## Philipp M Hildebrand: Policy implications of the financial crisis
Speech by Mr Philipp M Hildebrand, Vice-Chairman of the Governing Board of the Swiss National Bank, at the Conference, 'From Fragility to Stability', University of Geneva, 18 November 2009.
I would like to thank Jürg Blum and Rita Kobel for their valuable support in drafting this speech. I also want to thank Nicole Allenspach, Philipp Haene, and Signe Krogstrup for helpful comments and discussions.
* * *
## Introduction
It is an honour to speak at the University of Geneva this evening. I want to thank Rajna Gibson for her kind invitation.
A year ago, we were in the midst of a perfect financial storm. Following the bankruptcy of Lehman Brothers in mid-September, the imminent collapse of the global financial system became a distinct possibility. Since those dark days, the situation has manifestly improved. A bold and unprecedented global policy response has had its intended short-term effect. While important questions remain as to the sustainability of the global recovery, we are no longer staring into the abyss of financial and - potentially - economic chaos. The worst financial crisis for at least 70 years has not resulted in a new Great Depression.
On the other side of the ledger, the tangible and intangible costs of this crisis are enormous. The potential costs of the support measures taken - capital injection, asset purchases, and guarantees of bank debt - in the G7 countries together with Australia, the Netherlands, Spain and Switzerland amount to about twenty per cent of GDP in these economies. 1 Moreover, the consequences of the crisis in terms of job, output and wealth losses, unprecedented increases in public spending and declines in public revenues are enormous. In the end, the citizens of the individual countries involved will have to foot the bill, one way or another. Arguably the most important intangible cost, however, is the fact that governments had to issue an explicit guarantee for systemically relevant institutions, in effect executing a partial government take over of the financial system. 2
Going forward, the world of finance will not be the same anymore. It should therefore come as no surprise that at the national and international level intensive efforts are underway to increase the resilience of the financial system.
In recent months, these international reform efforts have included the presentation of many reports and recommendations. The Financial Stability Board (FSB) has published two comprehensive sets of recommendations to improve financial regulation. 3 Both have been endorsed by the G20 leaders. The public sector has not been alone in issuing
1 The outlays totalled eight per cent of GDP. These figures relate to government support measures introduced between September 2008 and June 2009. Source: 'An assessment of financial sector rescue programmes', BIS paper No 48 , July 2009. In Switzerland, the potential costs of capital injection and asset purchases amounted to about nine per cent of GDP. While the Swiss government has been able to sell its stake in UBS with a profit, the SNB exposure in illiquid assets taken over from UBS still amounts to approximately USD25 billion.
2 'We agree to take decisive action and use all available tools to support systemically important financial institutions and prevent their failure.', G7 Finance Ministers and Central Bank Governors Plan of Action, 10 October 2008.
3 The recommendations and principles to strengthen financial systems, published on 2 April 2009 (based on the 'Report of the Financial Stability Forum on Enhancing Market and Institutional Resilience', April 2008), and 'Improving Financial Regulation: Report of the Financial Stability Board to G20 Leaders', 25 September 2009.
recommendations. 4 Individual firms and industry groups, most notably the Institute of International Finance (IIF), have become increasingly vocal in stating what they consider to be appropriate and inappropriate changes to the global financial regulatory framework. 5
In short, good progress is being made on many issues. In particular, some of the largest banks have begun to reorient their business models and operating procedures. But many implementation challenges - in regulation as well as in business strategy - lie ahead of us. Moreover, given the proliferation of ideas put forward by both the public and the private sector, it is increasingly difficult to keep track of competing recommendations. And perhaps, most worryingly, the momentum for reform appears to have slowed, partly because markets have stabilised, partly because the public is losing interest and partly because, as Jerry Corrigan recently said, 'the subject matter of reform is so very complex and often controversial.' 6
In view of the complexity of the issue, allow me to lay out the regulatory reform priorities which the Swiss National Bank (SNB) believes we must focus on in our efforts to build a more robust financial system for the future. Fully in line with proposals by the FSB and the decisions already taken by the G20 leaders, we must pursue a dual-track approach to reform, combining preventive measures and resolution measures:
The preventive measures aim at reducing the probability of bank failures and systemic crises. Note that I deliberately use the word 'reduce'. Eliminating all instability would be neither feasible nor desirable.
The resolution measures aim at reducing the negative consequences in the event that failures or systemic crises occur - despite hopefully greater resilience in the future.
In what follows, it is useful to bear in mind the extraordinary situation we face here in Switzerland. In international comparison, our banking sector is very large relative to our economy. Currently, total assets of the Swiss banking sector exceed seven times GDP. Furthermore, the Swiss banking sector is highly concentrated. Based on total assets, the two big banks, UBS and Credit Suisse, have a market share of more than two-thirds. These exceptional orders of magnitude clearly necessitate prudent decision-making when determining the appropriate regulatory framework for banks in Switzerland.
## Preventive measures
Let me first turn to the preventive measures. These measures should focus on strengthening the shock absorbers of the financial system and thereby making the system more resilient. In the context of the most significant banks, this essentially means that they have to hold substantially more and better quality capital and liquidity.
## Strengthening capital requirements
As we all know, the crisis has revealed fundamental weaknesses in existing capital requirements.
4 Other important reports include 'Financial Reform: A Framework for Financial Stability' by the Group of 30, and 'The Global Financial Crisis: A Plan for Regulatory Reform', Committee on Capital Markets Regulation, May 2009.
5 'Restoring Confidence, Creating Resilience: An Industry Perspective on the Future of International Financial Regulation and the Search for Stability', Institute of International Finance, July 2009.
6 Speech by E. Gerald Corrigan, Managing Director at Goldman Sachs, at the Presidential Symposium, University of Rochester, 10 October 2009.
BIS Review 145/2009
To address these weaknesses, a considerable amount of work has already been done and is still underway in the FSB and, crucially, in a number of working groups of the Basel Committee on Banking Supervision (BCBS). 7 In line with, and in full support of these efforts, the Swiss authorities are convinced that a more robust capital framework needs to be built around the following features:
First, the amount and quality of capital need to be increased very substantially. In the medium term, this will be feasible without giving rise to drastic adjustments at banks that might be harmful to the overall economy. The key to this will be a sufficiently long time horizon and, crucially, more discipline in retaining earnings. The outlook for future earnings is by definition uncertain. Nonetheless, if we review the banks that have received public support, it is striking that many of them paid out more in dividends and share buybacks during the years preceding the crisis than they subsequently faced in losses.
Second, as a supplement to the risk-based capital requirements, a simple, commonly and coherently defined leverage ratio restriction needs to be introduced, of the sort already applied in a number of major financial centres. A leverage ratio prevents the build-up of excessive leverage and serves as a backstop to the complex, but fallible risk-based capital requirements. Fundamentally, anybody who claims that the leverage ratio is a useless or even a counterproductive complementary regulatory instrument is arguing that gross exposure doesn't matter. If there is one lesson we should learn from the crisis, it surely is that gross exposure manifestly matters a great deal.
Third, to address procyclicality, banks will have to build up capital buffers above the minimum requirements in good times. In difficult times, banks will be allowed to fall significantly below the target levels defined for good times. Allowing banks to draw down capital without violating any minimum requirements should help mitigate the potentially harmful effects of de-leveraging. 8
In Switzerland, compatible with these features, FINMA issued decrees imposing higher capital requirements on the two big banks at the end of last year. According to these decrees, the big banks will have to meet risk-weighted capital requirements that are double what they were before the crisis. As a complement, FINMA has introduced a leverage ratio. In good times, the capital base has to account for at least 5 per cent of the total adjusted assets. 9 To prevent a procyclical impact, these targets will apply as of 2013 at the earliest, giving banks enough time to recover from the crisis. Furthermore, the big banks will be allowed to temporarily fall short of these targets in bad times.
The Swiss experience suggests that the largest banks can adjust rapidly, even to ambitious new capital requirements, without causing collateral damage to the overall economy. To reduce uncertainty and to facilitate the planning process at banks, it is crucial that the ultimate target levels are clearly communicated, and that a sufficiently long phase-in period is provided for. This will enable banks to adjust to the new rules in an orderly fashion.
One interesting question is whether contingent capital, i.e., debt that is converted to capital if a particular event is triggered, can be devised in such a way as to enhance capital standards. There may well be a role for contingent capital, in effect defining in advance a debt-to-equity swap. In order for such a contingent convertible capital structure to be effective, it will be
7 It will thereby be important to carefully assess the combined effect of these changes. To this end, the Basel Committee will conduct a quantitative impact study in 2010, before determining the final calibration of the improved requirements.
8 To reduce the procyclicality of capital regulation, promoting more forward-looking provisions is an additional promising avenue to pursue. In this area, progress will depend a great deal on the accounting standard setters. The Basel Committee is working closely with accounting standard setters on this issue.
9 In the case of Switzerland, several adjustments are made to total assets. Most importantly, the domestic lending business is excluded.
crucial that the converted capital be equivalent to true, loss-absorbing core capital and incorporate a precise definition of the event that triggers conversion. The Basel Committee and the FSB have committed themselves to examining this further. 10
## More robust liquidity requirements
The crisis has also provided a number of important lessons regarding liquidity. On the whole, banks' liquidity holdings were clearly insufficient. This holds true for the quantity but, crucially, also for the quality of liquidity.
One of the explanations for these insufficient holdings of liquidity was that the stress scenarios considered by banks were far too optimistic. While secured funding remained the most stable source of refinancing, it was much less stable than banks and regulators had assumed. 11
Liquidity problems at individual banks have imposed considerable stress on the entire international system. To prevent a destabilization, central banks had to provide massive liquidity support. In some countries, this support has been provided against collateral of poor quality and even on an unsecured basis.
As in the area of capital, the Basel Committee is moving rapidly towards an international liquidity standard for banks. This standard should incorporate the following basic features to be effective:
The standard should reflect an adverse scenario, including a significant loss of confidence from depositors and a severe disruption of secured funding. Moreover, the standard should require banks to hold a buffer consisting of assets whose liquidity and value is robust to massive disruptions in the financial markets. Overall, the standard should materially improve banks' liquidity positions and make their funding more robust.
In Switzerland, liquidity requirements for the big banks are in the process of being comprehensively revised. In line with the features I have just outlined, the new regime will ensure that the big banks are able to cover their potential liquidity needs in the event of a widespread loss of market confidence. The new regulation will enter into force in spring 2010.
## Compensation
With the preventive measures I have just mentioned it will also be necessary to strengthen the incentives for prudent risk taking within financial institutions. Compensation systems must not remain the one-way street they were for many individuals and firms in the run up to the crisis. They must become risk aligned and long-term oriented. Banks and regulators must not let the returning competitive pressures for competent staff erode the momentum toward this goal. This is why we need a strong and broad international commitment to implement the FSB Principles for Sound Compensation and the corresponding FSB Implementation Standards at national levels, in a manner that is consistent across national borders. 12
10 See also Squam Lake Working Group on Financial Regulation, 'An Expedited Resolution Mechanism for Distressed Financial Firms: Regulatory Hybrid Securities', April 2009.
11 During the crisis, the volume and maturities of repo transactions decreased sharply.
12 A number of countries, including Switzerland, have already passed national norms in response to the FSB standards. The FSB will conduct a peer review by spring 2010.
BIS Review 145/2009
## Facilitating the orderly resolution of banking problems
While we can reasonably expect that preventive measures will strengthen the resilience of the financial system, we cannot ignore the problem of bank failures. Even with much improved shock absorbers, large and systemically important banks will again experience severe financial stress at some point in the future. Here we must accept that we still have not dealt with the fundamental problem that systemically important banks cannot currently be allowed to fail. The truth is that, if tomorrow morning a systemic institution were once again to find itself on the brink of failure, we would again face the terrible choice between either coming to its rescue at the expense of taxpayers or putting the stability of the financial system at risk. Moreover, as long as banks can expect public support, market discipline will always be undermined.
Moral hazard has been exacerbated by the fact that there is no longer any ambiguity about the willingness of the public sector to bail out large banks. In other words, we must acknowledge that the wide-ranging public support measures have aggravated the moral hazard problem associated with large and complex financial institutions.
Frankly speaking, we have no choice but to address the specific challenges of banks that are 'too big' or 'too interconnected' to fail. If we are committed to a market-based system, the financial system of the future must expose financial institutions of all sizes and structures to the ultimate test of the market place. The very definition of a market economy is that it must allow for failure as a sanction of excessive risk taking or managerial incompetence. In the event that large, systemically relevant financial firms face the threat of failure in a next crisis, the financial system of the future must allow for their orderly resolution. Such a system needs to ensure that failure of a large bank does not have serious negative consequences for the provision of financial services to the real economy.
There are several possible routes to achieve this goal. We need to examine all the available options with an open mind. The optimal policy is likely to combine different elements.
As I explained in the introduction to this speech, here in Switzerland the problem of 'too big to fail' is particularly pronounced. Accordingly, the FINMA and the SNB are jointly evaluating different avenues. The two big banks are now closely involved in this work. We welcome their active engagement. At the international level, the FSB and its member bodies are also addressing this problem at the request of the G20 leaders. Ladies and Gentlemen, it is imperative that a real solution to this problem be found.
At the forefront of our efforts to mitigate the 'too big to fail' problem must be an internationally agreed and orderly process to allow for the wind down of large, systemically important financial institutions in the event of a severe crisis.
Obviously, there are many technical and legal problems that have impeded any meaningful progress in reaching international coordination in this matter. But what has been missing is a bold and international political commitment to put in place a framework for the orderly resolution of large cross-border financial institutions. Provided we have such an unequivocal commitment, solutions will eventually emerge.
Needless to say, we must accept the reality that different national resolution regimes will continue to coexist. Let me be very clear on this: A full-fledged global resolution regime is a noble but seemingly unrealistic goal. We should not conclude from that, however, that the framework for cooperation across countries cannot be improved. One possibility is to work towards mutual recognition arrangements of compatible national resolution regimes. Recognising other resolution regimes helps each nation to adapt its own structures and processes in order to reduce frictions in crisis episodes.
We should require that firms more clearly delineate different functions within their internal structures and simplify those structures. This will help facilitate an orderly wind down in the event of a severe crisis. It will probably also enhance effective management and risk management in these firms. This is an important point. As the crisis has demonstrated, in a
number of cases it can be argued that firms were not only too big, but also too complex to be managed effectively.
Furthermore, we need a financial market infrastructure which - to the extent that this is possible - minimises the destabilising effects of a failure on the rest of the financial sector. A key element that cannot be emphasised enough is the ongoing effort to centralise the clearing and settlement processes in the derivative market. This enhances transparency and reduces counterparty risk. Opposition to infrastructure reform by those firms that have in the past benefited from complexity and lack of transparency must be firmly overcome.
At the same time, we need to keep in mind that central counterparties themselves can become single points of failure, with potentially devastating consequences. Central counterparties must therefore have sufficient financial resources to withstand the default of the largest participants. Operators and regulators should carefully evaluate the adequacy of the risk management models and financial resources of central counterparties. Only if central counterparties are financially sound can they play their critical role in preserving financial stability.
Finally, as an ultimate resort, we cannot ignore the need to consider measures that address the size of banks, indirectly or directly. 13 Admittedly, there are some benefits from size. But the empirical evidence suggests that these benefits are rather limited. Economies of scale, whereby larger amounts can be produced at lower average cost, already appear to have been exhausted in the case of relatively small banks. 14 Obviously, banking systems made up of only small banks are not immune to crisis episodes. But let us not make the mistake of using another problem as an excuse to shy away from the severe moral hazard problem associated with the largest banks. It is also sometimes argued that interconnectedness matters more than size when it comes to systemic importance. It is evident that both size and interconnectedness matter. Recent empirical evidence, however, indicates that systemic importance increases more than proportionately with size. 15
The debate about the plans to split up some very large European banks demonstrates the difficulty of measures aimed at size. At the same time, the debate illustrates that such measures are not impossible. The extent to which we are required to travel down this difficult path will depend a great deal on how much genuine progress we are able to make on the reforms I have tried to outline this evening.
## Competitiveness
All the measures to improve financial regulation which I have discussed tonight raise the delicate issue of banks' competitiveness. Of course, it is not the intention of any regulator to hurt banks. However, somebody has to bear the costs of preserving financial stability. Up to now, through the implicit and now explicit government guarantees, taxpayers were bearing a considerable part of these costs. By strengthening regulation, these costs are to be shifted to the banks. In economic terms, the new regulatory framework endeavours to internalise the externalities caused by banks.
13 See, for instance, Group of Thirty, Financial Reform: A Framework for Financial Stability, p. 59: 'To guard against excessive concentration in national banking systems, with implications for effective official oversight, management control, and effective competition, nationwide limits on deposit concentration should be considered at a level appropriate to individual countries'.
14 Cf. for example, Y. Altunbas et al., 'Efficiency in European banking', European Economic Review 45 , 2001, pp. 1931-1955.
15 Cf. N. Tarashev, C. Borio, and K. Tsatsaronis, 'The systemic importance of financial institutions', BIS Quarterly Review , September 2009, pp. 75-87.
BIS Review 145/2009
The inconvenient truth is that effective regulatory reform inevitably imposes some costs on banks. In the words of the late Milton Friedman, 'there is no such thing as a free lunch'. The current crisis of the global financial system was not an accident. It was the result of optimising behaviour on the part of market participants. Given the rules of the game, they were maximising short-term profit. It is therefore too simplistic to vilify bank managers. With respect to regulation, however, this implies that if we do not change the rules of the game, which guide this optimising behaviour, we will sooner or later end up in a similar situation again.
Costs are only one determinant of banks' competitiveness. Safety, soundness, and trustworthiness are other important factors. Hence, if we view competitiveness from a broader perspective, strengthening financial regulation ultimately has some benefits for banks, too. While some business areas may become less attractive, others will gain. In Switzerland, for instance, where wealth management forms the backbone of the banking industry, financial stability is a crucial factor for success. In this context, it is interesting and certainly promising to note that the two big Swiss banks now emphasise their strong capital position as an important competitive advantage.
Nevertheless, to be on the safe side regarding a level playing field, ideally all regulatory reform should be internationally coordinated. To this end, the SNB is investing very substantial resources to contributing actively to international regulatory reform projects, especially within the FSB and the Basel Committee. Indeed, a main raison d'être of these international bodies is to ensure a level playing field. We must accept, however, that not all countries are confronted with the same urgency for reform as we are in Switzerland. As I have already explained to you this evening, given the size and importance of our banking sector, our country is particularly vulnerable to a financial shock. Agreeing on an international common denominator of regulatory reform may turn out to be insufficient for the Swiss case. Ultimately, it will be a political decision to choose the level of risk Switzerland is willing to accept, both for the financial and for the real sector. Given the particular situation in Switzerland, higher-than-average regulatory standards are warranted.
## Conclusion
The worst of the crisis is behind us, and intensive efforts are underway to increase the resilience of the financial system. Banks are again generating profits. In some cases, these profits are very substantial and are clearly linked to the public support measures, many of which are still in place.
As the situation improves, complacency can easily become the rule of the game. We will forget the severity of the crisis and fall prey to lobbying by a powerful and recovering financial industry.
Ladies and gentlemen, we must not let this happen. The stakes are simply too high.
Equally, the most prominent banks should invest their efforts, not in resisting fundamental change, but in genuinely supporting the work to bring it about, in the interests of a more longterm and less volatile global banking and, ultimately, economic growth model.
If we fail to reform fundamentally the financial system and the position of the largest systemically relevant firms in that system, we will undermine one of the most basic mechanisms of any market economy, namely that of punishing failure. What is more, we will - to quote Mervyn King - bequeath to future generations a serious risk of enduring another crisis even worse than the one we have just experienced. 16 My biggest concern is that a future crisis on a larger scale could create such an immense regulatory and protectionist
16 Speech by Mervyn King to Scottish business organisations, Edinburgh, 20 October 2009.
BIS Review 145/2009
backlash, that the very market-based system, on which so much of our current welfare is based, will be threatened. In that sense, it is in the long-term interest of banks themselves to support the far-reaching efforts to make the financial system more resilient. If our reform efforts falter, we risk having to look back one day at the vast and unprecedented stabilisation efforts undertaken by so many governments and central banks and conclude that they were short-sighted. For how can we justify the costs of these efforts, which will have to be borne for at least a generation, without fundamentally addressing the flaws that got us into this terrible crisis in the first place?
Ladies and Gentlemen, there are many areas in financial regulation that can and should be improved. Given what is at stake, there is a need to prioritise. I have attempted to lay out to you today where the SNB puts those priorities and where it will focus its greatest efforts in the months and years to come. Together with FINMA, it is our duty to provide robust analysis and to issue sound recommendations that will help put our financial system on a more sustainable foundation.
In certain areas, like capital requirements, regulators can issue - and have issued - decrees intended to strengthen the resilience of the financial system. With respect to the 'too big to fail' problem, political decisions are required. The topic is crucial. The consequences of not addressing it adequately could be severe in the event of a future crisis. Politicians need to become acquainted with and get involved in this matter. The sooner, the better! 17
17 The Swiss government has recently mandated a group of experts to make proposals to limit the risk associated with large firms.
BIS Review 145/2009