How might the current financial crisis shape financial sector regulation and structure?

Keynote address by Mr Már Gudmundsson, Deputy Head of the Monetary and Economic Department of the BIS, at the Financial Technology Congress 2008, Boston, 23 September 2008.

BIS speech  | 
19 November 2008


The current financial crisis was triggered by increasing defaults on subprime mortgages and the turn of the housing cycle in the United States. However, it had deeper causes in the under-pricing of risk and debt accumulation in several countries during the period of low real interest rates and easy access to credit. The financial industry has responded by raising capital and reducing leverage. The public sector has provided massive liquidity support, injected capital and improved deposit insurance. Looking further ahead, wide-ranging reforms are underway aimed at increasing market and institutional resilience. It is an open question how wide the financial safety net will be cast. The future financial sector can be expected to be smaller and operate with higher capital and liquidity buffers than before the crisis.

Full speech

It is indeed both an honour and a pleasure for me to give a keynote address here today. My topic is how the current financial crisis might shape financial sector regulation and the structure of the financial system. When picking this topic several months ago, I did not anticipate the dramatic events of the last few weeks. These have put my question in a sharper focus. At the same time, these events might be beginning to provide some of the answers. To take an example, here in the United States we have seen major investment banks disappear as such and witnessed a significant extension of the financial safety net.

I will cover four main topics. First, I will say a few words about the origin and propagation of the crisis, which give us clues as to which aspects of regulation and structure of the financial sector might change going forward. Second, I will list the responses that we have seen so far from the financial industry itself and the public sector, at both the international and national levels. Third, I will discuss some of the key aspects of financial sector architecture and regulation that are currently under review. Finally, I will say a few words about the potential relevance of all this for the participants in this congress on financial technology.

The crisis

The current global financial crisis has now lasted more than a year, with no immediate end in sight. The crisis was triggered by increasing defaults on subprime mortgages and the turn of the housing cycle in the United States. Subsequently, the credit ratings of structured products, wholly or partly based on these mortgages, were significantly downgraded, raising uncertainty about the valuation of such products.

It was at this point that the banks at the centre of the financial system were hit much more speedily than most had envisaged before the crisis. Thus the drying-up of the market for asset-backed commercial paper created pressure on banks' funding liquidity.2 The reason was that the banks needed, for legal or reputational reasons, to provide their special purpose vehicles with liquidity or to bring them back onto their balance sheets. Thus, the banks needed to make more use of their own funding liquidity at the same time as their future liquidity needs were becoming both bigger and more uncertain. On top of this, they were becoming more uncertain about the creditworthiness of other banks, as they did not know where the exposure to the toxic subprime and structured product stuff was, or which banks might face problems because they would be forced into a distressed sale of assets due to a lack of funding liquidity. The result was a generalised hoarding of funding liquidity, which might have been rational from the standpoint of individual banks but was disastrous for the system as a whole.

This hoarding of funding liquidity made the crisis come to the fore in the drying-up of market liquidity3 in interbank money markets in the United States and in Europe on 9 August last year. This in turn prompted central banks in these regions to inject massive amounts of liquidity in order to stop money market interest rates from rising far above targeted levels.

We now know that this was only the beginning. What at first seemed mostly to be a US problem is now increasingly a global problem. What at first seemed to be valuation problems in specific segments of financial markets have turned into a broader-based downturn in asset prices. What at first seemed to be a liquidity problem has turned into major losses and writedowns of bank capital.

We are currently in a phase of this crisis where significant parts of the financial system in advanced economies are being forced to reduce their assets relative to capital, that is, to reduce what is called leverage. The reason is that the current level of leverage of many financial institutions implies a higher level of risk than they can manage in this environment of higher funding costs, increased volatility of most financial prices and more uncertainty. The deleveraging can take place through the raising of additional capital, which is currently becoming more difficult, or the disposal of assets and use of the proceeds to repay debt. However, a deleveraging of the whole financial sector, as distinct from individual institutions in normal market conditions, is a painful process involving asset price deflation and a lack of market liquidity.

The impairment of the wholesale money market along with higher funding costs and shorter available maturities has made many business models untenable. Those relying on short-term funding in wholesale money markets have been particularly vulnerable. This was the undoing of Northern Rock and contributed to the downfall of investment banks. One result of the decline of wholesale funding has been a significantly higher degree of competition for deposits, particularly in Europe.

The metamorphosis of the crisis from its initial stages to now is easier to understand when we realise that it had deeper causes than the faults in US subprime loan origination and the associated securitisation process. The crisis was preceded by a period of low real interest rates and easy access to credit, which fuelled risk-taking and debt accumulation. In the United States, it was the case both for households and for the financial sector itself. However, although the increased indebtedness of the US household sector was plain for everybody to see, the increased leverage of the financial sector was somewhat hidden. One reason was that the leverage was partly accumulating in what is now being called the shadow banking system. Another reason was that the focus on risk metrics like value-at-risk and the use of short time series as inputs allowed the low recent volatility of asset prices to mask the increase in leverage.

In the United States, easy credit conditions were made even more so by global current account imbalances and the willingness of foreign governments to finance the US current account deficit. Easy monetary policy in the aftermath of the bursting of the tech stock bubble in 2001 might also have contributed at the margin, although easy credit preceded it.

Last but not least, financial innovation contributed to debt accumulation. In particular, the originate-to-distribute model made it possible to originate loans - especially mortgages - to households, securitise them in large quantities, slice and dice them into differently rated tranches, and then sell them all over the world to both risk-averse and risk-seeking investors. The effect was that loan origination was less constrained by the balance sheet capacity of banks.

One result of this setup was that risk was apparently spread away from the institutions that are critical for the overall functioning and stability of the financial system, which should be good from the standpoint of financial stability. However, as it turned out, the distribution was less then met the eye, as the asset-backed securities were often held by special purpose vehicles closely associated with the banks originating them. Some commentators have for this reason called the arrangement "originate and pretend to distribute". Furthermore, as the value of structured products was potentially unstable and would become very uncertain at the first sign of stress and illiquidity in financial markets, what was distributed was not only risk, but also uncertainty and fear.

The upshot of all of this was the underpricing of risk. This underpricing was widely recognised in the central banking community, and by others, and was expected to result in significant repricing, which would in all probability be associated with lower asset values and a downturn in the credit cycle. What nobody knew, of course, was the timing of this repricing; nor did anyone anticipate the speed and ferocity of the change or the degree to which it would, in the first round, affect the core of the financial system.

Let me end the discussion of the crisis by looking at what I find a striking demonstration of the change in attitude towards two key risks, before and during the crisis, which are leverage liquidity risk. These were mostly ignored before the crisis but have since been the focus of markets, as you can clearly see if you look at Graphs 1 and 2.

Leverage ratios and CDS spreads
Liquidity ratios and CDS spreads

The responses

The responses to the crisis fall into two categories: current action and plans for future reforms.

The first line of action came from the financial industry itself. From the onset of the crisis until early September, major banks around the world had reported losses and writedowns in excess of $500 billion and raised capital to the tune of $350 billion. They also took steps to increase their liquid buffers in order to deal with current and expected balance sheet pressures and general uncertainty, as I mentioned earlier. Financial institutions are also taking steps to improve their risk measurement and risk management systems, although some would say that it is in a sense too late as the risky behaviour of the past probably will not be seen again for years to come. Finally, financial institutions are in the process of reducing their leverage and size, by both selling assets and reducing staff. The result will be a smaller financial sector, at least for a while.

The second line of action came from the central banks, which have in some cases taken extraordinary measures to provide liquidity to domestic money markets, help to liquefy illiquid bank assets, act as lenders of last resort to distressed financial institutions and facilitate cross-border liquidity management of banks through swap lines between central banks and through auctions of dollars in continental Europe, with Canada, Japan and the United Kingdom joining in last week. Central banks have understood that these actions would not solve the crisis. However, the measures were meant to deal with some of the symptoms, win time for financial institutions to deal with problem assets and contribute to a more orderly deleveraging process.4

The third line of action has come from the public sector more broadly. It has nationalised financial institutions that were deemed too big or too connected to fail, like Fannie, Freddie and AIG in the United States and Northern Rock in the United Kingdom. It has facilitated bank mergers, the latest being the takeover by Lloyds TSB of the United Kingdom's biggest mortgage lender, HBOS. The UK government also significantly improved the terms of deposit insurance for retail customers around the time of the Northern Rock episode.

At the same time as the financial sector and public authorities have been up to their knees in managing the crisis, there has been an increasing focus on reforms to the structure and regulation of the financial sector that would help to prevent and mitigate future crises. These deliberations have taken place at the international and national levels and in the public and private sectors.

The Financial Stability Forum, which was set up after the Asian crisis to oversee global financial stability, issued a report that, in its scope and authority, is probably the most important one regarding reform.5 It lists 67 recommendations on how to enhance market and institutional resilience. The report was endorsed by the G7. Global initiatives by the private sector include a report by the Institute of International Finance, published in its final version in July, and reports on counterparty risk management and risks associated with OTC derivatives and structured products.6 At the national level, reports have been issued, for example, in the United States and in the United Kingdom.7

Regulatory changes and the structure of the financial sector

Let me now list some of the changes to financial regulation and the structure of the financial sector that might result from this crisis; but these thoughts are informed by the actual responses so far and by some of the proposals contained in the reports that I just mentioned. But be clear that I say "might". There have been episodes where major changes in the direction of the financial sector were predicted but did not take place, although such misjudgments are probably less likely this time due to the severity of the financial crisis and the steps that have already been taken.

Let me start with potential changes to regulation and supervision, bearing in mind that the structure of the financial sector has always adapted to the regulations in place. In this connection, I should caution that I am discussing changes that will in most cases be implemented gradually and in some cases only after the crisis is over, as they are meant to dampen risk-taking. The system is currently in reverse mode. With these caveats, let me now highlight a few potential changes in the field of regulation.

The banking system will probably have to operate with higher capital buffers than prior to the crisis. There are three reasons for this. First, some of the loopholes revealed by the crisis will be filled. Second, there will be more volatility in the inputs to economic capital calculations as the crisis period enters the datasets. Third, there might be additional capital charges under Pillar 2 (supervisory review process). In addition, there is an active discussion taking place on the merits of introducing a limit on the leverage ratio of financial institutions, which, depending on where it is set, might call for more capital than otherwise at the peak of the next boom.

Regulators, as well as the financial industry itself, will be much more aware of liquidity risk than they were before the crisis, and liquidity buffers will therefore be higher than otherwise. However, there are several challenges involved, such as how to design relatively simple measures for funding liquidity risk and whether to impose minimum liquidity buffers.

It is becoming more widely recognised that the financial system has a natural tendency to amplify business cycles through a reinforcing feedback loop with the real economy and that certain aspects of accounting frameworks and capital regulation can increase this tendency. As an example, regulatory minima on capital and liquidity will turn procyclical once they are hit.8 There is therefore an active discussion under way on how these procyclical tendencies could be mitigated and even how certain elements of countercyclicality could be built into regulatory frameworks.9 The key issue here is how regulatory frameworks could lean against the build-up of risks and leverage during boom times - for example, through countercyclical capital and liquidity buffers that would be accumulated during booms and allowed to run down during periods of downturns and stress in the financial system. It will be interesting to see how this work will progress.

A key question going forward is how wide to cast the financial safety net composed of deposit insurance and access to emergency lending facilities at the central bank, and, by implication, how wide the reach of official regulation and supervision should be. Some have likened modern financial institutions to public utilities attached to a casino. It is the public utility that we care about, but if it is in the same institution as the casino, it will fail whenever the casino goes to the wall. In principle, we could mandate the split of the utility from the casino, regulate and supervise the former, and leave the latter to its own fate. Or we could extend regulation to the casino, as it governs the fate of the utility. It seems to me that, for various reasons, we are heading in the second direction. Thus, lightly regulated investment banks seem to be disappearing. One thing is clear, however: wherever the boundary of financial regulation is drawn will affect the structure of the financial industry, and the industry will try to find ways to circumvent the barriers.

Cross-border issues have been important during the crisis in terms of liquidity and the spread of toxic structured products around the world. Central banks have therefore been dealing with the cross-border strains in dollar funding markets, in particular through swap lines between central banks and the auction of dollar term liquidity by the ECB and the Swiss National Bank and, as of last week, the Bank of England, the Bank of Japan and the Bank of Canada. One question now is: what kind of arrangements should central banks have, if any, for facilitating the future cross-border liquidity management of banks? Among the options are swap lines and auctions of foreign liquidity, and the acceptance of foreign collateral, as some major central banks do already.

Let me now mention a few issues related to the potential evolution of the financial sector:

1. Higher capital and liquidity buffers and higher risk premia will entail a higher cost of capital and credit than before the crisis. That is not necessarily bad, as risk was underpriced.

2. Financial firms basing their business models on cheap access to funding in wholesale markets will either have to adapt or disappear. In the United States, the trend is currently towards universal banking. Competition for deposit financing will also be intense for a while.

3. On the product side, a premium will be put on simpler products because regulators and investors will remain sceptical of complex structured products and because management of financial firms will insist on understanding the products they offer, at least for a while!

4. The originate-to-distribute model remains to be fixed, and the interests of all the various players in the securitisation chain have to be better aligned. However, that does not mean that the model will disappear. At the fundamental level, the idea of distributing risk away from the institutions at the core of the financial system to investors that are willing and able to share in the risks is basically sound.

5. Finally, the financial sector will become smaller and less leveraged. That is the only way the sector can be returned to soundness and profitability in the environment that is likely to prevail in the post-crisis period. However, such retrenchment has to be seen against the earlier growth of the sector. Let me mention a few figures in this connection. From 1990 to 2006, the GDP share of the financial sector in the broad sense increased in the United States from 23% to 31%, or by 8 percentage points. During the same period, the increase in the GDP share was in excess of 10 percentage points in the United Kingdom but significantly less - around 6 percentage points - in both France and Germany. Graph 3 shows the development of the share of the financial sector in GDP for selected major advanced economies since the middle of the 1980s. The figures on profits are even more striking. For example, the financial services industry's share of corporate profits in the United States was around 10% in the early 1980s but peaked at 40% last year.

Share of the financial sector in GDP (in per cent)

Infrastructure, operational risk and data

Let me now say a few words about how all of this relates to what you are discussing at your conference here today. The first issue to note is that the infrastructure of markets and institutions has mostly worked well during the crisis, in spite of at times massive spikes in volatility and large volumes. This is particularly important, as one of the biggest risks to financial stability is malfunctioning of the payment and settlement system. There have also been significant concerns over operational risks in OTC derivatives markets, and you are discussing that issue here.

Regarding data: the call for more simplicity in both financial products and regulation might at first glance imply that data requirements will be less. However, even if simple devices like maximum leverage ratios are added to regulatory requirements - and nobody knows whether that will be the case or not - it will probably be in addition to Basel II, which is more data¬intensive than Basel I. Furthermore, the risks that remain will, it is hoped, be better monitored than before, and the macroprudential framework requires more data and analysis, at least for the supervisors. It is therefore hard to see that the demand for timely high-quality data will be reduced.

Final remark

Let me end by saying that the financial industry and its regulators face big challenges. The regulatory weaknesses that came to the fore during the crisis need to be fixed. This is also an opportunity to reduce the procyclical tendencies in the financial system and its regulatory framework. History shows that an efficient and stable financial system is key for sustainable economic growth. The question is how to strike the balance. An underregulated financial system is a risk to financial stability and economic growth. However, an overregulated financial system will not be efficient and will not give adequate support to innovation and growth. The challenges that policymakers face in this regard are therefore big.

Thank you very much.

1 Már Gudmundsson is the Deputy Head of the Monetary and Economic Department of the Bank for International Settlements. The views expressed are the author's and not necessarily those of the BIS.

2 By funding liquidity, I mean the ability of banks to meet payment obligations without significantly affecting daily operations or their financial condition.

3 By market liquidity, I mean the ability to trade an asset without having a major effect on its price.

4 On central bank liquidity measures during the turmoil and the associated policy issues, see Committee on the Global Financial System, "Central bank operations in response to the financial turmoil", CGFS Papers, no 31, Basel, July 2008.

5 See Financial Stability Forum, Report of the Financial Stability Forum on enhancing market and institutional resilience, Basel, 7 April 2008.

6 See Institute of International Finance, Report of the IIF Committee on Market Best Practices, July 2008, and Counterparty Risk Management Policy Group III, Containing systemic risk: the road to reform, 6 August 2008.

7 See US Department of the Treasury, Blueprint for a modernized financial regulatory structure,March 2008, and Bank of England, HM Treasury and FSA, Financial stability and depositor protection: strengthening the framework, January 2008.

8 BIS researchers have made major contributions to the analysis of the procyclicality of the financial system. See, for instance: Borio, "Towards a macroprudential framework for financial supervision and regulation?", BIS Working Papers, no 128, Basel, February 2003; Borio, Furfine and Lowe, "Procyclicality of the financial system and financial stability: issues and policy options", BIS Papers, no 1, 2001; and Borio and White, "Whither monetary and financial stability? The implications of evolving policy regimes", BIS Working Papers, no 147, February 2004.

9 See, for instance, Financial Stability Forum, Report on enhancing market and institutional resilience, 7 April 2008, and speeches by Ben Bernanke and Mario Draghi at the annual conference at Jackson Hole, Wyoming, sponsored by the Federal Reserve Bank of Kansas City, 22 August 2008.