This action might not be possible to undo. Are you sure you want to continue?
A World Economic Forum Report in collaboration with McKinsey & Company
The information in this report, or upon which this report is based, has been obtained from sources the authors believe to be reliable and accurate. However, it has not been independently verified and no representation or warranty, expressed or implied, is made as to the accuracy or completeness of any information contained in this report obtained from third parties. Readers are cautioned not to place undue reliance on these statements. The World Economic Forum and its project advisers, McKinsey & Company, undertake no obligation to publicly revise or update any statements, whether as a result of new information, future events or otherwise, and they shall in no event be liable for any loss or damage arising in connection with the use of the information in this report. The views expressed in this publication have been based on workshops, interviews and research, and do not necessarily reflect those of the World Economic Forum. World Economic Forum 91-93 route de la Capite CH-1223 Cologny/Geneva Tel.: +41 (0)22 869 1212 Fax: +41 (0)22 786 2744 E-mail: firstname.lastname@example.org www.weforum.org © 2010 World Economic Forum All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, including photocopying and recording, or by any information storage and retrieval system.
Preface Working Group Members Letter from the Working Group Executive Summary Introduction Chapter 1: Feast and Famine: The Uneven Rise of Credit Chapter 2: Criteria and Metrics for Sustainable Credit Chapter 3: Making Sustainable Credit a Reality: Challenges Ahead Chapter 4: Recommendations for Decision-Makers Conclusion Glossary References Technical Appendix Acknowledgements Project Team 5 7 9 11 19 21 31 41 51 59 61 63 67 79 81
More Credit with Fewer Crises | 3
Almost a year ago, when world leaders met at the World Economic Forum Annual Meeting in January 2010, the global economy was reeling from the worst financial crisis in half a century – a crisis attributed in large part to the failure of the financial system to detect and constrain pockets of excessive credit. Those gathered in Davos were determined to “rethink, redesign and rebuild” the institutions and practices that made the financial crisis possible. Some of the discussions also contained thoughts on recovery, highlighting previous pitfalls that needed to be addressed – and the use and availability of credit was among those issues most actively explored. In this spirit, the CEOs from many of the world’s leading financial institutions, those committed to the Industry Partnership programme of the World Economic Forum, endorsed a new initiative tasked with answering the following key questions: • What is a sustainable level of credit for the global financial system, a country or a given institution? • Where is credit most critical for economic development? • What actions can stakeholders take to ensure a healthy level of credit? To answer these questions, a World Economic Forum team has worked with many of its constituents and the active support of McKinsey & Company for much of the past year to build a comprehensive global credit model, analyse the relevant literature and seek the counsel of over 50 business, political and academic leaders around the globe. The team was steered by a working group of senior industry executives actively guiding its work. This report synthesizes the results of those efforts. Given its emphasis on the need for collaborative action to ensure sustainable credit, it is highly relevant to the theme of the World Economic Forum Annual Meeting 2011 – “Shared Norms for the New Reality”. The report strikes an optimistic tone, showing how such collaboration can help ensure that the credit needed to underpin global economic development is provided safely and responsibly. It proposes a new set of metrics to provide early warnings of excess lending, complementing the Annual Meeting’s focus on building a global risk response mechanism. Just as importantly, it also supports the Meeting’s focus on policies for inclusive growth, setting out mechanisms to address the credit blockages that hold back economic development – currently some 90% of small enterprises in developing markets lack access to credit. Finally, the report spells out the actions that regulators, policy-makers, and financial institutions can take to make sustainable credit a reality – echoing the emphasis the Annual Meeting will place on going beyond analysis and elaborating innovative solutions to global challenges. On behalf of the World Economic Forum, we wish to thank all who have contributed their time and expertise to this report, particularly the working group and the interview and workshop participants, Project Manager Isabella Reuttner, and our partners at McKinsey & Company. We all hope you find the report to be insightful and a helpful reference for best understanding the broad policy challenges associated with credit provision. Kevin Steinberg Chief Operating Officer World Economic Forum USA Gian Carlo Bruno Director and Head of Financial Services Industry World Economic Forum USA
More Credit with Fewer Crises | 5
Working Group Members
Viral V. Acharya Professor of Finance Leonard N. Stern School of Business, New York University Michael Bencsik Head of Europe, Middle East & Global Businesses Group Strategy and Planning HSBC Andreas Beroutsos Partner Eton Park Capital Management Fabrizio Campelli Managing Director, Head of Group Planning and Strategy Deutsche Bank Jayan Dhru Global Head of Financial Institutions Standard & Poors Michael Drexler Managing Director, Global Head of Strategy, Commercial Investment Banking and Wealth Management Barclays Shawn Miles Group Head of Global Public Policy MasterCard Dan Mobley Global Head of Government Relations Standard Chartered Charles Roxburgh Director McKinsey Global Institute Simon Samuels European Banks Research Barclays Capital Alexander Wolfson Managing Director and Head of Global Country Risk Management Citigroup From the World Economic Forum Gian Carlo Bruno Director and Head of the Financial Services Industries World Economic Forum USA
More Credit with Fewer Crises | 7
Letter from the Working Group
The financial crisis has thrown a harsh spotlight on the use of credit. Pockets of credit grew rapidly to excess – and brought the entire financial system to the brink of collapse. Yet, credit is the lifeblood of the economy, and much more of it will be needed to sustain the recovery and enable the developing world to achieve its growth potential. How, then, can the world’s growing demand for credit be met with greater stability and predictability – and fewer crises? This report is our contribution to answering that question. It takes a detailed look at the uses of credit in the real economy to define criteria for sustainable credit. And it presents a suite of tools that financial institutions, regulators and policymakers can use to gauge the sustainability of credit levels in both developed and developing economies. We are confident that these tools will provide clear yet nuanced early warnings of potential credit bubbles – and help decision-makers maintain credit within a defined band of sustainability, without unduly constraining growth. Just as important, the analysis and tools presented in this report highlight segments of the economy where credit shortages may stall growth. Indeed, a key aim of the report is to demonstrate credit’s role in the economic development cycle and help create mechanisms to ensure access to credit where it is needed to support growth. The report does not take a view on recent regulatory proposals, including from the Basel Committee on Banking Supervision (BCBS) and the Institute of International Finance (IIF), that may affect the supply of credit. Impact studies, most notably of Basel III, have already been concluded or are well underway elsewhere. Rather, this work supports the ongoing discussion on how to achieve sustainable levels of credit through collaboration between financial institutions (the engine of distribution), regulators (the keepers of stability) and policy-makers (those responsible for economic growth). It also has specific relevance to each of these stakeholders. For financial institutions, the report provides insights on the potential demand for credit over the next decade and on credit cycles over the past decade. And it provides guidance on what institutions can do today to ensure they meet credit demand responsibly in the years ahead. For regulators, the report proposes methodologies to monitor credit levels effectively, both within an economy and globally. It provides first thoughts on how to quantify credit contagion risk and offers a methodology to monitor this risk globally. For policy-makers, the report sets out approaches to identify credit segments at risk of overheating, as well as pockets of the economy that may be starved of credit. It proposes mechanisms to help address both problems. This report will have succeeded if it stimulates thinking and spurs discussion on how credit levels can be kept sustainable. Ultimately, though, we hope the findings and frameworks presented here will contribute both to increased stability for the financial system and to sustained economic development across the world.
More Credit with Fewer Crises | 9
To support economic development, global credit levels must grow substantially over the next decade. At the same time, public and private decision-makers must avoid a repeat of the credit excesses that recently brought the world financial system to its knees. Can the world’s growing demand for credit be met responsibly, sustainably – and with fewer crises? The answer, this report shows, is “yes”. But to achieve this goal, financial institutions, regulators, and policy-makers need more robust indicators of unsustainable lending, contagion risk, and credit shortages – and better mechanisms to ensure credit drives development. The report is intended as a contribution to building that toolkit. It has four parts: 1. Feast and Famine: The Uneven Rise of Credit 2. Criteria and Metrics for Sustainable Credit 3. Making Sustainable Credit a Reality: Challenges Ahead 4. Recommendations for Decision-Makers
1. Feast and Famine: The Uneven Rise of Credit
A detailed global credit model was developed to map historical credit volumes and forecast potential credit demand to 2020 – across the wholesale, retail and government credit segments in 79 countries, representing 99% of world credit volume. The precarious rise of credit, 2000-2009 Global credit stock doubled from US$ 57 trillion to US$ 109 trillion between 2000 and 2009, at a 7.5% compound annual growth rate. This expansion was spread fairly evenly between the government, wholesale and retail segments until 2009, when government lending rose sharply to fund the banking bailout and to support economic stimulus programmes (Exhibit i). By 2009, some 70% of the world’s US$ 150 trillion in financial assets was being used to fund credit. That 70% was intermediated almost equally between capital markets and banks. Developing markets’ credit stock grew the fastest – by US$ 15 trillion between 2000 and 2009, an annual rate of 13%. The increasing depth of developing countries’ financial systems during this period was a key contributing factor to this growth.
More Credit with Fewer Crises | 11
Credit growth over the decade was also fuelled by financial innovation and industry practice, including: • The rapid growth of securitization in the US and United Kingdom until 2007 – largely stalled since the crisis • The rise of leveraged lending, including the private equity boom between 2005 and 2007 • Rapid development of capital markets worldwide – total financing via government and wholesale corporate bonds doubled to nearly US$ 49 trillion between 2000 and 2009 • Low credit prices relative to risk as a result of competition among financial institutions for growth Between 2000 and 2009, the world economy grew at a healthy rate, and the world’s stock of credit outpaced GDP growth by less than 2 percentage points a year – not an unsustainable rate of leverage increase. Beneath this aggregate picture, though, there were pockets of overheating. Clear warning signs of credit excess in countries such as Ireland, Spain and Greece were largely ignored as borrowing continued to rise beyond sustainable levels. At the same time, other segments were credit-starved, including China’s retail segment, microfinance in India, and the small business sector worldwide. In developing markets, some 90% of small enterprises lack access to credit. Potential credit demand to meet forecast economic growth to 2020 The study forecast the global stock of loans outstanding from 2010 to 2020, assuming a consensus projection of global economic growth at 6.3% (nominal) per annum. Three scenarios of credit growth for 2009-2020 were modelled: • Global leverage decrease. Global credit stock would grow at 5.5% per annum, reaching US$ 196 trillion in 2020. To meet consensus economic growth under this scenario, equity would need to grow almost twice as fast as GDP. • Global leverage increase. Global credit stock would grow at 6.6% per annum, reaching US$ 220 trillion in 2020. Likely deleveraging in currently overheated segments militates against this scenario. • Flat global leverage. Global credit stock would grow at 6.3% per annum to 2020, tracking GDP growth and reaching US$ 213 trillion in 2020 – almost double the total in 2009 (Exhibit ii). This scenario, which assumes that modest deleveraging in developed markets will be offset by credit growth in developing markets, provides the primary credit growth forecast used in this report.
12 | More Credit with Fewer Crises
Rapid credit growth is forecast in developing markets, which will add almost US$ 50 trillion to their credit stock by 2020. China’s credit demand will lead global credit growth: it will require US$ 20 trillion more credit in 2020 than in 2009, with 80% of that growth going to the wholesale segment. In developed markets, including the large Western economies, most of the growth will come from the government segment. In North America alone, the value of government bonds is expected to grow by US$ 12 trillion to 2020. Deleveraging in overheated retail and wholesale segments of the developed world will be significant. The financing needs of infrastructure and green technology projects will form a significant part of credit demand growth.
2. Criteria and Metrics for Sustainable Credit
After interviewing more than 50 industry CEOs, central bankers, regulators and academics, the study defined four criteria for sustainable credit: • • • • Limited “hotspots”, or areas of excess credit where repayment and servicing prospects are at risk Transparent and manageable contagion risk in order to reduce system volatility Limited “coldspots”, or segments where growth is inhibited by a lack of access to credit Alignment with social goals, to ensure that both economic and welfare needs are met
Based on these criteria, the study developed a suite of new tools to monitor credit levels. Limited hotspots The study analysed 15 credit crises over the past quarter century and derived a dashboard of 12 “rules of thumb”, or quantified upper bounds, to identify potential hotspots in wholesale, retail and government credit (Exhibit iii).
Transparent and manageable credit contagion risk “Credit contagion risk” measures the likelihood and magnitude of the contagion that would arise from a hypothetical future credit crisis in a particular segment. In addition to the rules of thumb above, which measure a credit segment’s local sustainability, the study developed a framework to measure the connectivity between individual segments in different countries. There is potential to extend this methodology to create a global credit connectivity monitor covering most countries.
More Credit with Fewer Crises | 13
Limited coldspots When a country falls into the lowest quintile of credit penetration among countries in a similar development stage, this may be a first indicator of possible coldspots. Additional indicators of coldspots include large unbanked populations and shallow capital markets, which can create structural barriers to meeting credit demand. The study developed a credit “heatmap” to monitor both hotspots and coldspots across countries and segments. Alignment with social goals In the long run, the scale and distribution of credit is only economically sustainable if it meets broader social objectives. Credit is linked to such objectives during all stages of a country’s economic development. In early stages, credit is used to support family-owned businesses; next, it supports small and large corporations, and finally it is used to smooth consumption. However, if credit is to drive consistent economic development, three basic foundations must be in place: strong laws and legal enforcement; creditable and functioning institutions; and macroeconomic and fiscal discipline.
3. Making Sustainable Credit a Reality: Challenges Ahead
Using the methodology set out above, the study addressed three key questions that will determine whether the world’s growing demand for credit can be met sustainably. Are we at risk of future crises – and if so, where? Even though some economies will deleverage over the coming decade, the analysis projects a significant number of credit hotspots across the world in 2020 – including retail credit hotspots in countries representing almost half of global GDP. Government credit hotspots are projected for countries representing 13-14% of world GDP – although Western Europe will be more vulnerable. In wholesale credit, Asia and Western Europe will be the main drivers of hotspots in 2020 (Exhibit iv).
14 | More Credit with Fewer Crises
What is the future risk of contagion? The study mapped countries’ local credit sustainability against their degree of credit connectivity in 2009. Significant economies – including the United Kingdom, the US and Japan – pose a potential threat in terms of local sustainability, credit connectivity risk or both (Exhibit v). In the US, high levels of absolute credit stock, combined with extensive links to other major credit segments, drive considerable contagion risk. Analysis of countries in the Eurozone shows that Ireland, Portugal and Greece pose the highest threat to the zone’s stability. China’s credit contagion risk appears relatively low, driven by low foreign credit and import penetration, along with relatively low local sustainability risk. However, the second-order effects of a Chinese credit crisis, for example on supply chains, could be massive.
More Credit with Fewer Crises | 15
Will credit growth be sufficient to meet demand? Rapid growth of both capital markets and bank lending will be required to meet the increased demand for credit – and it is not assured that either has the required capacity. There are four main challenges. Low levels of financial development in countries with rapid credit demand growth. Future coldspots may result from the fact that the highest expected credit demand growth is among countries with relatively low levels of financial access. In many of these countries, a high proportion of the population is unbanked, and capital markets are relatively undeveloped. Challenges in meeting new demand for bank lending. By 2020, some US$ 28 trillion of new bank lending will be required in Asia, excluding Japan (a 265% increase from 2009 lending volumes) – nearly US$ 19 trillion of it in China alone. The 27 EU countries will require US$ 13 trillion in new bank lending over this period, and the US close to US$ 10 trillion. Increased bank lending will grow banks’ balance sheets, and regulators are likely to impose additional capital requirements on both new and existing assets, creating an additional global capital requirement of around US$ 9 trillion (Exhibit vi). While large parts of this additional requirement can be satisfied by retained earnings, a significant capital gap in the system will remain, particularly in Europe.
The need to revitalize securitization markets. Without a revitalization of securitization markets in key markets, it is doubtful that forecast credit growth is realizable. There is potential for securitization to recover: market participants surveyed by McKinsey in 2009 expected the securitization market to return to around 50% of its pre-crisis volume within three years. But to rebuild investor confidence, there will need to be increased price transparency, better data on collateral pools, and better quality ratings. The importance of cross-border financing. Asian savers will continue to fund Western consumers and governments: China and Japan will have large net funding surpluses in 2020 (of US$ 8.5 trillion and US$ 5.7 trillion respectively), while the US and other Western countries will have significant funding gaps. The implication is that financial systems must remain global for economies to obtain the required refinancing; “financial protectionism” would lock up liquidity and stifle growth.
16 | More Credit with Fewer Crises
4. Recommendations for Decision-Makers
The study recommends eight actions that financial institutions, regulators and policy-makers can take today to ensure sustainable credit levels for the future. a. Integrate the concepts of sustainable credit into the regulatory agenda. New liquidity and funding regulation can help reduce the frequency and intensity of credit hotspots and should be supported. However, decision-makers should guard against the risk that such regulation limits sustainable global credit growth or creates new credit coldspots. b. Create standardized government accounting practices to increase transparency and accurately assess sovereign finances. Governments should adopt uniform accounting standards so that a complete and transparent picture of each country’s financial resources and obligations is available. The challenge of creating and agreeing on such standards should be made part of the G20 agenda. c. Encourage responsible borrowing through financial education. An international, government-led initiative should undertake an impact study of existing financial education programmes worldwide, and then bring together government officials, education leaders and financial institutions to agree on an approach and implementation plan for improving financial literacy. d. Encourage financing of local coldspots through targeted mechanisms. Governments and banks should create targeted mechanisms to solve the well-documented problem of lending to SMEs in developing markets. In developed markets, they should establish a robust fact base on the extent to which SME lending is constrained, and develop innovative solutions to improve both the supply and the demand side. e. Task a single agency with monitoring global credit levels and system-wide credit sustainability. A single organization should be tasked with monitoring global credit levels so that the risks to financial institutions are accurately assessed. This organization should build on the credit sustainability metrics presented in this report and combine them with the Early Warning Exercise methodology developed by the FSB and IMF. f. Align banks’ risk appetite with sustainable credit criteria. Banks should ensure that their contribution to systemic risk is considered at the level of day-to-day credit and lending decisions. Regulators and supervisory bodies should recognize the contributions made to financial stability by banks that are aligned with sustainable credit principles. g. Drive innovation by financial institutions, developing new mechanisms that can safely meet future global credit needs. Governments could help kick-start securitization markets with targeted mechanisms such as a government-subsidized programme to securitize SME loans. Financial institutions should develop further mechanisms to grow balance sheet capacity safely and meet future worldwide credit needs – including measures to strengthen housing and environmental finance, and to integrate the unbanked into the banking system. h. Establish goals for efficient and deep capital markets by 2020 in developing economies. Governments, international agencies and the G20 should promote the strengthening of capital markets in developing countries by institutionalizing two fundamental capital market development goals: improving infrastructure to broaden participation by foreign firms and investors, and creating a sound institutional environment.
More Credit with Fewer Crises | 17
The rapid expansion of credit in recent decades has enabled unprecedented levels of economic development, business activity, home ownership and public sector spending. Yet, excess lending in some markets and sectors sparked a global crisis which brought the entire financial system to its knees. Not surprisingly, many commentators believe credit should be scaled back, even at the expense of economic growth. The analysis in this report suggests the opposite is true. There are major pockets of the world economy, particularly in developing markets, whose growth has been held back by credit shortages, even over the past 10 years. To unlock development in these areas, and to meet consensus forecasts of world economic growth, credit levels must grow substantially over the next decade. At the same time, public and private decision-makers must avoid a repeat of the credit excesses that have caused so much damage in recent years. Can the world’s growing demand for credit be met responsibly, sustainably – and with fewer crises? The answer, this report shows, is “yes”. But to achieve this goal, financial institutions, regulators and policy-makers need a far more robust toolkit to foresee and forestall potential credit crises. They need earlier, more in-depth indicators of unsustainable lending, contagion risk and credit shortages – and better mechanisms to ensure that credit drives development. The report, based on a nine-month collaboration between senior practitioners and experts, is intended as a contribution to building that toolkit. (See sidebar, “Study Methodology”.) It maps the potential growth in credit demand; proposes criteria to define and tools to track the sustainability of future credit growth; applies those tools to identify the segments most at risk of credit excess or shortage over the coming decade; and sets out actions to achieve a healthy balance of credit in the world economy. The report has four chapters: • Chapter 1: “Feast and Famine: The Uneven Rise of Credit”, shows how the steady expansion of credit over the past decade masked pockets of excess and shortage. It then forecasts the credit required to meet economic growth projections over the coming decade. • Chapter 2: “Criteria and Metrics for Sustainable Credit”, proposes a set of criteria by which sustainable credit can be defined and managed, and introduces a range of tools and methodologies to provide early indicators of areas of credit excess, contagion or shortage. • Chapter 3: “Making Sustainable Credit a Reality: Challenges Ahead”, applies these tools to identify segments at risk of excess credit, which could potentially lead to a crisis, as well as those in which credit shortages could stunt growth. • Chapter 4: “Recommendations for DecisionMakers”, sets out principles for action and potential measures that financial institutions, regulators and policy-makers can take to ensure sustainable credit for the future. The Technical Appendix provides further detail on the study’s methodology, fact base and data output.
Study Methodology To create a robust fact base for the study, a detailed global credit model was constructed to map credit volumes between 2000 and 2009, and to project potential credit demand to 2020. The model spans 79 countries representing 99% of credit volume (with approximated figures for the remaining countries). It disaggregates historic and projected lending by wholesale, retail and government segments, defined as follows: • Retail credit: All household credit stock (loans outstanding), including mortgages and other personal loans such as credit cards, auto loans and other unsecured loans • Wholesale credit: All corporate and SME credit stock (loans and bonds outstanding) • Government credit: All public sector credit stock (including loans and bonds outstanding) The model quantifies both capital market bonds and lending from financial institutions – the latter aggregated bottom-up from data at the sub-product level data, including on consumer finance, mortgages, leasing, and corporate and small and medium enterprise (SME) lending1. The model covers both commercial banks and non-banking financial institutions, includes credit held by both residents and non-residents in a given country, and calculates both non-securitized and securitized volumes. In addition, the project team – made up of financial services experts from the World Economic Forum and McKinsey & Company – interviewed more than 50 industry CEOs, rating agencies, central bankers, regulators and academics. (They are listed in the Acknowledgements.) Based on these expert perspectives, criteria and tools to define and track the sustainability of credit were developed – as were mechanisms and incentives to make sustainable credit a reality. Finally, the research was underpinned by an extensive review of the key academic and industry literature. (The References section provides a full list of study sources.) The report aims to build on this knowledge by providing a comprehensive, analytically robust fact base on historic and potential future credit growth, by bringing a practitioner’s perspective to bear on the complex problem of defining and measuring sustainable credit, and by setting out actionable recommendations.
Credit figures only include transactions involving the end-user, i.e. the final customer; credit such as interbank lending is excluded to avoid double counting.
More Credit with Fewer Crises | 19
Chapter 1: Feast and Famine: the Uneven Rise of Credit
This chapter looks back at the growth in global credit volumes over the past decade, and ahead to the next decade. First it shows why the expansion of credit since 2000 was unsustainable: not so much because the absolute levels of lending were too high, but because the aggregate picture masked pockets of both excess and shortage. Then it forecasts the potential credit expansion required to meet economic growth projections over the coming decade. This analysis demonstrates that historic credit growth has been highly uneven, and often volatile and unstable. It shows that, in several countries now in crisis, there were warning signs years before lending grew to excess – signs that were missed, or spotted but not addressed. But the analysis also makes it clear that, in the years ahead, continued strong credit growth will be required to support economic development. Ensuring that the credit flows of the future are efficient and sustainable will therefore require new approaches from financial institutions, regulators and policy-makers. The Precarious Rise of Credit, 2000-2009 From any angle, the growth of credit over the past decade has been rapid. Global credit stock doubled from US$ 57 trillion to US$ 109 trillion between 2000 and 2009, at a 7.5% compound annual growth rate (CAGR). This expansion was spread fairly evenly between the government, wholesale and retail segments until 2009, when government lending rose sharply to fund the banking bailout and to support economic stimulus programmes (Exhibit 1). Credit also grew rapidly relative to equities: whereas global credit stock was 1.7 times greater than the total market value of equity investments in 2000, by 2009 it was 3.7 times as large2. (It should be noted, however, that this ratio is not particularly high by historical standards: in the US, for example, the credit-to-equity ratio was above five for most of the 1970s and 1980s, and declined only with the onset of the 1990s equities boom.) By 2009, some 70% of the world’s US$ 150 trillion in financial assets3 was being used to fund credit (Exhibit 2). That 70% was intermediated almost equally between capital markets and banks.
Equity data is market value and includes investment in listed shares as well as directly owned equity stakes (global year-end data). Total size of equity market in 2000 = US$ 31 trillion; in 2009 = US$ 33 trillion. 3 Excludes wealth and assets outside financial intermediation: cash, gold and other commodities, non-investment real estate, etc.
More Credit with Fewer Crises | 21
Credit grew rapidly in every major region of the world. However, developing markets’ credit stock grew the fastest – by US$ 15 trillion between 2000 and 2009, an annual rate of 13% (Exhibit 3). The increasing depth of developing countries’ financial systems during this period was a key contributing factor to this growth, as was their greater openness to foreign direct investment (FDI). For example, the tenfold increase in India’s FDI between 2000 and 2008 was matched by a growth in credit stock from US$ 10 billion to US$ 50 billion. And the dramatic increase in Brazil’s credit bureau coverage, from less than one in 10 adults in 2005 to the majority of the population just four years later, coincided with a near-tripling of credit stock to almost US$ 60 billion in 2010.
22 | More Credit with Fewer Crises
Financial innovation and industry practice fuel credit growth Credit growth was also fuelled by financial innovation and industry practice. Consider the rapid adoption of securitization, the repackaging and sell-on of loans by banks to other investors. In the US, for example, the growth in securitization volumes outstripped that of retail credit between 2005 and 2007; in 2007, total securitization volumes had reached US$ 9 trillion, equivalent to almost two-thirds of the country’s total retail credit stock. Securitization was also a significant contributor to mortgage and consumer finance volume growth in the United Kingdom, growing fivefold from US$ 0.1 trillion in 2000 to US$ 0.5 trillion in 2006. The sub-prime crisis led to a sharp decline in new securitization issuances in the US and United Kingdom – by between 35% and 90%4 from 2007 to 2008, depending on segment, definitions and data sources. Government- and agency-backed securitization in the US is the only segment that has shown signs of recovery to date. Likewise, leveraged lending played a key role in wholesale credit growth. Lower interest rates, loosened lending standards, and increased use of the structure and distribute model by investment banks all led to a private equity boom between 2005 and 2007, when some of the largest leveraged buyouts (LBOs) in history took place. In 2007, leveraged lending levels reached approximately 7% of total wholesale credit volumes in both the US and the United Kingdom. However, the financial crisis brought the LBO market to a standstill: by 2009, leveraged lending had shrunk to less than 0.2% of total wholesale lending in these markets. A further important contributor to credit growth was the rapid development of capital markets. Total financing via capital markets – including both government and wholesale corporate bonds – doubled to nearly US$ 49 trillion between 2000 and 2009. Growth was particularly steep in the developing world, with the capital markets of Asia, the Middle East, Eastern Europe and South America all expanding at an annual rate above 12% for the decade. The Chinese wholesale bond market alone grew at 49% per annum from 2000 to 2007. A fourth contributor to credit growth was competition among financial institutions for growth, causing banks to expand their balance sheets significantly. In some segments, the race for growth led to low credit prices relative to risk. In large corporate lending, this underpricing was largely rational – lending at a loss led to profitable cross-selling of specialized finance, risk management products, investment banking, and the like. In sectors such as commercial real estate, however, lending was under-priced relative to the risks of these assets, as is now painfully clear in many commercial real estate markets (such as Ireland). And in the retail segment, it proved unsustainable and led to significant losses. In particular, the thin margins in mortgage lending left no buffer for default when US house prices began to plummet at the end of 2006. These financial industry innovations and practices should be seen alongside government’s significant role in driving rapid credit growth in the wholesale and retail segments. For example, the tax deductibility of interest payments in the US and elsewhere encouraged increased mortgage lending and led companies to favour debt over equity5. Moreover, the bank bailouts of recent years were a major, if unanticipated, contributor to the increase in governments’ own indebtedness. Credit growth compared with GDP growth On the face of it, the growth of credit between 2000 and 2009 might appear to have been unsustainable. But the rise in credit must be seen against the backdrop of economic growth. Between 2000 and 2009, the world economy was growing at a healthy rate – at 5.3% annually in nominal terms, or 2.2% in real terms. The rate of GDP growth between 2000 and 2008 was 6.2% nominal and 2.8% real. This means that the world’s stock of credit outpaced GDP growth by less than 2 percentage points a year – not a wide margin. In theory, there is nothing unsustainable about this picture: as long as credit grows broadly in line with economic growth, the credit is put to good use and borrowers can meet interest obligations and repay principal. At the global level, then, there was only a modest increase in leverage (the ratio of credit to GDP) between 2000 and 2009. However, the picture differed markedly for developing and developed economies. Even though the developing world’s credit stock grew fastest6, its GDP grew at an astonishing 11.3% annually from 2000 to 2009; its leverage
Different data sources use different approaches to define and quantify securitization. Even the broadest and most conservative measure, however, shows a large drop. International Monetary Fund. “Debt Bias and Other Distortions: Crisis-Related Issues in Tax Policy.” Prepared by the Fiscal Affairs Department. Approved by Carlo Cottarelli, 12 June 2009 6 Developing world defined as all countries and regions except North America, Western Europe and Japan.
More Credit with Fewer Crises | 23
therefore rose by only 1.8% a year over the decade. On the other hand, the developed economies in the Organisation of Economic Co-operation and Development (OECD) saw their annual GDP growth fall from 4.3% in 1990-2000 to 3.4% in 2000-2009. Their credit levels rose much faster: while OECD leverage grew at just 1.2% a year in 1990-2000, it grew almost three times faster, at 3.2% a year, over the following decade (Exhibit 4).
Segments of credit excess and shortage For the world in aggregate, this analysis indicates that overall growth in credit volumes and leverage was not unsustainable. Why, then, did things go so horribly wrong? As the sharp increase in developed world leverage suggests, the aggregate picture masked pockets of overheating. The methodology developed in this study, and set out in the next chapter, shows that there were clear warning signs of credit excess as early as 2006 in now crisis-struck countries such as Ireland, Spain and Greece – signs that were largely ignored as borrowing continued to rise beyond sustainable levels. For example, retail credit volume as a proportion of Ireland’s GDP doubled from 43% to 87% between 2000 and 2006 – a drastic rate of increase which should have flagged a problem. Instead, retail credit continued to rise, reaching 99% of GDP by 2008. Likewise, Spain’s wholesale credit volumes, driven by property development, doubled from 52% to 110% of GDP between 2000 and 2006 – clearly an unsustainable rate of growth. Again, warning signs of the impending crisis were missed, and wholesale credit volumes rose to 134% of GDP in 2008. There is controversy about the exact levels of Greek government debt over the past decade. According to Oxford Economics, Greece’s government borrowing stood at 88% of GDP in 2000 and rose to over 140% in 2006. Eurostat puts the figures at 115% in 2000 and 108% in 2006. Either set of figures would have provided a clear early warning of the country’s sovereign debt crisis in 2010. While some credit segments were overheating, however, others were credit-starved. Consider China’s lack of access to retail credit, which contributes to the country’s relatively low consumption7. China’s outstanding consumer credit as a proportion of GDP, at less than 15%, is the lowest of any of the 10 largest economies; the same is true of its consumption, at 35% of GDP8. It is notable that over 70% of the 4 trillion renminbi government stimulus package launched by China’s government in 2008 was directed towards infrastructure spending, while less than 10% went in support of consumption.
While there is a view that China’s low credit penetration is driven to a large extent by cultural factors, analysis shows that China’s investment-led model has skewed the economy towards industry and has made corporate investment too cheap. Inefficient investment into excess capacity has occurred at the cost of Chinese consumption. See McKinsey Global Institute, “If you’ve got it, spend it: Unleashing the Chinese consumer”, August 2009. 8 Calculated using data from Global Insight
24 | More Credit with Fewer Crises
Despite the recent turmoil around India’s microfinance industry9, small-scale lending in that country is a clear example of a credit-starved segment. Annual credit demand by the poor in India is estimated at US$ 15 billion, yet the current volume of microfinance barely tops US$ 1 billion; only about 5% of India’s rural poor have access to finance10. The microfinance sector lacks the infrastructure needed to ensure that credit is provided safely with credible repayment and servicing prospects. As recent events show, the political environment also creates barriers to microfinance meeting its full potential. Worldwide, small business is a key sector that lacks sufficient access to finance, with the gap in credit for micro, very small, small, and medium enterprises (MSMEs) estimated at between US$ 3.1 trillion and US$ 3.8 trillion11. These enterprises make a major contribution to innovation, long-term economic growth, income stability, and employment – in developed as well as developing countries. In Japan, for example, 99.8% of corporates are SMEs, which between them account for 70% of employment. In recent years, SMEs’ share of economic activity in OECD countries has increased relative to large companies. However, SMEs’ access to capital markets is restricted by informational barriers, transaction costs, and a perception of higher risk12. In developing markets, some 90% of MSMEs lack access to credit, despite the innovative small business credit models in place in many countries. Potential Credit Demand to Meet Forecast Economic Growth to 2020 The analysis of the past decade of credit growth suggests that decision-makers should be concerned with detecting pockets of excess credit growth – rather than constraining overall credit expansion. Indeed, the challenge of providing sufficient credit to segments with unmet demand warrants just as much attention as the need to avoid areas of excess. A segment-level forecast of potential future credit demand provides analytical underpinning for both these tasks. The study therefore built a regression model to forecast the global stock of loans outstanding from 2009 to 2020 – broken down by country as well as by retail, wholesale, and government segments – assuming the consensus forecast of a global economic recovery is correct. (See Sidebar, “A Model to Forecast Potential Credit Demand”.) A Model to Forecast Potential Credit Demand The model developed in this study calculates potential credit demand – the total level of demand for credit by country and segment, assuming that there is no restriction on supply over and above the restrictions in place prior to 2010. For the purposes of the model, credit stock is defined as the total outstanding lending amount in US dollars at year end. This includes both capital markets (bonds outstanding) and traditional loans. The following assumptions underpin the model’s projections: • Consensus projections are met for global economic growth – that is, nominal growth of 6.3% per annum in 20092020 13 • Macroeconomic indicators such as consumption, exports and GDP growth have the same effect on future credit supply and demand as they did in the past decade14 • A fixed US dollar exchange rate based on 2009 actual is used for all historic and projected credit stock • Interbank lending is excluded from all credit stock data to avoid double counting • Nominal credit stock values in all cases, in US dollars • No further major crises or large-scale currency debasing take place over the decade It is important to note that the model projects the annual growth of total global financial wealth (that is, credit and equity combined) to be almost one percentage point higher than GDP growth. Wealth has outstripped GDP for much of the past 15 years, and given the rapid growth of credit, equity and savings in developing markets, this trend is likely to continue in the decade ahead.
See, for example, “Discredited”, The Economist, 4 November 2010 http://www.economist.com/node/17420202?story_id=17420202&CFID=154289182&CFTO KEN=90408960 10 Basu, Priya and Pradeep, Srivastava. “Scaling up Mircrofinance for India’s Rural Poor”. World Bank Policy Research Working Paper 3646, June 2005 11 Goland, Schiff, and Stein, “Two trillion and counting: Assessing the credit gap for micro, small and medium-size enterprises in the developing world”, IFCMcKinsey & Company report, October 2010. MSMEs include formal and informal micro, very small, small and medium enterprises with less than 250 employees. This definition is broader than the one used by other reports, such as the CGAP report. 12 Seoul G20 Business Summit, “How can the small and medium enterprise sector be nurtured?” November 2010 13 The “consensus” GDP growth projection adopted for this analysis is Global Insight’s widely accepted forecast: nominal growth of 6.3% per annum between 2009 and 2020, equivalent to real GDP growth of 3.3% per annum over this period. Global Insight’s forecast is built on an aggregation of country-level forecasts. 14 The study’s forecast of potential demand therefore does not include new supply limitations that may enter the market, such as Basel III, or more expensive and limited securitization markets.
More Credit with Fewer Crises | 25
Three scenarios of future credit growth Three scenarios of credit growth for 2009-2020 were modelled using this methodology (Exhibit 5): 1. Global leverage decrease. This scenario assumes there will be modest global deleveraging, predominantly due to reduction in credit over a few years in overheated credit segments of developed markets, offset in part by continued credit growth in developing markets15. Under this scenario, overall global credit stock will grow at 5.5% per annum (0.8 percentage points below consensus GDP growth), reaching US$ 196 trillion in 2020 (Exhibit 6). Equity would need to grow almost twice as fast as GDP, bringing the ratio of credit to equity back down to its 2000 levels.
See McKinsey Global Institute, “Debt and deleveraging: The global credit bubble and its economic consequences”, January 2010
26 | More Credit with Fewer Crises
2. Global leverage increase. This scenario is based on the pure output of the credit demand model; in other words, it assumes the full potential demand for credit will be met. This is plausible if one expects rapid credit growth in developing markets, political pressure on banks to increase lending, and a trend to “de-equitize” by large companies. However, the likelihood of deleveraging in currently overheated segments militates against a global increase in leverage. Under this scenario, the world’s credit stock would grow at 6.6% per annum (0.3 percentage points faster than GDP) to reach US$ 220 trillion in 2020 (Exhibit 7). The ratio of credit to equity would reach 3.0 by 2020. 3. Flat global leverage. This scenario, which provides the primary credit growth forecast used in this report, assumes that global credit stock will grow at 6.3% per annum to 2020, tracking GDP growth. In this outlook, modest deleveraging in developed markets will be offset by credit growth in developing markets. Under this scenario, the world’s credit stock will reach US$ 213 trillion in 2020 – almost double the total in 2009 (Exhibit 8). The increased demand for credit would be split between capital markets (expected to make up US$ 39 trillion of the additional demand) and bank lending (US$ 66 trillion). The ratio of credit to equity is projected at 2.7 in 2020.
More Credit with Fewer Crises | 27
East and West, worlds apart The scenario of flat global leverage forecasts rapid credit growth in developing markets. These countries will add almost US$ 50 trillion to their credit stock by 2020, driven by rapidly expanding consumer and corporate financing (Exhibit 9). China’s credit demand will lead global credit growth: it will require US$ 20 trillion more credit in 2020 than in 2009, with 80% of that growth going to the wholesale segment. China’s 12th Five Year Plan reflects this demand, and includes plans to upgrade traditional industries, develop new strategic industries such as biotechnology and IT, and increase consumption16. (As noted above, Chinese retail credit is a currently significantly underpenetrated; steps to meet the pentup demand for consumer credit would only add to the rapid expansion of Chinese credit overall.) The rapid growth in credit in developing countries will be reflected in their growing share of lending from both capital markets and banks (Exhibits 10, 11). China alone will account for more than one quarter, or US$ 18 trillion, of the projected global increase in bank lending to 2020, with its total bank lending volumes more than quadrupling. In Africa, the Middle East, Eastern Europe and Latin America, the growth in bank lending will be almost as fast. The scenario of flat global leverage forecasts that in developed markets, including the large Western economies, most of the growth will come from the government segment. In North America alone, the value of government bonds is expected to grow by US$ 12 trillion to 2020. Even allowing for significant fiscal consolidation in some countries, developed economies will continue to run significant government deficits17. Deleveraging in overheated segments of the developed world will be significant under this scenario. For example, the leverage ratios of the US and United Kingdom retail segments will decrease by 2 and 5 percentage points respectively over the course of the decade. The leverage ratio for Spanish wholesale will fall by 13 percentage points over the same period. As a result, credit growth in these segments will trail GDP.
UBS Investment Research, China Focus, “Rebalancing, the Central Theme of the 12th Five Year Plan”, November 2010 The most conservative external sources were used for government debt leverage, taking into account announced or expected austerity measures. The projections of the following external sources were considered: Global Insight, Economist Intelligence Unit, OECD, IMF and Oxford Economics.
28 | More Credit with Fewer Crises
More Credit with Fewer Crises | 29
It is worth noting that the financing needs of infrastructure and green technology projects will form a significant part of credit demand growth. (See sidebar, “Financing Infrastructure and Green Technology”.) Financing Infrastructure and Green Technology Financing for infrastructure and green technology projects will contribute significantly to overall credit demand. The cumulative increase in global infrastructure spending between 2009 and 2020 is estimated at US$ 27 trillion to US$ 33 trillion18, funds that will be required to: • • • • Upgrade aging infrastructure in developed countries Meet the demands of urbanization and increase living standards in developing countries Build transportation infrastructure to facilitate growing international trade Achieve sustainable development goals
Although not all of this increased spending will require credit financing, it amounts to between 26% and 31% of the total credit demand increase to 2020 as projected by this study. Credit demand for infrastructure financing from both the public and private sector is likely to be substantial.
Seoul G20 Business Summit, “Closing the gap in infrastructure and natural resource funding”, November 2010
30 | More Credit with Fewer Crises
Chapter 2: Criteria and Metrics for Sustainable Credit
The experience of the past decade shows that, even when overall credit growth is at economically sustainable levels, country- and segment-level pockets can swing quickly into excess. At the same time, it is possible for key segments to remain credit-starved even in a world of credit abundance. Financial institutions, regulators and policy-makers are in search of approaches to help credit flow to the places it is needed to support economic growth – and to ensure that unsustainable credit booms are spotted and tamed before they spiral out of control. As a contribution to that quest, this chapter proposes a set of criteria to define “sustainable credit” – in which excess lending is avoided, contagion risk is transparent, and sufficient credit flows to the areas where it is needed to support growth and welfare. The chapter then introduces a suite of metrics to track credit sustainability at the segment level, raise early warnings of impending crises, and provide a fact base for interventions. These metrics include indicators to help identify both creditflooded “hotspots” and credit-starved “coldspots”; and a framework rooted in network theory to measure the credit connectivity of different segments and countries, allowing for more accurate monitoring of contagion risk. Sustainable Credit: A Challenge of Definition This study recognized it would be no easy task to come up with a definition of sustainable credit that was both specific enough to be meaningful and broadly accepted by public and private decision-makers. Before attempting a definition, therefore, the study canvassed the question – “What is a sustainable level of credit?” – in interviews with more than 50 industry CEOs, rating agency executives, central bankers, regulators and academics. Although there was a breadth of perspectives among these experts, there was also significant convergence. In particular, while a minority of interviewees believed current absolute levels of borrowing were unsustainable and would require substantial deleveraging, the general view was that the current global credit stock is at sustainable levels but should be rebalanced towards economically beneficial uses. In this view, the recent financial crisis was driven by ineffective monitoring and managing of information asymmetries, which drove misallocation of credit and excess contagion risk. Differing perspectives On some questions there was real debate. Some interviewees believed that the market mechanism was the most efficient long-term means of credit allocation. For example, Ivan Pictet, Founder and Managing Director of Pictet & Cie, argued that: “Government regulation is not the answer and will continue to distort markets as regulation lags market movement and will dampen growth and sustainability.” Others, however, argued that effective interventions in the market were needed. Neil Thomson, a Partner at private equity investment group Apax Partners, contended that: “While free markets work in theory, often asymmetric information and misalignment of incentives create distortions that need to be addressed. We see this frequently in credit markets.” There was also disagreement on which types of credit use should be prioritized in any attempt to define and promote sustainable credit: some emphasized the importance of large infrastructure projects such as mobile networks and power plants in driving development, while others believed increased lending to underserved market segments such as SMEs and low-income populations was key. There was substantial agreement, however, that transparency of information flows (for example, on borrower risk profiles), combined with the strength and capabilities of institutions (for example, at pricing risk), were crucial for a balanced allocation of credit. Many interviewees emphasized that if credit supply to underserved segments is to be increased, it will be important to address lenders’ perceived lack of profit opportunity, as well as barriers to entry such as supply chain complexity or lack of information on borrowers.
More Credit with Fewer Crises | 31
The experts interviewed provided several other valuable perspectives on what would make for sustainable credit. One was on how the contagion effects of credit could be reduced. Several people emphasized the centrality of improving the capacity of institutions (including regulators and rating agencies) to assess contagion risk – including the risk of exceptional “1 in 20” events. This was seen as particularly important, given that rational individual behaviour can be collectively foolish: in the recent crisis, firms were often decreasing their own risk but increasing total system exposure19. Objectives for sustainable credit The perspectives thus canvassed led the study to define a set of basic, overarching objectives for sustainable credit. First and foremost, credit should enable earlier investment and consumption than would otherwise have been achievable. In doing so, it should support growth – of the economy overall, or of earnings. Implicit in this definition is the assumption that future income growth will allow repayment; that there is confidence between the counterparties; and that the credit transaction enables both the creditor and the debtor to make a return, be it an economic return or welfare. Beyond these basic requirements, sustainable credit should also meet several mutually reinforcing economic and social objectives, including: • An acceptable level of confidence in repayment of the credit, both its interest and principal (that is, lending with acceptable risk criteria) • Credit-driven volatility and risk transfer should be set at an acceptable level; there should be transparency on where such risk transfer occurs; and the frequency and amplitude of credit-related crises should be kept at acceptably low levels • Credit allocation should be focused towards societies’ priorities, including delivering on economic development goals • Not permitting major shifting of credit burdens across generations Within these broad parameters, though, sustainable credit levels will differ significantly between countries and sectors, depending on a wide range of factors including their economic development profiles and the depth of their financial systems. The expert discussions emphasized, therefore, that the question of sustainable credit needs to be considered not just at the global level but also at a country or regional level and for each major sector within those countries – in particular, retail, wholesale and government borrowing. As Denis Bugrov, Chief Strategy Officer at Sberbank, pointed out: “It is difficult to generalize sustainability at a global level when disparities are different across markets and the economic environment is key to lending”. Moreover, interviewees pointed out that the definition of sustainable credit may vary over the course of the business cycle. Clear Criteria and Robust Tools for Sustainable Credit Consideration of these perspectives led the study to define four criteria for sustainable credit: 1. 2. 3. 4. Limited “hotspots”, or areas of excess credit where repayment and servicing prospects are at risk Transparent and manageable contagion risk in order to reduce system volatility Limited “coldspots”, or segments where growth is inhibited by a lack of access to credit Alignment with social goals, to ensure that both economic and welfare needs are met
Based on these criteria for sustainable credit, this study proposes a suite of new tools to measure and monitor credit levels, including: • A scorecard of indicators, or “rules of thumb”, to measure retail, government and wholesale credit and identify credit hotspots • A framework rooted in network theory with metrics to track credit connectivity risk • Heatmaps to track credit hotspots and indicate coldspots by segment and country These tools complement the Early Warning Exercise methodology developed by the Financial Stability Board (FSB) and the International Monetary Fund (IMF). (See sidebar, “In-depth Credit Metrics that Complement the Early Warning Exercise”.)
Prior to the recent crisis, many financial institutions had comparable risk exposures, based on similar funding structures, risk-management practices, and mitigation strategies. This led financial institutions to believe that their institutionally held risk was acceptable, while in actual fact systemic risk was building. When housing prices collapsed and subsequently the financial assets linked to mortgages, all financial institutions looked to close similar positions in a, at that time, very illiquid market, ultimately leading to extreme losses.
32 | More Credit with Fewer Crises
In-Depth Credit Metrics that Complement the Early Warning Exercise One of the G20’s first reactions to the financial crisis in November 2008 was to task the FSB and IMF to conduct an early warning exercise (EWE) to identify potential risks within the financial system, with the results to be shared confidentially with the International Monetary and Financial Committee annually. Recently, the IMF published a response to this request, with a detailed methodology to identify different types of risk20. This study endorses the EWE approach of developing metrics to help flag potential concerns in the economy. The EWE methodology assesses many different types of risk, including overall financial market volatility, asset bubbles and stock valuation. The metrics and tools presented in this report, however, focus solely and in detail on credit risk, looking at stock and flow measures as well as their derivatives (average annual change). This report’s in-depth global perspective on credit is complementary to the EWE, providing institutions and policy-makers with credit insights to inform their actions. The remainder of this chapter shows what each of the four sustainable credit criteria mean in practice, and profiles the tools proposed to measure and monitor credit against those criteria. 1. Limited hotspots For every country and credit segment (for example, US retail mortgages), an upper bound defines “hotspot” levels above which there is a material chance of a credit-related crisis. These hotspots can be caused by an over-exuberance of either demand or supply, or the interplay of both. Two factors determine these upper bounds: • Credible principal repayment prospects. The upper bound in this case is the maximum credit burden above which the credibility of principal repayment diminishes rapidly. It is measured in two ways: through current leverage ratios, and through the rate of change of leverage ratios. • Credible servicing prospects. The upper bound in this case is driven by income prospects to service interest payments and repayment of principal. It is measured through the annual repayment burden (interest and principal) – both the current credit repayment burden and the rate of change of the repayment burden. To determine warning indicators for hotspots, the study looked at the history of credit crises over the past 50 years, and quantitatively analysed 15 credit crises in detail, ranging from the 1985 Korean crisis to the recent government debt crises of 2009 (Exhibit 12). These crises were selected on the basis of three criteria: their scale was significant; they were rooted in unsustainable credit within the wholesale, retail or government segment (rather than in a general macroeconomic imbalance); and sufficient data was available. For each crisis, historic sustainability indicators were calculated (in line with the above definition of sustainable credit), both for the year of the crisis and for the eight years leading up to it. From these results, quantitative “rules of thumb” were derived for a range of defined sustainability indicators21.
International Monetary Fund, “The IMF-FSB Early Warning Exercise: Design and Methodological Toolkit”, September 2010 This study acknowledges the caution that has been expressed about such upper bounds: Alan Greenspan, for example, has argued that “rule of thumb” limits on credit, where the regulator hopes to anticipate a credit crisis, may be ill-advised. (“The Crisis”, Alan Greenspan, April 15, 2010. Paper written for the Brookings Institution.) However, it is submitted that the approach set out here to define and use rules of thumb is sufficiently nuanced to prevent its crude application.
More Credit with Fewer Crises | 33
Tools and Metrics: “Rules Of Thumb” to Identify Credit Hotspots This study has developed 12 rules of thumb22 to identify potential credit hotspots, covering wholesale and retail credit as well as government debt (Exhibit 13). These indicators examine both the stock and flow measures of credit (in order to measure both debt repayment and servicing prospects). They also examine the forward-looking derivatives of these two metrics (the forecasted average yearly change over the next five years); the static indicator as well as the derivative are critical for detecting possible pockets of unsustainable credit growth. Together, the rules of thumb can be used as a comprehensive “dashboard” of early warning indictors of potential credit hotspots.
Stock measures, which are often leverage ratios – retail, wholesale or government liabilities as a percentage of GDP, are the more commonly used indicators of indebtedness. Yet, flow and rate-of-change measures are just as important in detecting unsustainable credit patterns.
One of the rules of thumb – outstanding government credit exceeding 90% of GDP – has been proposed by Reinhart and Rogoff to denote the point at which government debt becomes a brake on economic growth.
34 | More Credit with Fewer Crises
As an example, consider the flow measure proposed here for retail credit. The rule of thumb is that a hotspot may be forming when the interest burden plus the average principal repayment on retail credit exceeds 20% of GDP. In Ireland, for instance, the interest burden plus average principal payment on retail credit was already at 21% of GDP in 2006, an early indicator of the potential unsustainability of its mortgage bubble; it reached to 25% in 2008 as the country approached crisis. The rate-of-change measures track the growth of both stock and flow. In the wholesale segment, for example, one rule of thumb is that credit growth is potentially unsustainable if wholesale credit as a proportion of GDP is projected to grow faster than 2 percentage points a year over the next five years. Likewise, wholesale interest payments plus average principal payment as a proportion of GDP could be unsustainable if they are forecast to grow faster than 2 percentage points a year over the next five years. In Spain in 2006, for instance, the projected five-year growth for wholesale credit as a proportion of GDP was 6 percentage points a year, an early indicator of the potential unsustainability of its commercial property development boom. Indeed, applying these rules of thumb to the growth of credit in key segments over the past decade shows that these indicators would have sounded early warnings about unsustainable credit growth in several countries that have recently faced credit crises (Exhibit 14). These warnings could have prompted more rigorous analysis of these hotspots, using a broader set of metrics.
Finally, it should be emphasized that these rules of thumb should be applied as early warning indicator thresholds; levels of credit in excess of the thresholds should trigger more detailed review. It is not appropriate to apply an absolute limit for credit sustainability, given the diversity in countries’ economic profiles. 2. Transparent and manageable credit contagion risk Beyond ensuring the economic sustainability of individual credit segments – in other words, local sustainability – the domino effects of credit crises must be understood and managed to ensure sustainability at the global level. There have been a range of attempts to define and assess contagion, typically focusing on implications for the real economy23. Having reviewed this literature, the study focuses on the connectivity between individual credit segments and countries. “Credit contagion risk”, as defined here, measures the likelihood and magnitude of the contagion that would arise from a hypothetical future credit crisis in a particular segment in a given country.
See, for example, Kaminsky, Reinhart, and Vegh, “The unholy trinity of financial contagion”, October 2003; IIF, “Systemic Risk and Systemically Important Firms”, May 2010; and Andrew Haldane, “Rethinking the Financial Network”, April 2009
More Credit with Fewer Crises | 35
To quantify the potential contagion risk of each credit segment in a particular country, two metrics are therefore required: • A measure of the local credit sustainability of individual segments within countries; the rules of thumb described above fulfil this role • A measure of the connectivity between individual segments in different countries; the approach for doing this is set out below To measure the connectivity between segments, this study proposes a credit connectivity framework based on the five principles of network design: i. Transparency and monitoring. Credit segments are assessed based on the extent and type of their linkages with other countries and credit segments, and on the extent to which this “network map” is communicated with market players and the public. Regulatory quality is a useful approximation for transparency and monitoring24. ii. Interconnection and modularity – a measure of the number and size of a segment’s connections to other credit segments. Interconnection in credit markets can be approximated by trade measures that spread credit crises directly. Import penetration is the best measure of connection, which transmits shocks from the originating crisis directly onto exporting countries. iii. Fault tolerance. This measure tests the extent to which a credit segment relies on external funding sources; the ability to isolate a single credit segment; and the availability of alternative credit sources in the event of weakness. For example, foreign capital’s share of a segment’s financial system – that is, its capital markets and bank lending volumes – shows the extent to which isolating that segment would be difficult. iv. Dimensioning and scalability – the ability to increase the size of the credit segment without increasing risk, by scaling up local funding or credit supply, or by ensuring that excess funding lines (current and future) are above required levels. This measure is approximated via the credit segment’s absolute size. v. Intervention – the extent of local or regional support and strength of contingency plans that is in place in case of credit segment weakness. This can depend, for example, on regulatory quality. Together with the rules of thumb set out above, which gauge the local sustainability of a given credit segment, this credit connectivity framework will allow decision-makers to quantify that segment’s level of contagion risk to other segments and countries. The results of such an exercise are set out in the next chapter. In this study’s view, the risk of contagion from one segment or country to another should be considered acceptable only if: • The economic benefit arising from the connectivity between the two segments (for example, through trade agreements) outweighs the contagion’s likely effect on the impacted segment’s local credit sustainability, and • In the event of weakness in any segment’s local credit sustainability, that segment can be isolated sufficiently well to minimize the economic cost of contagion to other segments and countries This approach is supported by previous work that has sought to understand connectivity and contagion risk in fields outside of financial services – including fisheries, forest fires, and immunology. (See sidebar, “Insights on Contagion from Other Disciplines”.) The credit connectivity framework proposed here can be used to assess a particular country’s overall level of connectivity to the global financial system, but also to map the country-to-country credit connectivity of individual nations – in other words, the extent to which any one country would be exposed to a credit crisis in any other country. There is potential to extend this model at a more granular level to create a global credit connectivity monitor covering a wide range of countries. The monitor would measure the connectivity of each country’s credit segments with segments in all or most other countries. The monitor would thus identify areas of concern and flag economic and credit concentration risks, allowing policy-makers to make informed trade-offs as to acceptable levels of contagion risk. The indicators used to calculate credit connectivity are set out in greater detail in the Technical Appendix.
Regulatory quality captures perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development. See Kaufmann, Daniel, Kraay, Aart and Mastruzzi, Massimo, “The Worldwide Governance Indicators: Methodology and Analytical Issues” (September 2010). World Bank Policy Research Working Paper No. 5430.
36 | More Credit with Fewer Crises
Insights on Contagion from Other Disciplines In addition to the use made of network theory in this report to illustrate and measure contagion risk, many approaches from other disciplines have been suggested to understand and mitigate the threat of contagion in the financial services industry25. Fisheries: The example of Chilean salmon has been used to emphasize the importance of diversity to avoid similar vulnerabilities. These salmon were subjected to similar treatments to avoid diseases. As a result they all developed the same vulnerabilities and were wiped out in large numbers by a single virus in 2008. This example underlines the value of contrarian approaches promoting diverse regulation to avoid symmetric vulnerability in the financial sector. Forest fires: The use of controlled fires to rejuvenate forests demonstrates that fires that wipe out parts of a forest can actually be a positive property of an ecosystem, as long as they are controlled and do not spread beyond the area that needs rejuvenation. This example is quoted as a rationale for introducing independent resolution frameworks for systemically important financial institutions (“living wills”) to allow bankruptcies of “unhealthy” banks without impacting the system. Immunology: The human body uses white blood cells as part of a pre-emptive system to avoid sickness to seek and destroy dangerous pathogens before they can spread. This example illustrates the opportunity for financial institutions to establish a pre-emptive system to screen for threats and raise warning signals early. 3. Limited coldspots “Coldspots” are segments where growth is inhibited by a lack of access to credit. Coldspots could be caused by market inefficiencies such as underdeveloped financial markets, a lack of transparency and weak institutions. Conversely, there is evidence that the existence of adequate credit supply increases GDP growth through a range of transmission mechanisms, including increased entrepreneurship and increased access to schooling26. Brown University economics professor Ross Levine, interviewed for this study, emphasized that: “Credit supply and financial development can increase economic growth by adding liquidity and diversification to the market and by generating a return from credit flowing to the best investment.” Experts assert that there are “structural coldspots” in developing countries based on structural lack of access to credit and a wider lack of financial development27. As the next chapter shows, potential credit demand will grow sharply over the coming decade in many countries with relatively low levels of financial development, pointing to the possibility that such coldspots could be a significant problem in the years ahead. One marker of a country’s financial development is the proportion of the population with bank accounts; in many of the fastest growing credit markets, very large numbers of people remain “unbanked”, creating a serious structural barrier to meeting credit demand. Additionally, a lack of sophistication in capital markets coldspots can results in coldspots. Denis Burgov of Sberbank cited Russia as an example: “A general lack of understanding of risk leads to a binary view of good versus bad credit risk, so pricing and lending can be too strict.” Inter-country comparisons provide a further clue to the existence of credit coldspots. Where a country falls in the lowest quintile of credit penetration among countries in a similar development stage, this may be a first indicator of possible coldspots.
See “Rethinking Risk Management in Financial Services – Practices from other domains”, World Economic Forum, April 2010 World Bank Policy Research Working Paper, Asli Demirguc-Kunt and Ross Levine: Finance, Financial Sector Policies and Long-Run Growth, 2008 27 International Monetary Fund Working Paper, Enrique Gelbard and Sergio Leite: Measuring Financial Development in Sub-Saharan Africa, 1999
More Credit with Fewer Crises | 37
Tools and Metrics: Heatmaps to Track Credit Hotspots and Coldspots by Segment This study proposes a credit “heatmap” to monitor both hotspots and coldspots across countries and segments. Hotspots are those segments that fall above the rule of thumb on any of the four criteria, while coldspots are identified by grouping the countries according to their development stage and looking towards the lowest quintile of credit stock. Additional metrics to help define coldspots include the size of the unbanked population and the size and depth of capital markets; these are discussed in the next chapter. Exhibit 15 shows an example of such a heatmap for 2006. The next chapter discusses how credit heatmaps can be used in today’s context.
38 | More Credit with Fewer Crises
4. Alignment with social goals In the long run, the scale and distribution of credit is only economically sustainable if it also meets society’s broader social objectives. Credit is linked to social objectives during all stages of a country’s economic development. In early stages of development, credit is used to support family-owned businesses; next, it supports small and large corporations; and finally it is used to smooth consumption. Muhammad Yunus, founder of Grameen Bank, goes so far as to say that credit is a human right, and adds:
“If we are looking for one single action which will enable the poor to overcome their poverty, I would focus on credit.”28 To illustrate how smart allocation of credit can drive economic development goals, consider the example of the Wirtschaftswunder (“economic miracle”) of post-war West Germany. Some US$ 1.6 billion of credit issued under the European Recovery Plan was steered towards export-focused industries such as iron and steel, and to essential infrastructure such as energy and transportation. As growth picked up, credit was steadily deployed to new sectors such as manufacturing, as well as to housing. Strong institutions and policies – such as high depreciation allowances and other tax concessions for investment – underpinned West Germany’s rapid economic growth during this period. This study’s analysis shows that the credit mix and the level of leverage in any given country are correlated strongly with its level of economic development (Exhibit 16). Countries at an early stage of development are characterized by very low levels of leverage and a credit mix skewed towards the retail sector. As GDP rises, so leverage levels rise too, and wholesale credit plays a much greater role as industry expands. Wealthy countries with sophisticated financial systems see much higher
Exhibit 16 : Credit stock levels during economic development
Credit penetration by segment and in groups of countries with similar overall penetration level, Credit / GDP, Percent 2009
GDP per capita, USD 000 120 Percent of total credit penetration by segment Government Wholesale Retail Low growth countries 1 Medium growth countries 1 High growth countries 1
100% = 247 100% = 219 30 32
100% = 298 46
100% = 182
100% = 145
100% = 97
Denmark Australia Finland Germany Sweden US France Singapore New Zealand Israel Hong Kong Portugal Austria Canada Ireland Netherlands Belgium UK Italy Spain Greece Japan
100% = 55 100% = 29
Taiwan Croatia Czech Latvia Turkey Hungary Estonia Lithuania Poland Brazil Russia Mexico Malaysia Argentina Romania Bulgaria South Africa Thailand Bosnia Tunisia Colombia Egypt Ukraine India Morocco Belarus Peru Indonesia China 0 Nigeria Kenya Philippines 120 0 40 80 160 Macedonia Vietnam Pakistan Slovakia
Leverage, total credit / GDP, Percent
1 High growth = Real GDP CAGR > 4.5% from 2000 -2009, Medium growth = Real GDP CAGR > 3% from 2000 -2009, Low growth = Real GDP CAGR < 3% from 2000 -2009
Yunus, Muhammad. Keynote address delivered at the 85th Rotary International Convention held in Taipei, Taiwan, 12 June 1994
More Credit with Fewer Crises | 39
levels of leverage, and substantial credit penetration in all segments – retail, wholesale and government. However, once leverage exceeds 250% of GDP, the correlation with development begins to fall away and government accounts for the lion’s share of lending.
Policies, incentives and mechanisms to direct credit to meet social goals should be carefully designed to minimize economic distortions and avoid unintended consequences. For example, regulatory mechanisms intended to make the system safer may have the unintended consequences of forced subsidiarization, trapping liquidity and limiting trade finance – ultimately lowering output and slowing economic growth. Finally, it should be emphasized that, if credit is to drive consistent economic development, a set of basic legal, institutional and policy foundations must be in place. Without this underpinning, there is every likelihood that credit will be squandered on economically and socially unproductive uses. There are three fundamental requirements: • Strong laws and legal enforcement. Contracts should have a creditable legal backing to support the belief of repayment; the bankruptcy resolution process for failed ventures should be orderly, efficient and predictable; and a system of property rights should be in place to ensure credit is backed by collateral, assets are protected and there is a system of ownership. The growth stories of Brazil and Chile in recent decades illustrate the value of a strong legal structure: both countries benefited from stable legal and tax frameworks for long-term investment projects. • Creditable and functioning institutions. The economy should have relatively sophisticated private financial institutions and banks; an independent central bank should be able to make decisions on currency and interest rates without constraint of political pressure; and institutions and organizations should suffer from minimal fraud and corruption. Peru’s stunted economic development during the 1980s and 1990s illustrates the high cost of weak institutions and endemic corruption: an estimated 15% of gross income was paid out in bribes29. • Macroeconomic and fiscal discipline. Government should support a healthy macroeconomic environment; and government should save money when revenues are growing. Numerous crises in recent history demonstrate the importance of this tenet.
Peruvian economist Hernando de Soto points to lack of institutions and laws that protect property and create capital as a factor limiting economic growth and the under penetration of capital. Additionally, the scarcity of creditable institutions causes a drag on the productive economy, as in Peru where widespread corruption caused a large extra-legal manufacturing sector.
40 | More Credit with Fewer Crises
Chapter 3: Making Sustainable Credit a Reality: Challenges Ahead
How, then, does one apply the sustainable credit criteria and tools discussed above? This chapter uses the methodology set out in this report to highlight the major challenges that stand in the way of achieving sustainable, productive credit growth in the decade ahead. In particular, it addresses three key questions that will determine whether the world’s growing demand for credit can be met sustainably: • Are we at risk of future crises – and if so, where? • What is the future risk of contagion? • Will credit growth be sufficient to meet demand? Are We at Risk of Future Crises – And if so, Where? To answer this question, the study applied the rules of thumb set out in the previous chapter to projected credit demand over the coming decade. The answer is unequivocal: even though some economies will deleverage over this period, the analysis reveals a potential new set of credit hotspots. The heatmap tool introduced in Chapter 2, when applied across nations and segments for 2009, highlights a range of countries in which there are potential credit hotspots (Exhibit 17).
More Credit with Fewer Crises | 41
Credit hotspots on the horizon This study projects a significant number of credit hotspots across the world in 2020 – whether the scenario modelled is one of increased, decreased or flat global leverage (Exhibit 18). A country is identified as potential hotspot if it violates any one of the four rules of thumb for the segment, as described in Chapter 2. The sustainability of retail credit could be a global problem in 2020: in countries representing almost half of global GDP, retail credit levels are expected to be above one or more of the rules of thumb. By contrast, government credit hotspots are projected for a much smaller set of countries, between them representing 13-14% of world GDP. At a regional level, however, Western Europe will remain vulnerable to unsustainable government borrowing: countries representing 31% of the region’s GDP are expected to exceed one or more of the rules of thumb for government credit. In wholesale credit, Asia is expected to be the main driver of hotspots in 2020, with countries representing more than half of the region’s GDP projected to be above a rule of thumb for wholesale borrowing. Western Europe also appears at risk of unsustainable credit levels in the wholesale segment.
As discussed in the previous chapter, the rules of thumb are indicators that should spark a deeper investigation into the potential for a particular segment to become a credit hotspot. Some countries and segments are better able to sustain high levels of credit than others. One determining factor is a country’s competitiveness. As Exhibit 19 suggests, more competitive countries are able to sustain larger ratios of credit to GDP, as competitiveness facilitates higher growth for the future. If highly indebted but less competitive countries are to improve their sustainability, they will need to increase their competitiveness or decrease leverage, or both.
42 | More Credit with Fewer Crises
Mechanisms to limit hotspots Lessons from the past help identify mechanisms to address hotspots in the retail, wholesale and government sectors. Canada’s resilience during the 2007-08 financial crisis is an instructive case. Experts attribute this in part to the stable structure of the banking system, which is dominated by five large, heavily regulated banks that operate under very tight risk controls. For example, the governor of the Bank of Canada ruled in 2008 that bank-owned mortgages would be eligible for insurance only if they had a loan-to-value ratio around or below 70%. As a consequence, banks were very strict with their lending. One key factor that drives unsustainable borrowing is a lack of financial understanding among consumers (and often among wholesale and government borrowers as well). According to a 2007 survey of US credit card holders, over half of the respondents said they had learned “not too much” or “nothing at all” about finance at school30. In Slovakia, a World Bank report noted that many consumers borrowed from unregistered consumer credit companies at effective annual interest rates of 120-229% when bank loans at a rate of 14% were available; while in Croatia, many co-signers of loans did not realize they would be responsible for paying the debt of a friend or family member in the case of default31. Financial education by governments and financial institutions is therefore a further intervention to mitigate against consumers taking on credit they are unable to repay. The next chapter sets out specific recommendations for financial institutions, regulators and policy-makers to detect and limit future hotspots.
Niall Ferguson, The Ascent of Money: A Financial History of the World (Penguin, 2009) Rutledge, Susan. “Consumer Protection and Financial Literacy: Lessons from Nine Country Studies”, The World Bank, February 2010
More Credit with Fewer Crises | 43
What Is the Future Risk of Contagion? Once hotspots have been identified in particular segments of individual countries, one must ask: which of those hotspots have high levels of connectivity with other segments, and thus pose a serious contagion risk? To answer this question, the study mapped countries’ local credit sustainability against their degree of credit connectivity in 2009. Exhibit 20 plots the results of this analysis: negative values for credit sustainability risk or credit connectivity risk indicate that a country is less risky than average, while positive values indicate it is riskier than average. (This exhibit provides a snapshot of countries’ contagion risk at the end of 2009; to track the evolution of contagion risk, the tools profiled here should be used on an ongoing basis.) This analysis generated some stark findings. Significant economies – including the United Kingdom, the US and Japan – pose a potential threat in terms of local sustainability or credit connectivity risk, or both. US credit levels are well documented: its federal deficit reached almost 10% of GDP in 2009, while total US consumer indebtedness in 2010 exceeded US$ 11 trillion, equivalent to almost US$ 50,000 per person32. These levels of absolute credit stock, combined with extensive links to other major credit segments, drive considerable contagion risk. Japan, on the other hand, appears locally less sustainable but with low credit connectivity risk.
This analysis also shows that some higher growth credit segments, such as South Africa, appear underpenetrated and relatively safe. China’s credit contagion risk also appears relatively low, driven by low foreign credit and import penetration, along with relatively low local sustainability risk. It should be emphasized, though, that the contagion risk framework presented in this report focuses on the first order effects of a credit crisis; China may be exposed on non-credit
Federal Reserve Bank of New York. “Quarterly Report on Household Debt and Credit”, November 2010
44 | More Credit with Fewer Crises
dimensions such as the rate of asset price appreciation. A significant credit crisis in China would in all likelihood have enormous knock-on effects on the world economy through connectivity beyond credit channels, for example by triggering sudden shifts in the balance of trade and gaps in supply chains, or by destabilizing the market environment. While these effects may be part of a broader notion of contagion, this analysis aims to underscore the effects of credit contagion in particular. Applying the analysis to the Eurozone produced some important insights. The Eurozone as a whole appears stable, but a nuanced view of individual countries reveals potential contagion risk obscured by the aggregate picture (Exhibit 21). Ireland, Portugal and Greece pose the highest threat to the Eurozone’s stability, while Italy and Spain appear less risky due to their relatively lower interconnectedness. The size effect of Italy and Spain, however, must not be underestimated. While their direct connectivity is relatively low due to low foreign asset penetration, second order effects from their size, the market environment and other factors may have broader economic consequences, not represented by this credit connectivity analysis.
The network framework, set out in the previous chapter, can help decision-makers identify appropriate mechanisms for reducing connectivity risk – including monitoring cross-border capital flows to ensure transparency, and introducing contagion risk insurance products to strengthen fault tolerance. Specific mechanisms to limit contagion risk are proposed in the next chapter. Will Credit Growth Be Sufficient to Meet Demand? As Chapter 1 made clear, the fastest growing demand for credit is shifting from developed to developing economies where structural gaps in credit supply, including underdeveloped financial markets, could result in more coldspots. Rapid growth of both capital markets and bank lending will be required to meet this demand – and it is not assured that either has the required capacity.
More Credit with Fewer Crises | 45
The challenge of meeting new credit demand will be exacerbated by the current freeze – and expected slow recovery – of securitization markets in the wake of the recent global financial crisis. Moreover, large-scale, cross-border financing will be needed if credit demand is to be met. (For example, Asian savers will continue to fund Western consumers.) “Financial protectionism” could hamper these flows. In short, there will be significant challenges in channelling credit to where it is needed. Low levels of financial development in countries with rapid credit demand growth To identify segments where there may be significant credit coldspots in 2020, the study mapped countries’ projected credit demand growth against their Financial Access Score – a measure that forms part of the World Economic Forum’s Financial Development Index33. This analysis shows that future coldspots may result from the fact that the highest expected credit demand growth is among countries with relatively low levels of financial access – including Argentina, Brazil, China, India, Indonesia, Mexico, Pakistan, Russia, South Africa, Thailand, Turkey, Ukraine and Vietnam. In many of these countries, a high proportion of the population is unbanked; worldwide, there are currently 2.5 billion adults without bank accounts. Integrating this population into the banking system will be a significant challenge. As noted in Chapter 1, increased demand for credit will be split between capital markets and bank lending, with capital markets expected to fill US$ 39 trillion of the additional demand to 2020, and bank lending a further US$ 66 trillion. The distribution of this new demand reveals that the highest growth in credit demand will be in countries with currently less developed capital markets, and low levels of banking penetration (Exhibits 22, 23). By implication, there is significant risk of structural credit coldspots appearing in some of these countries.
Total index calculated for 55 countries based on scores in “Factors, Policies and Institutions”, “Financial Intermediation” and “Capital Availability and Access”. For access metric: Min score: 2.14 (Venezuela) Max. score: 5,19 (Australia), Median: 3.44. Not available for Portugal and Greece.
46 | More Credit with Fewer Crises
Challenges in meeting new demand for bank lending By 2020, some US$ 27.7 trillion of new bank lending will be required in Asia excluding Japan (a 265% increase from 2009 lending volumes) – US$ 18.7 trillion of it in China alone. The 27 EU countries will require US$ 13 trillion in new bank lending over this period, and the US US$ 9.8 trillion. Will banks have the capacity to meet this massive new demand? Increased bank lending will increase the size of banks’ balance sheets and regulators are imposing additional capital requirements on both new and existing assets. These two effects together will create an additional global capital requirement of around US$ 9-9.5 trillion, with different effects in each region (Exhibit 24). While large parts of this additional requirement can be satisfied by retained earnings, a significant capital gap in the system will remain – which will make meeting increased demand for bank lending more challenging.
More Credit with Fewer Crises | 47
It appears that banks in the US will face the least difficulty in meeting the additional demand for bank lending – assuming US securitization markets for mortgage lending continue to fund future mortgage originations at similar rates as they do currently. (Between 65% and 90% of mortgage originations are financed through securitization, shifting US$ 6 trillion of increased bank lending off balance sheet). Under the assumption that the Basel III reforms will increase target tier 1 capital ratios to 11% in the industry, US banks would need to grow their tier 1 capital by US$ 1.2-1.6 trillion by 202034. The analysis projects that the US banking industry as a whole would be able to raise a large part of this additional capital thanks to approximately US$ 1.0-1.1 trillion in forecasted retained earnings, assuming a dividend payout ratio of 20.6% after 2012, in line with historical averages. However, if securitization levels dropped below 60-65% of mortgages, the US banking system would face an additional capital shortfall. Maintaining this rate of securitization may pose a challenge in itself. The challenge of holding sufficient capital appears tougher for European banks, which will need to provide US$ 13 trillion in additional bank lending to meet demand between 2009 and 2020. In the EU, limited use of securitization – covering only 10% of residential mortgages in 200735– prevents banks from moving lending off balance sheet, leading to a wider impact of increased capital requirements. Even if they channelled retained earnings towards funding the capital shortfall (assuming a continued dividend payout ratio of 50%, in line with historical averages), banks would only be able to cover just above US$ 1.7 trillion of the US$ 3.8 trillion in forecasted additional capital. This leaves a capital shortfall of US$ 2.1 trillion, needed to meet additional credit demand while providing sufficient capital against current assets. For Asia, a more preliminary top-down calculation shows that most of the capital increase to fund existing assets and US$ 28 trillion in extra bank lending should be covered by retained earnings on aggregate. Assuming a 20% increase in equity to lending ratios, Asian banks would need to increase their capital base within 10 years by about US$ 4 trillion – assuming the absence of securitization – of which US$ 2.4 trillion would be required for China and US$ 1.6 trillion for the rest of Asia excluding Japan36. The study’s initial estimates based on trend-line forecasts suggest that Asian banks will require US$ 300 billion beyond full retained earnings of US$ 3.7 trillion, assuming dividend payout rates in line with individual countries’ historical values (for example, 23% for China). It must be noted that this is no more than a high-level estimate. A 20% increase in the equity to lending ratio may seem very conservative, given relatively high capital levels held by Asian banks and the low share of securities held on their balance sheets. However, a lack of insight into the quality of Asian banks’ current capital makes it difficult to estimate the required capital increase to meet Basel III requirements. It should be emphasized that these figures do not mean banks will necessarily have to fill the full gaps identified above solely through raising new capital. Given the assumptions laid out above, there is expected to be a capital shortfall in the system that may constrain banks’ ability to supply sufficient amounts of credit to meet growing demand. However, there are many levers that can be used to bridge this gap and so provide sufficient credit to meet demand. In addition to an increase in bank capital, the gap can be filled through securitization, other innovative and traditional capital market products, different regulatory requirements, or a combination of these levers. In Europe in particular, greater development and innovation in capital markets could have a profound impact in closing the capital gap; in 2009 only 47% of European credit was mediated through capital markets, against 69% in the US37. The need to revitalize securitization markets Securitization, a key mechanism that helped credit grow over the last decade, has been profoundly damaged. As a result of lax lending standards and the sub-prime crisis, new securitization issuances in the US and United Kingdom – which represent some 80% of global securitization volumes – fell by between 35% and 85% from 2007 to 200838. In the United Kingdom, and in the US asset-backed securities market, volumes have since remained low. Only in the US mortgagebacked securities market has the decline been limited – and then only because of the sharp increase in government’s role in the market. (Conforming loans that qualify for government and agency backed securitization made up 99% of all newly issued US securitization in 2009, up from 64% in 200739.) Two significant challenges must be met if securitization markets are to be revived. The first is to restore demand for the segments driven primarily by private label securitization by rebuilding investors’ confidence. The barrier is the current (very low) risk appetite of investors and lack of transparency about risks. The current regulatory reform proposals include, quite
Analysis assumes trading book assets grow in line with forecast growth in banking book assets 11% target tier 1 ratio includes minimum Basel III requirements and conservation buffers as well as an additional cushion banks will hold as an additional buffer, also to absorb any additional regulation (e.g. “too big to fail” surcharges, countercyclical buffers, national discretion). 35 American Securitization Forum, SIFMA, Australian Securitisation Forum, European Securitisation Forum, “Restoring Confidence in the Securitization Markets”, 3 December 2008 36 The high-level analysis for Asia focuses only on bank lending. The addition of trading book assets could widen the shortfall. 37 The main drivers of the relative underdevelopment of European capital markets appear to be wholesale credit (32% of wholesale credit mediated through capital markets in 2009, versus 67% in the US) and especially retail credit, as a result of lower securitization volumes in Europe. (In 2007, before the crisis, just 13% of Europe’s retail credit was mediated through capital markets, versus 49% in the US.) 38 Depending on segments, definitions and data sources. The broadest definition (Inside Mortgage Finance) still found a gap of 35% in mortgage backed securities. Dealogic and Thomson Reuters, which track only the face value (as opposed to the collateral) of non-retained tranches (as opposed to tranches that are not resold by the issuer) of securitization, find much larger gaps. 39 Inside Mortgage Finance, using a broad definition of securitization (See footnote nr. 39).
48 | More Credit with Fewer Crises
rightly, a requirement that issuing banks hold a portion of the risk on their balance sheets; while they align incentives and increase transparency, these proposals leave little room for a return to the scale of the previous “originate to distribute” model. This will dampen any recovery of securitization levels and could result in emerging coldspots in areas that previously had access to credit – though in some cases too much – such as US sub-prime. The second challenge is to maintain momentum in the segments where demand relies increasingly on government and agency backed securities. Current debate around the US government’s role in the securitization market and the possible restructuring of the government-sponsored enterprises (GSEs) such as Fannie Mae and Freddie Mac paints an uncertain picture about the future role of governments in securitization. Given the current reliance of US securitization markets on government guarantees, it is essential that any exit is planned in a way that will fill the void with a functional market and avoid halting the supply of credit. Without a revitalization of securitization markets, it is doubtful that forecast credit demand can be met in some key markets. There is certainly potential for securitization to recover. In a 2009 McKinsey survey, market participants expected the securitization market to return to around 50% of its pre-crisis volume within three years40. At the same time, the survey respondents were emphatic that the market was in need of drastic improvement in some key respects, including increased price transparency, better data on collateral pools, and better quality ratings. As one US issuer warned: “Restoring confidence in the credit rating agencies will take a long, long time.” The importance of cross-border financing This study’s forecasts for credit demand over the next decade underline how individual countries will need access to international sources of credit. One enduring hallmark of the global financial system is that Asian savers will continue to fund Western consumers and governments. Under a scenario of flat global leverage, China and Japan will have large net funding surpluses in 2020 (of US$ 8.5 trillion and US$ 5.7 trillion, respectively), while the US and other Western countries will have significant funding gaps (Exhibit 25).
Survey of 250 global market participants (e.g. issuers, dealers, investors, rating agencies, lawyers, accountants)
More Credit with Fewer Crises | 49
The study modelled several different assumptions about savings rates over the coming decade – for example, China or Japan reducing savings by 50% – and found in all cases that a need for substantial cross-border financing would remain. The obvious implication is that financial systems must remain global in order for economies to obtain required refinancing. Decision-makers will need to guard against “financial protectionism”, which has the potential to lock up liquidity and stifle growth. In this regard, it is worth noting that, although conventional theory holds that mature economies should have higher surpluses that can be used to promote global growth by investing in the emerging economies, the opposite is the case today. Many mature economies, notably the US, are running sustained deficits funded by high-growth developing economies such as China.
50 | More Credit with Fewer Crises
Chapter 4: Recommendations for Decision-Makers
What, then, are the actions that financial institutions, regulators and policy-makers can take today to ensure sustainable credit levels for the future? Based on the analysis and toolkit presented in this report, this chapter proposes a set of core principles to guide interventions, details eight specific recommendations for interventions these stakeholders can make to achieve sustainable credit over the decade ahead, and identifies which stakeholders should play the lead implementation role for each recommendation. Core Principles for Action This study proposes five principles upon which action for sustainable credit should be based: • Preference for market-based solutions. Interventions should be implemented only if a market-based solution is not able to achieve the same desired results • Cost/ benefit evaluation. Evaluation of interventions, incentives and mechanisms needs to include potential negative externalities. Expected costs and benefits need to be transparent • “Sunset” period for incentives. Interventions need to be restricted to a defined time window, after which they should be re-evaluated based on the following questions: Is the intervention still needed, or would a market-based mechanism now work? Is the purpose of the intervention still desired? Is it still effective? • Enforced accountability. Incentives for all stakeholders (originators, creditors, debtors, regulators) need to be aligned as much as possible to avoid moral hazard. Parties involved need to be accountable for their actions in the long-run • Level playing fields. Countries need maximum alignment on incentive structures in order to avoid arbitrage opportunities arising from inconsistent tax or regulatory structures Recommendations The study proposes a set of actions that financial institutions, regulators and policy-makers can take to promote sustainable credit levels in the years ahead. These actions are aimed both at detecting and preventing pockets of excess credit, and at ensuring sufficient credit flows to segments and countries whose growth is held back by a shortage of credit. The recommendations are: 1. Integrate the concepts of sustainable credit into the regulatory agenda 2. Create standardized government accounting practices to increase transparency and accurately assess sovereign finances 3. Encourage responsible borrowing through financial education 4. Encourage financing of local coldspots through targeted mechanisms 5. Task a single agency with monitoring global credit levels and system-wide credit sustainability 6. Align banks’ risk appetite with sustainable credit criteria 7. Drive innovation by financial institutions, developing new mechanisms that can safely meet future global credit needs 8. Establish goals for efficient and deep capital markets by 2020 in developing economies The remainder of this chapter sets out each of these recommendations in turn. 1. Integrate the concepts of sustainable credit into the regulatory agenda Recommendation: New liquidity and funding regulation can help reduce the frequency and intensity of credit hotspots and should be supported. However, decision-makers should guard against the risk that such regulation limits sustainable global credit growth or creates new credit coldspots. Several initiatives currently on the regulatory agenda should contribute to more sustainable credit. Notably, the likely result of Basel III will be that banks hold close to 11% tier 1 capital41, including the new regulatory minimum and a buffer. If applied consistently and globally, these increased capital levels could help reduce the frequency and intensity of credit hotspots, keep credit levels below the rules of thumb proposed in this report, and improve local credit sustainability.
This is not the same as “core tier 1” capital or “tier 1 common”, which only includes common stock and retained earnings. This broader measure includes other forms of capital that qualify for tier 1. Target ratios for core tier 1 capital are likely to hover around 8-9%.
More Credit with Fewer Crises | 51
However, there are several risks that decision-makers should guard against as new regulation is designed and implemented: • National regulators implementing capital requirements beyond what is proposed by Basel III should bear the “burden of proof” that additional regulation will not limit future economic growth. Regulation should also be adopted globally to avoid increased systemic instability by way of regulatory arbitrage. • Additional capital and liquidity standards should be assessed carefully to avoid overly negative impact on institutions that do not have alternate sources of financing, such as small and medium enterprises (SMEs) that do not have access to corporate bond markets. • Regulation should not unfairly punish parts of the economy that are critical for global growth. Trade finance, in particular, is a crucial enabler for SMEs to expand their businesses beyond national borders, and is characterized by very low default rates. (The ICC-ADB Trade register recorded fewer than 1,200 defaults out of 5.2 million transactions in 2009; the average transaction term was 115 days.) Yet, during the 2008-09 crisis, 10-15% of the decline in global trade resulted directly from lower trade finance liquidity as banks reduced trade finance to shore up their capital positions42. For the purposes of capital and liquidity regulation, trade finance is often grouped within products that have very different characteristics, such as corporate lending and off-balance sheet items. New regulation should be thought through carefully to avoid unduly restricting trade finance. • Today, banks’ credit models primarily use backward-looking credit analysis. Over the next few years, then, these models will be programmed with information from the crisis as a matter of default, which will have the effect of raising capital costs for many segments. For segments where more credit is needed and the prognosis is positive, banks should work with regulators to incorporate forward-looking metrics into their models. A forward-looking assessment based on an industry perspective and a detailed view on the viability of individual business models will improve growth prospects for borrowers as well as opportunities for lenders. • Liquidity and funding regulation needs to strike a careful balance between making individual credit “cells” more stable and creating future credit coldspots. Many might argue that a fully locally funded subsidiary would be a safer node in the overall network. However, given continued global imbalances between assets and liabilities, liquidity and funding regulation should take account of its potential to create trapped pools of liquidity, limit sustainable global credit growth, or create new credit coldspots. • Reforms to date have focused primarily on increased capital levels and have focused much less on underlying credit rating disciplines and the resulting risk-weighted assets (RWAs) in the banking book. If regulatory attention were to turn more directly to minimizing hotspots, future reforms would focus on creating a stronger correlation between RWAs and risk (for example, by discriminating more intelligently between RWAs for low- and high-LTV mortgages), as well as on keeping a close watch on how banks apply through-the-cycle ratings. This more granular evaluation of credit risk would do more to address future hotspots than additional blanket capital increases. • Lastly, stricter regulation on the traditional banking system should not lead to increased credit activity through the unregulated shadow banking system. 2. Create standardized government accounting practices to increase transparency and accurately assess sovereign finances Recommendation: Governments should adopt uniform accounting standards so that a complete and transparent picture of each country’s financial resources and obligations is available. The challenge of creating and agreeing on such standards should be made part of the G20 agenda. The recent sovereign debt crisis has highlighted and amplified the risks of substantial government debt. Although there are still many unknowns, the analysis in this report, based on best available data sources, suggests the next credit crisis is just as likely to be a government as a private debt crisis. Confidence in government debt has been undermined by the lack of transparency caused by inconsistency in government accounting standards, the lack of a consolidated balance sheet including all debt management units, and the treatment of complex financial products. Because banks have very large exposure to sovereign risk (more than ever, given new liquidity buffers put in place after the crisis), uncertainty related to government debt has an even greater potential to undermine the international financial system.
Seoul G20 Business Summit, ““Revitalizing World Trade: Towards a sustainable and equitable recovery”, November 2010
52 | More Credit with Fewer Crises
The IMF has laid out guidelines for debt management in order to reduce financial vulnerability. The guidelines highlight lessons learned in the areas of transparency and accountability, legal and institutional frameworks, debt management strategies, and the maintenance of an effective debt market. In addition to such guidelines, the IMF and the World Bank have advocated strengthening international accounting standards. While countries may be hesitant to change their accounting practices, this report recommends that governments adopt uniform standards so that a complete and transparent picture of each country’s financial resources and obligations is available. This is an urgent necessity, as many current government accounting practices undermine transparency, consistency and confidence. In particular: • Calculations of government debt often do not include off-budget obligations required for entitlement spending such as pensions and healthcare • Some governments use off balance sheet vehicles in the form of loans to state-guaranteed banks that do not appear in the state budget • Others have been known to sell securitizations backed by uncollected social security contributions to decrease government debt With transparent and consistent government accounting, stress tests on sovereign finances would provide a better understanding of how a country’s financial health would change under different assumptions – for example, changes in healthcare costs, longevity or interest rates. Implementing this recommendation in reality would undoubtedly be difficult and complex. It is therefore proposed that the challenge of creating standardized government accounting practices is made part of the G20’s agenda. 3. Encourage responsible borrowing through financial education Recommendation: An international, government-led initiative should undertake an impact study of existing financial education programmes worldwide, and then bring together government officials, education leaders and financial institutions to agree on an approach and implementation plan for improving financial literacy. A lack of financial understanding on the part of borrowers can be a key contributing factor to unsustainable credit growth. For example, many corporate and government borrowers are unsure about what constitutes acceptable debt levels. A lack of financial literacy on the part of retail borrowers played a significant role in the US sub-prime crisis, and puts credit sustainability at risk in many other markets. This report recommends that the financial education programmes in place in many countries be expanded and improved on an urgent basis. Governments, banks and community organizations should identify innovative ways to engage citizens to increase financial literacy, including: • Developing a financial education curriculum in schools where appropriate • Making financial education a part of state welfare programmes • Finding “teachable moments” to provide highly targeted financial training right before a particular financial activity (such as providing mortgage counselling before a borrower takes on a home loan) • Measuring the success of financial education programmes Governments will need to determine how best to improve the financial literacy of their populations, and should therefore commit to undertaking an impact study of various existing financial education programmes worldwide. They should then host a summit for government officials, education leaders and financial institutions to agree on the best approach for improving financial literacy and to establish an implementation plan and timeline. First steps have been made with the establishment of the OECD’s International Gateway for Financial Education, which has a mandate to perform research on financial education programmes, to disseminate data collected, and to provide a forum of exchange about best practices. This report also urges individual financial institutions to take the initiative to strengthen the financial literacy for their own current and potential customers. One North American bank, for example, has instituted a financial literacy strategy aimed at young customers which involves online courses, planning tools for students, financial advice, blogs and a programme based on Second Life.
More Credit with Fewer Crises | 53
4. Encourage financing of local coldspots through targeted mechanisms Recommendation: Governments and banks should create targeted mechanisms to solve the well-documented problem of lending to SMEs in developing markets. In developed markets, they should establish a robust fact base on the extent to which SME lending is constrained, and develop innovative solutions to improve both the supply and the demand side. As discussed in previous chapters, addressing credit coldspots is a critical driver of growth. Internationally, these coldspots include infrastructure investment and SME financing. As this report shows, consumer credit also plays a significant role in development as economies mature; today, there are potential retail credit coldspots in several key markets, notably China. Mechanisms are needed to help ensure that credit flows to areas such as these that are crucial for economic growth – rather than being tied up in less productive credit uses. For example, a significant part of the assets tied up in commercial real estate in recent years could be more productive if it were invested in current credit coldspots. SME financing is a coldspot that has been the focus of considerable international attention. This report endorses the view of the G20 that SMEs are critical for economic growth and that SME finance must be a key priority if credit is to drive development43. This is an issue that is fundamental in both developing and developed markets. In developing markets, financing for SMEs is a difficult challenge. This is partly due to crowding out of SMEs from bank lending where capital markets are underdeveloped. In addition, there is often a problem of market failure; mutually beneficial lending relationships between banks and SMEs are not pursued due to information asymmetry or lack of transparency. Multiple stakeholders can help develop mechanisms to solve the problem of lending to SMEs in developing markets: • Governments can help by facilitating the relationships between banks and SMEs through an improved environment that grants stability and transparency; they can act as a bridge between banks and SMEs by promoting transparency. For example, governments can set up credit bureaus and effective audit mechanisms for increased transparency, or create an asset registry to provide backing to banks that lend to SMEs. However, governments should be careful not to distort markets by picking winners or providing blanket guarantees. • Banks must engage with the SME opportunity, and clearly flag barriers that prevent them from doing so. SME lending (with appropriate risk management) is profitable, and by default banks should be willing proponents. • Other actors, notably players with large existing infrastructure, such as retailers, mobile phone operators and logistics firms, can complement the banking system in the provision of payments and innovative forms of financing. However, this should not lead to the formation of a potentially unstable shadow finance system beyond regulatory scope. SME financing in developed markets is just as critical for economic growth and is a topic of current debate between national governments (which assert SMEs are credit-starved) and banks (which tend to argue that it is principally a demand rather than a supply problem). This debate must be informed by a far more robust fact base and more innovative solutions than is currently the case. Solutions for SME financing in developed markets could encompass some of the innovative approaches currently operating in the SME space, such as peer-to-peer lending. However, more will need to be done to bring “industrial quality” to such channels, and to ensure they are sufficiently regulated so that both borrowers and investors are protected. Ideally, mechanisms to promote access to credit by SMEs would be administered via a targeted scheme. One example of such a scheme is the Industrial and Commercial Finance Corporation (now “3i”), set up by the Bank of England and the major British banks in 1945 to provide long-term equity funding for small and medium-sized enterprises immediately after the Second World War. (It must be noted that equity schemes may be easier to implement than credit schemes due to the positive incentive from the possible up-side of equity.)
Seoul G20 Business Summit, ““How can the small and medium-sized enterprise sector be funded and nurtured”, November 2010
54 | More Credit with Fewer Crises
5. Task a single agency with monitoring global credit levels and system-wide credit sustainability Recommendation: A single organization should be tasked with monitoring global credit levels so that the risks to financial institutions are accurately assessed. This organization should build on the credit sustainability metrics presented in this report and combine them with the Early Warning Exercise methodology developed by the FSB and IMF. The Early Warning Exercise (EWE) developed by the FSB and the IMF, and discussed in Chapter 2, was a response to the G20’s recognition that lack of transparency within the financial system is a problem. The EWE is useful for central governments because it flags vulnerabilities in the global economic and financial system – analysis which is confidentially shared with the International Monetary and Financial Committee in order to provide policy-makers with options for response. Financial institutions need access to similar information to identify the role that they play in creating systemic risk. It is therefore proposed that a specific organization be tasked with monitoring credit levels so that the risks to financial institutions are accurately assessed. This organization should build on the credit sustainability metrics presented in this report and combine them with some of the methods in the EWE. For example, metrics that test the repayment and servicing prospects of debt could be incorporated into a vulnerabilities analysis as described in the EWE’s “Crisis Risk Models Methodology”, which helps identify additional thresholds for each country. The constituent parts of the analysis could be performed by individual banks, or alternatively by a single agency that represents the financial services industry as a whole. While the first option would foster innovation by allowing banks to compete to continuously improve their credit monitoring process, it would also be less efficient and could cause confusion if the monitoring produced conflicting results. This report therefore proposes that a single agency be tasked by the major financial institutions to monitor credit levels to increase transparency in and across markets. Whether such an agency is part of the existing multinational framework (for example, the IMF or the World Bank), or formed in the private sector, is up for debate. On the one hand, the agency would have to connect the individual microlevel analyses suggested in this report with macro-economic risk assessments such as those conducted by the IMF or World Bank already – suggesting that placing it inside an existing multinational organization is more efficient. On the other hand, the agency would need to link closely into the risk functions of private sector players (and be seen as beneficial rather than an additional regulatory burden). Legitimacy is also a consideration: the creation of a new entity might address some of the perceived legacy surrounding existing multinational organizations. Examples such as MarkIt, a company that provides impartial pricing data on credit instruments to trading firms (many of which are shareholders), show that such entities can be set up successfully within the private sector. Finally, it should be emphasized that such an agency needs to be remote from the influence of political goals – the recent calls for rating agencies to avoid “destabilizing messages” on sovereign credit should underline this point44. 6. Align banks’ risk appetite with sustainable credit criteria Recommendation: Banks should ensure that their contribution to systemic risk is considered at the level of day-to-day credit and lending decisions. Regulators and supervisory bodies should recognize the contributions made to financial stability by banks that are aligned with sustainable credit principles. Most banks set their risk appetite based on risks that impact primarily or solely their own institutions. Banks generally do not consider their own contribution to systemic risk, and only partially consider their exposure to systemic risk. This report recommends that banks should incorporate systemic risk considerations into their risk appetite statement for two reasons: • There is a strong business case for including, or improving, the notion of systemic risk as an exposure that shapes banks’ risk appetite decisions. Moreover, introducing systemic risk monitoring will allow banks not only to track hotspots but also to detect coldspots – and hence new business opportunities. • Ethical standards should lead banks to consider the risk their lending imposes on the system. Ignoring this risk has put the financial system under threat in the past – and could do so again in the future. To incorporate systemic risk into their risk appetite statement, banks need to: • Define, understand and quantify systemic risk in order to set appropriate targets. A single agency to monitor credit, coupled with application of the tools presented in this report, can provide an external view of the systemic risk that banks need to consider.
More Credit with Fewer Crises | 55
• Operationalize targets to ensure that the concept of systemic risk is considered at the level of day-to-day credit and lending decisions. For example, banks can use thresholds to avoid contributing to and being exposed to hotspots. Such thresholds could take the form of limits on lending into hotspots, and can be cascaded down the organization to ensure that they are operationalized. In turn, regulators and supervisory bodies should recognize the contributions made to financial stability by banks that are aligned with sustainable credit principles. For example, such recognition could take the form of capital relief (similar to the advanced model treatment under Basel II/III). 7. Drive innovation by financial institutions, developing new mechanisms that can safely meet future global credit needs Recommendation: Governments should help kick-start securitization markets with targeted mechanisms such as a government-subsidized programme to securitize SME loans. Financial institutions should develop further mechanisms to grow balance sheet capacity safely and meet future worldwide credit needs – including measures to strengthen housing and environmental finance, and to integrate the unbanked into the banking system. Increased demand for credit worldwide will provide both a challenge and an opportunity for banks. Long-term project financing, for infrastructure and other key uses, is critical for meeting economic and social goals. Meeting these long-term financing needs in the current funding-constrained environment will be almost impossible without innovation that brings back mechanisms allowing maturity transformation and risk transfer. As discussed in Chapter 1, securitization was one of the principal drivers of growth over the last 20 years, but its role as a contributor to the crisis is widely known and securitization volumes are a fraction of what they were five years ago. Revitalizing securitization markets in a safe manner is a critical step in meeting credit demand. Some institutions, including the ECB, SEC and FDIC, are focusing on ways to make these markets safer by increasing investor transparency, restoring confidence in credit rating agencies, and aligning the interests of originators and investors through retention requirements. In addition to these measures, governments could help kick-start securitization markets with targeted mechanisms. For example, a government subsidized programme to securitize SME loans could help free up balance sheet capacity. However, if such a scheme is implemented, it must have a clear sunset period to avoid long-term market distortions. Securitization is one element of the innovation in capital markets that is needed to close the projected capital shortfall set out in Chapter 3. Such innovation is particularly important in Europe, where capital markets’ role in mediating retail and wholesale credit lags the US, and in the developing world (see Recommendation 8). There are many other innovative mechanisms that financial institutions could bring to the challenge of growing balance sheet capacity safely and meeting future worldwide credit needs. Potential examples include: • Housing finance: The use of covered bonds as a more sustainable means of financing than securitization; and reduced equity or equity-sharing approaches such as limited equity co-ops, community land trusts, and deedrestricted housing. • Environmental finance: Clean energy for renewable energy projects, green bonds and debt-for-nature swaps that relieve country’s debt burdens in return for commitments to nature conservation projects. The UN’s programme on Reducing Emissions from Deforestation and forest Degradation in developing countries (REDD) could offer a good model for such instruments. Providing access to credit for those who are unbanked – currently 2.5 billion adults worldwide – will be a key element of meeting the world’s future credit needs. Efforts to serve the unbanked must both advance social goals and be economically sustainable for providers. To achieve such an outcome, it will be especially important that banks define and enforce sound lending practices. Banks may also need to develop innovative new ways to interact with customers45. Finally, it should be emphasized that financial innovations such as those discussed above should not lead to the formation of a potentially unstable shadow finance system beyond regulatory scope. 8. Establish goals for efficient and deep capital markets by 2020 in developing economies Recommendation: Governments, international agencies, and the G20 should promote the strengthening of capital markets in developing countries by institutionalizing two fundamental capital market development goals: improving infrastructure to broaden participation by foreign firms and investors, and creating a sound institutional environment.
Chaia, Goland and Schiff, “Counting the world’s unbanked”, McKinsey Quarterly 31 March 2010
56 | More Credit with Fewer Crises
Just as important as the steps decision-makers should take to forestall credit crises, determined action is required to increase access to credit in areas where it is needed to support economic growth – particularly in developing markets. Strengthening capital markets is one key way to do so. Corporate credit markets in developing economies are small and often inefficient, making it difficult to finance credit needs locally. Larger local corporates struggle to access global capital markets and “crowd out” smaller corporate credit needs. Faced with forecasted wholesale credit growth rates above GDP, many developing economies will have to grow their capital markets considerably and sustainably. This report recommends that governments and international agencies promote this process by institutionalizing two fundamental capital market development goals – goals that could form part of the G20’s agenda. First, infrastructure must be improved to broaden participation by foreign firms and investors, and reduce red tape in issuance and listing: • Efficient primary markets are needed to establish functioning yield curves • A sound legal framework with clear and enforced laws and rights is needed to help decrease the perceived regulatory risk of doing business in developing economies, as well as reducing the instability of transactions • A reliable clearing and settlement system is needed to ensure smooth transactions Second, a sound institutional environment must be created: • Credit rating agencies must be in place to provide the required information about risks • Financial institutions must develop industry and functional expertise and in-depth regional knowledge • Where no alternative exists, expertise should be imported – talented individuals could be recruited or seconded from abroad. The focus in this case should be to transfer knowledge to local players as soon as possible Implementing the Recommendations – Roles for Stakeholders To implement each of the eight recommendations set out above, the study has identified: • One or more “leader” stakeholders, who will initiate the recommendation and drive it through to implementation • “Enabler” stakeholders, who will support the leader(s) by providing technical assistance or making the environment more conducive for change • “Collaborator” stakeholders, who will engage with the leader(s) and enablers to shape implementation and provide feedback on progress These implementation roles are set out in Exhibit 26.
More Credit with Fewer Crises | 57
The financial crisis not only shook the foundations of the world economy, it also suggested to many observers that overall credit levels were unsustainably high and would need to be scaled back. The analysis in this report provides a different perspective: credit demand will grow strongly in the decade ahead, and meeting this demand is not only sustainable in principle, but also essential if the world is to meet its economic development goals. What drove the crisis were not excessive absolute levels of credit but a misallocation of credit, with some segments overheating while others faced serious credit shortages. As this report shows, there is every danger that this imbalance – and with it significant numbers of credit hotspots and coldspots – will remain unless leaders in the private and public sectors take decisive action. Achieving sustainable credit growth will require tough choices on the part of world leaders. However, the action leaders have taken to bring stability to the financial system in the aftermath of the crisis provides a sound platform for sustainable credit. A commitment to transparency, fact-based analysis, collaboration, and innovation by industry players, regulators, and policy-makers will help ensure that credit flows where it is needed to support growth, and that pockets of excess lending are spotted and contained before they are allowed to escalate into crises. The steps leaders take today can pave the way for credit to meet its full potential for improving the state of world.
More Credit with Fewer Crises | 59
ABS. Asset backed securities: Securities whose value and income payments are derived from and collateralized by a specified pool of small, illiquid underlying assets and their cash flows, for example credit card receivables, auto loans and student loans Credit depth of information index. From the World Bank. Measures rules affecting the scope, accessibility, and quality of credit information available through public or private credit registries. The index ranges from 0-6, where higher values indicate the availability of more credit information, from either a public registry or a private bureau, to facilitate lending decisions Disposable income. The amount of income left to an individual after taxes have been paid, available for spending and saving EE. Eastern Europe EWE. Early warning exercise, analysis performed by the IMF and FSB FSB. Financial Stability Board Financial access score. From the World Economic Forum, the score for one of the seven pillars that make up the World Economic Forum Financial Development Index. It encompasses financial access for both commercial and retail financing, and includes 11 different metrics for each country. The score is an index scoring each country relative to the others in a group of 55 FDI – Financial Development Index. From the World Economic Forum, the FDI is an index that measures the quality of 55 countries’ financial systems on a relative scale. It measures each country’s performance across 120 measures, clustered along seven “pillars” in three major clusters: “Factors Policies and Institutions”, “Financial Intermediation” and “Financial Access” Flow measure. Measures the ability to service debt, calculated as the interest burden and average principal repayment over income. Income is equal to household disposable income, GDP or budget revenue in the retail, wholesale and government segments respectively GCI – Global Competitiveness Index. From the World Economic Forum, the GCI is an index that measures competitiveness for 133 countries on a relative scale. It measures each country’s performance across a several measure, clustered along 12 pillars: institutions, infrastructure, macroeconomic stability, health and primary education, higher education and training, goods market efficiency, labour market efficiency, financial market sophistication, technological readiness, market size, business sophistication, and innovation Government credit. All public sector credit stock (including loans and bonds outstanding) Government interest burden. Annual cost of government debt, including interest payments on bank loans to government entities and yield paid on government bonds GSEs – Government-sponsored enterprises. For the purpose of this report: entities originally created by the US government to provide liquidity, stability and affordability to the US housing and mortgage markets. Originally state owned, many, such as Fanny Mae, are now owned by private shareholders. They are primarily active on the secondary mortgage markets, by buying loans from banks and selling tranches of aggregated pools in the secondary markets. IMF. International Monetary Fund Leverage. Credit as a percentage of GDP LBO – Leveraged buyout. A takeover where the acquiring company secures a controlling interest in the target company using significant amounts of borrowed money. The assets of the target company are often used as collateral for the borrowed money
More Credit with Fewer Crises | 61
LTV – Loan to Value. A lending risk assessment ratio that lenders examine before approving a mortgage. Typically, assessments with high LTV ratios are generally seen as higher risk and, therefore, if the mortgage is accepted, the loan will generally cost the borrower more to borrow. LTV Ratio = Mortgage amount divided by appraised value of the property MBS – Mortgage backed securities. Special type of asset backed security whose value and income payments are derived from and collateralized by a specified pool of mortgage loans and their cash flows ME. Middle East NA. North America OECD. Organisation of Economic Co-operation and Development Penetration. Calculated as credit over GDP PFA – Personal financial assets. Represents total household assets including, deposits, stocks, pensions and other investments PFL – Personal financial liabilities. Represents total household credit stock outstanding Retail credit. All household credit stock (loans outstanding), including mortgages and unsecured loans Retail interest burden. Annual cost of household debt, including interest payments on loans ROW. Rest of world RWA – Risk weighted assets. Risk weighted assets are a percentage of the bank’s assets weighted by risk. Regulators require banks to hold a certain amount of capital against RWAs to absorb potential losses. In the calculation of RWA, each asset class is given a different weight (expressed as a percentage of its face value) according to its riskiness. As a result, an asset of low value but with high risk may have the same RWA value as an asset of very high value but low risk implying that they should require the same amount of capital reserve to absorb potential losses SME. Small or medium enterprise Stock measure. For retail, personal financial liabilities over GDP; outstanding wholesale credit (incl. loans and bonds) over GDP; outstanding government credit (incl. loans and bonds) over GDP WE. Western Europe Wholesale credit. All corporate sector credit stock (including loans and bonds outstanding) contains both large corporates and SMEs Wholesale interest burden. Annual cost of corporate and SME debt, including interest payments on bank loans
62 | More Credit with Fewer Crises
American Securitization Forum, SIFMA, Australian Securitisation Forum, European Securitisation Forum, “Restoring Confidence in the Securitization Markets”, December 3, 2008. Bae, Kee-Hong and Vidhan Goyal. Creditor Rights, Enforcement and Bank Loans. In The Journal of Finance, 2006, 64: 823-860. Bank of England Financial Stability Report. June, 2010. London: Bank of England. Bank of International Settlements, A Framework for Assessing the Systemic Risk of Major Financial Institutions (2009). Bank of International Settlements, The Future of Public Debt: Prospects and Implications (2010). Basu, Priya and Pradeep Srivastava. Scaling up Mircrofinance for India’s Rural Poor. In World Bank Policy Research Working Paper, 2005. Bayoumi, Tamim and Ola Melander. Credit Matters: Empirical Evidence on US Macro-Financial Linkages. In IMF Working Paper, 2008. Beck, Thorsten, Felix Rioja and Neven Tomov Valeve. Who Gets the Credit? And Does It Matter? In CEPR Discussion Papers, 2009. Beck, Thorsten and Asli Demirguc-Kunt. Small and Medium-Size Enterprises: Access to Finance as a Growth Constraint. In Journal of Banking & Finance, 2006, 30: 2931-2943. Beck, Thorsten, Asli Demirguc-Kunt and Ross Levine. Law and Firm’s Access to Finance. In NBER Working Papers, 2004. Besley, Timothy. Nonmarket Institutions for Credit and Risk Sharing in Low-Income Countries. In Journal of Economic Perspectives, 1995, 9:115-127. Buttari, Juan. Subsidized Credit Programs: The Theory, The Record, The Alternatives. In USAID Evaluation Special Study, 1995. Cecchetti, Stephen, Alfonso Flores-Lagunes and Stefan Krause. Financial Development, Consumption Smoothing and Reduced Volatility of Real Growth. 2006. Chaia, Alberto, Anthony Goland and Robert Schiff “Counting the world’s unbanked”, McKinsey Quarterly, 31 March 2010. Chetty, Raj and Adam Looney. Consumption Smoothing and the Welfare Consequences of Social Security in Developing Economies. In Journal of Public Economics, 2006, 90:2351-2356. Claessens, Stijn and Enrico Perotti. Finance and Inequality: Channels and Evidence. In Journal of Comparative Economics, 2007. Classens, Stijn. Access to Financial Services: A Review of the Issues and Public Policy Objectives. In Policy Research Working Paper Series, 2005. De Nederlandsche Bank, Credit and Economic Recovery (2009). de Rato, Rodrigo. Economic Growth and Financial Market Development: A Strengthening Integration. 2007. [Speech] Demiquc-Kunt, Asli. Finance and Economic Development: Policy Choices for Developing Countries. In World Bank Policy Research Working Paper, 2006. Demirguc-Kunt, Asli and Ross Levine. Finance, Financial Sector Policies and Long-Run Growth. In World Bank Policy Research Working Paper, 2008.
More Credit with Fewer Crises | 63
Driscoll, Gerald and Lee Hoskins. Property Rights: The Key to Economic Development. In Policy Analysis, 2003. Federal Reserve Bank of New York, Quarterly Report on Household Debt and Credit, November, 2010. New York. Ferguson, Niall, The Ascent of Money: A Financial History of the World (Penguin, 2009). Ferrucci, Gianluigi and Adrian Penalver. Assessing Sovereign Debt Under Uncertainty. In Financial Stability Review, 2003. Financial Services Authority, Bank Regulation, Capital and Credit Supply: Measuring the Impact of Prudential Standards (2009). Financial Services Authority, The Turner Review: A Regulatory Response to the Global Banking Crisis (2009). Financial Stability Board, International Monetary Fund: The Financial Crisis and Information Gaps (2009). Financial Stability Board, Reducing the Moral Hazard Posed by Systemically Important Financial Institutions (2010). Gartner, Five Dimensions of Network Design to Improve Performance and Save Money (2009). Gelbard, Enrique and Sergio Leite. Measuring Financial Development in Sub-Saharan Africa. In International Monetary Fund Working Paper, 1999. Goland, Anthony, Robert Schiff, and Peer Stein. Two trillion and counting: Assessing the credit gap for micro, small and medium-size enterprises in the developing world, IFC-McKinsey & Company report, 2010. Greenwood, Jeremy, Juan Sanchez, and Cheng Wang. Quantifying the Impact of Financial Development on Economic Development. In The Federal Reserve Bank of Richmond Working Papers, 2010. Greenspan, Alan. “The Crisis”, Paper for the Brookings Institution, 2010. Haldane, Andrew. Rethinking the Financial Network. April, 2009. [Speech] Haldane, Andrew. The $100 Billion Question. March, 2010. [Speech] Härle, Philipp, Erik Lüders, Theo Pepanides, Sonja Pfetsch, and Thomas Poppensieker. “Basel III: Its impact, how banks might respond and the challenges of implementation” McKinsey & Company, 2010. INSEAD, Crisis and Consumption Smoothing (2009). Institute of International Finance, Systemic Risk and Systemically Important Firms: an Integrated Approach (2010). Institute of International Finance. Interim Report on the Cumulative Impact on the Global Economy of Proposed Changes in Banking Regulatory Framework, June, 2010. Washington DC. International Monetary Fund, The IMF-FSB Early Warning Exercise: Design and Methodological Toolkit, September, 2010. Washington DC: IMF. International Monetary Fund, Debt Bias and Other Distortions: Crisis-Related Issues in Tax Policy, June, 2009. Washington DC. Jorion, Philippe and Gaiyan Zhang. Good and Bad Credit Contagion: Evidence from Credit Default Swaps. In Journal of Financial Economies, 2006. Kaminsky, Graciela, Carmen Reinhart, and Carlos Vegh. The Unholy Trinity of Financial Contagion. In Journal of Economic Perspectives, 2003, 17:51-74.
64 | More Credit with Fewer Crises
Kaufmann, Daniel, Aart Kraay, and Massimo Mastruzzi. The Worldwide Governance Indicators: Methodology and Analytical Issues. World Bank Policy Research Working Papers, 2010. Krueger, Daron and Fabrizio Perri. Understanding Consumption Smoothing Evidence from the US Consumer Expenditure Data. In Journal of the European Economic Association, 2005, 3:340-349. Levine, Ross. Financial Development and Economic Growth: Views and Agenda. In Journal of Economic Literature, 1997, 35:688-726. Lowen, Cara and Donald Morgan. The Credit Cycle and the Business Cycle: New Findings Using the Loan Officer Opinion Survey. In Journal of Money, Credit, and Banking, 2006, 38:1575-1597. McKinsey Global Institute, Debt and Deleveraging: The Global Credit Bubble and it’s Economic Consequences (2010). McKinsey Global Institute, If You’ve Got It, Spend It: Unleashing the Chinese Consumer. McKinsey Global Institute (2009). Morduch, Jonathan. Income Smoothing and Consumption Smoothing. In Journal of Economic Perspectives, 1995, 9:103-114. Qian, Jun and Philip Stahan. How Laws and Institutions Shape Financial Contracts: The Case of Bank Loans. In The Journal of Finance, 2007, 62:2803-2834. Rajan, Raghuram and Luigi Zingales. Financial Dependence and Growth. In The American Economic Review, 1998, 88: 559-586. Rajan, Raghuram. Has Financial Development Made the World Riskier? In NBER Working Papers, 2005. Reinhart, Carmen M. and Kenneth S. Rogoff. This Time is Different: Eight Centuries of Financial Folly. Princeton: Princeton University Press, 2009. Rutledge, Susan. Consumer Protection and Financial Literacy: Lessons from Nine Country Studies. In World Bank Working Paper, 2010. Second OECD Conference of Ministers Responsible for Small and Medium-sized Enterprises (SMEs). Promoting Entrepreneurship and Innovative SMEs in a Global Economy: Towards a More Responsible and Inclusive Globalisation. Istanbul, Turkey. June 2004. Seoul G20 Business Summit, How can the small and medium enterprise sector be nurtured? (Seoul, 2010). Seoul G20 Business Summit, Closing the gap in infrastructure and natural resource funding (Seoul, 2010). Seoul G20 Business Summit, ““Revitalizing World Trade: Towards a sustainable and equitable recovery”, November 2010. Systemic Risk and Systemically Important Firms. May, 2010. Washington DC: Institute of International Finance. UNESCAP, Implementing the Monterrey Conference (2005). UBS Investment Research, China Focus, “Rebalancing, the Central Theme of the 12th Five Year Plan”, November 2010.
More Credit with Fewer Crises | 65
This Technical Appendix provides further detail on study’s methodology, fact base, and data output, in three sections: 1. Data and forecasts for the global credit model. This section provides detail on the inputs and methodology used to develop the credit model (as set out in the Introduction and Chapter 1). 2. Approach to develop rules of thumb. This section details the approach adopted in developing rules of thumb to identify potential credit hotspots (as set out in Chapter 2). It describes the analysis of historical credit crises from which the rules of thumb was derived. 3. Credit connectivity framework – approach and application. This section details the approach to develop the credit connectivity framework (as set out in Chapter 2), as well as its application in measuring country-to-country credit connectedness and the contagion risk of individual credit segments 1. Data and Forecasts for the Global Credit Model As set out in the Introduction, a detailed global credit model was constructed to map credit volumes between 2000 and 2009, and to project potential credit demand to 2020. The model spans 79 countries representing 99% of credit volume (with approximated figures for the remaining countries). It disaggregates historic and projected lending by wholesale, retail, and government segments, defined as follows: • Wholesale credit: all corporate and SME credit stock (loans and bonds outstanding) • Retail credit: all household credit stock (loans outstanding), including mortgages, and other personal loans such as credit cards, auto loans, and other unsecured loans • Government credit: all public sector credit stock (including loans and bonds outstanding) Within these segments, the model also details historic and projected growth for particular credit products, by year – for example, SME lending within the wholesale segment and mortgage lending within the retail segment. Sources used to develop the model included: • Global banking pools – a financial institution database was developed to provide lending stock data across the 79 countries analysed • The McKinsey Global Institute’s global asset database, which includes projections for outstanding bonds to 2020 • Expert sources including Global Insight’s macroeconomic forecasts The model quantifies both capital market bonds and lending from financial institutions – the latter aggregated bottom-up from data at the sub-product level data, including on consumer finance, mortgages, leasing, and corporate and small and medium enterprise (SME) lending. Credit figures only include transactions involving the end-user, i.e. the final customer; credit such as interbank lending is excluded to avoid double counting. The model covers both commercial banks and non-banking financial institutions; includes credit held both by residents and non-residents in a given country; and calculates both non-securitized and securitized volumes. For the credit growth forecasts to 2020, the model calculates potential credit demand – the total level of demand for credit by country and segment, assuming that there is no restriction on supply over and above the restrictions in place prior to 2010. For the purposes of the model, credit stock is defined as the total outstanding lending amount at year end in 2009 US dollars. Fixed exchange rates are modelled to allow a more direct assessment of changes in credit demand, allowing easy comparability while avoiding distortions that would be created by currency fluctuations. This includes both capital markets (bonds outstanding) and traditional loans. The following assumptions underpin the model’s projections: • Consensus projections are met for global economic growth – that is, nominal growth of 6.3% per annum in 2009-2020 • Macroeconomic indicators such as consumption, exports, and GDP growth have the same effect on future credit supply and demand as they did in the past decade. (The study’s forecast of potential demand therefore does not include new supply limitations that may enter the market, such as Basel III, or more expensive and limited securitization markets.) • A fixed US dollar exchange rate based on 2009 actual is used for all historic and projected credit stock • Interbank lending is excluded from all credit stock data to avoid double counting • Nominal credit stock values in all cases, in US dollars
More Credit with Fewer Crises | 67
The “consensus” GDP growth projection adopted for this analysis is Global Insight’s widely accepted forecast: nominal growth of 6.3% per annum between 2009 and 2020, equivalent to real GDP growth of 3.3% per annum over this period. Global Insight’s forecast is built on an aggregation of country-level forecasts. For each country, the model provides credit demand forecasts based on: • Macroeconomic forecasts for 2009-2020, built off 14 drivers detailed by Global Insight. These drivers include real and nominal GDP, private consumption, and investment; nominal disposable income, exports and imports; population; unemployment; inflation; and money market and exchange rates. (Exhibit A1 details key macroeconomic indicators for 2009-2020 for selected countries and regions.) • Independent forecasts of credit penetration levels based on regressions using historic credit volumes for wholesale, retail, and government credit products. The model’s forecasts for government credit were validated against a range of existing projections – including government budgets, and forecasts by the OECD and the Economist Intelligence Unit (EIU).
2. Approach to Develop Rules of Thumb As Chapter 2 describes, the study developed 12 rules of thumb to identify potential credit hotspots, covering wholesale and retail credit as well as government debt (Exhibit A2). These indicators examine both the stock and flow measures of credit (in order to measure the prospects for both debt repayment and servicing). They also examine the forward-looking derivatives of these two metrics (the forecasted average yearly change over the next five years); the static indicator as well as the derivative are critical for detecting possible pockets of unsustainable credit growth. Together the rules of thumb can be used as a comprehensive “dashboard” of early warning indictors of potential credit hotspots. The study’s approach to identifying these rules of thumb was built on an analysis of credit crises over the past 50 years – including detailed quantitative analysis of 15 credit crises over the past 25 years, ranging from the 1985 Korean crisis to the recent government debt crises of 2009. These 15 crises included five wholesale, five retail, and five government credit crises. They were selected on the basis of three criteria: • The scale of the crisis was significant • The crisis was rooted in unsustainable credit within the wholesale, retail or government segment (rather than in a general macroeconomic imbalance) • Sufficient data was available on an historic basis for the country and segment in which the crisis occurred, for all or most of the local credit sustainability indicators used in this study
68 | More Credit with Fewer Crises
For each crisis, historic sustainability indicators were calculated (in line with the definition of sustainable credit set out in Chapter 2), both for the year of the crisis and for the eight years leading up to it. From these results, quantitative rules of thumb were derived for a range of defined sustainability indicators. The rules of thumb were then back-tested against time periods when no crisis occurred. Exhibit A3 lists the 15 crises analysed in order to derive the rules of thumb, along with the principal drivers of each crisis.
More Credit with Fewer Crises | 69
In developing the rules of thumb on the rate of change in credit levels, the study reviewed historic data of relative leverage development to identify periods when the sustainability of credit was threatened. This exercise demonstrated that an acceleration in the growth of credit stock as a proportion of GDP provides a useful early warning indicator of potential credit unsustainability. Crises with the highest relative yearly increase in credit stock as a proportion of GDP were selected for further analysis. As an example, Exhibit A4 shows how this analysis for the US led to the study’s focus on the 2008 mortgage crisis and the 1986 savings and loans crisis.
Based on the analysis of these crises, the study built a database of metrics from which to identify the best candidates for the rules of thumb. A total of 19 measures were developed and tested against each of the 15 credit crises analysed, for the eight years prior to each crisis. This exercise allowed the study to select, as rules of thumb, the metrics that would have best predicted those crises (Exhibit A5).
70 | More Credit with Fewer Crises
Exhibits A6-A11 show the results of the analysis of the 15 selected historical credit crises against the measures that were ultimately chosen as rules of thumb for each of the wholesale, retail, and government segments.
More Credit with Fewer Crises | 71
72 | More Credit with Fewer Crises
More Credit with Fewer Crises | 73
Each of the rules of thumb was back-tested against actual credit growth in each of the wholesale, retail and government segments across a range of countries from 2000 to 2009. This exercise generated a number of “false positives” – instances of credit growth that exceeded the upper bounds of the rules of thumb, but which did not lead to manifestations of credit unsustainability (Exhibits A12-A14). This finding underlines the point made in Chapter 2: that the rules of thumb should be applied as no more than early warning indicator thresholds, and that levels of credit in excess of the thresholds should trigger more detailed review. It is not appropriate to apply an absolute limit for credit sustainability, given the diversity in countries’ economic profiles. Rather, a range of measures needs to be considered alongside the rules of thumb in order to determine the sustainability of credit growth for a particular country or segment. Such measures could include country competitiveness, and (for the retail segment) the loan-to-value ratio on recent mortgage origination.
74 | More Credit with Fewer Crises
3. Credit Connectivity Framework – Approach and Application Chapter 2 emphasizes that, beyond ensuring the economic sustainability of individual credit segments – in other words, local sustainability – the domino effects of credit crises must be understood and managed to ensure sustainability at the global level. It therefore sets out a framework to measure the connectivity of individual segments. Together with an assessment of each segment’s local sustainability, this credit connectivity analysis allows decision-makers to quantify the potential contagion risk of each credit segment in each country. It should be emphasized that the analysis presented in this report focuses on the first order effects of a credit crisis in a particular country and segment – that is, the risk that a credit crisis in one country could trigger credit crises in other countries. This report does not assess the second-order effects of a credit crisis beyond credit channels – that is, the effects of a credit crisis in one country on factors such as the balance of trade and the market environment for other countries. While these effects may be part of a broader notion of contagion, this analysis aims to underscore the effects of credit contagion in particular. Chapter 2 proposes a framework to measure credit connectivity, based on five principles of network design: 1. Transparency and monitoring 2. Interconnection and modularity 3. Fault tolerance 4. Dimensioning and scalability 5. Intervention Exhibit A15 defines each of these principles and sets out the criteria by which connectivity within network nodes can be measured under each principle. Exhibit A16 goes further, applying these criteria to define metrics that measure the credit connectivity of a particular credit segment to the global financial system. These metrics can be used to measure the general risk of a hypothetical crisis in a particular country spreading to other credit markets via first-order effects. Paired with the metric of local sustainability, a measure of credit connectivity can help assess the contagion risk posed by a specific country to the financial system (as shown in Chapter 3 – see Exhibits 20 and 21 in the main body of the report).
More Credit with Fewer Crises | 75
76 | More Credit with Fewer Crises
These metrics can also be applied to measure the country-to-country credit connectivity more specifically. As an example, Exhibit A17 defines the metrics that could be used to assess which countries would be most directly affected by a hypothetical credit crisis in the United Kingdom. Exhibit A18 shows the output of this hypothetical analysis.
E xhibit A18 : Impac t of a hypothetic al UK credit c ris is in 2009 on other c ountries
Highest quintile All other quintiles
B Interc onnec tion and modularity Members hips in ec onomic as s oc iations in c ommon with UK C ount
C F ault toleranc e
C ountries Aus tralia Aus tria B elgium B razil C anada Denmark F inland F rance G ermany G reece Ireland Italy J apan Netherland P ortugal S pain S weden S witzerland U.S .
E xports to UK /total ex ports P ercent
B ank loans to UK /total bank loans to foreign c ountries P ercent
C redit inves ted in UK /total c redit holdings P ercent
E quity inves ted in UK /total equity holdings P ercent
T otal Z-s core
Note: Dimens ions A and D are not relevant since they apply in equal meas ure to all countries
More Credit with Fewer Crises | 77
There is potential to extend this model at a more granular level to create a global credit connectivity monitor covering a wide range of countries. The monitor would measure the connectivity of each country’s credit segments with segments in all or most other countries. Exhibit A19 provides an illustrative example of the output of such a monitor.
E xhibit A19 : Illus trative G lobal c redit c onnec tivity heatmap
Defined as low/medium ris k Defined as high ris k
78 | More Credit with Fewer Crises
This publication is a synthesis of ideas of many individuals from financial services and other domains. The Sustainable Credit project team would like to thank everyone involved for contributing so generously their time, energy and insights. The project team would also like to offer its special gratitude to the members of the Working Group and to the Senior Project Advisors for their contributions through workshops, idea generations, and document reviews. Their dedication was critical to the shaping of this report. The members of the Working Group are listed at the beginning of the report. Senior Project Advisors to the study included: • Michael Drexler, Managing Director, Global Head of Strategy, Commercial Investment Banking and Wealth Management, Barclays • Paal Weberg, Director, McKinsey & Company • Susan Lund, Director of Research, McKinsey Global Institute In addition, the project team would like to thank all workshop and interview participants for contributing their insights and time. These individuals were (in alphabetical order): • Peter Bacilie, Vice-Chairman, Investment Banking, JP Morgan Securities • Claudio Borio, Deputy Head of the Monetary and Economic, Department and Director of Research and Statistics, Bank for International Settlements • Tab Bowers, Director, McKinsey and Company • Lowell Bryan, Director, McKinsey and Company • Denis Bugrov, Chief Strategist, Sberbank • Pat Butler, Director, McKinsey and Company • Stephen Cecchetti, Head of the Monetary and Economic Department, Bank for International Settlements • Richard Cooper, Economics Professor, Harvard University • Howard Davies, Director, London School of Economics and Political Science • Richard Dobbs, Director, McKinsey and Company • Ismail Douiri, Co-Chief Executive Officer, Attijariwafa Bank • Vijay D’Silva, Director, McKinsey and Company • Wilson Ervin, Senior Advisor, Credit Suisse • Steven Fries, Chief Economist, Shell • Paul Frischer, Executive Managing Director, Research and Real Estate Strategies, Newmark Knight Frank • Tony Goland, Director, McKinsey and Company • Barry Gosen, CEO, Newmark Knight Frank • Olivier Hamoir, Director, McKinsey and Company • Philipp Härle, Director, McKinsey and Company • Yasuchika Hasegawa, CEO, Takeda Pharmaceuticals • Francois Henrot, Managing Director, Rothschild & Cie • David Hunt, Director, McKinsey and Company • Subir Lall, Division Chief of Asia Pacific, International Monetary Fund • Jean Lemierre, Advisor to the CEO, BNP Paribas • Ross Levine, Professor of Economics, Brown University • Patick McKenna, Managing Director and Regional Chief Risk Officer • Americas, Deutsche Bank • Sally Ng Siok-Leng, Managing Director, Research Department, China International Capital Corporation • Ergun Ozen, President and Chief Executive Officer, Garanti Bank • Efrat Peled, Chairman and CEO, Arison Investments • Ivan Pictet, Managing Director, Pictet & Cie
More Credit with Fewer Crises | 79
• • • • • • • • • • • • • •
Raghuram Rajan, Financial Economics Professor, University of Chicago Booth Ken Rosen, CEO of Rosen Consulting and former Professor of Real Estate at University of California Berkeley Susan Rutledge, Researcher of Financial Literacy at the World Bank Robert Shiller, Arthur M. Okun Professor of Economics, Yale University Luis Serven, Macroeconomist, World Bank Hanns-Peter Storr, Managing Director and Chief Credit Officer, PBC & PWM, Deutsche Bank John Sununu, former US Senator for New Hampshire Zubin Taraporevala, Director, McKinsey and Company Neil Thomson, Partner, LBO financing, Apax Partners Atsumasa Tochisako, CEO, MFI-Corp Philip Turner, Head of the Secretariat Group, Monetary and Economics Department, Bank of International Settlements Christian Upper, Head of Financial Markets, Bank for International Settlements Steve Wechsler, President of National Association of Real Estate Investment Trusts Harald Wilhem, Chief Financial Officer, Airbus
80 | More Credit with Fewer Crises
Core project team (in alphabetical order) • Katelyn Donnelly Consultant, McKinsey and Company • Anna Marrs Partner, McKinsey and Company • Max Neukirchen Partner, McKinsey and Company • Isabella Reuttner Project Manager, Financial Services Industries, World Economic Forum USA Contributors (in alphabetical order) • Luca Beltrami • Miklos Dietz • Geeta Gupta • Elizabeth Holt • Edward Knapp • Marc Lien • Miklos Radnai • Cornelius Vogel
Production • Writer: Colin Douglas • Editor: Nancy Tranchet, World Economic Forum • Creative Design: Kamal Kimaoui, World Economic Forum Floris Landi, World Economic Forum From the World Economic Forum Financial Services Team (in alphabetical order) • Yvonne Betlem, Associate Director • James Bilodeau, Associate Director • Donald Curry, Intern • Trudy Di Pippo, Associate Director • Lisa Donegan, Community Manager • Nadia Guillot, Senior Coordinator • Ibiye Harry, Project Manager • Alexandra Hawes, Coordinator • Abel Lee, Associate Director • Tom Watson, Project Manager
More Credit with Fewer Crises | 81
The World Economic Forum is an independent international organization committed to improving the state of the world by engaging business, political, academic and other leaders of society to shape global, regional and industry agendas. Incorporated as a not-for-profit foundation in 1971, and headquartered in Geneva, Switzerland, the Forum is tied to no political, partisan or national interests. (www.weforum.org)
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.