Weaknesses revealed by the market turmoil: where do we go from here?
Keynote address by Mr Malcolm D Knight, General Manager of the BIS, at the Chatham House event: "The New Financial Frontiers", London, 29 April 2008.
The financial market turmoil has revealed two paradoxes. The first is that after several years of high profits the global banking sector was thought to be well capitalised, whereas, in fact, banks' actual capital buffers and provisioning were much less robust than they had seemed. Indeed, widely used VaR indicators of risk underestimated the rise in underlying position-taking during the period of low volatility. The second paradox is that elements of massive illiquidity occurred in certain financial market segments within a context of overall excess liquidity worldwide. Three areas are central to understanding these paradoxes and addressing the weaknesses that have been revealed in the global financial system: risk management in financial institutions; the originate-to-distribute business model of the large banks; and the coordination of financial regulation and supervision across financial institutions, markets and national borders.
Good morning. It is a pleasure and an honour to address this distinguished group. I would like to thank DeAnne Julius and Chatham House for inviting me to share some views on the current turmoil in financial markets and what can be done to make the global financial system more robust.
Over recent months our staff at the BIS have been deeply involved in addressing the ongoing dislocation in key financial markets, both analytically, to determine its causes and what should be done about it, and in the context of our own banking business. That's because the basic objective of the BIS is not only to foster cooperation among central banks, but also to help to strengthen the global financial system. It is for these reasons that the BIS hosts the secretariats of international groupings that are working to monitor financial markets and develop norms and standards of financial stability. These groups include the Financial Stability Forum, the Basel Committee on Banking Supervision, the Committee on the Global Financial System, the Markets Committee, the International Association of Insurance Supervisors and the International Association of Deposit Insurers.
Based on our own work and the work of the Secretariats we host in Basel, I would say that there are three key elements that have played a critical role in the recent turmoil and where major improvements need to be made. These key elements are: risk management in financial institutions; the originate-to-distribute business model of the large banks; and the coordination of financial regulation and supervision across financial institutions, markets and national borders.
Main features of the turmoil
Why are these three areas so important? As you know, the turmoil that has disrupted financial institutions and markets since the summer of 2007 has been characterised by significant and unexpected tensions in the major (uncollateralised) interbank money markets. We have also come to realise that these problems are occurring largely because there was a higher degree of leverage in the financial system than one could have guessed from looking solely at the balance sheets of regulated banking institutions. As severe underlying asset quality problems came to the fore in several market segments, the withdrawal from risk and the deleveraging process gathered strength and became a storm.
Several aspects of the financial turmoil we have been experiencing should not have taken us by surprise - indeed, the central banking community had been warning for some time about the dangers of excessive risk-taking. These features include: the sudden sharp repricing of risk; the consequent turn of the credit cycle after years of excessive compression of the yield spreads on risky assets; the simultaneous evaporation of market liquidity and funding liquidity; and the fact that banks did not prove immune to the turmoil in credit markets. But other aspects have come as a real surprise: the sheer intensity and speed with which the turmoil hit the large internationally active banks; the major role played by off-balance sheet special purpose vehicles - what some observers have called the securitisation crisis - in the shadow banking system; and the apparent inadequacy of banks' capital cushions.
In a nutshell, two paradoxes have been revealed by the turmoil. The first is that after several years of high profits the global banking sector was thought to be well capitalised; whereas, in fact, banks' actual capital buffers and provisioning were much less robust than they had seemed. Indeed, widely used VaR indicators of risk underestimated the rise in underlying position-taking during the period of low volatility. The second paradox is that elements of massive illiquidity occurred in certain financial market segments (such as those for complex structured products) within a context of overall excess liquidity worldwide.
I think that three key elements - failures in risk management, shortcomings in the implementation of the originate-to-distribute business model, and limitations in the coordination of regulation and supervision across jurisdictions and sectors - are central to understanding these two paradoxes and to addressing the weaknesses that have been revealed in the global financial system. This, indeed, was very much the focus of the recent report of the Financial Stability Forum's Working Group on Enhancing Market Functioning and Resilience, of which I am a member and which presented its conclusions to the G7 Finance Ministers and central bank Governors in Washington on 11 April.
Risk management by financial institutions
Risk management is the first area where substantial progress can and must be made. We have learned repeatedly - and often painfully - that financial and banking crises are characterised by the failure to adhere to basic risk management principles, especially in times of financial innovation. Indeed, the past two decades provide vivid examples of fundamental inadequacies in risk management by financial institutions that triggered financial crises. One could recall, among many other examples, the crisis related to Latin American debt to commercial banks that culminated in 1982 when Mexico was no longer able to service its debt; various asset price bubbles such as the one in Japan in the late 1980s; the Asian financial crisis of 1997-98; Russia's sovereign debt default in 1998; and the LTCM bailout in the same year.
Inadequate risk management has also been at the heart of the current turmoil. Take, for example, the US subprime mortgage market. Those at the front end of the securitisation chain received fees to originate mortgages. They had little incentive to assess the borrowers' capacity to repay because they knew that someone else would buy and hold these mortgages. Banks at the centre of the securitisation process focused on the profits earned by distributing these instruments. Those closer to the end of the securitisation chain placed too much trust in the due diligence of originators and packagers, the judgments of the credit rating agencies, and the capacity of modern technology and diversification to manage financial risks - they had no credit monitoring capacity. This toxic combination of excess lending with an obvious failure to adhere to fundamental and sound risk management standards not only produced significant losses in mortgage portfolios; it also tainted an asset class that was key to the broader securitisation and credit distribution process.
There were also large risk control failures in the management of firm-wide concentrations in the US housing sector and related structured products, such as collateralised debt obligations (CDOs), that were based on residential subprime mortgages. Rather than having a clear understanding of their exposures to these asset types prior to the turmoil, many banks had to scramble after the crisis hit to understand and aggregate their exposures. Financial firms then found themselves exposed to major unexpected problems: the pervasive concentration of their assets in subprime mortgages; the liquidity facilities they had extended to their sponsored vehicles, such as special investment vehicles (SIVs) and asset-backed commercial paper conduits; and their own investments in the paper issued by these vehicles.
What should be done to enhance risk management? A key point is that greater reliance on risk transfer technologies means that banks are more vulnerable to the exposures that remain on their balance sheets when market liquidity seizes up. Financial institutions have to incorporate such effects into their stress tests and set adequate contingency plans. And, of course, banks' senior managers have to pay appropriate attention, even if they consider that the stress tests simulated by their risk managers are too severe to be plausible.
Liquidity risk management is another area where some fundamental shortcomings should be corrected. Increased reliance on market liquidity means that problems in financial markets can adversely affect banks' funding liquidity very quickly. Moreover, banks' growing role in providing backstop liquidity facilities to their sponsored special purpose vehicles, such as SIVs and conduits, raises the risk of the unexpected retention, or the reintegration, of assets on their balance sheets in times of stress and illiquidity in the securitisation market. Indeed, these were among the most poorly identified and managed risks in the whole securitisation process. It is now clear that institutions need to stress-test their funding capabilities by assuming market-wide disruptions and freezes, not just firm-specific events. They must capture potential liquidity and capital requirements more effectively in their contingency plans and in their economic capital measures.
And, of course, there is a critical need for risk management systems to measure counterparty credit risk exposures, whether on- or off-balance sheet. To this end, financial institutions need to carefully consider whether non-standard collateral is sufficiently robust, and whether the levels of collateral haircuts are appropriate.
The originate-to-distribute business model
The operation of the originate-to-distribute business model - in which large international banks originate credit, securitise the loans and thus take the credit exposure off their balance sheets - is a second area that needs strengthening. This model is, in my view, basically a good one. When it functions properly, it has the capacity to disperse risk widely, allocate capital efficiently, diversify the revenue streams of banks and foster financial innovation in the system. But the model has not been working well and has to be enhanced. The complexity and lack of transparency of new structured credit products make them more difficult to value and hedge, and can also severely impair their liquidity, which - as we saw recently - dried up almost immediately once tensions in the credit market developed.
A key issue is to ensure that financial innovation is accompanied by a concomitant improvement in risk management in the originate-to-distribute process. This is particularly relevant for primary intermediaries: the development of too many complex and investor-specific illiquid products inevitably leads to the creation of thin markets that are ineffective, if not problematic, in times of stress. In particular, greater attention should be focused on the impact of new structured financial instruments, which have complicated the lending process by making it less transparent and have rendered more difficult the identification of the ultimate holders of the risk.
There are other problems brought on by the way the originate-to-distribute model has been implemented. As I have already noted, significant failures in risk management have been inherent in the way this business model has been implemented (including incentives to originate loans imprudently, inadequate monitoring of borrowers and credit risk, etc). All in all, the ability of originators to shed risk through asset securitisation, along with the over-reliance on asset diversification, has led to the global dispersion of low-quality credit.
And, of course, many investors were attracted by the much wider yield spreads on triple-A rated subprime mortgage-backed securities and CDOs relative to other triple-A paper. They relied primarily on external credit ratings when making their investment decisions, and did not question why one type of triple-A rated product was paying a higher return than others. The purchasers of structured products forgot - or ignored - the basic relationship between risk and return, which was sending the message that they needed to conduct more rigorous due diligence before investing in exotic financial products.
These elements indicate that the originate-to-distribute business model needs to be strengthened in several ways:
- by realigning the incentives among the originators and other participants in the securitisation chain;
- by enhancing the transparency of the risks inherent in structured products; and
- by improving the usefulness and transparency of credit ratings so that investors can more appropriately apply the necessary market discipline.
These elements should help in addressing current weaknesses, while still allowing for financial innovation. To restore adequate incentives in the system I would caution against a heavy-handed regulatory approach and advocate market-based solutions, as these are more likely to be effective.
Cross-border coordination of the regulation and supervision of financial firms
Now let me turn to my third area of interest. This relates to the regulation and supervision of financial institutions, particularly the need for further advances in fostering coherence and cooperation across national financial jurisdictions, institutions and markets. Regulatory and supervisory frameworks are often best viewed through the prism of market turmoil, so what insight can we gain from this current period of stress?
First, the turmoil has highlighted the need for ensuring close cooperation among policymakers and financial supervisors with their different mandates and in different jurisdictions. The FSF's recent report is a prime example of how such cooperation can be enhanced. Representatives from more than 30 national authorities, international financial institutions and international regulatory and supervisory groupings convened to analyse the causes and weaknesses that led to the turmoil. Many of the FSF's recommendations for increasing the resilience of markets and institutions transcend national borders and cut across financial sectors. For example, the Basel Committee on Banking Supervision and the International Organization of Securities Commissions are working closely together to develop capital requirements for certain credit exposures held by banks and securities firms.
More must be done to further develop a deeper dialogue and closer cooperation among public authorities. Among central banks themselves, current disruptions in interbank markets have already resulted in closer cooperation: for example, on a number of recent occasions, they have found ways to better coordinate their liquidity management operations. But market monitoring and communication have also been greatly enhanced among central banks and supervisory authorities in the past few months: the recent Northern Rock and Bear Stearns cases have shown that the challenges faced by monetary policy authorities and supervisors are clearly intertwined.
The turmoil has also confirmed the need for robust, consistently applied global standards of prudent risk management and financial stability. This is particularly the case of the Basel II Framework, which could have alleviated some of the weaknesses revealed by the current crisis if it had already been in place in key jurisdictions. This framework should be promptly implemented, as rightly stressed by the FSF in its report.
And the turmoil has underlined the importance of effective liquidity risk management for cross-border and foreign currency operations. This issue is related to the increasing integration of financial markets, as can be seen from the sharp rise in the volume and speed of cross-border flows. As a result, many financial institutions have increased their international business and their dependence on the smooth functioning of international financial markets. Moreover, with large cross-border flows, liquidity disruptions originating in one country can pass quickly across different markets and settlement systems. The problem is that liquidity may not be fully transferable across borders and currencies, particularly in times of market stress. For instance, a bank can face foreign currency liquidity needs that cannot be covered by its (national) provider of liquidity. Another challenge is that each national regulator requires its own financial institutions to hold sufficient liquidity for its local operations. Thus, an important consideration in conducting cross-border operations is the need to understand fully the supervisory and regulatory practices within each jurisdiction. To this end, the Basel Committee, which as I said is hosted by the BIS and whose work is fully supported by us, continues to review current practices so as to strengthen banks' liquidity risk management and improve global supervisory practices.
Let me conclude. No one knows how long the present deleveraging process will take, what its precise dynamics will be and what feedback effects it may have on the real economy. We do know, however, that this deleveraging will have to run its course, and that it will be accompanied by deflation in asset prices until a new equilibrium is found. As history has shown, this process is often painful. This may even be more relevant in the present turmoil, because current disturbances in credit markets interact with the uncertainty created by two major disequilibria in real sector markets. One is housing markets in the United States and a number of other countries. A second is the persistence of large external current account imbalances. In such circumstances, it is of utmost importance to radically strengthen our modern financial systems. This will take time. But my point today is that there are three avenues where we can focus our immediate efforts.
First, improving risk management systems. Shortcomings related to underwriting, credit concentration management, hedging techniques, stress testing and liquidity risk management all played significant roles in contributing to and exacerbating the turmoil. Second, strengthening the originate-to-distribute model. This model remains basically a good and important element in financial intermediation, but it needs to be enhanced with better incentives, greater transparency and a more efficient use of credit ratings. Third, developing cross-border and cross-sectoral cooperation in regulation and supervision. There is a growing need for consistency between the banking sector and other financial sectors, as well as for a general level of consistency across countries. Implementing the Basel II Framework and the recent recommendations of the Financial Stability Forum will go a long way towards achieving this important goal.