International governance for the prevention and management of financial crises

Dinner speech by Mr William R White, Economic Adviser and Head of Monetary and Economic Department of the Bank for International Settlements, at the Bank of France International Monetary Seminar on 'Liquidity Crisis, Capital Crisis?', Paris, 10 June 2008.


Financial crises come in many shapes and sizes, but are increasingly international in their scope. Crisis prevention to date has largely been based on a bottom-up approach which tries to identify vulnerabilities with respect to each of the major pillars that make up the international financial system: financial institutions, markets and the supporting infrastructure. Recently, however, there has been growing interest in a complementary top-down approach (a "macrofinancial stability framework") which focuses on leaning against credit excesses which can lead to destabilising behaviour in both the real and financial sectors. Since crisis management and crisis resolution (where unmanageable debts are restructured) face so many difficulties at the international level, it is concluded that, as Churchill said of democracy, crisis prevention seems the best of all the bad alternatives available.

1. Introduction

Let me begin by thanking Gérard Béduneau for the opportunity to contribute to this Bank of France International Monetary Seminar. The BIS has both participated in and contributed to them for some years, and we have found both experiences useful. Gérard has asked me to talk briefly tonight about international governance arrangements for the prevention and management of financial crises. In effect, he wants me to address the issue of whether the current international financial architecture is adequate. My answer must be that it is not. Financial crises are generally enormously costly, and we have been faced with them repeatedly over the course of the last few decades. Indeed, we are currently in the midst of a crisis which, being at the heart of the global financial system, could turn out to be the worst of them all. What have been the shortcomings of the international governance process, and how might they be improved?

Evidently, financial crises are the by-product of liberalised financial systems. Let me state up front that the answer is not simply to re-regulate and to repress the financial system, as was the case in many countries for many decades after the Second World War. Rather, what has to be done is to find ways to address shortcomings in the current system, while retaining the bulk of the benefits provided by liberalised economic and financial systems. In effect we are still looking for Keynes's "Third Way", a balance of forces to stop an inherently good capitalist system from destroying itself in the end through an inherent tendency to excess. In my talk tonight, I will focus mostly on the shortcomings of the current system, rather than suggesting concrete solutions. As a matter of logic, solutions can be proposed only after problems or shortcomings have been identified.

My presentation will be in four parts. First, let me say a few words about different kinds of international financial crises. The fact that they can be quite different in their origins and dynamics makes preventing and coping with them much more complicated. In a second part, I will talk about international steps to prevent crises, and then in a third part about how to manage them. Finally, let me say a few words about crisis resolution, which I take to mean procedures either for reducing debt (for that is normally at the heart of the problem) to manageable levels, or for invoking bankruptcy proceedings so that the economic and financial system can restart unencumbered.

It is important to note that all public policy actions of the sort I will be talking about involve some degree of moral hazard. This should not be an impediment to action. The secret is to rely on the least hazardous and costly policies ex ante, to avoid the inevitable recourse to the most hazardous (and costly) ex post.

2. Various kinds of financial crises, each with an international dimension

I define an international financial crisis as one likely to have important international implications. It is the fact that these spillover elements exist, and that they can be very damaging, that provides the justification for saying there should be some element of international governance to both prevent and deal with such crises. Recognition of such interrelationships goes back at least as far as the establishment of the gold standard, which was a system designed to ensure that international trade imbalances were re-equilibrated before they got totally out of hand. The Bretton Woods agreements constituted another attempt to manage our inherent interdependencies.

It is useful as an analytical device to distinguish between various kinds of international financial crises. However, it should also be noted that crises can start in one way and then quickly transform themselves into something quite different. In addition, the classifications below are by no means watertight. The upshot is that every crisis will be similar to others in certain respects but different in others. This said, consider the following classifications, along with references to previous financial crises that clearly had international implications:

  • Operational disruptions and institutional insolvencies: the problem of large and complex international institutions. Think of what might have been the implications had Bear Stearns been allowed to fail in a disorderly way.
  • Short-term market disruptions in which markets seize up: markets are mostly international (unlike financial institutions). Think LTCM and the recent financial turmoil, especially in the interbank term market.
  • Medium-term misalignments and systemic vulnerabilities: sovereign debt crises and credit-driven bubbles in the private sector which eventually turn to bust. In most cases these crises were also characterised by large trade deficits and substantial international capital inflows. Think of the Mexican (1994) and Asian (1997) crises and the US housing and structured product crisis which is still playing itself out.
  • Contagion across countries and markets: almost international by definition. Think of how the Russian default led to the demise of LTCM and then the Brazilian crisis, to say nothing of the speed with which the US subprime crisis manifested itself in the essential paralysis of the interbank term market.

Moreover, the ongoing structural change towards securitisation (markets are international), consolidation (big banks are international) and globalisation (by definition) means the spillover effects from initially localised crises are getting bigger and bigger. By the same token, the need for an effective international governance process, run by all those likely to be affected by such crises, is also growing.

3. Crisis prevention

International efforts directed at crisis prevention to date have been largely focused on the health of the financial system, and have been based on the principle of national sovereignty. That is to say that the procedures suggested for international implementation have been agreed to by committees of national "experts", working to produce "a level playing field". Their recommendations, however, have only the moral force of "soft law", at least until incorporated into domestic legislation.

This work is essentially based on a bottom-up approach which tries to ensure the good health of the various pillars that make up the international financial system. In this regard, various groups of national experts set standards for their constituents. Three of the longest-standing are hosted by the BIS. They are the Basel Committee on Banking Supervision (banks), the Committee on the Global Financial System (markets) and the Committee on Payment and Settlement Systems (infrastructure). More recently, the International Association of Insurance Supervisors and the International Association of Deposit Insurers (both also Basel-based), and a wide variety of other cooperative bodies, have also devised standards, many of which have been designated as of particular importance by the Financial Stability Forum. In general, the IMF leaves standard-setting to such bodies, although both the Fund and other international institutions do also suggest standards occasionally. The IMF has in recent years carried out Financial Stability Assessments in most member countries. These not only help to spread and enforce such standards, but the Fund then iterates with the standard setters when they see shortcomings in what has been suggested. This process seems to have worked very well to date.

All of the groups mentioned above also do surveillance work directed at identifying shortcomings in the system and the possible impact of structural changes. For example, the CGFS had already published two studies laying out the dangers associated with credit risk transfer instruments before the recent crisis broke. Unfortunately, the risks identified were not thought significant enough by the private sector to elicit any real changes in behaviour. More recently, at the invitation of the G8, a working group of the Financial Stability Forum has made a large number of specific recommendations in the wake of the crisis, as to how revealed deficiencies in the financial system might be addressed.

In recent years, there has been growing interest in a more top-down approach that emphasises the macroeconomic roots of most national and international economic crises. The need for sensible fiscal policies has been understood for many years, with the IMF often playing the role of enforcer, particularly in emerging market economies. This was based on the observation that so many of the crises in the 1960s and 1970s were associated with fiscal excess. Similarly, in the light of more recent crises, growing attention is now being paid to how credit growth fosters procyclicality in the economic and financial system, through its influence on asset prices and spending behaviour. This top-down approach notes that "booms" commonly end in "busts", often made worse by serious impairment of the financial system. This commonly, although not universally, then leads to credit rationing effects that provide still more negative feedback to the real economy. Harkening back to the work of Hyman Minsky, this approach also notes that apparent liquidity crises are commonly based on an underlying concern about counterparty risk. This concern builds up over time as credit standards slip during the "boom", but then manifests itself suddenly in a "Minsky moment" of recognition and recoil. The "bust" is on.

To deal with such problems at the national level, suggestions have been made (by me and my BIS colleagues in particular) for the establishment of a new macrofinancial stabilisation framework. This would involve paying much greater attention to systemically important institutions, and to the interactions between financial institutions and markets which render virtually everything endogenous and ultimate outcomes highly uncertain. The macroprudential and monetary policy instruments used by regulators and central banks respectively would also be used in a much more symmetric way over the cycle, contributing to the pursuit of both price stability and financial stability. Finally, the central banks and regulators would have to have much more profound interactions than at present, both to assess the build-up of problems and what might be done to resist them.

To date, however, these heightened macroeconomic concerns have not elicited any serious attempts to establish an international framework for trying to resist procyclicality. On the one hand, this might seem seems odd given the accumulating evidence that problems which arise in one country are often reflected in others. In particular, efforts to peg exchange rates to countries in the "boom" stage of the cycle will generally import similar symptoms to the pegging country, though here perhaps the simplest answer is to cut the peg. On the other hand, international cooperation can only happen if the important countries share the same assessment (paradigm) of the underlying problem. At the moment, there is no such shared assessment, with the Europeans and Japanese being generally more inclined to worry about the monetary and credit cycle, but the United States and some other English-speaking countries rather less concerned. The US authorities, in particular, have been adverse to the suggestion of resisting "bubbles", preferring for various plausible reasons to clean up the debris afterwards.

Further, among those who share a similar view of the underlying problem, there is still no agreement as to when or how countervailing action might be taken, even at the national level. For example, views differ as to whether it would be technically possible to gross up (under Pillar 2) the capital requirements estimated under Pillar 1 of Basel II to reflect system-wide variables. At the international level, these kinds of problems become even more difficult. Finally, there is the "will to act" problem. Policymakers at the national level have explicit responsibilities and accountabilities, yet for all sorts of reasons are still often hesitant to tighten. This would be even more the case in international fora where no explicit international mandate exists. This said, John Taylor has recently suggested an increased degree of international monetary cooperation to resist global inflationary trends, noting that many domestic authorities treat food and energy prices as "external" even though they are fully "internal at the global level". This kind of thinking gives hope for enhanced international cooperation with respect to other kinds of macroeconomic problems as well.

This evidently leads on to the question of whether more authority should be shifted from the national sovereigns to some international agency that did have an explicit mandate for crisis prevention. Suggestions have been made that the BIS should become a kind of "super-regulator". Suggestions have also been made that the IMF's role in crisis prevention be enhanced. In particular, the Fund should more actively encourage exchange rate changes to avoid exchange rate crises. While a far cry from suggestions around the time of Bretton Woods that the Fund take overall responsibility for the growth of the world money supply, it goes in the direction of increasing the IMF's powers. It has also been suggested in recent years that the Fund play a bigger role in assessing when countries' debt positions (both internal government and net external debt) have become dangerously high. None of these suggestions has so far come to much.

Why has there been so little movement on this front? First, there is the difficult issue of legalising the mandate. Changes to the Articles of established institutions like the IMF and BIS are not easy and would take ages. Creating new institutions might be even more difficult. But perhaps the most difficult issue is that the countries exercising their sovereign power simply do not wish to give it up. This applies to small countries (think of the "chairs and shares" issue at the IMF) as well as large. In particular, the United States still wields a significant degree of influence behind the scenes, at the IMF and the World Bank in particular, which it does not wish to see diluted. As changes in economic and political realities impose themselves, in particular the shift of economic power and influence towards emerging market economies, there might eventually be a greater willingness to rethink what all our international institutions are for, and also how they should properly be governed.

4. Crisis management

Crisis management should be preceded by steps (and well in advance) which facilitate it. At the national level, these would include: evaluating the adequacy of deposit insurance schemes, setting up "shelf banks" to allow the crucial functions of banks to be maintained even if the bank stops operating under its original charter, ensuring that the authorities have adequate powers to do what needs to be done, Memoranda of Understanding between the relevant authorities, and finally war games played by those who would actually manage problems in real time. Recently in the United Kingdom, a highly sophisticated and advanced country by anyone's standards, shortcomings in each of these regards became all too evident and had to be quickly addressed.

Taking the international dimension into account, each of the above steps should be evaluated to ensure some rough consistency, or at least no obvious inconsistencies, between the practices of the largest countries at least. A preliminary evaluation indicates that much of the preparatory work to facilitate management of an international financial crisis has not yet been carried out. In particular, domestic legislation in some countries still precludes sharing information that both home and host supervisors and their respective central banks would need in such a situation.

International cooperation to manage financial crises as they occur has taken diverse forms depending on the character of the crisis. For sovereign crises, including those of the "boom-bust" sort noted above, the IMF has always been at the centre, although it has not generally applied a single crisis management template in every case. In part, this reflects the fact that all crises are different to some degree, but it also seems to reflect the influence of important members in guiding the decisions to be taken. These "idiosyncratic" responses of the Fund eventually led to it being subject to increasing criticism. In the Mexican crisis, and even more so in Asia, the Fund's support for troubled countries reached (for the first time) many multiples of the affected countries' IMF quotas. This led to a big debate among the major members of the Fund as to whether this trend should be rolled back or not. In the event, so many countries have now built up massive foreign exchange reserves that the need for liquidity support from the Fund has now been greatly attenuated.

As for other forms of crises, the obvious kind to highlight in current circumstances would be those characterised by short-term disruptions in financial markets. Since the freezing-up of the interbank term market last August, a very wide range of steps to restore liquidity have been taken in a highly cooperative way by central banks in the major financial sectors. However, these measures have not in fact been as successful as many had hoped, since that market still does not function properly. Conversely, and perhaps due in part to central bank actions, a number of other markets which were for a time dysfunctional do seem to have begun to function more normally. At the present juncture, these measures are thought to have at least "bought time" to evaluate the seriousness of the prospective losses at a number of the large international financial institutions and to determine what needed to be done in response.

Whether this "time" has been well used remains debatable. In spite of a call by the Financial Stability Forum for an international template to deal with "fair value" losses related to the subprime mortgage market in the United States, there is still an enormous amount of uncertainty about the international comparability of some of these valuations. This also leaves open the question of the comparability and the adequacy of the capital of the large, internationally active banks. In this context, two problems arise. On the one hand, the uncertainty may lead to an inadvertent degree of regulatory forbearance in closing down affected institutions in an orderly way; in effect, insolvent banks might for a time be deemed illiquid. Historically, forbearance of this sort leads to far larger losses in the end. On the other hand, this same uncertainty might lead private sector counterparties to decide to cut previous relationships; the sort of behaviour that led to the shockingly sudden demise of Bear Stearns.

It is worth noting that, in cases where crisis management is rendered difficult for whatever reason, there will always be a temptation to ease the tensions by lowering interest rates. This has happened many times in the past in virtually all the major developed countries. In most instances, such policies have in fact proved effective in maintaining financial stability and restoring growth prospects. The logical problem, however, is that more credit cannot forever be the answer to problems brought on by excessive credit growth in the first place. The eventual outturn must then be a still larger crisis. This logic leads to the conclusion that the price of not improving both crisis prevention and crisis management could eventually prove to be very high.

5. Crisis resolution

I have defined crisis resolution as processes either to lower debts to sustainable levels or to invoke bankruptcy procedures. Evidently this raises the same question (illiquid or insolvent?) just noted above. Moreover, how can one judge whether there would be residual value from writing off debt and how much debt reduction would be required to do the trick? Valuing complex structures has received a lot of attention recently, but the problems run far deeper than that.

For a sovereign debtor, for example, a sustainable debt level will depend on assumptions about prospective economic growth rates and the levels of other government expenditures (on both human capital and infrastructure) required to support such growth. The evolution of the global economy and terms of trade shocks will also be important considerations affecting the answer. For banks, assumptions about losses and profits going forward, and in turn the viability of the chosen business strategy, will all be crucial inputs for determining what needs to be done to the institution to either revitalise it or kill it. These assumptions will also be crucially dependent on what happens to important macroeconomic variables affecting corporate and household bankruptcies. For a current example, the viability of many banks, not just in the United States but worldwide, now depends on when and at what level house prices stabilise in the United States. Also relevant is the question of how many homeowners will put their house keys in the mail (jingle mail) in the interval, or declare bankruptcy themselves.

In recent years, new financial developments have made the problem of achieving an effective debt reduction even more complicated. In contrast to earlier sovereign debt crises (like Mexico in 1994 and, to a lesser degree, Asia in 1997), there are now millions of troubled borrowers (in particular US households) and a myriad of lenders. Moreover, given the growth of credit risk transfer instruments, the interests of investors are no longer aligned in preferring "half a loaf" to "no loaf" and seeking to avoid bankruptcies. On the contrary, those who hold debts against which they have taken out default insurance actually gain in the event of default. In sum, orderly private sector workouts are going to be even more difficult going forward.

As for invoking bankruptcy procedures, a process which applies to private sector entities and not sovereigns, there are again major problems at the international level. Not least, there is no international bankruptcy law nor other agreed procedures for winding down a large, complex international financial institution. This has been known since the winding down of BCCI in the early 1990s, but not much has been done about it. This means that, in such an event, national authorities are very likely to ring-fence the assets in the hands of their nationals, thus impeding an orderly international solution. As to why there has been no action to resolve this issue, national authorities in the big financial centres wish to preserve their option to ring-fence, and thus there are no important champions for change.

6. Conclusions

Our current circumstances are what they are, and we must manage and resolve the current crisis as best we can. I leave it to others in other fora to suggest how best to do this. However, along with these challenges come opportunities. History indicates that it is precisely in such circumstances that the political will can be mustered to make changes that will be helpful in preventing similar crises in the future. While being careful not to move too hastily and imprudently, measures to improve our capacity to prevent crises should also be encouraged at the present moment. This is all the more so given the difficulties I have just described in improving our capacity to manage and resolve crises. Crisis prevention may then, like democracy, be the best of all the alternatives available.