is based on a review of each banks financial
available. If neither of these is reported, a manual review of qualitative information in bank financial reports and websites.
statements is used.
Proportion
Sample Assets (billions of U.S. dollars) of Sample Number of Firms in the Sample
Europe Commercial Assets Europe Commercial
Countries Total Domestic Focused Global Banks Others (percent) Total Domestic Focused Global Banks Others
France 7,785 2,475 5,309 0 3,648 4,136 22.3 7 4 3 0 2 5
United Kingdom 6,697 2,177 0 4,520 6,648 49 19.2 13 10 0 3 12 1
Germany 4,452 2,180 595 1,677 2,483 1,969 12.7 22 17 4 1 9 13
Spain 3,767 1,583 0 2,184 3,601 166 10.8 15 13 0 2 11 4
Italy 2,651 1,744 906 0 2,063 588 7.6 21 20 1 0 10 11
Switzerland 2,476 612 95 1,769 1,978 498 7.1 20 16 1 3 8 12
Netherlands 1,866 276 1,590 0 940 926 5.3 7 5 2 0 4 3
Sweden 1,545 318 1,227 0 1,506 39 4.4 7 4 3 0 5 2
Denmark 825 331 494 0 803 22 2.4 6 5 1 0 5 1
Austria 733 514 220 0 307 426 2.1 14 13 1 0 5 9
Belgium 543 253 290 0 543 0 1.6 5 4 1 0 5 0
Norway 346 346 0 0 321 26 1.0 6 6 0 0 4 2
Greece 340 340 0 0 340 0 1.0 5 5 0 0 5 0
Portugal 314 273 40 0 214 99 0.9 6 5 1 0 3 3
Ireland 289 289 0 0 188 101 0.8 4 4 0 0 3 1
Others 299 278 21 0 106 193 0.9 14 12 2 0 9 5
Total 34,929 13,990 10,788 10,150 25,690 9,239 100 172 143 20 9 100 72
31
GLOBAL FINANCIAL STABILITY REPORT: Getting the Policy Mix Right
lower bank profitability even after cyclical conditions due to a high number of branches or staff (Figure1.22,
improve.16 panel5).
To understand which structural features may be The causes of overbanking can vary from country to
creating the greatest impediments, it is important to country; examples include a banking system with assets
assess the sources of weak profitability. Figure1.22, that are large for the economy it serves, a long weak
panel5, shows how return on assets can be decom- tail of banks with low buffers, or too many banks with
posed into revenues, costs, and loan loss provisions. a regional focus and narrow mandate. These features
The range of revenues, costs, and provisions for the can result in concentrated lending opportunities and
domestic banks in our sample in 2016 is shown less scalable lending, or a high number of branches
in Figure1.23, panels3 and 4. Interestingly, some relative to the assets in the banking system that add to
domestic banks with weak return on assets have a costs and reduce operational efficiency.
relatively high preprovision operating profit. This The strength of the structural factors in each
suggests that profitability is largely being affected by country varies across domestic banking systems. In
the provisioning that they need to undertake to build countries where most of the sample domestic banks
buffers against the nonperforming loans on their bal- perform poorlyshown in Figure1.23, panel6, where
ance sheets. Other domestic banks in the sample with both return on assets and the variation in returns are
weaker preprovision operating profits may be facing lowsystem-wide impediments to profitability are
different structural challenges affecting revenues and more likely.
costs, as discussed below. Table1.4 shows a number of system-wide metrics
to highlight the aspects of overbanking in different
Overbanking in Systems Should Be Reduced systems.
One main structural challenge is overbanking. The size of banking systems is illustrated by the
There is no common definition of overbanking. The ratio of local bank claims in a system relative to
European Systemic Risk Board has used this term to GDP.
describe excessive growth in the European banking The degree of concentration in a banking system
system, and the European Central Bank has said that can be suggested by a number of measures, includ-
overbanking and overcapacity create intense compe- ing bank assets per credit firm, the number of banks
tition and affect bank profitability.17 Here, the term operating in a country, and an index of concentra-
overbanking refers to the variety of structural factors tion (Herfindahl).
that lead to an overly large banking sector that affects Cost pressures reflect many factors, and in reality
the profitability of the banks in the system. Overbank- the structural drivers of revenues and costs are inter-
ing can affect revenuespossibly owing to too many twined; for example, a high number of branches
banks chasing too few profitable and sound lending and staff can be a by-product of having too many
opportunities, compressing pricing and marginsand banks in the system. These operational efficiencies
can affect costs and operational efficiencypossibly are illustrated by the level of assets per branch and
per employee.
16TheOctober 2016 GFSR includes an analysis of the impact of In addition, system structure may have an impact
a cyclical recovery on European bank profitability and finds that this
on profitability. Banking systems with a high propor-
would not be sufficient to fully restore profitability.
17European Systemic Risk Board 2014 highlights the ratio of tion of savings or cooperative banks, Landesbanken,
banking system assets to GDP as an important metric in identi- and policy or state-owned banks may face additional
fying overbanking. A European Central Bank speechResolving pressure on revenues.18 Figure1.23, panel5, shows
Europes NPL Burden: Challenges and Benefits, by Vitor Constan-
cio (February 3, 2017)notes that in addition to the resolution that the domestic cooperative and savings banks, devel-
of nonperforming loans, bank profitability is also challenged by
high costs and overbanking (https://www.ecb.europa.eu/press/key 18Savings banks are institutions whose primary purpose is to
/date/2017/html/sp170203.en.html). Another European Cen- channel savings deposits, particularly by providing local or regional
tral Bank speechWelcome Address at the First Annual ESRB banking services to small and medium-sized enterprises. Cooperative
Conference, by Mario Draghi (September 22, 2016)notes that banks are similar institutions that are owned by their customers.
overcapacity and the ensuing intensity of competition affect bank Landesbanken are public banks in Germany that are owned by
profitability (https://www.ecb.europa.eu/press/key/date/2016/html regional authorities. Policy or state-owned banks are owned by
/sp160922.en.html). governments.
opment and policy institutions, Landesbanken, and the Netherlands, and Spain, in particular, have
state-owned banks in our sample tended to have lower seen larger percentage reductions in branches and
overall return on equity than other sample domestic employees. Rationalizing branches, so that the ratio
banks in 2016. of deposits to branches of each sample bank at least
Overall, no single structural factor clearly explains reaches the European average, could reduce operating
profitability concerns across a range of countries. A expenses by about $23billion overall, equivalent to
number of features in a system may hurt institutions 23percent of after-tax profits for the banks consid-
pricing and other behavior that then put downward ered here.19
pressure on the profitability of other banks operating Both business pressures and labor market rigidities
in the same country. Each country has a unique mix can inhibit banks ability or incentives to restructure
of structural features that may impact profitability. For more quickly and aggressively. For many banks, high
example, the French banking sector is large relative restructuring costs reduce up-front earnings, effectively
to the economy and has a high share of savings and precluding banks from making the cuts needed to
cooperative banks. The banking systems in Austria and become more efficient. Likewise, many branches may
Germany have a large number of banks, low concen- have operating leases that run for a number of years,
tration, and a large share of savings and cooperative preventing the realization of short-term savings from
banks. In Italy, Portugal, and Spain there is a large closing branches. Demographic factors can also affect
number of branches or staff relative to banking assets a decision to maintain branches because older popu-
(there is also a large number of banks and low concen- lations tend to prefer banking in person, rather than
tration in Italy). over the Internet.
There has also been progress in tackling other
More Progress Needs to Be Made in Systems with the structural features. For example, Spain underwent a
Biggest Challenges substantial consolidation in 200912, accompanied by
Some banking systems have also been reduc-
ing costs by cutting excess capacity (Figure1.24, 19This calculation is based on 159banks out of the 172-bank
panels1 and 2). Banking systems in Denmark, sample, representing about 98percent of sample assets.
Figure 1.24. European Banking System Actions to Reduce provides recommendations for actions in several
Costs European countries.
Actions to reduce branches vary signicantly ...
The Burden of Nonperforming Loans Still Needs to
1. Overbranching and Reduction in Branches Be Reduced
400
NLD The euro area as a whole has made progress in alle-
350 se
s pon viating the burden of nonperforming loans on balance
300 e re
ctiv sheets. The formation of new problem loans has slowed
Deposits per branch
ea
(millions of euros)
250 Mor
GBR as the economy has started to recover, write-offs have
200 picked up, and sales of nonperforming loans have
150 SWE increasedcumulative 201016 sales now total about
AUT BEL
100 40percent of the peak level of impaired loans in the
ITA DEU ESP
50 GRE euro area.20
FRA PRT
0 Resolving problem loans should bring real ben-
0 10 20 30 40 50 60
efits. Institutions that have dealt adequately with
Reduction in branches, 200815 (percent)
nonperforming loans should also need to provision
... as do cuts in headcount. less in the future. Banks in Ireland and Spain, in
2. Change in Branches and Headcount particular, have made good progress in reducing
0 nonperforming loans from peak levels. This followed
FRA
a recognition of the systemic size of the problem,
Change in number of employees,
5 o nse
ve resp coupled with firm action to address the overhang,
10 acti DEU AUT
200815 (percent)
e
Mor ITA including through asset management companies, a
PRT
15 BEL strategic approach to restructuring banks, and gov-
20 ernment recapitalization support. But relatively little
NLD GBR
25 reduction, relative to peak levels, has occurred in two
DNK ESP
of the countries with the highest nonperforming loan
30
ratios, Italy and Portugal (Table1.5), and further
35
60 50 40 30 20 10 0 progress needs to be made (shown by the recom-
Change in number of branches, 200815 (percent) mendations in Table1.6). For example, it could take
about six years on average for the countries across
the euro area to resolve the burden of impaired assets
Sources: European Central Bank; and IMF staff calculations.
Note: Data labels in the gure use International Organization for Standardization at current write-off rates and new bad debt forma-
(ISO) country codes. tion rates21though the pipeline of loan transac-
tions suggests that sales of bad assets could pick up,
particularly in Italy.
While actions are being taken to address the debt
reforms to strengthen governance. There has been some
overhang, a number of structural barriers are still
consolidation in the German banking system, and
blocking the disposal of nonperforming loans. Ineffi-
Landesbanken continue to deleverage, with institutions
cient legal frameworks can impede loan recovery and
downsizing balance sheets, increasing their focus on
require banks to provision more. Several of the larger
core business activities, and closing subsidiaries or
distressed asset markets reportedly continue to suffer
foreign operations. In Italy, two large popolare banks
from poor information quality, which lowers buyers
have merged, and reforms to the cooperative bank
reservation prices. The characteristics of loan portfolios
sector, aiming to strengthen governance, have also been
are structurally unattractive in some countriesit is
legislated.
harder for investors to price portfolios consisting of
More progress needs to be made to tackle prof-
itability challenges in the banking systems with the
20Data for sales of nonperforming loans are estimated from data
biggest challenges. The specific actions needed will
in Deloitte 2017 and Pricewaterhouse Coopers 2016.
vary according to the mix of structural factors affect- 21IMF 2016 reached a similar conclusion on the length of time to
ing the profitability in each banking system. Table1.6 resolve nonperforming loans in the euro area.
small, heterogeneous loans to small and medium-sized European G-SIBs have strengthened their capitaliza-
enterprises with collateral of uncertain value than to tion and liquidity positions and are in the process of
price portfolios of homogenous unsecured loans. But restructuring business models by cutting back balance
while these technical issues are important, insufficient sheets and reorganizing businesses. They have also
buffers at banks to absorb additional losses recognized made good progress in writing off legacy assets. But
on sales of bad debts at market prices continues to profitability remains a challenge for many of these
be an impediment. Therefore, the lack of progress banks and virtually none of the European G-SIBs are
on resolving nonperforming loans also reflects weak currently able to approach the profitability of their
earnings and insufficient generation of capital and U.S. peers (Figure1.25, panel1). Of those that have
provisioning buffers. comparable preprovision profitability, several continue
to be hampered by continued high provisions, which
Systemically Important Banks May Also Be Affected lowers their return on assets. But many banks have
by System-wide Problems poor preprovision profit margins and thus require
These system-wide challenges are not only a prob- further restructuring of their business models to
lem within countries: they can affect the profitability improve core profitability (Figure1.25, panel2). While
of global systemically important banks (G-SIBs) European G-SIBs have been making efforts to cut costs
in Europe as well. These institutions are finding it by reorganizing their businesses, these efforts have had
difficult to keep up with their global competitors, varying degrees of success (Figure1.25, panel3).
and in some cases this is partly due to the profitabil- Market pricing of G-SIBs shows differences in
ity problems they are facing in their home countries. investor perceptions of European banks (Figure1.25,
The extent of this domestic impact will depend on panel4). Higher price-to-book ratios and lower credit
the exposure of G-SIBs to their home economies. default swap spreads indicate market conviction that
Although this exposure varies significantly across business models are already robust. In contrast, lower
banks, domestic business represents on average about equity market valuations and higher spreads suggest
half of European G-SIBs total assets and about that investors believe further progress is needed to
40percent of total revenues.22 strengthen business models. Addressing system-wide
22Domestic assets and revenues range from about 10percent
of the total to about 95percent of the total across European financial statements, Bloomberg L.P., SNL Financial, and IMF
G-SIBs, excluding Standard Chartered, based on data from bank staff calculations.
problems together with efforts to address business ment.23 Measures such as the EU Bank Recovery
models would work best together in resolving prof- and Resolution Directive and Total Loss Absorbing
itability challenges, therefore enhancing systemic Capacity rules should limit spillovers from banks to
resilience. sovereigns.24 But it will take some time to build up
sufficient bail-inable liabilities and address bank and
The Sovereign-Bank Nexus Could Reemerge system-wide weaknesses, implying that severing the
The combination of weak profitability in both bank-sovereign nexus remains a work in progress.
domestic banks and G-SIBs, lack of access to private More recently, government bond spreads have risen
capital, and a large stock of unresolved problem loans in France and Italy, and they remain at high levels in
has the potential to reignite systemic risks in some Portugal. This likely reflects a combination of con-
economies. Weaknesses in the Italian and Portuguese
banking systems led to a widening in bank credit 23IMF 2015c discusses these issues in more detail.
default swap spreads in 2016 (Figure1.26, pan- 24The Bank Recovery and Resolution Directive establishes rules
els1 and 2). These banking risks led, in turn, to a within the European Union for recovery and resolution of banks,
including the resolution of nonviable banks, through the bail-in of
rise in associated sovereign spreads through market some creditors, and rules establishing a minimum amount of bail-
concerns about contingent liabilities for the govern- inable instruments (8percent of total liabilities).
BPCE )
0.5 osts
Revenue
HSBC ING BNP
CS
UCG
e q ual c
0.4 es
CIT BARC DB venu 2
0.3 pro
t
r o t (re
BARC WFC High NOR RBS
SG ting
p
BNP BOA pro
t CA pera
0.2 DB CA SG ING
JPM
Mid o v i s ion o 1
t r
0.1 RBS HSBC pr o prep
UBS BPCE NOR MS GS Low Zero
0.0 CS 0
0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Preprovision operating prot Operating costs
... that have been unevenly tackled ... ... as suggested by market indicators.
3. Change in Costs and Headcount, 201115 4. Bank Price-to-Ratios and Five-Year Credit Default Swap
Spreads, March 2017
10 180
SAN DB UCG 160
Change in headcount (percent)
(basis points)
NOR CS 120
SAN
20 SG CA RBS BNP 100
MS GS
30 UCG 80
CIT BOA JPM
RBS WFC 60
40 HSBC UBS
ING 40
50 20
40 20 0 20 40 60 80 100 120 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Change in operating costs/assets (basis points) Price-to-book ratio
Sources: Financial statements and management presentations from banks listed in the note; Bloomberg L.P.; SNL Financial; and IMF staff calculations.
Note: In panels 1 to 3, conduct costs have been removed. GSIB = global systemically important bank; BARC = Barclays; BNP = BNP Paribas; BOA = Bank of America;
BPCE = Groupe BPCE; CA = Crdit Agricole; CIT = Citigroup; CS = Credit Suisse; DB = Deutsche Bank; GS = Goldman Sachs; HSBC = HSBC Holdings; ING = ING
Bank; JPM = JPMorgan; MS = Morgan Stanley; NOR = Nordea; RBS = Royal Bank of Scotland; SAN = Santander; SG = Socit Gnrale; UBS = UBS Group; UCG =
Unicredit; WFC = Wells Fargo.
cerns about higher political risks and government debt portfolios. These wholesale funding and trading risks
burdens. There is a risk that higher sovereign spreads would be especially problematic if financial conditions
could spill back to the banking sector. First, sovereign were to tighten sharply.
downgrades could increase bank wholesale funding
costs and reduce the amount of assets that banks have Brexit25 further Complicates Challenges for System
available as acceptable collateral. Second, although Efficiency and Financial Stability
banks have generally reduced their holdings of local Box1.3 assesses the potential financial stability and
government bonds, some institutions continue to hold cost implications of Brexit, albeit with a high degree of
a significant amount of these bonds on their balance
sheets and could face mark-to-market losses on the 25Brexit refers to the June 2016 U.K. referendum result in favor of
bonds held in trading books and available-for-sale leaving the European Union.
Figure 1.26. Bank and Sovereign Nexus ating any too-big-to-fail concerns. To determine weak
links in banking systems with significant asset quality
1. Sovereign and Bank Spreads, by Country
(Basis points) challenges, consideration could be given to targeted
Bank Sovereign asset quality reviews for banks that have not undergone
600
Portugal such an exercise. Regulators should then take action to
500 resolve unviable institutions to remove excess capacity
400 from banking systems. Authorities should also focus
on removing system-wide impediments to profitabil-
300 Italy ity. The precise prescription, however, will vary across
200 Spain countries (Table1.6).
France
100
Banks have the primary responsibility for developing
sustainable earnings by tackling business model prob-
0 lems. While there is no single business model that will
2014 15 16 17 2014 15 16 17 2014 15 16 17 2014 15 16 17
work for all, banks should continue to restructure their
2. Sovereign and Bank Spreads, 201517 business to enhance returns and invest in technology
(Basis points)
to increase medium-term efficiency. But supervisors
300
also have a role to play. Encouragingly, authorities
Portugal
250 are increasingly emphasizing the examination of bank
business models in their supervisory frameworks. Both
Sovereign spreads
200
the Single Supervisory Mechanism, in its Supervisory
150 Italy
Review and Evaluation Process, and the U.K. Pruden-
100 Spain tial Regulation Authority are taking a forward-looking
approach to assessing the sustainability of bank busi-
50
France
ness models. If any banks are reacting to profitability
0
0 100 200 300 400 500 challenges by taking greater risks, authorities should
Bank spreads consider macroprudential or other regulatory measures
to reduce the probability of future problems.
Further action is needed to fully resolve the burden
Sources: Bloomberg L.P.; and IMF staff calculations.
Note: The gure shows mean ve-year senior credit default swaps for banks from of nonperforming loans.26 A number of initiatives
each country and ve-year sovereign spreads to Bunds. Panel 1 shows a 10-day have been undertaken, which should help speed
moving average. Panel 2 shows the evolution of sovereign and bank spreads from
2015 (dot) to 2017 (arrow). up bad debt disposal. The European Central Bank
has published guidance to banks on how to tackle
nonperforming loans.27 In Italy, two Atlante funds
have been set up by a group of financial institutions
uncertainty about the final outcomes of negotiations. and banking foundations, and the authorities have
In particular, London is susceptible to losing some established a public guarantee on senior tranches of
of its predominance as a global financial center, with securitized bad loans. Several countries have also put
attendant costs related to the loss of economies of scale in place reforms to legal frameworks to help alleviate
in conducting financial activities. Regulatory challenges the process of resolving problem loans. Accounting
and complexities may also increase, although lower standards (International Financial Reporting Stan-
concentration in one center may bring diversification dard 9) should also ensure greater forward-looking
gains to financial stability. provisioning when phased in and may change the
dynamics of loss recognition by making banks more
More Comprehensive Efforts Are Needed to Address proactive. But supervisors should ensure that banks
System and Business Model Challenges adopt ambitious, time-bound strategies for the dis-
Banks should seek out opportunities to increase posal of nonperforming loans. Authorities should also
weak revenues and reduce high operating costs. Any 26See, for example, European Central Bank 2016; IMF 2015a,
consolidation should also go hand in hand with gov- 2015b, and 2016; and Jobst and Weber 2016.
ernance reforms, where needed, and should avoid cre- 27European Central Bank 2017.
encourage the development of a market for problem Completing the regulatory reform agenda is vital
loans. To help erode bank-sovereign links, consid- to ensure that weaknesses are addressed and to reduce
eration should be given to reducing the thresholds uncertainty. In particular, it is important to finalize an
for the direct recapitalization of viable banks under agreement on the Basel III package of reforms, includ-
the European Stability Mechanism and a common ing the revision of the standardized approach to the
deposit insurance scheme should be established in calculation of risk-weighted assets and boundaries to
the euro area. the use of internal models to assess risks (Box1.2).
Box 1.1. Could Fragilities in Offshore Dollar Funding Exacerbate Liquidity Risk?
Global liquidity risks could be amplified by the currency resulting imbalance in the supply and demand of
mismatch between non-U.S. banks assets and liabilities, offshore dollars has led to a persistent premium in the
especially if U.S. interest rates were to increase sharply price to swap local-currency funding into dollars via
and the dollar were to appreciate. Risks would be greatest foreign exchange swaps, known as the cross-currency
for those banking systems that are highly dependent on swap basis.3
short-term dollar funding for long-term assets. After having steadily widened over 201416,
cross-currency swap bases have narrowed consid-
In recent years, monetary policy divergence between erably since late 2016 (Figure1.1.1). It is unclear
the United States and other economies has led what has reduced the cost of dollar funding over this
some non-U.S. banks to accumulate higher-yielding recent period, though several factors point to greater
foreign-currency assets at a pace that has exceeded availability of dollar fundingmost notably a modest
their funding in those currencies. In many cases, U.S. pickup in U.S. prime money fund assets, greater
dollardenominated assets have outpaced the supply demand from other investors less affected by regula-
of U.S. dollar funding via deposits, certificates of tory balance sheet constraints (for example, corpora-
deposit, commercial paper, and other sources.1 At the tions, offshore money funds, private liquidity funds),
same time, regulatory changes and money fund reform and central bank efforts to provide larger backstops.
have limited the supply of U.S. dollar funding.2 The
cker Rulewhich prohibits firms from engaging in proprietary
trading activities in foreign exchange forwards and swaps; and
1McGuire and von Peter 2012. (3)over-the-counter derivatives reformwhich increased the
2These include, for instance, bank regulatory reforms, notably capital and minimum margin requirements for cross-currency
adjustments to (1)the capital ratiothe cross-currency swap swap bases.
basis has been more volatile since the crisis, and greater volatility 3Under no-arbitrage conditions, the cost of funding in dollars
increases a banks value-at-risk measure, which in turn affects the should be equal to the combined cost of funding in a foreign
risk-weighted assets calculation and capital charges; (2)the Vol- currency and swapping the funds for dollars.
20
20
40
Negative interest rates
Sovereign debt crisis
60
U.S. MMF reform
Subprime crisis
80
100
2007 08 09 10 11 12 13 14 15 16 17
Sources: Bank for International Settlements; Bloomberg L.P.; and IMF staff estimates.
Note: Weights are based on daily average foreign exchange swap turnover versus the U.S. dollar for the euro, Japanese
yen, British pound, and Swiss franc. MMF = money market fund.
Percent
believed to be held in U.S. banks or U.S. investments. Maturity gap
Based on what happened after the previous repatria- 1 $2.9 trillion 5.0
tion tax holiday in 2004 when U.S. companies repa- 0
triated $362billion, tax incentives under a corporate 4.5
1
tax reform could lead to repatriation of a significant
2 4.0
portion of U.S. dollar assets. Other administrative 2007 08 09 10 11 12 13 14 15 16
measures, such as bank ring-fencing, have the potential
to increase frictions in the supply of dollar funding, ... while emerging market banks reliance on short-
thus increasing the cost, and may lead to a more frag- term foreign-currency funding has been steady.
mented offshore dollar market. 2. Emerging Market Banks
Advanced economy banks, in particular, have 1.0 5.0
become reliant on cheap short-term foreign-currency 0.8 4.5
funding for their long-term foreign-currency assets 0.6 4.0
Trillions of U.S. dollars
Percent
0.2
and long-term foreign-currency liabilitieshas nearly Maturity gap 2.5
0.0
$0.3 trillion 2.0
doubled to $2.9trillion. As a percentage of total 0.2 1.5
assets, the maturity gap grew from 4.4percent to a 0.4 1.0
high of 6.1percent in November 2015. 0.6 0.5
Banks are vehicles for maturity transformation, and 0.8 0.0
interest rate risk is an intrinsic part of banking. Banks 2007 08 09 10 11 12 13 14 15 16
also actively manage foreign exchange and interest
rate risk via derivative hedges. However, hedging
Sources: Bloomberg L.P.; IMF, International Financial
introduces counterparty risk and does not eliminate Statistics database, Monetary and Financial Statistics
rollover risk. When local-currency assets come under database; and IMF staff calculations.
funding stress, the local central bank can usually Note: The foreign-currency maturity mismatch is the
difference between long-term foreign-currency assets
provide almost limitless liquidity to banks via tempo- and long-term foreign-currency liabilities. Assets include
rary funding transactions. But when funding strains 50 percent of other deposits, 50 percent of securities,
arise for foreign-currency-denominated assets, local other loans, 50 percent of equities, insurance, derivatives,
trade credit, other accounts receivable residential, and
central banks can provide liquidity only from their accounts receivable. Liabilities include 50 percent other
finite foreign-currency reserves or by tapping foreign deposits ex-broad money, 50 percent of securities, other
exchange swap facilities and credit lines with other loan liabilities, insurance, derivatives, trade credit liabilities,
other accounts payable residential, and accounts payable.
official institutions. If offshore dollars were to become In panel 1, advanced economies include Denmark, Finland,
a scarcer resource, the resulting frictions could lead France, Germany, Greece, Ireland, Israel, Italy, Japan,
banks to reduce their global footprint or to increase Korea, Luxembourg, Netherlands, Norway, Portugal, Spain,
and Sweden. In panel 2, emerging markets include Brazil,
Bulgaria, Chile, Colombia, Egypt, Hungary, Indonesia,
Malaysia, Mexico, Nigeria, Philippines, Poland, South Africa,
4Borio and others 2016. Thailand, and Turkey.
Figure 1.3.1. Measures of Financial Linkages between the United Kingdom and the
European Union
(Percent)
0 20 40 60 80 100
Sources: (1) Bruegel; (2) Bank for International Settlements, as of April 2016; (3) Oliver Wyman; (4) Ofce of National
Statistics; (5) Bruegel; and (6) PricewaterhouseCoopers.
Note: EU = European Union; FDI = foreign direct investment; FX = foreign exchange; IPO = initial public offering; OTC =
over the counter; UK = United Kingdom.
Summary
A
dvanced economies have experienced a prolonged episode of low interest rates and low growth since
the global financial crisis. From a longer-term perspective, real interest rates have been on a steady
decline over the past three decades. Despite recent signs of an increase in long-term yields, particularly
in the United States, the experience of Japan suggests that an imminent and permanent exit from a
low-interest-rate environment need not be guaranteed. A combination of slow-moving structural factors, notably
population aging and slower productivity growth common to many advanced economies, could conceivably gener-
ate a steady state of lower growth and lower nominal and real interest rates in these countries.
What would be the consequences for the financial sector of such a scenario? This chapter examines this question,
abstracting from the role of monetary policy and from temporary effects. The chapter argues that the persistence
of a prolonged low-interest-rate environment would present a considerable challenge to financial institutions. Over
the long term, the scenario would entail significant changes to the business models of banks, insurers, and pension
funds and the products offered by the financial sector.
In such an environment, yield curves would likely flatten, lowering bank earnings and presenting long-lasting
challenges for life insurers and defined-benefit pension funds. If bank deposit rates cannot drop (significantly)
below zero, bank profits would be squeezed even further. Smaller, deposit-funded, and less diversified banks would
be hurt most, which could increase the pressure to consolidate. As banks reach for yield at home and abroad, new
financial stability challenges may arise in their home and host markets. These hypotheses are supported by the
experience of Japanese banks.
Low growth and aging populations would likely lower credit demand by households and firms and increase
household demand for liquid bank deposits and transaction services. Consequently, in this scenario, domestic
banking in advanced economies may generally evolve toward provision of fee-based and utility services.
Pension arrangements and the products and business models of life insurers would also likely change signifi-
cantly in the long term. In this scenario, defined-benefit pension plans provided by employers would tend to
become less attractive relative to defined-contribution plans, which offer more portability. Rising longevity would
likely boost the demand for health and long-term care insurance. Demand for guaranteed-return, long-term sav-
ings products offered by insurers could be expected to weaken, while that for passive index funds offered by asset
management firms would likely grow.
Policies could help ease adjustment to such an environment. Prudential frameworks would need to provide
incentives to ensure longer-term stability instead of falling prey to demands for deregulation to ease the short-term
pain. For banks, policies should help facilitate smooth consolidation and exit of nonviable institutions, while limit-
ing excessive increases in risk taking and ensuring that the too-big-to-fail problem does not worsen. Implementing
economic solvency requirements that encourage life insurers to undertake necessary adjustments to their business
models would be vital. Surveillance and regulation of asset management activities would become more important
as this industrys share in the financial sector grows.
JorgeChan-Lau, Qianying Chen, Gee Hee Hong, Mitsuru Katagiri, from that of low real rates, but recent experience, particularly in
Oksana Khadarina, John Kiff, Frederic Lambert, Sheheryar Malik, Japan, has been marked by both low nominal and low real rates.
WinMonroe, Nobuyasu Sugimoto, and Kai Yan, under the general 4Various other studies have examined the effects of temporary
guidance of Gaston Gelos and Dong He. Breanne Rajkumar and monetary policy measures under low interest rates. For a recent
Annerose Wambui Waithaka provided editorial assistance. paper analyzing the effects of negative interest rate policies on mon-
1The concept of natural rates was introduced by Wicksell (1936). etary transmission and bank behaviors, see IMF 2017. The study
Holston, Laubach, and Williams 2016 present evidence of falling finds that these policies have not had major side effects on bank
natural interest rates in a number of advanced economies. profits, payment systems, and market functioning.
Figure 2.1. Interest Rates, Term Spreads, and Volatility in Advanced Economies
Real interest rates have been decreasing over the past three Nominal interest rates have fallen.
decades.
1. Real Short-Term Interest Rates, 19832015 2. Selected Three-Month Treasury Bill Yields, 19992015
(Percent) (Percent)
10 7
United States United Kingdom
8 Germany Japan 6
6 5
4
4
3
2
2
0
1
2 0
United States United Kingdom
4 1
Germany Japan
6 2
1983 86 89 92 95 98 2001 04 07 10 13 15 1999 2001 03 05 07 09 11 13 15
4 Germany Japan
1.0
3
0.8
2
0.6
1
0.4
0
1 0.2
2 0.0
1999 2001 03 05 07 09 11 13 15 United States United Kingdom Germany
Sources: Thomson Reuters Datastream; IMF, World Economic Outlook database; and IMF staff calculations.
Note: Term spreads in panel 3 are dened as the difference in yield between a 10-year government bond and a three-month Treasury bill. Low-for-long periods
in panel 4 are dened as those when the 10-year yield was less than 2 percent.
interest rates on financial intermediation and financial butions do shed light on them. First, it provides a new
stability. What is the long-term impact on profits and analytical framework to help understand the behavior
solvency of financial institutions? How does it depend of the term structure of interest rates in an equilibrium
on their business models? Will the existing menu of with low natural rates of interest. This is important
financial products and services survive? How will these given the relevance of the slope of the yield curve for
circumstances change the relative importance of banks, the profits and solvency of different types of financial
insurers, pension funds, and asset managers in the institutions. Second, it extends a standard model of
financial system? In taking this approach, the chapter bank profitability to such an equilibrium to assess the
seeks to examine the long-term implications of the impact on banks according to their business models,
proposed scenario and its underlying structural drivers and compares the insights with Japans experience.
for financial intermediation. Third, it empirically assesses the impact of low interest
While not aiming to offer definitive and exhaustive rates on banks profits, distinguishing between situa-
answers to these questions, the chapters novel contri- tions when interest rates are expected to remain low
for a long time and other periods. Fourth, it discusses To the extent that population aging and rising lon-
implications for insurers and pension funds, simulating gevity are key forces behind the scenario, there are
alternative portfolio choices and discussing the viability likely to be major changes to demand for banking
of typical pension and insurance products in the and insurance products. Aging would likely reduce
low-natural-rate equilibrium. Fifth, it offers a discus- household demand for credit and increase demand
sion of how such a scenario affects households asset for transaction services from banks. In combination
allocations and the role of asset managers in financial with increased longevity, it would likely increase
intermediation. Sixth and last, it discusses potential demand for health and long-term care insurance,
implications for financial stability. with ambiguous implications for life annuities.
The main findings for this scenario are as follows: Retail demand for asset management products
The yield curve would be flatter compared to an would continue to grow, in particular for passive
equilibrium with higher rates and growth. modes of index investing targeted at minimizing
Although lower interest rates may boost banks earn- management fees.
ings in the short term, they hurt profitability in the Pressure on smaller banks would lead them to
steady state once they fall below a particular positive consolidate among themselves or with larger banks.
threshold. Smaller, geographically undiversified, depos- Credit demand would likely be lower in this sce-
it-funded banks would be hurt most in such a scenario. nario given an aging population and lower produc-
Tail risk exposure could increase.5 Banks tend to tivity growth. Domestic bank lending would likely
adopt different strategies in reaching for yield, shrink, focusing more on small businesses and less
depending on their business models. Smaller, depos- on households and large firms. Business models in
it-funded banks typically take on more interest rate advanced economies would tend to evolve toward
risk by increasing the duration of bond portfolios. fee-based and utility banking services.
Large banks are likely to increase risk exposures in Insurers would likely cede some of their savings
foreign countries that offer higher returns (in par- business to asset managers and banks over the long
ticular, emerging market economies) and rely more term. The reason for this shift is that, at low rates,
heavily on wholesale funding markets to do so. their guaranteed products are relatively less attractive.
Life insurers and pension funds would face a long- Although insurers may respond, in part, by switching
lasting transitional challenge to profitability and their focus to unguaranteed savings products, they
solvency, which is likely to require additional capital. could face tough competition from asset managers.
This challenge arises because some of them would find Health and long-term care businesses would likely
it difficult to meet cash outflows on large stocks of grow strongly as people age and live longer.
existing liabilities contracted in past periods of higher The pooled management of household life cycle risks
interest rates by only altering asset portfolios. More- would likely decline more rapidly. Employers could be
over, many of their other business lines may struggle expected to increasingly move away from defined-ben-
to show profit in the tepid growth environment. efit and toward defined-contribution pension plans,
although the pace and extent of this transition may
All of this would likely result in major changes vary significantly across advanced economies.
in the long term to household demand for financial
products and asset allocation, the menu of services the The key policy challenge in this scenario would be
financial sector offers, and the relative role of institu- to successfully balance multiple objectives, including
tions versus markets in financial intermediation.6 the following:
For banks, providing a legal and regulatory frame-
5Risk taking may arise due to competitive pressures, nominal
work that facilitates smooth consolidation should
return targets, or risk shifting in response to lower interest rates,
go hand in hand with efforts to limit excessive
among other factors.
6The discussion of the potential long-term impact of the scenario risk taking in an environment with lower expected
on financial intermediation seeks to take into account the interrelation returns and avoid a worsening of the too-big-to-fail
across different sectors and key drivers. However, it is not based on problem. This includes containing incentives to
a formal general equilibrium model, and does not aim to capture all
potential accompanying factors, such as changes in labor supply (includ- increase exposure to tail risk from widening maturity
ing changes in retirement ages), regulations, or social safety nets. mismatches, higher wholesale funding, and foreign-
currency exposures. A similar challenge would be returns increase when economic shocks reduce other
to reap benefits from banks higher engagement sources of income, investors would be willing to pay a
in emerging market economies while containing premium to hold the bond (a negative term premium).
potential new financial stability risks in home and If bond returns decline in tandem with other sources
host countries. of income, investors require a premium to be paid to
Providing incentives to undertake necessary business them (a positive term premium).
model adjustments (life insurers) and contain When the equilibrium rate of economic growth is
gambling for resurrection (certain pension funds) high and nominal and real rates are not close to zero
would be key in this scenario. This would strengthen (normal economy), the model implies an upward slop-
the case for implementing economic solvency ing nominal yield curve (Figure 2.2, panel 1).8 When
requirements that ensure recognition of the costs of inflation goes up, incomes fall and bond returns decline
guarantees and options embedded in insurance and due to the central banks policy response of raising inter-
pension products. est rates. Because bonds worsen the impact of inflation
Surveillance and regulation of asset management shocks on incomes, bond term premiums are positive.
activities would become even more important as this The key distinguishing feature of the low-for-long
industrys share of the financial system grows. In economy is a zero lower bound on short-term nominal
particular, further strong growth of index invest- interest rates. It is assumed that the central bank can-
ing could entail new financial stability challenges. not, or will not, lower policy interest rates below zero,
Closing significant data gaps would also be essential which prevents it from responding by cutting interest
to allow for effective macroprudential surveillance of rates in response to negative (noninflationary) shocks
this sector. to real income.9 This means that bond returns remain
resilient in the face of such shocks in a low-for-long
economy compared with what happens in a normal
The Term Structure of Interest Rates economy, which results in lower term premiums and
Thissection discusses the shape of the yield curve in an flatter yield curves (Figure 2.2, panel 2).
economy with very low natural rates. The slope of the The decline in term premiums at the zero lower
yield curve is important for the financial system, since it bound can also be interpreted as a consequence of
affects all financial institutions that tend to have maturity investors perceiving a lower risk of holding long-term
mismatches between their assets and liabilities. The section securities. Once short-term interest rates are near the
summarizes insights from a new model that applies and zero lower bound and are expected to stay there for the
extends the techniques of existing consumption-based foreseeable future, their sensitivity to macroeconomic
asset pricing models to incorporate a zero lower bound on news drops because central banks reaction functions
nominal interest rates.7 are constrained.10 In such a situation, investors are
more willing to hold long-term bonds, lowering the
The spread between the yield on a longer-maturity term premium.
bond and the short-term interest rate is the sum of
two components. These are the market expectations of 8The results, as depicted in Figure 2.2, correspond to a parameter-
how the short rate will evolve between today and the ization of the model described in Annex 2.1. These results are robust
to modeling endowment and inflation shocks as a joint process cali-
maturity date of the longer-term bond, and the bonds
brated through a vector autoregression based on data from Germany,
term (risk) premium. Around a steady state in which Japan, the United Kingdom, or the United States.
the short rate is at its long-term equilibrium level, the 9Strictly speaking, it is sufficient for there to be an effective, pos-
slope of the yield curve is driven entirely by the sign sibly negative, lower bound on nominal short-term interest rates so
long as it is close to zero. See Vials, Gray, and Eckhold 2016 for a
and magnitude of (nominal) bond term premiums. discussion of effective lower bounds for monetary policy rates.
A simple way to understand the term premium 10The flattening of yield curves due to compression in term premi-
on a long-term bond is that it reflects the degree to ums is a robust result across term structure models with a zero lower
bound. Nakata and Tanaka (2016) and Gourio and Ngo (2016)
which bond returns provide insurance against shocks investigate the term premium at the zero lower bound in a New
to other sources of an investors income. If bond Keynesian asset pricing model developed by Campbell, Pflueger, and
Viceira (2012). In their models, however, the zero lower bound is
a temporary phenomenon following a crisis rather than a persistent
7Annex 2.1 contains details of the model and the literature. element of a low-for-long economy.
Figure 2.2. Term Premiums in Economies with Normal versus futurebut this favorable impact dissipates the
Low Natural Rates longer interest rates remain low. Empirical studies
covering banks in the United Kingdom (Alessandri
1. Normal Economy and Nelson 2012) and the United States (English,
(Percent)
2.0 van den Heuvel, and Zakrajsek 2012) show the
1.5 existence of separate channels for short- and medium-
1.0
term effects of interest rate changes on banks interest
margins, profits, and equity valuations. Banks tend
0.5
to lose profitability from longer-lasting drops in
0.0
interest rates in direct proportion to how much they
0.5
engage in maturity transformation and make use of
1.0 deposit funding. However, falling interest rates boost
1.5 bank profits and equity values in the short term due
2.0 to gains in the value of collateral, valuation gains
0 2 4 6 8 10
Term (years) on mark-to-market assets, and lower default risk on
loans repriced to lower interest rates.11 Banks appear
2. Low-for-Long Economy to respond to falling rates by increasing risk taking
(Percent) through higher leverage.12
2.0
This literature does not provide guidance on several
1.5
questions of interest in a low-for-long economy.
1.0 What is the long-term impact on profits when banks
0.5 operate in such an environment? Does this impact
0.0 strengthen as interest rates go ever lower? Are some
0.5 bank business models especially affected? Are signifi-
1.0 cant changes to the market structure of the banking
1.5 industry likely? This chapter addresses these issues
2.0 using a three-pronged approach. First, the section
0 2 4 6 8 10 provides a new theoretical model of banking in a low-
Term (years) for-long economy. Next, the insights of this model
are applied to interpret the experience of Japanese
Source: IMF staff calculations, based on the model described in Annex 2.1. banks over the past decade. The section concludes
with an empirical examination of the impact on bank
profitability and equity values and how these depend
on banks business models.
Banking with Low Natural Rates of Interest
This section augments the literature in two ways. First,
it shows that with an unchanged yield curve, even 11Brunnermeier and Sannikov (2016) demonstrate that the
permanently lower interest rates need not affect banks adverse short-term impact of an increase in interest rates can be
earnings. Second, it clarifies how a zero lower bound on amplified through liquidity spirals (deteriorating net worth increases
deposit rates generates pressure on bank interest margins bank risk aversion, which lowers the market value of assets and
lending volumes) and disinflationary spirals (the safe-asset value of
and profits in an equilibrium with a low natural rate. cash increases).
These insights are applied to study the experience of 12This is consistent with theoretical findings of DellAriccia,
Japanese banks since 2000 and the wider cross-country Laeven, and Marquez (2014). DellAriccia, Laeven, and Suarez
(forthcoming) find that U.S. banks risk taking responds similarly to
experience. The analysis also explains how the impact changes in interest rates induced by monetary policy. Focusing on
of this low-natural-rate equilibrium depends on bank the impact of unconventional monetary easing in the United King-
business models. dom, the United States, and the euro area in recent years, Lambert
and Ueda (2014) find that it is associated with deterioration of bank
credit risk and delayed balance sheet repair. Chodorow-Reich (2014)
Previous studies have clarified that negative interest does not find evidence of increased risk taking by U.S. banks in
rate shocks increase bank profits in the immediate response to unconventional monetary policies.
Insights from Theory the Bank of Japan adopted the zero interest rate policy
A simple model of banking is explored to show how bank in the early 2000s, with the exception of the extraor-
profits evolve in a low-rate equilibrium.13 dinary period of 200708. Long-term interest rates
have also been low since the early 2000s and recently
Bank profits fall significantly in a low-for-long econ- declined further, particularly after the Bank of Japan
omy if deposit interest rates are subject to a zero lower adopted policies of quantitative and qualitative mon-
bound (Figure 2.3; Box 2.1).14 Banks interest margins etary easing in 2013 and of negative interest rates in
are (almost) independent of the level of market interest 2016.
rates if they can flexibly adjust loan and deposit rates Econometric analysis of the drivers of bank net
in response to changes in steady-state market interest interest margins supports the predictions of the theo-
rates. Once deposit rates hit the zero lower bound, retical model (Figure 2.4). An assessment of the behav-
banks can no longer maintain spreads between loans ior of Japanese banks asset returns, funding costs,
and deposits, reducing net interest income under lower and market interest rates demonstrates that banks
equilibrium market interest rates. interest margins have fallen primarily in response to
Several implications ensue for the business models of the narrowing of funding spreads once deposit rates hit
different types of banks. Banks able to operate interna- the zero lower bound in the mid-2000s.16 Although
tionally increase their exposure to countries where rates market interest rates have remained close to zero since
of return remain favorable, notably emerging market the 1990s, deposit rates first approached the zero
economies. They can be expected to increase reliance lower bound in the mid-2000s. Bank net interest
on wholesale funding in foreign currency (within exist- margins then gradually and steadily fell, particularly
ing regulatory limits) to finance this expansion. More for regional and small regional cooperative financial
generally, banks that raise a larger proportion of their institutions known as shinkin banks. Japanese banks
funding from capital markets will be less susceptible to have not introduced negative deposit rates or charged
the squeeze in interest margins and incomes induced additional fees, such as account maintenance fees, on
by the zero lower bound. Scale efficiencies in managing deposits even in the face of almost zero deposit spreads
deposits would imply incentives for consolidation. At (Bank of Japan 2011).17
the same time, scale efficiencies in the costs of manag- The relative performance of Japanese banks across
ing wholesale funding would mean that larger banks business models also confirms the theoretical predic-
will be more inclined to seek this form of financing. tion that resilience to the low-for-long steady state
improves with diversification (Figure 2.5). Smaller,
domestically oriented, deposit-dependent regional and
Lessons from Japan shinkin banks have sought to counter the compression
The Japanese economy over the past decade provides of net interest margins primarily through expansion
the closest real-world approximation to a steady state or adjustment of their domestic balance sheets. When
with low growth and natural rates. The insights from benefits to this strategy declined, they engaged in cost
the theoretical model can thus be weighed against the cutting and consolidation. Large internationally active
experience of Japanese banks over this period.15 Japan
has faced low interest rates for more than a decade. 16The analysis uses an error-correction model in the spirit of Gam-
Short-term interest rates have been close to zero since bacorta (2008). The model assumes that asset returns and funding
costs are in a stable relationship with market interest rates in the
long term, and that deviations from this relationship shrink gradu-
13The model abstracts from the decrease in bank earnings due to ally in the short term. Moreover, the long-term relationship changes
yield curve flattening, focusing instead on a new mechanism that has depending on the level of market interest rates. The parameters
not been explored in the existing literature. Brunnermeier and Koby governing the long-term relationship and the short-term dynamics
(2016) explore a model with similar features to examine limits to are simultaneously estimated for a panel of Japanese banks.
monetary policy. 17It is important to focus on the past decade when examining
14The existence of an effective lower bound friction on deposit the evolution of bank net interest margins and net interest income.
rates is sufficient to generate this result for interest rate levels around First, deposit rates hit the zero lower bound only at the start of this
and below this lower bound. period. Second, earlier hits to Japanese banks profits in the period
15Box 2.2 describes the experience of U.S. banks, which shares of low interest rates were the result of losses during the banking
some, but not all, characteristics of Japanese banks adaptation to the crisis, which had very different origins (Caballero, Hoshi, and
prolonged low-interest-rate environment. Kashyap 2008).
Deposit spreads are squeezed at low rates ... ... compressing margins and prots.
1. Rates 2. Ratios
(Percent) (Percent)
4.5 8
Lending rates Return on equity
4.0 Net interest margin 7
Market rates
3.5 Deposit rates Return on assets 6
3.0
5
2.5
2.0 4
1.5 3
1.0 2
0.5 1
0.0 0
2.0 1.5 1.0 0.5 0.0 0.5 2.0 1.5 1.0 0.5 0.0 0.5
Growth rate of the economy Growth rate of the economy
(percent) (percent)
Banks respond by expanding lending abroad ... ... to maintain margins and prots.
5. Domestic and Foreign Loans 6. Net Interest Margins
Domestic loans (Percent)
0.5 Net interest margins (without foreign loans) 2.5
Foreign loans
0.4 Net interest margins (with foreign loans)
0.4 2.0
0.3
0.3 1.5
0.2
1.0
0.2
0.1 0.5
0.1
0.0 0.0
2.0 1.5 1.0 0.5 0.0 0.5 2.0 1.5 1.0 0.5 0.0 0.5
Growth rate of the economy Growth rate of the economy
(percent) (percent)
banks, on the other hand, have sought to expand the Figure 2.4. Japan: Evolution of Bank Net Interest Margins in
diversification in their income sourcing. This strategy Normal and Low-for-Long Settings
(Percent)
has been more effective, and these banks have faced
little pressure to cut costs or to consolidate. Asset returns and funding costs normally adjust proportionally as
Assets and earnings: Almost all the growth in the interest rates fall ...
major banks assets can be accounted for by the 1. Normal Steady State
3.5
increase in international loans and securities, Asset returns
through both foreign branches and mergers with 3.0 Five-year rate
and acquisitions of foreign entities. The major 2.5 Funding costs
banks have expanded their fee businesses outside 2.0
Japan, including in emerging marketsfor exam-
1.5
ple, through the coordination of syndicated loans.
Consequently, the share of income from interna- 1.0
uity financing.
1.8
Operational costs: Regional and shinkin banks have
cut these costs substantially by rationalizing their 1.6
branch networks in the face of lower profitability.
This is in contrast to the major banks, which have 1.4
kept operational cost ratios almost flat for the past
1.2
two decades. NIM (ve-year rate drop: 2 to 1 percent)
Consolidation has enhanced the effectiveness of NIM (ve-year rate drop: 1 to 0 percent)
1.0
strategies to maintain profits in the low-for-long
environment. Consolidation can raise profitabil-
ity by both cutting fixed operational costs and by Sources: Fitch Connect; and IMF staff calculations.
Note: NIM = net interest margin.
increasing the banks monopolistic power in deposit
and loan markets. Recently, regional banks have
pursued consolidation by forming financial groups
to enhance their profitability.
Figure 2.5. Japan: Banks Adaptation to the Deposit Rate Lower Bound Period
Smaller banks have taken more interest rate risk. Large banks have expanded abroad.
1. Average Asset Maturity, 200015 2. Share of Foreign Business of Major Banks, 200615
(Years) (Percent)
4.0 35
Major banks
3.5 Fee and commission income 30
Regional banks
3.0 Net interest income on loans
Shinkin banks 25
2.5
20
2.0
15
1.5
10
1.0
0.5 5
0.0 0
2000 05 10 15 2006 07 08 09 10 11 12 13 14 15
Smaller banks have also cut costs ... ... in part, by closing branches.
3. Operational Cost Ratios, 200015 4. Number of Branches, 200515
(Percent of total earning assets) (Index; 2005 = 100)
1.8 106
Major banks Major banks
Regional banks Regional banks 104
Shinkin banks Shinkin banks
1.4 102
100
1.0 98
96
0.6 94
2000 05 10 15 2005 10 15
Sources: Bank of Japan; Fitch Connect; Japanese Bankers Association; and IMF staff calculations.
Alternative strategies have different risk implications. Cross-Country Experience with Prolonged Low Interest
Major banks have maintained net interest margins and Rates18
profits at the cost of higher cross-border market and Impact of Low-for-Long Episodes on Bank Profits
counterparty risk. In particular, given the growing share of
A cross-country analysis aims to compare, with other
wholesale foreign currency funding used by major banks,
periods, bank profitability at times when interest rates
the adverse impact of a tightening in these markets could
are low and are expected to remain low for the foresee-
be large (Chapter 1 of the October 2016 Global Financial
able future. The approach uses a combination of criteria
Stability Report [GFSR]). Already, the costs of funding in
to demarcate these two types of periods. The first is that
this market have risen significantly due to market friction
the short-term yield is below 1percent. The second is
(Avdjiev and others 2016). Shinkin banks have increased
that the on-the-run, 10-year nominal bond yield is
interest rate risk by extending the average maturity of
lower than the historical average of short-term policy
domestic bonds, but risk-adjusted returns have nonethe-
less increased modestly, given unusually low inflation and
interest rate volatility during the past decade. 18Details of the empirical framework are in Annex 2.2.
interest rates.19 The reason for applying a double-thresh- Figure 2.6. Prolonged Low Interest Rates and Bank Prots
old criterion is that it is typically satisfied only when
both economic growth and nominal and real interest Bank prots are signicantly lower under prolonged low interest rates.1
rates have been low for a considerable timeeven if the 1. Return on Equity
dip in these measures initially resulted from an eco- (Percent)
14
nomic downturn or macro-financial instability.20 The
analysis also explores how the impact on profits depends 12
on banks business models (Table 2.1).21 10
8
Profits Are Lower in Periods of Prolonged Low
Interest Rates 6
individual balance sheet indicators and on an approach in which Sources: Bloomberg L.P.; Fitch Connect; IMF, Monetary and Financial Statistics
a statistical model combines these multiple indicators to classify a Manual; Thomson Reuters Datastream; and IMF staff calculations.
1
Prots in panel 1 are dened as the return on equity (ROE) averaged across all
banks business strategy. The statistical (clustering) model is based on
banks across all years in the sample horizon (overall) or for all years in periods
Roengpitya, Tarashev, and Tsatsaronis 2014. with prolonged low and normal interest rates.
22Reported results are robust to controlling for the time-varying 2
Impact measures represent the deviations in ROE from the average bank ROE
intensity of macroprudential policies, notably including enhanced (in percent of average bank ROE) in periods of prolonged low rates after a
prudential rules for banks in recent years. Stronger macroprudential one-standard-deviation change from the average value of each bank business
policies are estimated to soften future profitability of banks but have model characteristic during such episodes. Only statistically signicant results
an insignificant contemporaneous effect. are shown.
Table2.2. Bank Profitability and Equity Values in Periods of Normal and Prolonged Low Interest Rates
Dependent Variable: Return on Equity Sign Dependent Variable: Equity Price Return Sign
Explanatory Variables Explanatory Variables
Prolonged-Low-Rate Period1
Term Structure
Three-Month Interest Rate
Term Premium (normal period) n.s. Surprise on Monetary Policy Announcement Dates in
Three-Month Interest Rate (prolonged low rates) + Normal Times
Term Premium (prolonged low rates) + Surprise on Monetary Policy Announcement Dates in +
Bank Characteristics (prolonged low rates)2 Prolonged-Low-Rate Periods
Size +
Leverage
Deposit Funding Share
Loan-to-Asset Ratio +
Controls Controls
Macro Controls Macro Controls
Market Return
Estimation Method Bank FE, Estimation Method Bank FE
time FE
Source: IMF staff calculations.
Note: The table shows the signs of the coefficients of regressors in the cross-country panel regressions of bank profits and daily equity returns that are statistically
significant at least at the 10 percent level. Further details about regressions, variable definitions, and data sources are in Annex 2.2. FE = fixed effect; n.s. = not
significant at the 10 percent level of significance.
1Periods of prolonged low interest rates are defined as described in the chapter.
2Denotes the sign and significance of bank business model characteristics in periods of prolonged low interest rates.
Resilience to Episodes of Prolonged Low Rates conditions affect banks franchise values. A linear factor
Depends Significantly on Banks Business Models model is used to estimate the impact of changes in
Banks that are smaller, rely more on deposit funding, forward interest rates immediately following monetary
and have fewer lending opportunities tend to experience policy announcements in periods of normal and pro-
a significantly bigger dent in their profits (Figure 2.6, longed low interest rates.23 Daily stock returns around
panel2; Table 2.2). For example, a one-standard-devia- the dates of monetary policy decisions are analyzed
tion increase in the size of a banks balance sheet signifi- to ensure that, to the extent possible, the equity price
cantly tempers the damage from prolonged low interest changes do not reflect the release of other relevant
rates by raising bank profits an estimated 67percent rel- information on future economic conditions and bank
ative to the sample average for such periods. By contrast, profitability.
a one-standard-deviation increase in the share of deposit Monetary easing surprises affect bank equity returns
funding and in the share of loans in the asset portfolio are differently in normal times compared with periods of
associated, respectively, with estimated bank returns lower prolonged low interest rates (Table 2.2). In normal
by 14percent and higher by 22percent than the sample times, unexpected monetary easing could generate
average for such periods. Clustering the banks by business expectations of higher economic activity and asset
model confirms these results. Large, internationally more returns, fewer nonperforming loans, and higher spread
diversified, wholesale-funded banks tend to outperform income on fixed-rate assetsall of which increase
other types of banks when interest rates are low for a long expectations of future bank profits. Monetary easing
time. Their estimated average profit is 2.2percentage surprises should, therefore, boost bank equity returns
points higher than that of deposit-funded domestic banks in normal times. During episodes of prolonged low
with small lending portfolios, which have the lowest interest rates, however, lower forward rates in response
estimated average profits during such episodes. to monetary policy decisions are more likely to imply
bad news for economic conditions and bank earnings.
How Do Bank Equity Values Respond to Changes in
Expectations Regarding a Low-for-Long Scenario?
23Monetary policy events are used only as exogenous shocks that
Changes in stock returns are used to measure how provide new information about how long interest rates will remain
changes in market expectations of future economic low and hence about the impact on banks future profits.
They should, therefore, lower equity returns.24 Estima- Figure 2.7. Impact of Forward Rate Surprises on Bank
tion results confirm this intuition. Equity Returns
(Percent)
Larger, more diversified, and more-wholesale-funded
banks are less sensitive to monetary policy news during 120
Normal periods Prolonged low rates
periods of prolonged low rates (Figure 2.7). This out-
100
come may reflect the markets recognition of such banks
greater ability to adapt to changing domestic economic 80
prospectswhich corresponds both to theoretical
60
prediction and to the experience of Japanese banks.
In contrast, for smaller, deposit-funded, domestically 40
oriented banks, the response of equity returns confirms
their greater sensitivity to bad news about the domestic 20
economy during prolonged low rates.
0
20
The Evolution of Banking over the Long Term
40
In a scenario of low natural rates, some consolidation
in the banking industry is likely in the long term. Small 60
deposit-funded banks that are less internationally diver- All Business Business Business
model model model
sified tend to suffer the largest hit to profitability. Even- 1 2 3
tually, consolidation could result through the merger of
smaller banks or of midsize banks with smaller banks,
Sources: Bloomberg L.P.; Fitch Connect; IMF, Monetary and Financial Statistics
and industry concentration could rise through the exit Manual; Thomson Reuters Datastream; and IMF staff calculations.
of nonviable institutions. Merged banks would have Note: The gure depicts the estimated impact of a 1 percentage point surprise
decrease in forward interest rates, occurring on monetary policy announce-
lower average operational costs, be more diversified, and ment dates, on the daily equity returns of banks relative to the estimated
have greater market powerall of which may mean less sample average impact of such surprises when interest rates are normal. For
incentive to take excessive risks. The resulting industry example, the far-left bar is the relative magnitude of the estimated impact on
banks daily equity return of forward rate surprises during normal periods, and
structure could be more efficient and stable.25 is equal to 100 percent. Only statistically signicant impact estimates are
Tail risk exposure is expected to increase. Over the depicted as nonzero values. Business models are as dened in Table 2.1.
Further details of the methodology are in Annex 2.2.
medium term, banks, especially those that are smaller
and less diversified, may actively seek longer maturities
for their assets. Although less interest rate volatility
in the scenario softens the risk implications of such a than retail deposits. This would be particularly true for
strategy, a large positive interest rate shock can mean larger banks, because the low-for-long environment
significant losses. Banks would also feel pressure provides strong incentives to use capital market financ-
to increase, within regulatory limits, their share of ing, especially for international expansion. Such a
wholesale funding, a more volatile source of financing development may affect prospects for financial stability
in their home and host countries, depending on the
modality of expansion.26
24More precisely, it would reflect the expectation of a lower net
Demographic factors, low productivity growth,
present value of future bank profits, even though the short-term
impact of monetary easing could still be positive in such a period and advances in financial technology will likely cause
(though lower when deposit rates are at their zero lower bound). significant shifts in banks business lines under this
25Some of the efficiency losses from consolidation, including
scenario. When the population ages, especially in
higher funding costs for nonfinancial firms and reduced relation-
ship banking for small and medium-sized enterprises, would be a context of reduced income growth, demand for
balanced by the gains from more rational branch networks and lower household loans falls, and deposits tend to rise (Imam
operational costs. Stability benefits may be significant, particu- 2013). Aging will also increase demand for transaction
larly if forces for consolidation are not strong for the large banks,
preventing a worsening of the too-big-to-fail problem. In practice,
bank mergers do not always achieve the desired scale economies, and 26For a comparison of the stability implications of cross-border
can be fraught with difficulties in integrating participating banks lending and expansion through subsidiaries, see Chapter 2 of the
infrastructures and cultures. April 2015 GFSR.
growth and interest rates, guaranteed rates of return are ment base salary from a high share of final salary to a lower career
average share. Moreover, the system has removed the guarantee, but
possible only if they are reset significantly lower.27 not the aspiration, to indexed pension payments. This allows for a
collective approach to asset management, so that active participants
can continue to benefit from equity investments suitable to their
27The
remainder of this section does not aim to capture all factors age and retirees continue to enjoy indexation and incur less risk of
that may be relevant to the long-term evolution of pension arrange- benefit cuts (Ponds and van Riel 2007).
with insurers focusing more on the unit-linked busi- progressively falls, the value of health insurance peaks
ness segment. only at a very advanced age, and that of long-term
Population aging and rising longevity should raise care insurance progressively rises. Population aging and
the demand for life annuities, but countervailing increased longevity could, therefore, give a boost to
forces may exist (Yaari 1965; Turra and Mitchell new products that automatically replicate the life-cycle
2004; Davidoff, Brown, and Diamond 2005; De profile of an optimal package of insurance, eliminating
Nardi, French, and Jones 2010; Lockwood 2012). the need for potentially costly active rebalancing by
Where social safety nets are not sufficiently gen- households.
erous, longer life spans could increase demand for
precautionary savings and liquid assets to cover out- A Difficult and Long-Lasting Transition
of-pocket health expenses in retirement. At very low The challenge for insurers and pension funds is
rates of interest, administrative costs of managing the medium-term impact of prolonged low interest
annuity portfolios may tip relative returns in favor rates on profits and solvency. Their assets are often
of bonds and demand deposits.30 Finally, a continu- of significantly shorter duration than their liabilities.
ing switch from defined-benefit to defined-contribu- Given the lower interest rates and flatter yield curves
tion pensions in such a scenario may also contribute of the scenario, they will be forced to reinvest assets
to reducing annuity demand if very low take-up at significantly lower rates of return much earlier than
rates of voluntary annuitization (as in the United their higher, fixed-rate obligations terminate. Can they,
States) continued to prevail.31 The combined effect without assuming significantly greater risk, adjust their
of these forces on annuity demand is ambiguous. asset portfolios to meet cash flow obligations incurred
Life insurers may increasingly seek to expand into in an environment of higher growth and interest rates?
so-called unit-linked products on the savings side, If not, what other options do they have to safeguard
where investors bear the risk of asset price vola- solvency?
tility. These products make up a significant share
of insurer business in such countries as Australia, Insurance Companies
Belgium, Canada, Ireland, Sweden, the United Not all insurers face this transitional challenge.
Kingdom, and the United States.
Nonlife insurance businesses, whose liability duration
is short and whose main income source is profits from
However, it is unclear what fundamental advan-
underwriting, are relatively unaffected. By contrast,
tages insurers have in offering these products. Insurers
long-term, guaranteed-payout businesses are especially
ability to compete for household savings through these
vulnerable because when interest rates fall, a negative
products will increasingly depend on how they stack
duration gap boosts the present value of a companys
up against retail investments offered by asset managers.
long-term liabilities much more than it boosts the
If the tax advantages currently enjoyed by unit-linked
present value of its assets. Other factors are options
products disappear, a portion of household savings
offered to policyholders that increase insurer losses
could shift over to funds offered by asset managers. when interest rates are low, and the difficulty of raising
Demand for health and long-term care insurance premiums due to competition and high price elasticity
and for new products may increase significantly. of demand for their savings products (Swiss Re 2012;
Koijen, Van Nieuwerburgh, and Yogo (2016) clarify Koijen and Yogo 2015).
that as households age, the value of life insurance
Defined-Benefit Pensions
30For example, demand for liquid assets has risen in Japan in a
context of population aging and prolonged low interest rates (Suzuki Defined-benefit pension funds with substantial
2005). vested obligations suffer most in a low-for-long
31The experience of Chile suggests that a transition to defined-
environment. Because expected life spans after
contribution pensions may also increase voluntary purchases of
deferred life annuities (Rocha, Morales, and Thorburn 2008). retirement are long, projected pension obligations
In Chile, pension reform resulted almost exclusively in defined- can be seen as a large portfolio of long-term nomi-
contribution plans starting in the early 1980s, and the annuity nal bonds (real bonds, if the pension contract offers
industry subsequently expanded as workers in the new system
reached retirement ageabout 60percent of retired workers opt for indexation) with coupon payments corresponding to
an annuity instead of a phased withdrawal option. normal interest rates. Pension plan sponsors would
Figure 2.8. Asset Allocation of Pension Funds and Can Existing Product Lines Be Maintained by
Insurers, 2015 Changing Asset Allocation?
(Percent)
Adopting a liability-driven investment strategy is
Excluding Japan and the Netherlands, pensions place less than a recommended as an effective way for life insurers and
third of funds in bonds.
pension funds to hedge against economic risks. For
1. Pension Funds
Cash Bonds Equities Other
a life insurer or a mature or closed defined-benefit
100 pension plan, liability-driven investment would entail
90 finding a bond portfolio whose duration is similar
80 to the bond-portfolio-like structure of its liabilities
70
(Figure 2.8).33
60
50 Life insurers and defined-benefit pension plans tend
40 to enter a period of low interest rates with reduced
30 economic capital buffers (insurance companies) or a
20
higher funding gap (pension funds).34 This situation
10
0 significantly complicates financial risk management for
Canada Japan Netherlands United United these institutions. On the one hand, portfolio decisions
Kingdom States will need to continue to be guided by considerations of
Life insurers consistently invest a majority of their portfolios in minimizing the adverse impact of market risk, in par-
bonds.
ticular future interest rate volatility. This will call for an
2. Life Insurance Companies
asset portfolio of bonds with cash flow characteristics
Deposits and currency Bonds Equities Loans Other
100 to match cash outflows. On the other hand, given
90 the wider funding gap, institutions have an incentive
80 to generate returns on assets that exceed returns on
70
60
liabilities in a sufficient amount to close the gap. This
50 may call for riskier portfolios with a heavier weight on
40 equities and alternative assets.
30 Can these institutions recover solvency margins and
20
close funding gaps through changes to asset allocation,
10
0 and, if so, how long would that take? A scenario
Canada Germany Japan Netherlands United United simulation examines an underfunded defined-bene-
Kingdom States
fit pension fund faced with the choice of alternative
Sources: Bank of Japan; Centraal Bureau voor de Statistiek; Deutsche
portfolios of fixed-income and other assets; that is,
Bundesbank; GDV, Statistical Yearbook of German Insurance; Statistics evaluating the trade-off between the time it will take
Canada; U.K. Ofce of National Statistics; U.S. Federal Reserve; and IMF each portfolio to return it to fully funded status and
staff calculations.
Note: In panel 1, pension funds include dened-benet and dened- the solvency risk entailed.35
contribution plans. In panel 2, for the Netherlands and the United Recovering adequate solvency margins by changing
Kingdom, the numbers exclude assets invested through mutual funds in
separate accounts. asset allocation appears feasible only by taking poten-
tially unacceptable levels of risk (Box 2.3). The sim-
ulation shows that the volatility risk life insurers and
be hard-pressed to find a duration-matched risk-free defined-benefit pension funds would need to absorb is
bond portfolio to deliver the required cash flow in a very high. This would either deter them from ven-
low-for-long economy.32
33The share of equity investments in the asset portfolios of pen-
sion funds may reflect the degree to which beneficiaries can rely on
32In contrast, sponsors of defined-benefit pension plans with a alternative sources of retirement income.
majority of actively employed, younger participants have several 34The reason is the presence of significant negative duration gaps,
other options to actively manage the (future) accumulation of pen- as described earlier. A defined-benefit pension plan is said to have
sion obligations. These options may render a smoother adjustment a funding gap when the present value of its assets is less than the
to such an equilibrium, including raising the retirement age and present value of its projected benefit obligations.
grandfathering current arrangements and subsequently reducing the 35The simulation adapts the analytical approach of Leibowitz,
replacement rate and removing indexation. Kogelman, and Bader (1995) and Leibowitz and Bova (2015).
turing into such portfolios or entail the risk of falling these institutions will have to make a fresh investment
afoul of regulatory constraints. For example, prudential of equity capital to cover part of the loss.
regulation of insurers prevents significant reach for Insurers can attempt to expand the scale of their
yield across broad asset classes or across risk catego- nonlife and protection businesses to generate earn-
ries within fixed income (Becker and Ivashina 2015). ings and cover some of the loss from their savings
Many regulators risk-based capital requirements for business. Other than health insurance, though,
insurance companies comprise high capital charges it is unclear whether, in the low-growth environ-
for risky investments, including equity, non-invest- ment with an aging population, they can achieve
ment-grade bonds, real estate, and alternative invest- the necessary business growth. The largest firms in
ments (Table 2.3). Expected returns on those assets the life insurance sector may gain market share if
may not compensate for the higher (regulatory) capital financial difficulties drive some of these insurers out
charge. This may explain why search for yield in the of business.37
insurance sector so far has been moderate (Chapter 3 Since many firms defined-benefit pensions are
of the April 2016 GFSR). In the case of defined-ben- mature or closed, and pension obligations are large
efit pensions, regulatory reform for corporate plans in relative to their businesses, the variation in the
the United States has resulted in tough penalties for plans net values due to market volatility increasingly
underfunding, which also discourages excessively risky drives companies financial results. Transferring these
investment strategies. Public pension plans organized pension obligations, or at least their financial risk, to
on a defined-benefit basis in the United States are an insurers after recapitalizing the plans to close their
important exception: regulatory and accounting rules funding gaps is an attractive option and has boosted
may encourage so-called gambling for resurrection growth of the market for pension risk transfers. At
incentives, especially in an environment of low returns the level of the aggregate population, the mortal-
on safe assets.36 ity risk business provides insurers a natural hedge
The preceding analysis makes clear that asset allo- against longevity risk. Pension risk transfers may
cation changes alone cannot adequately address the represent a market-efficient arrangement under
solvency challenge posed by negative cash flows on the which nonfinancial firms close out defined-benefit
current portfolio of liabilities. This means that, in the plans and sell them to insurers at actuarially fair
medium term, insurers and sponsors of defined-benefit prices. Regulation could play an important role in
pensions must find a way to capitalize their losses. A this area by facilitating such transactions.
number of options are potentially available, including
those discussed below. However, it seems likely that The severity of the transitional challenge portends
large business model adjustments in the life insurance
industrys long-term-savings businesses in the medium
36This discussion presumes that current regulatory rules remain
term. Lower and less flexible guarantees on returns
stable even under the chapters scenario. The analysis does not can be expected. Insurers may be given the option to
formally examine the strength of gambling for resurrection incentives
in a low-for-long economy highlighted in the literature (Antolin,
Schich, and Yermo 2011) because such incentives reflect a more 37Japans long experience with low interest rates has led to
complex combination of regulatory and accounting factors. See, for supervisory intervention in the case of seven insurers whose losses on
example, Addoum, van Binsbergen, and Brandt 2010 for the case of existing stocks of guaranteed return liabilities proved impossible to
U.S. corporate plans, and Andonov, Bauer, and Cremers 2016 for absorb, even though the firms had reduced guarantee levels on new
U.S. public plans. contracts.
adjust guarantees at regular intervals to reflect evolving United States tend to be intermediated into both
market conditions. Regulation can play an important equities and bonds via mutual fundsmore than
role in encouraging a switch to more sustainable busi- for defined-benefit plans, in which direct invest-
ness models. This switch will inevitably occur in part ments are more common (Broadbent, Palumbo, and
as a result of new regulatory and accounting regimes Woodman 2006).
requiring economic valuation of portfolios and full As discussed in the preceding section, insurers may
recognition of the economic costs of long-term guar- lose clients to investment funds.
antees. Implementation or introduction of legal and Finally, as explained earlier, financial technology
regulatory requirements for reduction and adjustment could drive up the share of market funding of non-
of costly guarantees and options would support such financial firms, particularly large firms, with direct
a switch. bank lending focusing more on small businesses.
growth. Standard & Poors 500 index, see, for example, Chen, Norohna and
39The relationship between investment in equities and demo- Singal 2004; and Kasch and Sarkar 2011. Barberis, Shleifer, and
graphic structure is significantly richer (Goyal 2004). A higher Wurgler (2005) discuss the role of nonfundamental factors in driving
dependency ratio would, all else equal, reduce investment in equities, market betas of stocks of firms included and deleted from the
but this would be attenuated or even reversed if the middle-age share Standard & Poors 500 index and quantitatively assess their relative
of the population rose at the same time. significance.
have shown that widespread index investing could Figure 2.9. U.S. Mutual Fund Expense Ratios and Growth of
ultimately result in detachment of asset returns from U.S. Index Funds
information regarding fundamentals, hence thwarting
Fees charged by active funds are signicantly higher than those
price discovery (Barberis and Shleifer 2003; Wur- charged by index funds.
gler 2010; and Sullivan and Xiong 2012).41 Finally, 1. Expense Ratios of Actively Managed Funds and Index Funds
benchmarking may have a detrimental impact on (Basis points)
150 Actively managed bond funds
price discovery in additional ways. For example, Index bond funds
it appears to motivate even sophisticated investors Actively managed equity funds
to overweight high-beta assets (Baker, Bradley, and Index equity funds
100
Wurgler 2011).
Three important financial stability issues stem
from the rising share of asset managers and index 50
funds in financial intermediation in a low-for-long
economy. First, as emphasized in earlier reports,
stronger oversight of, and liquidity risk management 0
by, mutual funds are needed, especially if investors 2000 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15
widely available options for automatic enrollment in corporate pension plans in the United States would
employee defined-contribution plans. safeguard the solvency positions of these institutions
Prudential authorities would need to contain from further erosion.
incentives arising in a low-for-long scenario that may Surveillance and regulation of asset management
increase exposure to tail risk. Banks may respond to activities will become even more important if this
incentives in this environment with wider maturity industrys share of the financial system continues to
mismatches, higher leverage, or more wholesale fund- grow. Further strong growth of the sector can contribute
ing (within regulatory limits). Insurance and pension to financial stability, but also entails new challenges. For
regulators that have not yet introduced economic example, if passive index investing becomes preeminent,
solvency requirements would need to implement such price discovery could be hampered and markets could
regulations as soon as practical. Public pension funds become more prone to swings in sentiment. More
in the United States are allowed to discount liabilities generally, as emphasized in earlier reports (Chapter 3 of
at expected rates of return on their asset portfolios. the April 2015 GFSR; Chapter 2 of the October 2015
They have taken advantage of this opportunity by GFSR), closing significant data gaps and implementing
aggressively investing in risky assets, with negative adequate macroprudential rules to address risks, such as
financial results (Andonov, Bauer, and Cremers 2016). those related to liquidity mismatches, are essential for
Aligning liability discounting rules with those for effective surveillance and to contain systemic risk.
in countries where the loan-to-deposit ratio is far below 1, as in on deposit rates is to introduce a preference for cash relative
Japan, RM can be interpreted as the rate of return on government to deposits that is a function of their relative rates of return
bonds. (Drechsler, Savov, and Schnabl 2016).
3Since profits are measured before dividend distribution, 6In practice, leverage constraints will eventually force banks to
returns to equity are added back in. raise capital or decline further deposit inflows.
Box 2.2. How Have U.S. Banks Reacted to the Low-Interest-Rate Environment?
The recent experience of the United States does not Figure 2.2.1. Bank Earnings and
lend itself to direct conclusions about the scenario Noninterest Income since 2007
considered here. Nonetheless, reviewing the response 1. Return on Assets by Bank Business Model
of U.S. banks to the prolonged period of very low (Percent) GSIBs
interest rates may provide valuable additional insights 1.5 DSIBs
Other commercial banks
into how banks may adapt to such circumstances.
After a significant dip around the Lehman Brothers 1.0
bankruptcy, bank profitability in the United States has
0.5
returned to precrisis levels (Figure 2.2.1, panel 1). A
range of adaptation strategies are evident across banks 0.0
of different sizes.
A common strategy is the increased focus on fee- 0.5
based businesses and trading. The share of noninterest
1.0
income in banks total income has risen compared 2007 09 11 13 15
with the precrisis period (Figure 2.2.1, panel 2).
The increase ranges from 5to 10percentage points
depending on bank size and business model, with 2. Fees, Net Gains on Assets, and Trading Revenue
(Percent of total income)
the largest increase observed for global systemically
important banks. In particular, selected components of 45
Precrisis Low-interest-rate period
40
noninterest income, such as fees, net capital gains, and 35
trading revenue, have grown significantly during the 30
low-interest-rate period. 25
20
Banks have also increased the maturity of their 15
assets, potentially seeking, as far as possible, to 10
conserve interest margins from lending and bond 5
0
investing.1 Interestingly, banks that least successfully
GSIBs DSIBs Other
increased earnings from fees and trading are also the commercial
ones that most aggressively pursued this strategy. In banks
60
55
50
45
2009 10 11 12 13 14 15
Y
Annex 2.1. Term Premiums under a Low-for- ___t_ = * 1t gt
Yt 1
Long Scenario42
that are only partially indexed to inflation. Inflation
This chapters model derives from consumption- shocks and income shocks are independently Gaussian,
based asset pricing models, extending them to envi- ,t~N(0, ) and log(gt )~N(0, g). The central banks
ronments in which steady-state growth, inflation, and policy reaction function,
interest rates are very low and nominal interest rates
{ }
g
( g *)
Rt = max b(bt b *) +(_
*t) _
gt
are subject to a zero lower bound.43 R *, ,
An endowment economy model of asset pricing in
is subject to an effective lower bound, . The sensitiv-
the spirit of Deaton (1991) is adapted to accommodate
ity of the central banks policy response to growth and
an incomplete market with only nominal bonds, no
inflation shocks is g > 0; > 1. A fiscal risk pre-
borrowing constraints, an exogenous inflation process,
mium, b, is assumed to be negative to ensure against
an endowment process partially indexed to inflation,
explosive paths of capital accumulation by households.
and nominal interest rates determined by a modified
In particular, it is assumed that
Taylor rule. The household receives an endowment
Yt at time t, which it may allocate to consumption Ct *g *
R * =____,
+
or savings through nominal bonds Bn,t, in which n,t
denotes a term of n-periods at date t.44 Subject to the in which the value of is chosen so that the average
period t budget constraint, equilibrium value of real (government) debt outstand-
ing is maintained at b *. The model is solved following
Bt 1 +n > 1Qn,t
Bn,t
1 +Yt
the approach of Caldara and others (2012).
B A steady state with a low natural rate of interest
= __t +n > 1Qn,t
Bn,t
+Ptct ,
Rt
close to the zero lower bound has flatter yield curves
households solve the following function: and compressed term premiums relative to a steady
1
____
state with higher growth, inflation, and interest rates
Vt =maxc {ct 1 + Et (Vt + 1 1 ) 1
}
1 1
,
t (Annex Figures 2.1.1 and 2.1.2).
in which Rt, Qn,t
, and Pt are the nominal bond In a normal economy, an inflation shock elicits
return, long-term bond price, and the price level, a corresponding change in real rates because of the
respectively, in period t; denotes the inverse of the strong policy response of the central bank (Annex
intertemporal elasticity of substitution; and the Figure 2.1.1). Moreover, inflation persistence, cen-
coefficient of risk aversion. The household optimiza- tral bank policy reaction, and partial indexation of
tion problem is situated in the context of exogenous endowments ensure that real savings, real incomes,
inflation shocks described by and expected lifetime utility move in a direction
opposite from that of inflation and real interest rates,
log t = log t 1 +(1 )log *+ ,t, and hence in the same direction as bond prices.
and income shocks gtaround their steady-state values Accordingly, in this economy, households lifetime
* and g *. Households receive nominal endowments utility moves positively with bond returns, which
implies positive term premiums and a positively
42The author of this annex is Mitsuru Katagiri. sloped yield curve.
43Many factors play a role in determining the slope of the yield In a low-for-long economy around the zero lower
curve, including the covariation between household consumption bound, central banks constrained ability to respond
growth and inflation (Piazzesi and Schneider 2007), the hedge
provided by bonds against other asset returns (Campbell, Sunderam, to inflation shocks means that real rates now move
and Viceira 2016), and the ability and willingness of arbitrageurs to in a direction opposite from that of inflation shocks.
execute risky, profitable trades in bond markets (Vayanos and Vila In turn, through the same transmission channels
2009; Greenwood and Vayanos 2010). The empirical literature on
as above, this generates negative comovement of
the measurement of term premiums has also advanced significantly,
for example, based on the affine term structure models developed expected lifetime utility and bond returns, which
by Adrian, Crump, and Moench (2013) and Abrahams and others lowers nominal and real term premiums in this econ-
(2016) for the United States and applied by Malik and Meldrum omy relative to an economy with higher equilibrium
(2016) for the United Kingdom.
44Uppercase letters denote nominal values, lowercase real values, levels of growth, inflation, and interest rates (Annex
and starred variables steady-state values. Figure2.1.2).
Annex Figure 2.1.1. Impulse Responses outside of a Annex Figure 2.1.2. Impulse Responses in Low-for-Long
Low-for-Long Scenario Economies
(Percentage points) (Percentage points)
Under persistent ination shocks ... Under persistent ination shocks ...
1. Ination 1. Ination
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Quarters Quarters
... real rates remain elevated ... ... real rates remain lower ...
2. Real Interest Rate 2. Real Interest Rate
0.2 0.0
0.1
0.2
0.1 0.3
0.4
0.5
0.0 0.6
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Quarters Quarters
Annex 2.2. Cross-Country Evidence of Prolonged sions are clustered into the same group, following
Low Interest Rates Impact on Banks45 Roengpitya, Tarashev, and Tsatsaronis 2014, and three
This annex discusses the data and the empirical methodol- group-types of business models are estimated and
ogy used to analyze how periods of low interest rates affect assigned one bank at a time.
bank profitability as measured by realized profits and The exercise covers an unbalanced panel of almost
expected future profits, as reflected in banks equity price 17,000banks in eight advanced economies, using
returns. annual data from 1990 through 2015. Only banks
with end-of-year statements are included.48 The esti-
The Impact on Bank Profitability mation incorporates bank-level and time-level fixed
effects.49
Bank profits, typically measured by return on equity,
The baseline results are robust to a number of
are analyzed using the following regression (for bank i
perturbations of this benchmark specification, includ-
in country j in year t):
ing alternative definitions of bank profits (return on
Profit ijt = i + Macrojt + lowjt + 1Shortratejt assets); inclusion of other bank business characteris-
+ 2Shortratejt lowjt + 3TPjt tics; alternative definitions of periods of prolonged
+ 4TPjt lowjt + 1Businessmodelijt low interest rates; lagged values of bank business
+ 2Businessmodelijt lowjt + ijt, model characteristics, controlling for the scope and
intensity of macroprudential policies and for concen-
in which Profit is measured by return on equity; Macro
tration in the banking industry; and incorporating a
is a vector of macroeconomic control variables, such
lagged dependent variable. A dynamic panel regres-
as consumer price index inflation, credit growth, and
sion was initially implemented resulting in a find-
GDP growth; and low is a dummy for periods with
ing of insignificant year-to-year persistence of bank
prolonged low rates of interest, defined as years when
returns, which argued for dropping the lagged depen-
the 10-year, on-the-run spot rate on government bonds
dent variable and reporting results of a cross-country
is less than the historic in-sample average of the mon-
panel regression.
etary policy interest rate, and the three-month govern-
ment bond or bill interest rate is less than 1percent.
For Japan, the threshold for the 10-year spot rate is The Impact on Bank Equity Price Return
2percent;46 Shortrate is the three-month interest rate;
The general specification can be written as follows:
TP denotes the term premium, based on Wright 2011;
and Businessmodel represents the indicators of banks EquityPriceReturnijt
business models.
Two approaches are used to characterize banks = +marketreturnjt + 0surprisejt + 1surprisejt
business models. First, several balance sheet indica- MP_normaltime jt + 2surprisejt MP_lowjt
tors are considered individually, including size (total +conditioningvariablejt + ijt,
assets), leverage (assets-to-equity ratio), the deposit
funding ratio, the loans-to-total-assets ratio, and the in which the dependent variable EquityPriceReturn
share of trading assets in total assets. Second, business is the daily change in equity prices (in loga-
models are constructed for each bank using a cluster- rithm); marketreturn denotes the daily change in
ing method. The business models are defined by three country-specific stock market indices, capturing the
features: size, deposit funding ratio, and loan-to-asset overall market return (in logarithm); surprise denotes
ratio.47 Banks that are similar in these three dimen- the unexpected change in market expectations of
45The authors of this annex are Qianying Chen and Kai Yan. 48Canada, Finland, France, Germany, Japan, the Netherlands, the
46Since Japan was in an environment of policy rates of less than United Kingdom, and the United States. The country coverage is
2percent for most of the time in the sample, the historical average subject mainly to data availability of the term premium.
of policy rates is considered inappropriate for defining the ceiling of 49Incorporating country and time fixed effects eliminates, as
a period of low interest rates. expected, the effect of changes in the term structure of interest
47Data on the geographic distribution of bank incomes could not rates on bank profits in periods with prolonged low interest rates.
be included because it was available only for a small subsample of However, the estimated sensitivity of the impact depending on
banks and skews the country and size distributions relative to the bank business model characteristics is robust to inclusion of these
overall sample of banks. fixed effects.
future short-term interest rates, defined as the change on that day. Assuming that there are no other major
in the country-specific nine-year-ahead one-year-for- announcements on the same day, these interaction
ward rate; MP_low is the dummy for monetary policy terms ensure the exogeneity of the interest rate shock.
announcement dates in periods with prolonged low The analysis relies on daily data spanning 2000
rates, while MP_normaltime represents the announce- through 2016, covering banks in 16advanced econo-
ment dates in other periods. The period of prolonged mies.51 Details of variable definitions and data sources
low rates is defined as the time when the 10-year are provided in Annex Table 2.2.1. Only banks whose
government bond yield is less than 2 percent, a level stocks are traded with sufficient frequency are included
when the real rate adjusted by inflation target is at zero in the analysis.
in many countries.50 Endogeneity may appear when including the sur-
The interaction terms surprisejt MP_normaltimejt prise in the regression, because other economic news
and surprisejt MP_lowjtmeasure the market surprises that changes the expectations of forward rates may
on the expected future short-term rate on the mon- also directly affect the equity price return. The missing
etary policy announcement days. This is either the variable of other news in the residual may be correlated
surprise triggered by the news about a change in the with the surprise and result in biased estimation.
monetary policy stance or a correction of previous Therefore, additional robustness checks are conducted.
expectations when there is no change in the policy An event study regression was run, covering only the
dates of the monetary policy announcements, and also
50In defining periods of prolonged low rates, the second threshold a daily frequency regression with an alternative surprise
applying to short-term interest rates (in the profit regression) was measure extracting the component in surprise that is
not applied in this regression to avoid the noise introduced by the orthogonal to the market return, which is taken to
volatile movement of daily short-term market interest rates. As part
represent news that affected interest rate expectations,
of robustness exercises, two alternative definitions were also exam-
inedperiods when the forward rate was less than the in-sample but not the equity price return directly. Both of these
average of the monetary policy rate and when the shadow policy rate checks confirm that the main results are robust.
deviated from the actual policy rate. However, using the first of these
alternative definitions does not work well with the Japanese data
because interest rates were also low in the 1990s, and the second 51Australia, Belgium, Canada, Denmark, Finland, France,
definition was problematic: it identified periods of prolonged low Germany, Ireland, Italy, Japan, the Netherlands, Spain, Sweden,
rates only with periods of negative interest rates. Switzerland, the United Kingdom, and the United States.
Macroeconomic
Consumer Price Index Inflation Year-over-year growth of consumer price index, percent IMF, International Financial
Statistics database
Credit-to-GDP Ratio Private sector credit in percent of GDP Bank for International Settlements
Real GDP Growth Year-over-year growth of GDP, constant prices IMF, World Economic Outlook
database
Three-Month Interest Rate Typically central bank bill/Treasury bill yield or interbank offered Haver Analytics
rate
Term Premium Term premium estimated based on Wright 2011 IMF, Global Financial Stability
Report, October 2016
Ten-Year Government Bond Yield On-the-run 10-year government bond yield (from yield curve Thomson Reuters Datastream
values for constant maturity)
Monetary Policy Rates Short-term interest rates represent the monetary policy stance Haver Analytics
in a country
Financial Market
Equity Price Return Log difference of equity prices
Market Return Difference of overall country-specific equity price indices
VIX Chicago Board Options Exchange Market Volatility Index Bloomberg L.P.
Oil Price West Texas Intermediate crude oil spot price Bloomberg L.P.
Source: IMF staff.
Risk-Taking Channel: Evidence from the United States. Lambert, Frederic, and Kenichi Ueda. 2014. Effects of Uncon-
Journal of Finance. ventional Monetary Policy on Banking Sector Soundness.
Drechsler, Itamar, Alexi Savov, and Philipp Schnabl. 2016. The IMF Working Paper 14/152, International Monetary Fund,
Deposit Channel of Monetary Policy. New York University Washington, DC.
working paper, New York. Leibowitz, Martin L., and Anthony Bova. 2015. Pension Fund
English, William B., Skander J. van den Heuvel, and Egon Flatlines and Recovery Rates. Portfolio Strategy, Morgan
Zakrajsek. 2012. Interest Rate Risk and Bank Equity Valu- Stanley Research, December 8.
ations. Finance and Economics Discussion Series 201226, Leibowitz, Martin L., Stanley Kogelman, and Lawrence N.
Board of Governors of the Federal Reserve System, Washing- Bader. 1995. Return Targets and Shortfall Risks: Studies in
ton, DC. Strategic Asset Allocation. Burr Ridge, IL: Irwin Professional
European Systemic Risk Board (ESRB). 2016. Macroprudential Publishing.
Policy Issues Arising from Low Interest Rates and Structural Lockwood, Lee M. 2012. Bequest Motives and the Annuity
Changes in the EU Financial System. Frankfurt. Puzzle. Review of Economic Dynamics 15 (2): 22643.
Freixas, Xavier, and Jean-Charles Rochet. 2008. Microeconomics Malik, Sheheryar, and Andrew Meldrum. 2016. Evaluating the
of Banking. Cambridge, MA: MIT Press. Robustness of U.K. Term Structure Decompositions Using
Gambacorta, Leonardo. 2008. How Do Banks Set Interest Linear Regression Methods. Journal of Banking and Finance
Rates? European Economic Review 52 (5): 792819. 67: 85102.
Gordon, Robert J. 2014. The Demise of U.S. Economic Nagel, Stefan. Forthcoming. The Liquidity Premium of Near
Growth: Restatement, Rebuttal and Reflections. NBER Money Assets. Quarterly Journal of Economics.
Working Paper 19895, National Bureau of Economic Nakata, Taisuke, and Hiroatsu Tanaka. 2016. Equilibrium Yield
Research, Cambridge, MA. Curves and the Interest Rate Lower Bound. Board of Gov-
Gourio, Francois, and Phuong Ngo. 2016. Risk Premia at the ernors of the Federal Reserve System Finance and Economics
ZLB: A Macroeconomic Interpretation. Unpublished. Discussion Series 2016-085, Washington, DC.
Goyal, Amit. 2004. Demographics, Stock Market Flows, and Piazzesi, Monika, and Martin Schneider. 2007. Equilibrium
Stock Returns. Journal of Financial and Quantitative Analysis Yield Curves. In NBER Macroeconomics Annual 2006, Vol.
39 (1): 11542. 21, edited by Daron Acemoglu, Kenneth Rogoff, and Michael
Greenwood, Robin, and Dimitri Vayanos. 2010. Price Pressure Woodford, 389472. Cambridge, MA: MIT Press.
in the Government Bond Market. American Economic Ponds, Eduard H. M., and Bart van Riel. 2007. The Recent
Review: Papers and Proceedings 100 (2): 58590. Evolution of Pension Funds in the Netherlands: The Trend to
Holston, Kathryn, Thomas Laubach, and John C. Williams. Hybrid DB-DC Plans and Beyond. Working Paper 20079,
2016. Measuring the Natural Rate of Interest: International Boston College Center for Retirement Research, Boston, MA.
Trends and Determinants. Working Paper 201611, Federal Rachel, Lukasz, and Thomas Smith. 2015. Secular Drivers of
Reserve Bank of San Francisco, San Francisco, CA. the Global Real Interest Rate. Bank of England Working
Imam, Patrick. 2013. Demographic Shift and the Financial Paper 571, London.
Sector Stability: The Case of Japan. Journal of Population Rocha, Roberto, Marco Morales, and Craig Thorburn. 2008.
Ageing 6 (3). An Empirical Analysis of the Annuity Rate in Chile. Journal
International Monetary Fund (IMF). 2017. Negative Interest of Pension Economics and Finance, 7 (1): 95119.
Rate PoliciesInitial Experiences and Assessments. IMF Roengpitya, Rungporn, Nikola Tarashev, and Kostas Tsatsaro-
Policy Paper, Washington, DC. nis. 2014. Bank Business Models. BIS Quarterly Review
Jagannathan, Ravi, and Narayana R. Kocherlakota. 1996. Why (December).
Should Older People Invest Less in Stocks than Younger Peo- Sullivan, Rodney N., and James X. Xiong. 2012. How Index
ple? Federal Reserve Bank of Minneapolis, Quarterly Review Trading Increases Market Vulnerability. Financial Analysts
(Summer): 1123. Journal 68 (March/April): 7084.
Kasch, Maria, and Asani Sarkar. 2011. Is There an S&P 500 Summers, Lawrence H. 2014. U.S. Economic Prospects:
Index Effect? Federal Reserve Bank of New York Staff Report Secular Stagnation, Hysteresis and the Zero Lower Bound.
484, New York. Business Economics 49 (2): 6573.
Koijen, Ralph S. J., Stijn Van Nieuwerburgh, and Motohiro Suzuki, Wataru. 2005. Determinants of the Upsurge in M1
Yogo. 2016. Health and Mortality Delta: Assessing the Wel- Demand in Japan: Evidence from Household-Level Analysis.
fare Cost of Household Insurance Choice. Journal of Finance Tokyo Gakugei University Empirical Analysis of Economic
71 (2): 9571009. Institutions Discussion Paper Series 81, Tokyo.
Koijen, Ralph S. J., and Motohiro Yogo. 2015. The Cost of Swiss Re. 2012. Facing the Interest Rate Challenge. Sigma 4.
Financial Frictions for Life Insurers. American Economic Turra, Cassio M., and Olivia S. Mitchell. 2004. The Impact
Review 105 (1): 44575. of Health Status and Out-of-Pocket Medical Expenditures
on Annuity Valuation. University of Michigan Retirement Wright, Jonathan. 2011. Term Premia and Inflation Uncer-
Research Center Working Paper 2004086, Ann Arbor, MI. tainty: Empirical Evidence from an International Panel.
Vayanos, Dimitri, and Jean-Luc Vila. 2009. A Preferred Habitat American Economic Review 101 (4): 151344.
Model of the Term Structure of Interest Rates. NBER Work- Wurgler, Jeffrey. 2010. On the Economic Consequences of
ing Paper 15487, National Bureau of Economic Research, Index-Linked Investing. In Challenges to Business in the
Cambridge, MA. Twenty-First Century: The Way Forward, edited by W. T. Allen,
Vials, Jos, Simon Gray, and Kelly Eckhold. 2016. The Broader R. Khurana, J. Lorsch, and G. Rosenfeld. Cambridge, MA:
View: The Positive Effects of Negative Interest Rates. https American Academy of Arts and Sciences.
://blog-imfdirect.imf.org/2016/04/10/the-broader-view-the Yaari, Menahem E. 1965. Uncertain Lifetime, Life Insurance
-positive-effects-of-negative-nominal-interest-rates. and the Theory of the Consumer. Review of Economic Studies
Wicksell, Knut. 1936. Interest and Prices. English Translation of 32 (2): 13750.
the 1898 manuscript, Royal Economic Society, London.
Summary
H
ow much influence do countries retain over their domestic financial conditions in a globally integrated
financial system? This question has recently been attracting increased interest in policy and academic
circles alike. Financial conditions broadly refer to the ease of obtaining finance, and measuring them
can be valuable for appraising the impact of policy and economic prospects.
Greater financial integration can complicate the management of domestic financial conditions in several ways.
First, policymakers may need to take external factors into greater consideration when pursuing domestic objectives.
Second, global financial integration may make it harder for domestic policymakers to control financial conditions
at homefor example, it may hamper the transmission of monetary policy.
This chapter examines the evolving importance of common global components of domestic financial conditions.
It develops financial conditions indices (FCIs) that make it possible to compare a large set of advanced and emerg-
ing market economies. It finds that a common component (global financial conditions) accounts for about 20 to
40 percent of the variation in countries domestic FCIs, with notable heterogeneity across countries. Its impor-
tance, however, does not seem to have increased markedly over the past two decades.
Global financial conditions loom large, but evidence suggests that, on average, countries still appear to hold
sway over their own financial conditionsspecifically, through monetary policy. Nevertheless, the rapid speed at
which foreign shocks affect domestic financial conditions may also make it difficult to react in a timely and effec-
tive manner, if deemed necessary. Given that global financial conditions tend to account for a greater fraction of
FCI variability in emerging market economies, these countries, in particular, should prepare for the implications of
global financial tightening. Governments can promote domestic financial deepening to enhance resilience to global
financial shocks. In particular, developing a local investor base, as well as fostering greater equity- and bond-market
depth and liquidity, can help dampen the impact of external financial shocks.
emerging market economiesin itself a contribution ities that leave domestic financial conditions sensitive
to the literature. to external shocks. When disruptive outflows threaten
These are the chapters highlights: financial stability, capital flow management measures
The new FCI measures appear to signal downside could have a temporary role, as noted in IMF 2016.
risks to GDP well. In particular, economic con- By promoting financial deepening, countries can help
tractions are more clearly associated with a preced- protect against global financial shocks. Specifically,
ing change in financial conditions in contrast to developing a local investor base (both banks and non-
expansions. banks) can help soften the blow of financial shocks.
A single factor, global financial conditions, appears
to account for a large share of variation in domestic
financial conditions around the world. This factor An Overview of Financial Conditions
moves in tandem with the U.S. FCI and measures This section examines the concepts surrounding financial
of global risk, such as the Chicago Board Options conditions, their transmission across countries, and their
Exchange Volatility Index (VIX). measurement.
There is no conclusive evidence, however, that this
global factor has gained significant influence over Financial Conditions: Main Concepts
the past two decades. Financial conditions generally refer to the ease
Financial linkages (such as cross-country invest- of obtaining financing. The literature offers several
ments) are the most reliable indicator of global complementary definitions of financial conditions.
financial conditions influence on local FCIs. At For instance, Hatzius and others (2010) define them
the same time, greater financial development can as the current state of financial variables that influ-
reduce the sensitivity of domestic FCIs to global ence economic behavior and thereby the future of the
financial shocks. economy, while Carlson, Lewis, and Nelson (2012)
About 20 to 40 percent of the variation in domestic connect them to price and nonprice costs of credit.
FCIs across countries can be attributed to global This chapter focuses on a notion of domestic finan-
financial conditions, with domestic factors account- cial conditions that seeks to gauge the costs, condi-
ing for the rest. However, the importance of global tions, and availability of domestic funds to the local
financial shocks for domestic financial conditions economy. In addition to interest rates and asset price
varies notably across countries. Importantly, mone- valuations, financial conditions are influenced by risk
tary policy shocks account for about 15 percent of appetite and, for example, agents willingness to hold
the variation across countries with flexible exchange illiquid assets.
rates, suggesting that amid exposure to external fac- Financial conditions play a central role in the trans-
tors, changes in the monetary policy stance still can mission of monetary policy to the broader economy.
matter for domestic financial conditions. In particular, monetary policy influences the rest of the
economy mainly by altering financial conditions, and
Even with a sizable impact from global financial the transmission channels can be classified into two
shocks, evidence suggests that, on average, countries broad categories:
appear to be able to influence their own financial con- The first comprises the traditional, or New
ditions. In particular, the analysis indicates that they Keynesian, channels of monetary policy. The
generally have scope to use monetary policy. However, emphasis is on changes in (short-term) policy rates
given that local financial conditions react more rapidly and how expectations about those changes alter
to global financial shocks than to changes in domestic longer-term rates and thereby consumption and
policy rates, timely policy responses may often be diffi- investment decisions. Effects on trade through
cult. Emerging market economies, in particular, clearly exchange rate movements also belong to the list of
need to guard against the risks associated with sharp traditional channels.
changes in global financial conditions. Countries can The second category predominantly comprises
resort to other policies to protect themselves against imperfections in credit supply arising from institu-
destabilizing shocks. For example, macroprudential tional constraints on financial intermediaries and
measures can contain potentially lingering vulnerabil- from informational asymmetries (Boivin, Kiley, and
Mishkin 2010). Examples include the balance sheet The extent to which global factors affect domes-
channel (Bernanke and Gertler 1989; Kiyotaki and tic financial conditions is a question this chapter
Moore 1997), the bank capital channel (Van den attempts to decipher. Accordingly, the FCIs consist
Heuvel 2002), and the risk-taking channels (Adrian of domestic financial variables such as corporate,
and Shin 2011; Adrian and Boyarchenko 2012), as interbank, and term spreads; equity and house price
discussed in greater detail in Adrian and Liang 2016 returns; equity return volatility; and credit growth.
and Chapter2 of the October 2016 Global Finan- An attempt is made to purge the FCIs of (con-
cial Stability Report (GFSR). temporaneous) macroeconomic conditions. That
way, in principle, it is possible to assess how much
Many of these nontraditional monetary transmis- unwarranted global financial shocks affect domestic
sion channels feature both incomplete markets and financial conditions.
heterogeneous agents, which lead to differences in the
pricing of risk over time. As a result, the risk-free rate
is not an adequate statistic for funding costs or for The Transmission of Financial Conditions across
assessing the impact of monetary policy on the real Countries
economy.2 FCIs thus aim to distill information from a Financial conditions can be transmitted across coun-
broad array of financial variablesincluding measures tries through different channels. A significant strand
of risk taking and various kinds of financial frictions of the literature has focused on the degree of mone-
ideally capturing the prevalence of credit constraints tary independence in setting interest rates. A central
and the magnitude of external financing premiums. principle guiding monetary policy in open economies
FCIs can only capture some measure of average is the so-called Mundell-Fleming trilemma. It states
funding costs, although different agents may face large that policymakers can seek to achieve only two out of
variations in funding costs and conditions. Naturally, the three following objectives: (1) fixed exchange rates,
as financial systems evolve, the most relevant variables (2) free international capital mobility, and (3) mone-
for tracking financial conditions may change. tary autonomy.3 However, financial conditions can be
Empirically, measures of financial conditions can transmitted across countries through other mechanisms
be more helpful in predicting economic activity than as well, in ways that usually cannot be fully offset by
indicators of current and past real economic activity. movements in exchange rates (Obstfeld 2015). In fact,
Studies, including Hatzius and others 2010 and Koop exchange rate movements also typically induce changes
and Korobilis 2014, argue that FCIs are good pre- in financial conditions in small open economies, and
dictors of future economic activity. Likewise, Adrian, can be sizable (Kearns and Patel 2016). Changes in
Boyarchenko, and Giannone (2016) show that FCIs financial conditions can further spill over from orig-
are particularly useful in flagging future economic inating countries to other economies through several
contractions. interrelated channels. For example, changes in credit
Financial conditions are driven only partly by policy. volumes and other types of capital flows can have pow-
Changes in uncertainty about the exposures of major erful cross-border effects. Another transmission channel
financial players, shocks to the net worth of borrow- works through comovements in risk premiums, which
ers not triggered by policy actions, runs on financial can affect collateral valuation and thereby borrowing
institutions, changes in risk perception, and shifts in constraints (Obstfeld 2015).
investor sentiment triggered by idiosyncratic events can Global financial integration can complicate the
all influence access to funding in an economy. management of domestic financial conditions in at
least two distinct ways. First, as countries integrate
2As underscored by Dudley (2010), financial conditions are more into the global economy, policymakers may need
explicitly taken into account in the conduct of monetary policy. In to take external factors into greater consideration when
the United States, he notes that this is evident in the transcripts of pursuing domestic objectives. However, this complica-
the Federal Open Market Committee meetings and minutes going
back more than a decade. Even before the global financial crisis,
Bernanke (2007) highlighted links between financial conditions and 3Broadly consistent with the predictions of the trilemma, studies
growth. More recently, Yellen (2016) drew attention to the relevance have typically found that greater exchange rate flexibility does
of financial conditions for the economic outlook and the stance of provide some degree of flexibility in steering short-term interest rates
monetary policy. (Klein and Shambaugh 2015; Obstfeld 2015).
tion does not, by itself, imply that countries lose their control over domestic financial conditions. Financial
ability to steer their domestic financial conditions.4 conditions that move together across countries may be
Second, global financial integration may indeed make a natural reflection of comovement in fundamentals
it harder for domestic policymakers to control domes- because of greater trade and financial integration and
tic financial conditionsfor example, by hampering could, therefore, be optimal from a domestic stand-
the transmission of monetary policy or limiting the point. For example, for a globally integrated economy
effectiveness of prudential policies. The speed at which whose business cycle is highly correlated with the rest
foreign shocks affect local financial conditions also of the world, raising domestic interest rates in response
makes it difficult to react in a timely and effective to a rise in world interest rates may be the best deci-
manner.5 In particular, the efficacy of financial stability sion from a domestic perspective. But some changes
policies can be weaker in an open economy (Schoen- in financial conditions have nothing to do with
maker 2013).6 macroeconomic factors and may arise from financial
Various studies suggest that financial conditions frictions (including changes in investor sentiment, the
around the world are heavily influenced by global effects of herd behavior, risk management constraints,
factors. Building on earlier work by Calvo, Leiderman, or regulations). Conceptually, in an extreme case,
and Reinhart (1996), many studies emphasize the empirically domestic financial conditions being pre-
important role of global push factors, such as the dominantly influenced by such spillovers not driven by
VIX, as drivers of financial variables (see, for example, fundamentals (and therefore likely to be undesirable)
Bruno and Shin 2013; IMF 2014a; Fratzscher 2012; would suggest a lack of control by policymakers. The
Baskaya and others 2017). Miranda-Agrippino and reason is that policymakers will most likely attempt to
Rey (2015) argue that prices of risky assets (equities, counteract such shocks. Accordingly, these spillovers
corporate bonds) across countries can be summarized still featuring prominently in domestic financial condi-
by a single global factor, the global financial cycle, tions would be an indication that policymakers do not
which is driven by U.S. monetary policy shocks. have the tools to react in an effective or timely manner
Therefore, as argued by Rey (2016), U.S. monetary to offset them. Empirically, the distinction between
policy shocks spill over and affect domestic financial fundamentals-driven versus other types of spillovers is
conditions even in inflation-targeting economies not easy to derive (see Disyatat and Rungcharoenkitkul
with large financial markets. Longstaff and others 2016 for an effort in this regard). This chapter seeks to
(2011) find that three factors account for more than address this issue by focusing on measures of financial
50 percent of the variation in credit default swap conditions that are purged of macroeconomic funda-
spreads across countries, and Adrian, Stackman, and mentals, acknowledging the difficulties and limitations
Vogt (2016) estimate a highly significant price of risk inherent to such an endeavor.
that forecasts global stock and bond returns as a non-
linear function of the VIX.7 Constructing Financial Conditions Indices across
Evidence of global factors greater influence, how- Advanced and Emerging Market Economies
ever, is not by itself proof that policymakers are losing Previous studies have constructed FCIs mainly for
4For
selected advanced economies, using various meth-
example, as trade becomes more important, monetary policy
may work more through exchange rates and net exports and less ods, each with its strengths and limitations. FCIs
through its effects on domestic demand. are unobservable (latent) variables that are estimated
5Global financial integration could also worsen the trade-offs
using a wide range of financial variables so as to best
authorities face when pursuing financial stability objectives along
with more standard macroeconomic stabilization goals (Obstfeld reflect the financial conditions faced by domestic end
2015). This is because greater openness to international financial users, such as firms and households. The literature
markets would likely diminish the effectiveness of macroprudential has concentrated primarily on developing FCIs for
tools, which would suffer more from leakage problems (IMF, Finan-
the United States and, occasionally, for major econo-
cial Stability Board, and Bank for International Settlements 2016).
6According to the financial trilemma put forward by Shoen- mies of the Organisation for Economic Co-operation
maker (2013), only two of the following three goals can be achieved and Development (Box3.1). However, previous
simultaneously: (1) national autonomy over financial policies, (2) studies have not developed a consistently estimated
international financial integration, and (3) financial stability.
7See also Kennedy and Palerm 2014; and Bekaert and others set of FCIs for both major advanced and emerging
2016, among many others. market economies.
This chapter goes beyond existing studies and Boivin and Ng (2006) emphasize that including more
develops FCIs for major advanced and emerging data does not always yield better results.
market economies. For the purposes of this chapter, In this chapter, the choice of variables used for the
latent FCIs are extracted from an array of finan- construction of the FCIs is guided by two consid-
cial variables while taking account of growth and erations, one conceptual and the other practical.
inflation. In particular, a time-varying parameter Conceptually, since the chapter focuses on how global
factor-augmented vector autoregression model based factors affect financial conditions in domestic markets,
on the work of Koop and Korobilis (2014) is used to variables measuring the ease of access to finance on
estimate the FCIs.8 This method jointly considers the international markets are not included.11 With regard
dynamic interactions of the FCI and macroeconomic to practical considerations, the choice of variables
fundamentals, and has two notable advantages. First, should be consistent across countries and reflect as
the method aims to purge the FCI of the effects of many segments of the financial system as possible.
macroeconomic conditions.9 Although empirically dif- Accordingly, the FCI should include the equity, hous-
ficult, conceptually this purging is desirableideally, ing, bond, and interbank markets so as to capture the
the estimated FCIs would therefore entail primar- various channels through which monetary and macro-
ily exogenous shifts in financial conditions that are prudential policies can influence the broader economy.
distinct from the endogenous reflection of macroeco- Following the literature, the financial variables used
nomic fundamentals. Second, because the parameters include various interest rates and spreads (for example,
are allowed to change, the model can account for the changes in longer-term interest rate, corporate, inter-
evolving relationships between macroeconomic and bank, and term spreads), asset price returns (equity
financial variables over time.10 and house price returns), equity return volatility, and
In principle, the range of possible financial variables credit growth. Where available, survey-based infor-
to include in an FCI is vast. In practice, however, only mation (lending standards) can provide additional
a few studies use a large array of financial variables. For information about financial frictions (Annex 3.1).
example, the Organisation for Economic Co-operation Naturally, as the structure of, and products in, finan-
and Development develops FCIs for six major advanced cial systems evolve, the variables most relevant for
economies using seven variables. Even for the United tracking financial conditions may change. This chapter
States, the Kansas City Financial Stress Index is based estimates comparable monthly FCIs for 43 advanced
on 11 variables. Although Hatzius and others (2010) and emerging market economies during 19902016,
use up to 45 variables, and the Federal Reserve Bank of depending on data availability.
Chicago uses more than 100 in its U.S. factor models,
roeconomic conditions. However, financial variables can also reflect 11For financially open economies, financial conditions encompass
expectations of future macroeconomic developments. The FCIs are the ease of access to funding in both the domestic jurisdiction and
not purged of these expectations about the future in the baseline across borders. When firms rely more on international markets for
estimations to the extent that these expectations cannot be captured funding, global factors are expected to have a larger direct impact on
by the past and current behavior of macroeconomic variables. This their financing conditions. For the purposes of this chapter, the more
is an issue common to all FCIs. As a robustness check, professional indirect channel is considered, whereby global factors are potentially
forecasts of macroeconomic variables were considered as controls in a driver of domestic financial conditions. Similarly, the exchange
the case of the United States (based on data availability), which did rate is not included in the FCI. As mentioned earlier, exchange-
not result in any material changes to the FCI (consistent with Koop rate movements may influence domestic financial conditions, for
and Korobilis 2014). example, by altering the net worth of borrowers and thereby their
10Another advantage of the time-varying parameter factor-aug- terms of access to finance. The analysis aims at measuring these indi-
mented vector autoregressive model (TVP-FAVAR) is that the rect effects. Including the exchange rate directly in the FCI would
time-varying parameters help account for changes in (policy) regimes overstate the influence of global conditions on domestic financial
and, for example, financial-accelerator-related dynamics. Similarly, conditions, for example, in economies where exchange rate move-
the TVP-FAVAR recognizes that financial shocks in various periods ments have little effect on domestic financial conditions or where
can be transmitted to the real economy with varying intensity. they effectively serve as an insulating buffer.
Financial Conditions Indices: Selected Countries Figure 3.1. United States: Financial Conditions Indices,
19912016
Given its central role in the global financial system, (Standard deviations)
the United States is a natural starting point for appraising
the usefulness of the FCIs developed here. In addition, Estimated U.S. nancial conditions seem to reect key nancial
events well.
because many FCIs have been developed for the United
1. IMF Financial Stress Index versus Financial Conditions Index
States, several benchmarks can facilitate comparisons
7 IMF FSI FCI
across complementary approaches. It is reassuring that
6
the pattern of the U.S. FCI developed in this chapter
5
closely tracks counterparts developed by the IMF and 4
other institutions, such as the Federal Reserve Banks 3
of Chicago and Kansas City during 19902016 (Fig- 2
ure3.1).12 At the same time, the fluctuations in the 1
FCI appear to capture key U.S. financial events quite 0
well.13 After a period of relative tranquility in the early 1
2
1990s, financial conditions tightened as stock markets,
Jan. 1991
Jun. 92
Nov. 93
Apr. 95
Sep. 96
Feb. 98
Jul. 99
Dec. 2000
May 02
Oct. 03
Mar. 05
Aug. 06
Jan. 08
Jun. 09
Nov. 10
Apr. 12
Sep. 13
Feb. 15
Jul. 16
in particular, were rattled by the collapse of Long-Term
Capital Management, a hedge fund, in 1998. The FCI
remained elevated because of the dot-com crash in 2000,
when stock market declines were led by the technology 2. Financial Conditions Indices
sector. Then around 2002, the demise of accounting firm Chicago Fed FCI Kansas City Fed FSI FCI
7
Arthur Andersen and the bankruptcy of telecommunica-
6
tions corporation WorldCom (the largest in U.S. history 5
at the time), among other events, resulted in tighter 4
financial conditions. After a period of favorable condi- 3
tions, the global financial crisis broke out in 2008, result- 2
ing in an unprecedented spike in the FCI. More recently, 1
the FCI has been on a gradual uptrend, although still 0
1
indicating broadly accommodative conditions.
2
The FCIs in selected small open economies seem
Jan. 1991
Jun. 92
Nov. 93
Apr. 95
Sep. 96
Feb. 98
Jul. 99
Dec. 2000
May 02
Oct. 03
Mar. 05
Aug. 06
Jan. 08
Jun. 09
Nov. 10
Apr. 12
Sep. 13
Feb. 15
Jul. 16
to reflect their financial histories well. For instance, in
Russia, the FCI tightened dramatically during 1998 as
a consequence of the acute financial distress experi-
enced by the country at the time, with the degree of Sources: Federal Reserve Bank of Chicago; Federal Reserve Bank of Kansas City;
IMF, Global Data Source database; and IMF staff estimates.
tightening outpacing that encountered 10 years later Note: Higher values indicate tighter-than-average nancial conditions. The IMF FSI
during the global financial crisis (Figure3.2). By con- aims to identify episodes of acute nancial stress, when nancial intermediation is
impaired, similar to the Kansas City Fed FSI. The Chicago Fed FCI summarizes U.S.
trast, financial conditions in Korea were tighter during nancial conditions in money markets and debt and equity markets and in the
the global financial crisis than they were during the traditional and shadow banking systems (see Box 3.1 for details). FCI = nancial
Asian financial crisis (199798). Likewise, for Chile, conditions index; Fed = Federal Reserve Bank; FSI = nancial stress index.
which includes a discussion of different methods for constructing FCIs. sponding IMF FSIs. Interbank and corporate spreads, equity return
13Positive (negative) values of the FCI indicate that financial volatility, and changes in house prices are at the top of the list of
conditions are tighter (looser) than on average, which corresponds, the underlying financial variables contributing to countries FCIs.
for example, to higher-than-average (lower-than-average) corporate This result is broadly consistent for advanced and emerging market
spreads and lower-than-average (higher-than-average) credit growth. economies and in line with those in Hatzius and others 2010.
Figure 3.2. Selected Advanced and Emerging Market Economies: Financial Conditions Indices
(Standard deviations)
Major nancial developments appear to be captured by the nancial conditions indices across selected economies.
1. Russia 2. Korea
5 5
4 4
3 3
2 2
1 1
0 0
1 1
2 2
1997 2001 05 09 13 16 1997 2001 05 09 13 16
3. Chile 4. Netherlands
5 4
4 3
3
2
2
1 1
0
0
1
1
2
3 2
1999 2002 05 08 11 14 16 1997 2001 05 09 13 16
Financial Conditions and GDP Growth 3 percent).15 These findings confirm and extend the
The financial conditions indices developed in conclusions of Adrian, Boyarchenko, and Giannone
this chapter tend to signal downside risks to GDP. (2016). They show that the lower quantiles of
Domestic FCIs are significant predictors of future GDP growth (recessions) are more closely linked to
GDP growth across countries; however, this relation- financial conditions than upper quantiles (economic
ship changes depending on the state of the business expansions) for the United States.
cycle (Figure3.3). In particular, the inverse relation- Historical examples highlight the predictive power
ship between FCIs and future GDP growth is stron- of FCIs for future economic downturns. Two dates
ger for economic contractions (the lower percentiles are considered as an illustration: the second quar-
of the growth distribution) than for expansions (the ter of 2006 and the third quarter of 2008, broadly
upper percentiles of the growth distribution). For corresponding to the precrisis expansion and the
example, at the one-year-ahead horizon, the nega- onset of the global financial crisis, respectively. Fig-
tive coefficient at the 10th percentile (when growth ure3.4 shows the conditional distribution of growth
is well below percent) is about three times as 15A one standard deviation increase in the FCI (corresponding to
large in absolute terms relative to the coefficient tighter financial conditions) is associated with a 0.4 percentage point
corresponding to the median (when growth is about decrease in median future GDP growth at a one-year horizon.
Figure 3.3. Future GDP Growth and Financial Conditions: Figure 3.4. Probability Distributions of One-Year-Ahead GDP
Quantile Regressions Growth
(Percentage points) (Probability)
Financial conditions indices can ag downside risks to growth. Financial conditions improve the ability to predict future economic
downturns.
0.0
1. 2006:Q2
Unconditional historical distribution
0.2 Distribution conditional on GDP
0.8 Distribution conditional on GDP and FCI
0.7 Realized value
0.4
0.6
0.5
Probability
0.6
0.4
0.8 0.3
0.2
1.0 0.1
0
1.2 4 2 0 2 4 6 8
GDP growth (percent)
1.4
2. 2008:Q3 Unconditional historical distribution
5 10 25 50 75 90 95
Distribution conditional on GDP
Percentile Distribution conditional on GDP and FCI
Realized value
Source: IMF staff estimates. 0.30
Note: The gure shows the sensitivity of future growth to nancial conditions at
various quantiles. For all countries in the sample, growth at the one-year-ahead 0.25
horizon across selected quantiles is regressed against countries nancial
conditions indices. 0.20
Probability
0.15
0.10
one year ahead based on two empirical forecasting
0.05
models: one in which current and past growth rates
are used as predictors and one that augments the 0.00
8 3 2 7
first model by including FCIs. The idea is to gauge GDP growth (percent)
the extent to which additional information from the
FCIs helps improve forecast accuracy. Based on the Source: IMF staff estimates.
Note: The gure displays conditional probability distributions of one-year-ahead
information available as of the second quarter of real GDP growth for a panel of 43 countries based on quantile regressions. In
2006, the model with the FCIs attributes approxi- particular, it includes unconditional historical distributions and two conditional
distributions (of growth) based on two forecasting models that use either growth
mately a 45 percent probability to the actual growth or growth and FCIs to predict future growth (in 2007:Q2 or in 2009:Q3). The gure
outturn in one year (6 percent), which is more than also includes realized values (average actual GDP growth). FCI = nancial
conditions index.
twice the probability generated by the model that
uses only growth rates (Figure3.4). The distributions
using information up to the third quarter of 2008
differ to an even greater extent. The long left tail The Evolution of Financial Conditions around the World
in the distribution associated with the model with Three global factors seem to capture the dynamics
the FCIs (as opposed to the simple forecast model) of financial conditions across countries. A statistical
assigns a higher probability to economic downturns, dynamic factor model is used to generate multiple
more starkly signaling the actual GDP contraction unobservable (latent) factors that summarize the main
in the third quarter of 2009. FCIs appear to contain patterns across countries financial conditions. Although
valuable information about the future state of the these factors can be subject to various interpretations, an
economy and can be particularly useful in flagging interesting story emerges. It appears that the financial
downside risks to economic activity. conditions around the world can be summarized by three
Figure 3.5. Three-Factor Model Based on Financial Figure 3.6. Single Factor versus Principal Component
Conditions Index, 19952016 Analysis, 19952015
(Standard deviations) (Standard deviations)
Evidence suggests that global nancial conditions move in lockstep
Financial conditions around the world seem to be characterized by with the U.S. nancial conditions index and the VIX.
three global factors.
Global nancial conditions (factor model)
Global nancial crisis factor Emerging market factor
Global nancial conditions (PCA)
Euro area factor
6 U.S. nancial conditions 70
7
VIX (right scale)
6 5 60
5 4
50
4 3
40
3 2
2 30
1
1 20
0
0
1 10
1
2 0
2
Jan. 1995
Mar. 96
May 97
Jul. 98
Sep. 99
Nov. 2000
Jan. 02
Mar. 03
May 04
Jul. 05
Sep. 06
Nov. 07
Jan. 09
Mar. 10
May 11
Jul. 12
Sep. 13
Nov. 14
Jan. 16
Jan. 1995
Mar. 96
May 97
Jul. 98
Sep. 99
Nov. 2000
Jan. 02
Mar. 03
May 04
Jul. 05
Sep. 06
Nov. 07
Jan. 09
Mar. 10
May 11
Jul. 12
Sep. 13
Nov. 14
Jan. 16
factors, which can be characterized by the three main for this predominance is that the U.S. dollar takes center
historical crisis episodes over the past two decades. In stage as an international currency with important roles
particular, there seems to be an emerging market factor, in invoicing, issuance of financial assets, and commodity
a euro area factor, and a global financial crisis factor trading, among others (see also IMF 2014a).
(Figure3.5). Although each factor spikes during the A sizable share of fluctuations in countries financial
global financial crisis, the emerging market and euro area conditions is attributable to global financial shocks. On
factors also depict markedly tighter financial conditions average, global financial conditions account for about
during the late 1990s and around 2012, respectively. 30 percent of the variation in financial conditions across
Nevertheless, a single global factor adequately sum- countries, and though not shown, reaches almost 70
marizes financial conditions across countries. Such a percent in several economies (Figure3.7). As would be
factor is consistent with the notion of a global financial expected, the proportion of FCI variability explained
cycle discussed in Miranda-Agrippino and Rey 2015. by the three-factor model is larger than its single-factor
This single factor (the global financial factor or global counterpart and is greater than 40 percent.17 Relative
financial conditions) closely tracks the movements in the to emerging market economies, it appears that financial
U.S. FCI and the VIX (Figure3.6).16 This is in line with conditions in small open advanced economies are more
Reys (2013) argument that global financial conditions synchronized with global financial conditions.
are strongly driven by the United States, the key country
in the international monetary system. Part of the reason 17These magnitudes are larger than those in Miranda-Agrippino
and Rey (2015), for example, who report that a measure of global
16The average correlation between the U.S. FCI and the two mea- financial conditions accounts for about 21 percent of the variation
sures of global financial conditions and the VIX is 82 percent. across risky asset prices.
Figure 3.7. Variance Accounted for by One- and Three-Factor Figure 3.8. Variance Attributable to Global Financial
Models Conditions, 19952016
(Percent) (Percent)
An appreciable fraction of uctuations in countries nancial conditions The share of countries FCI variability accounted for by global nancial
is attributable to global nancial conditions. conditions does not appear to display a pronounced upward trend.
35 80
30 70
25
60
20
50
15
10 40
5 30
0
20
All
Emerging market
economies
Emerging
Asia
Emerging
Europe
Latin
America
10
Jan. 1995
Apr. 96
Jul. 97
Oct. 98
Jan. 2000
Apr. 01
Jul. 02
Oct. 03
Jan. 05
Apr. 06
Jul. 07
Oct. 08
Jan. 10
Apr. 11
Jul. 12
Oct. 13
Jan. 15
Apr. 16
Source: IMF staff estimates.
Note: The gure displays the share of countries FCI attributable to global factors Source: IMF staff estimates.
based on a one- or three-factor modelspecically, the R-squared from panel Note: The gure displays how the share of countries FCI variability
regression models in which countries FCIs are regressed on measures of global attributable to global nancial conditions changes over time. Specically, it
nancial conditions based on the one- or three-factor model. Simple averages are presents the total variance explained by the rst principal component across
used. FCI = nancial conditions index. countries FCI using either a 36- or 60-month rolling window. FCI = nancial
conditions index.
However, no clear evidence indicates that the impor- an upward trend from the end of the 1990s through the
tance of global financial conditions has been markedly global financial crisis, but then decline notably.20
increasing over the past two decades. The share of varia-
tion across FCIs accounted for by global financial condi- Country Characteristics and Sensitivity to Global
tions displays some cyclical patterns, especially during the Financial Conditions
global financial crisis, but portrays a broadly flat trajectory
Country characteristics are likely to influence how sen-
when viewed over the past 20 years (Figure3.8).18 These
sitive domestic financial conditions are to global financial
developments may reflect that the effect of greater finan-
shocks. Given the prominence of the United States in
cial linkages across countries has been partly offset by
the international monetary system, the U.S. FCI is taken
financial deepening that has been taking place in paral-
as a proxy for global financial conditions, based on the
lel.19 Although FCIs encompass various asset classes, these
findings discussed earlier.21 Key country characteristics
patterns are consistent with Bekaert and others (2016),
who document that equity return correlations display 20Carrieri, Chaieb, and Errunza (2013) argue that emerging mar-
bon 2002type adjustments, which correct for heteroscedasticity. ping confirm the importance of U.S. FCIs relative to other FCIs across
As an example of an additional robustness exercise, the average R2 countries. The U.S. FCIs provide more statistically significant informa-
statistics based on 36- and 60-month rolling regressions of countries tion about future FCIs in other countries than do other financial cen-
FCIs on global factors reveal broadly similar patterns. ters (including Germany, Japan, and the United Kingdom), with an
19Chapter2 of the April 2017 World Economic Outlook finds that average p-value of 7 percent. Analysis using convergent cross-mapping
the relative importance of external financial conditions for emerging (which complements Granger causality using nonlinear methods as
market and developing economies medium-term growth outcomes described in Sugihara and others 2012) suggests that U.S. FCIs reduce
has increased over time. prediction errors to the greatest extent across countries. Although
considered include financial linkages with the United general, and deeper financial (equity, bond) market
States (foreign direct investment, banking, and portfolio), depth in particular, are associated with an attenuated
financial openness and development, institutional quality, impact of global financial shocks on domestic FCIs.25
and the exchange rate regime (see Forbes and Chinn This echoes the conclusions in Chapter2 of the
2004; Aizenman, Chinn, and Ito 2015; and Sahay and April 2014 and April 2016 GFSRs, which find that
others 2015). For example, the expectation is that FCIs a larger domestic investor base and deeper banking
of countries that are more financially open and that systems and capital markets can increase the resilience
feature stronger financial linkages with the United States of emerging market economies to external financial
should be more sensitive to global financial conditions. shocks. Trade linkages to the United States do not
Conversely, countries with strong institutional and policy seem to matter, although trade relationships with the
frameworks as well as deep financial markets should dis- rest of the world appear to play a rolepossibly, this
play less sensitivity (Chinn and Ito 2007; Alfaro, Kalem- variable captures other factors such as indirect finan-
li-Ozcan, and Volosovych 2008; Brando-Marques, cial linkages. No clear pattern emerges regarding the
Gelos, and Melgar 2013; Chapter2 of the April 2014 exchange rate regime and capital account openness
GFSR).22 Given that an attempt has been made to purge results that are broadly consistent with Aizenman,
the FCIs of macroeconomic drivers, real economic link- Chinn, and Ito 2015.26 These findings are generally in
ages (such as trade ties) should not be among the deter- line with evidence that exchange rate flexibility allows
minants that help explain the influence of U.S. financial for considerable independence at the short end of the
conditions on local FCIs. Exchange rate regimes may term structure, but less so when it comes to broader
not matter very much for the transmission of financial measures of financial conditions, including, for exam-
conditions across countries because financial conditions ple, longer-term rates (Obstfeld 2015).27
work through various channels that typically cannot be
fully counterbalanced by exchange rate movements alone
(Obstfeld 2015). In what follows, the chapter investigates Can Countries Manage Domestic Financial
the extent to which FCIs across countries are correlated Conditions amid Global Financial Integration?
with the U.S. FCI, using a panel of small open advanced This section quantifies the relative share of fluctuations
and emerging market economies. It explores how the var- in countries domestic financial conditions explained by
ious country characteristics discussed earlier strengthen or global financial conditions and domestic monetary policy.
weaken this correlation.23 It finds that despite the importance of global finan-
Financial linkages are most closely associated with cial shocks, evidence suggests that monetary policy still
the extent to which FCIs are influenced by global accounts for a notable share of the variation in domestic
financial conditions. In particular, FCIs in countries financial conditions.
with stronger financial linkages (proxied by foreign
direct investment) with the United States tend to ment tends to be more permanent, it captures financial linkages
be more synchronized with global financial condi- better than portfolio and bank linkages.
25Along with the overall and financial markets indices developed by
tions (Table3.1).24 Greater financial development in Sahay and others (2015), the financial markets depth subindex tends
to be statistically significant and robust across specifications. This sub-
index includes measures of equity and bond market size and liquidity.
the U.S. FCI is taken as a proxy for global financial conditions, U.S. 26Recall that in contrast to a general measure of capital account
financial conditions may also be affected by financial developments openness, a more specific measure of financial integration as cap-
in other advanced and emerging market economies; see, for example, tured by foreign direct investment linkages with the United States is
Chapter2 of the April 2016 GFSR. statistically significant.
22It is sometimes argued that more liquid markets are more exposed 27Regarding the role of exchange rate regimes, recall that financial
to sell-offs by foreign investors. However, as discussed in Sahay and conditions can be transmitted across countries through various
others 2014, although some emerging market economies with rela- channels that typically cannot be fully offset by exchange rate move-
tively deeper and more liquid financial markets were strongly affected ments. Furthermore, relative to the sample in this chapter, which
during the taper tantrum in 2013, their more-developed financial considers 43 advanced and emerging market economies, studies that
markets subsequently facilitated the needed adjustment. find that exchange rate flexibility does confer monetary autonomy
23This is done by including the interaction between the U.S. FCI use larger sets of countries (for instance, Obstfeld [2015] considers
and the various country characteristics in the regressions (Annex 3.3). 70 countries) that are much more heterogeneous in composition
24Portfolio linkages matter too (and bank linkages to an even (and include low-income countries and other countries with a variety
lesser extent), but results of their importance are not as robust across of exchange rate regimes, which helps uncover the potential role
various specifications. It may be that because foreign direct invest- exchange rate flexibility can play).
Table3.1. Determinants of the Sensitivity of Domestic Financial Conditions to Global Financial Shocks
Variable Expected Sign Estimated Sign Significance
Direct Effect of U.S. FCI + + ***
Interaction with:
FDI Linkages with the United States + + **
Portfolio Linkages with the United States +
Banking Linkages with the United States +
Trade Linkages with the United States + +
Trade Openness + + **
Financial Openness + +
Exchange Rate Flexibility +
Financial Development **
Rule of Law
Source: IMF staff estimates.
Note: This table summarizes panel regressions in which countries' domestic FCIs are regressed against a measure of global financial conditions (U.S. FCI),
various country characteristics, and their interactions. Regressions include country fixed-effects terms, and standard errors are clustered at the country level.
See Annex 3.3 for details on baseline specifications. FCI = financial conditions index; FDI = foreign direct investment.
*** p < 0.01, ** p < 0.05, * p < 0.1.
5
28Initially,a parsimonious panel vector autoregression (VAR) model
is used, in which the variables are ordered as follows: U.S. FCI, indus- 10
trial production growth, inflation, domestic FCI, and the change in the 1 3 5 7 9 11 13 15 17 19 21 23
domestic monetary policy rate (all shocks identified using a Cholesky
decomposition); the results are robust to using the level of the variables Months
(Annex 3.4). The results do not change when exchange rate terms
are added into the panel VAR as an additional endogenous variable. Source: IMF staff estimates.
The results are also robust to inclusion of global industrial production Note: The gure displays the impulse response functions and 90 percent
condence bands of domestic nancial conditions indices to global nancial
growth, commodity prices, and a measure of global interest rates (prox-
or domestic monetary policy shocks for countries in the sample with exible
ied using several U.S. shadow rate measures) as exogenous controls. exchange rates. It is based on a panel vector autoregression model. See
The average responses from VAR models estimated for individual Annex 3.4 for details.
countries result in broadly similar findings. Complementary methods
of identifying the monetary policy shocks are discussed later in this
section. See also He and McCauley 2013; Chen, Mancini-Griffoli, and
Sahay 2014; Chen and others 2015; and Kose and others 2017.
Figure 3.10. Share of Domestic Financial Conditions Index Figure 3.11. Share of Domestic Financial Conditions Index
Fluctuations Attributable to Global Financial and Monetary Fluctuations Attributable to Global Financial Conditions
Policy Shocks (Frequency)
(Percent)
The importance of global nancial conditions varies across countries.
A notable share of domestic FCI uctuations can be attributed to global 8
nancial and domestic monetary policy shocks.
40 6
35
5
Frequency
30
4
25
20 3
15 2
10
1
5
0
0 13 17 21 25 29 33 37 41 45 50 54 58
Global nancial Monetary policy Domestic Other domestic
shocks shocks nancial factors Variance decomposition (percent)
conditions
Source: IMF staff estimates.
Source: IMF staff estimates.
Note: Histogram intervals on the x-axis vary because of rounding.The gure
Note: The gure displays the share of domestic FCI uctuations accounted for by
displays the share of uctuations in domestic nancial conditions attributable to
global nancial, domestic monetary policy, or domestic nancial condition shocks,
global nancial shocks based on vector autoregression models estimated
and shocks associated with other domestic factors for countries in the sample
individually for all countries in the sample. See Annex 3.4 for details.
with exible exchange rates. It is based on the panel VAR model or on VAR models
estimated individually for each country. See Annex 3.4 for details. FCI = nancial
conditions index; VAR = vector autoregression.
A considerable share of domestic FCI fluctuations is to monetary policy account for, on average, about 40 percent and
12percent of countries domestic FCI variations, respectively. The VAR
attributed to global financial conditions and domes-
model contains U.S. FCI, industrial production growth, inflation,
tic policy rates. On average, about 21 percent of the domestic FCI, and the change in domestic monetary policy. Robustness
variation in domestic FCIs across small open econo- exercises that control for global growth and commodity prices and, for
mies with flexible exchange rates is attributed to global instance, various lag lengths, yield broadly similar results. The variance
decompositions are statistically significant at the 10 percent level.
financial shocks (Figure3.10). This implies that the 30These results are based on the country-by-country VAR
Figure 3.12. Selected Advanced Economies: Response of Financial Conditions Index to Monetary Policy Shocks
(Standard deviations)
Country case studies highlight the inuence of domestic monetary policy on domestic nancial conditions.
1. Australia 2. New Zealand
GK: FCI GK: FCI
GK: FCI (upper and lower bound) GK: FCI (upper and lower bound)
Cholesky: FCI Cholesky: FCI
0.15 Cholesky: FCI (upper and lower bound) Cholesky: FCI (upper and lower bound) 0.04
0.10
0.02
0.05
0.00
0.00
0.02
0.05
0.10 0.04
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35
Months Months
3. Norway 4. Sweden
GK: FCI GK: FCI
GK: FCI (upper and lower bound) GK: FCI (upper and lower bound)
Cholesky: FCI Cholesky: FCI
0.10 Cholesky: FCI (upper and lower bound) Cholesky: FCI (upper and lower bound) 0.08
0.08 0.06
0.06
0.04
0.04
0.02
0.02
0.00
0.00
0.02 0.02
0.04 0.04
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35
Months Months
Sources: IMF, Global Data Source database; and IMF staff estimates.
Note: The gure displays the impulse response functions and 90 percent condence bands of domestic FCIs to domestic monetary policy shocks for
countries using two complementary methods to identify the monetary policy shocks. FCI = nancial conditions index; GK = Gertler and Karadi.
Moreover, in line with intuition, the results indicate measures of monetary policy shocks. In line with
that fluctuations in global financial conditions are the methodology traced out by Gertler and Karadi
associated with a greater share of FCI variability in (2015), who build on Grkaynak, Sack, and Swanson
countries that are relatively more financially integrated 2005, among others, unexpected changes in bond
with the rest of the world, and these differences are yields on central bank policy announcement dates
greater for emerging market economies. are used to measure policy surprises. Such shocks are
A closer look at relevant case studies reinforces derived for Australia, New Zealand, Norway, and
these results. The identification of shocks can be Swedenfour small open advanced economies with
especially difficult in the context of the VAR mod- floating exchange rate regimes and relatively deep
els used in the chapter, particularly for monetary financial markets. In each of the country cases shown
policy. Because precisely identifying monetary policy in Figure3.12, VAR models using these better-iden-
shocks is challenging, recent studies have devel- tified monetary policy shocks yield results similar to
oped methods that help better pinpoint exogenous those examined earlier, lending further credence to
Figure 3.13. Share of Domestic Financial Conditions Index of domestic financial conditions attributed to global
Fluctuations Attributable to Global Financial Shocks before financial conditions appears to be broadly stable over
and after the Global Financial Crisis the two periods (Figure3.13).32
(Percent)
of these shocksis statistically significant at the 5 percent level. statistically different at the 95 percent level.
shocks (IMF 2014b). Likewise, there may be circum- shocks. In particular, developing a local investor base
stances that warrant a temporary role for capital flow that encompasses both bank and nonbank financial
management measures (IMF 2016). intermediaries, as well as fostering greater equity and
Governments should prioritize domestic financial bond market depth and liquidity, can help dampen the
deepening to enhance resilience to global financial impact of external financial shocks.
each variables relative impact on GDP. For Canada, a relatively Korobilis 2014; Brave and Butters 2011; Hakkio and Keeton
open economy, the exchange rate received a weight about one- 2009; Carlson, Lewis, and Nelson 2012; Kliesen, Owyang, and
third that of the policy rate. Vermann 2012; Oet and others 2011.
2Financial stress indices (FSIs)which should not be confused 4See Illing and Liu 2003; Davis, Kirby, and Warren 2016;
with financial soundness indicatorscan be constructed with Moccero, Darracq Paries, and Maurin 2014; Guichard, Haugh,
similar variables and methods as FCIs. FSIs aim to identify and Turner 2009; Hollo, Kremer, and Lo Duca 2012; Dattels
episodes of acute financial stress, when financial intermediation and others 2010; Schler, Hiebert, and Peltonen 2016.
is impaired (extreme events are typically considered outright 5Exceptions include Brando-Marques and Perez-Ruiz forth-
crises). In practice, FCIs and FSIs can display similar dynamics in coming; Gumata, Klein, and Ndou 2012; and Kara, Ozlu, and
part because they can include similar financial variables (such as Unalmis 2012; for Chile, South Africa, and Turkey, respectively.
selected spreads) and because they may be constructed with similar 6Chapter4 of the October 2008 World Economic Outlook and
methods. For the United States, the patterns of the Kansas City Cardarelli, Elekdag, and Lall 2011 develop FSIs for 17 advanced
FSI resemble those of the FCIs developed by the other Federal economies, and Chapter4 of the April 2009 World Economic Out-
Reserve Banks (for example, Chicago and St. Louis) where all indi- look and Balakrishnan and others 2009 for major emerging market
ces capture the accommodative conditions before and the sharp economies. Osorio, Unsal, and Pongsaparn 2011 develop FCIs for
tightening in conditions during the global financial crisis. 13 selected Asian economies; see also IMF 2015.
Global-Level Variables
VIX Chicago Board Options Exchange Market Volatility Index Bloomberg L.P.
Global Real GDP Growth PPP-weighted average of real GDP growth IMF, World Economic Outlook database
Global Industrial Production Growth PPP-weighted average of industrial production growth IMF, Global Data Source database
(continued)
particular events, as discussed in the chapter. However, integration (trade and financial openness), linkages to
over the full sample, the variance gain offered by the the United States (foreign direct investment, banking,
two regional factors is limited (about 10 percentage portfolio and trade), exchange rate flexibility, financial
points on average), which suggests that the largest development, and rule of law. Additional controls (Z)
share of common dynamics across countries is actually include global variables (commodity price inflation and
driven by a single global factor, which moves in lock- global growth) and domestic variables (growth, inflation,
step with the U.S. FCI. and current account balance).39 The model includes
country fixed effects, and standard errors are clustered
at the country level. Results are generally robust to
Annex 3.3. Panel Regression Analysis37 alternative specifications, such as the inclusion of lags of
The effect of country characteristics on the sensi- the global driver and alternative measures of domestic
tivity of countries domestic financial conditions to macroeconomic conditions including growth expec-
U.S. financial conditions is estimated using a panel tations based on Consensus Economics forecasts (see
regression model. The specification is based on other Annex Table3.3.1 for baseline results).40
studies in the literature that analyze the relationship
between domestic financial variables (for instance,
stock returns and sovereign bond yields) and a global Annex 3.4. Panel Vector Autoregression
driver (typically proxied by the Chicago Board Analysis41
Options Exchange Volatility Index).38 The sample The study of the transmission of domestic monetary
covers 39 advanced and emerging market economies policy and global financial conditions to domestic
from 1991 to 2016. Countries that could be main financial conditions is based on a panel vector autore-
drivers of global financial conditions (Germany, gression (VAR) model. The system includes the U.S.
Japan, United Kingdom, United States) are excluded. financial conditions index (FCI), growth, inflation,
The model estimated is the following:
39All variables except the global driver are lagged to mitigate
CIit = i + 1FCItUS
F + 2CCHARit-1 + 3FCItUS
endogeneity concerns.
40FCIs are by construction standardized at the country level
CCHARit-1 + 4Zit-1
+ it, (A3.3.1)
(to aggregate information from the multiple financial variables).
in which FCI denotes domestic financial conditions, and This implies that a one-standard-deviation change in the FCI can
correspond to different changes in, for example, corporate spreads in
country characteristics (CCHAR) include measures of different countries, which could bias estimation. At the same time,
robustness analysis based on individual financial markets (including
37The authors of this annex are Nicolas Arregui and Dulani corporate spreads and equity returns) confirms the dampening role
Seneviratne. of financial development (see Chapter2 of the April 2014 GFSR).
38See, for instance, Bowman, Londono, and Sapriza 2015; Chap- 41The authors of this annex are Nicolas Arregui, Luis Brando-
ter2 of the April 2014 GFSR; Passari and Rey 2013; and Rey 2013. Marques, and Romain Lafarguette.
domestic FCI, and the change in domestic monetary compare results according to countries financial open-
policy. Growth is measured by industrial production, ness, an analogous exercise is conducted splitting the
and inflation is computed using the consumer price sample into two groups based on their relative capital
index. Monetary policy is measured with a mone- account openness (as measured by the Chinn-Ito
tary-policy-related interest rate (usually a central bank index). Results are generally robust to alternative lag
discount rate or a short-term money market rate). The specifications and to the inclusion of global industrial
sample consists of 25small open economies with flexi- production growth, commodity prices, and a measure
ble exchange rate regimes and uses monthly data from of global interest rates (proxied using several U.S.
2001 to 2016. The panel VAR is estimated with four shadow rate measures) as exogenous controls. The
lags using Pesaran, Shin, and Smiths (1999) mean results do not change when exchange rate terms are
group estimator, which is consistent in the presence of added into the panel VAR as an additional endoge-
dynamic heterogeneity. Impulse responses are drawn nous variable. The VAR models estimated individually
from Cholesky decompositions under the assumption for each country use the same set of variables and are
that domestic interest rates move last and U.S. FCI robust to the inclusion of global controls including
moves first. All standard errors are estimated using a commodity prices and world industrial production
nonparametric bootstrap and 1,000 replications. To growth.
Davis, E. Philip, Simon Kirby, and James Warren. 2016. The Area, Japan, United Kingdom and United States. OECD
Estimation of Financial Conditions Indices for the Major Economics Department Working Paper 677, Organisation for
OECD Countries. OECD Economics Department Working Economic Co-operation and Development, Paris.
Paper 1335, Organisation for Economic Co-operation and Gumata, Nombulelo, Nir Klein, and Eliphas Ndou. 2012. A
Development, Paris. Financial Conditions Index for South Africa. IMF Working
Disyatat, Piti, and Phurichai Rungcharoenkitkul. 2016. Finan- Paper 12/196, International Monetary Fund, Washington, DC.
cial Globalisation and Monetary Independence. BIS Paper Grkaynak, Refet S., Brian Sack, and Eric Swanson. 2005. The
88, Bank for International Settlements, Basel. Sensitivity of Long-Term Interest Rates to Economic News:
Doz, Catherine, Domenico Giannone, and Lucrezia Reich- Evidence and Implications for Macroeconomic Models.
lin. 2011. A Two-Step Estimator for Large Approximate American Economic Review 95 (1): 42536.
Dynamic Factor Models Based on Kalman Filtering. Journal Hakkio, Craig S., and William R. Keeton. 2009. Financial Stress:
of Econometrics 164: 188205. What Is It, How Can It Be Measured, and Why Does It Mat-
Dudley, William C. 2010. Comments: Financial Conditions ter? Federal Reserve Bank of Kansas City, Economic Review (2): 5.
Indexes: A Fresh Look after the Financial Crisis. Remarks at Hatzius, Jan, Peter Hooper, Frederic S. Mishkin, Kermit
the University of Chicago Booth School of Business Annual L. Schoenholtz, and Mark M. Watson. 2010. Financial
U.S. Monetary Policy Forum, New York, February 26. Conditions Indexes: A Fresh Look after the Financial Crisis.
, and Jan Hatzius. 2000. The Goldman Sachs Financial NBER Working Paper 16150, National Bureau of Economic
Conditions Index: The Right Tool for a New Monetary Policy Research, Cambridge, MA.
Regime. Global Economics Paper 44, Goldman Sachs, New He, Dong, and Robert N. McCauley. 2013. Transmitting
York. Global Liquidity to East Asia: Policy Rates, Bond Yields, Cur-
Forbes, Kristin J., and Menzie D. Chinn. 2004. A Decompo- rencies, and Dollar Credits. BIS Working Paper 431, Bank
sition of Global Linkages in Financial Markets over Time. for International Settlements, Basel.
Review of Economics and Statistics 86 (3): 70522. Hollo, Daniel, Manfred Kremer, and Marco Lo Duca. 2012.
Forbes, Kristin J., and Roberto Rigobon. 2002. No Contagion, CISS-A Composite Indicator of Systemic Stress in the
Only Interdependence: Measuring Stock Market Comove- Financial System. ECB Working Paper 1426, European
ments. Journal of Finance 57 (5): 222361. Central Bank, Frankfurt.
Fratzscher, Marcel. 2012. Capital Flows, Push Versus Pull Fac- Illing, Mark, and Ying Liu. 2003. An Index of Financial Stress for
tors and the Global Financial Crisis. Journal of International Canada. Working Paper 2003/14, Bank of Canada, Ottawa.
Economics 88 (2): 34156. Ilzetzki, Ethan, Carmen M. Reinhart, and Kenneth S. Rogoff.
Freedman, Charles. 1994. The Use of Indicators and of the 2017. Exchange Rate Arrangements Entering the 21st Cen-
Monetary Conditions Index in Canada. In Frameworks tury: Which Anchor Will Hold? NBER Working Paper 23134,
for Monetary Stability: Policy Issues and Country Experiences, National Bureau of Economic Research, Cambridge, MA.
edited by Tomas J. T. Balino and Carlo Cottarelli. Washing- International Monetary Fund (IMF). 2014a. Global Liquidity
ton, DC: International Monetary Fund, 45876. Issues for Surveillance. IMF Policy Paper, Washington, DC.
Gauthier, Celine, Christopher Graham, and Ying Liu. 2004. . 2014b. Staff Guidance Note on Macroprudential Pol-
Financial Conditions Indexes for Canada. Working Paper icyDetailed Guidance on Instruments. IMF Policy Paper,
22, Bank of Canada, Ottawa. International Monetary Fund, Washington, DC.
Gertler, Mark, and Peter Karadi. 2015. Monetary Policy . 2015. Financial Conditions in Asia: How Accommo-
Surprises, Credit Costs, and Economic Activity. American dative Are They? Regional Economic Outlook: Asia and Pacific,
Economic Journal: Macroeconomics 7 (1): 4476. Box1.4, International Monetary Fund, Washington, DC.
Gilbert, Paul D., and Erik Meijer. 2005. Time Series Factor . 2016. Capital FlowsReview of Experience with the
Analysis with an Application to Measuring Money. Working Institutional View. IMF Policy Paper, Washington, DC.
Paper, University of Groningen. , Financial Stability Board, and Bank for International
Gilchrist, Simon, and Egon Zakrajek. 2012. Credit Spreads Settlements. 2016. Elements of Effective Macroprudential Pol-
and Business Cycle Fluctuations. American Economic Review icies: Lessons from International Experience. Note to the G20,
102 (4): 1692720. August 2016, International Monetary Fund, Washington, DC.
Goodhart, Charles, and Boris Hofmann. 2001. Asset Prices, Kamin, Steven B. 2010. Financial Globalization and Monetary
Financial Conditions, and the Transmission of Monetary Pol- Policy. International Finance Discussion Paper 1002, Federal
icy. Paper presented at Conference on Asset Prices, Exchange Reserve Bank, Washington, DC.
Rates, and Monetary Policy, Stanford University, Stanford, Kara, A. Hakan, Pinar Ozlu, and Deren Unalmis. 2012. Finan-
CA, March 3. cial Conditions Indices for the Turkish Economy. CBT
Guichard, Stphanie, David Haugh, and David Turner. 2009. Research Notes in Economics 1231, Central Bank of the
Quantifying the Effect of Financial Conditions in the Euro Republic of Turkey, Ankara.
Kearns, Jonathan, and Nikhil Patel. 2016. Does the Financial Osorio, Carolina, D. Filiz Unsal, and Runchana Pongsaparn.
Channel of Exchange Rates Offset the Trade Channel? BIS 2011. A Quantitative Assessment of Financial Conditions in
Quarterly Review (December), Bank for International Settle- Asia. IMF Working Paper 11/170, International Monetary
ments, Basel. Fund, Washington, DC.
Kennedy, Mike, and Angel Palerm. 2014. Emerging Market Passari, Evgenia, and Hlne Rey. 2015. Financial Flows and
Bond Spreads: The Role of Global and Domestic Factors the International Monetary System. Economic Journal 125
from 2002 to 2011. Journal of International Money and (584): 67598.
Finance 43: 7087. Pesaran, Hashem M., Yongcheol Shin, and Ron P. Smith. 1999.
Kiyotaki, Nobuhiro, and John Moore. 1997. Credit Cycles. Pooled Mean Group Estimation of Dynamic Heterogeneous
Journal of Political Economy 105 (2): 21148. Panels. Journal of the American Statistical Association 94
Klein, Michael W., and Jay C. Shambaugh. 2015. Rounding (446): 62134.
the Corners of the Policy Trilemma: Sources of Monetary Primiceri, Giorgio. 2005. Time Varying Structural Vector
Policy Autonomy. American Economic Journal: Macroeconom- Autoregressions and Monetary Policy. Review of Economic
ics 7 (4): 3366. Studies 72: 82152.
Kliesen, Kevin L., Michael T. Owyang, and E. Katarina Ver- Rey, Hlne. 2013. Dilemma Not Trilemma: The Global
mann. 2012. Disentangling Diverse Measures: A Survey of Financial Cycle and Monetary Policy Independence. Paper
Financial Stress Indexes. Federal Reserve Bank of St. Louis presented at Global Dimensions of Unconventional Monetary
Review 94 (5): 36997. Policy Symposium, Jackson Hole, WY, August 24.
Koop, Gary, and Dimitris Korobilis. 2014. A New Index of . 2016. International Channels of Transmission of Mon-
Financial Conditions. European Economic Review 71: 10116. etary Policy and the Mundellian Trilemma. IMF Economic
Kose, M. Ayhan, Csilla Lakatos, Franziska Ohnsorge, and Marc Review 64 (1): 635.
Stocker. 2017. The Global Role of the U.S. Economy: Link- Sahay, Ratna, Vivek Arora, Athanasios V. Arvanitis, Hamid
ages, Policies, and Spillovers. CEPR Discussion Paper 11836, Faruqee, Papa M. NDiaye, and Tommaso Mancini-Griffoli.
Centre for Economic Policy Research, London. 2014. Emerging Market Volatility: Lessons from the Taper
Lane, Philip R., and Gian Maria Milesi-Ferretti. 2007. The Tantrum. IMF Staff Discussion Note 14/09, International
External Wealth of Nations Mark II: Revised and Extended Monetary Fund, Washington, DC.
Estimates of Foreign Assets and Liabilities, 19702004. Sahay, Ratna, Martin Cihak, Papa M. NDiaye, Adolfo Barajas,
Journal of International Economics 73 (2): 22350. Ran Bi, Diana Ayala, Yuan Gao, Annette Kyobe, Lam Nguyen,
Longstaff, Francis A., Jun Pan, Lasse H. Pedersen, and Kenneth Christian Saborowski, Katsiaryna Svirydzenka, and Seyed Reza
J. Singleton. 2011. How Sovereign Is Sovereign Credit Risk? Yousefi. 2015. Rethinking Financial Deepening: Stability and
American Economic Journal: Macroeconomics 3 (2): 75103. Growth in Emerging Markets. IMF Staff Discussion Note
Matheson, Troy D. 2012. Financial Conditions Indexes for 15/08, International Monetary Fund, Washington, DC.
the United States and Euro Area. Economics Letters 115 (3): Schoenmaker, Dirk. 2013. The Financial Trilemma. Economic
44146. Letters 111 (1): 5759.
Miranda-Agrippino, Silvia, and Hlne Rey. 2015. World Asset Schler, Yves Stephan, Paul Hiebert, and Tuomas A. Peltonen.
Markets and the Global Financial Cycle. NBER Working 2016. Coherent Financial Cycles for G-7 Countries: Why
Paper 21722, National Bureau of Economic Research, Cam- Extending Credit Can Be an Asset. https://ssrn.com
bridge, MA. /abstract=2539717 or http://dx.doi.org/10.2139/ssrn.2539717.
Moccero, Diego Nicolas, Matthieu Darracq Paris, and Laurent Sugihara, George, Robert May, Hao Ye, Chih-hao Hsieh, Ethan
Maurin. 2014. Financial Conditions Index and Identifica- Deyle, Michael Fogarty, and Stephan Munch. 2012. Detect-
tion of Credit Supply Shocks for the Euro Area. International ing Causality in Complex Ecosystems. Science 338: 496500.
Finance 17 (3): 297321. Swiston, Andrew J. 2008. A U.S. Financial Conditions Index:
Obstfeld, Maurice. 2015. Trilemmas and Trade-Offs: Living Putting Credit Where Credit Is Due. IMF Working Paper
with Financial Globalization. BIS Working Paper 480, Bank 08/161, International Monetary Fund, Washington, DC.
for International Settlements, Basel. Van den Heuvel, Skander J. 2002. Does Bank Capital Matter for
Oet, Mikhail V., Ryan Eiben, Timothy Bianco, Dieter Gramlich, Monetary Transmission? Economic Policy Review 8 (1): 25965.
and Stephen J. Ong. 2011. The Financial Stress Index: Iden- Yellen, Janet L. 2016. Current Conditions and the Outlook for
tification of Systemic Risk Conditions. Federal Reserve Bank the U.S. Economy. Speech at the World Affairs Council of
of Cleveland, Working Paper 1130, Cleveland, OH. Philadelphia, Philadelphia, PA, June 6.
From Great Depression to Great Financial Integration in Latin Breaking the Oil Spell: The Gulf
Recession: The Elusive Quest for America: A New Strategy for a Falcons Path to Diversification
International Policy Cooperation New Normal $40. 2016. Paperback
$27. 2017. Paperback $25. 2017. Paperback ISBN 978-1-51353-786-3. 212pp.
ISBN 978-1-51351-427-7. 229pp. ISBN 978-1-51352-024-7. 177pp.