> > > >
Presented by Charles I. Plosser, President and Chief Executive Officer
Federal Reserve Bank of Philadelphia
For the New York University Conference "Restoring Financial Stability: How to Repair a Failed System"
March 6, 2009
PDF version (53 KB, 7 pages)
The financial crisis has underscored the need to reform the regulatory and supervisory architecture of our financial system. The importance of this undertaking, and doing it properly, cannot be overstated — nor should it be rushed. In general, I believe making major policy reforms in the "heat of battle" risks adopting policies that have unintended consequences. Such "quick fixes" may inadvertently hamper market competition or innovation and create conditions that provide the foundation of the next crisis. Moreover, without careful analysis, hasty reform may fail to address the heart of the problem. We must make sure that the major regulatory reforms we put into place will be as agile and dynamic as the financial system itself. The regulatory structure must foster financial stability without stifling financial innovation.
New York University's collection of thoughtful white papers on restoring financial stability covers many facets of regulatory redesign and makes an important contribution to the debate. Today, I would like to highlight three important considerations as we undertake the serious task of redesigning our financial regulatory structure. These involve addressing the too-big-to-fail and too-interconnected-to-fail issue; developing systematic and transparent regulatory approaches for dealing with firms that pose systemic risk; and addressing the Federal Reserve's future role in ensuring financial stability.
In the current crisis, we have seen that financial firms that become too big or too interconnected to fail pose serious problems for financial stability and for regulators.1 Due to the complexity and interconnectivity of today's financial markets, the failure of a major counterparty has the potential to severely disrupt many other financial institutions, their customers, and other markets. We have also seen that market discipline won't work when creditors and counterparties believe they are never at risk. The belief that regulators will bail out creditors of financial firms creates moral hazard that leads to poor risk-taking decisions and undermines the incentives for creditors to monitor these firms. Moreover, it creates incentives for financial firms to become too large and too complex to fail in order to exploit the implicit government guarantees.
During the past year, regulators were repeatedly confronted with the unpalatable choice of either permitting large financial firms to enter bankruptcy without an adequate resolution mechanism or taking unprecedented actions to limit potential systemic risks. The financial problems at Bear Stearns, AIG, and Lehman Brothers elicited different responses from government, which contributed to uncertainty. Arguably, this uncertainty in itself became a source of systemic risk.
To mitigate these risks, we must find a credible way to resolve these uncertainties and eliminate the need for ad hoc rescues by government agencies and the Fed in particular. One wrong-headed approach would be to erect a battery of new regulatory restrictions in an attempt to drive the probability of failure to zero. Such an approach would generate large supervisory costs, stifle innovation, and result in regulatory arbitrage as markets work to evade the regulations. This type of arbitrage was a contributor to the current financial crisis.
Instead, we must recognize that failures are an inevitable consequence of a dynamic financial system. Reforms should seek to reduce the cost of failures of systemically important financial firms, which would enable regulators to allow such firms to fail when appropriate. Market participants, believing such failures are possible, would exercise greater market discipline and help prevent financial firms from getting into trouble in the first place. An important step toward reducing the cost of financial firm failure is designing better bankruptcy procedures for financial firms, with particular rules for firms with systemically important operations. Regulators already have a procedure for systemically important commercial banks. The FDIC has the authority under FDICIA (the FDIC Improvement Act) to resolve a large bank failure by operating a bridge bank for a limited time, thereby reducing disruptions to the financial system while resolving the bank's problems. The bridge bank authority requires the FDIC to pursue the least cost resolution once systemic risks have receded. This means that common shareholders lose their investments and haircuts are imposed on uninsured creditors.
We need to develop a similar resolution procedure for nonbank financial firms that are systemically important. The existing bankruptcy law, which seeks to maximize the payoffs to the firm's creditors, makes no provisions for systemic considerations. Therefore, we need a resolution mechanism for financial firms that explicitly addresses the tradeoff between reducing financial disruptions and minimizing the costs to taxpayers.
Such a resolution mechanism would contain procedures for separating out the systemically important components of a financial firm. Of course, this would require regulators to identify these segments ex ante, an undertaking that requires planning.
The mechanism should also include rules for determining the timing of payments for uninsured counterparties and creditors of the systemically important segment. To reduce the potential for systemic disruptions, the resolution mechanism should provide for speedier restructuring of claims than is typical in bankruptcy. The restructuring rules should also be known to market participants to reduce uncertainty. Here, we might take another leaf from the FDIC's playbook for handling bank failures. The FDIC has often provided expedited payment for uninsured creditors, in contrast to the automatic stay in regular bankruptcy. These payments, which involve a haircut based on historical recoveries, mitigate the threat of a run, reduce the costs of failure for the bank's claimants, and impose market discipline.
Several open questions remain in the design of an effective resolution mechanism for financial firms. First, we must be more specific about what we mean by systemic risk. I believe this is harder than it sounds, but it has profound implications for the types of firms and activities that require oversight. Second, we must develop a systematic and dynamic way to determine which financial firms are or will become systemically important. And third, since many of these firms operate globally, we must work with our global counterparts to ensure international coordination of our resolution mechanisms. While these international issues can become quite complex, they should not stand in the way of developing an improved resolution mechanism for financial firms here in the U.S.
Of course there is another, perhaps oversimplified, solution to the too-big-to-fail problem and that is, if an institution is too big to fail — shrink it. Under this approach no firm would need to be saved due to the systemic risks inherent in its scale of operations. Unfortunately, in the current environment, consolidation in the financial industry is moving things in the wrong direction, and imposing size limits on financial firms would sacrifice efficiencies of scale. A better solution is perhaps to increase the costs on firms that pose systemic risks so that they will choose to limit these risks.
As I mentioned earlier, government policy actions during the past year have created some uncertainty about how nonbank financial failures would be handled, and arguably, this uncertainty contributed to the stress in the markets.
I have long advocated the benefits of conducting policy in a systematic way.2 I believe this approach applies to both monetary policy and policy toward promoting financial stability. Being systematic about when we intervene in financial markets, and in what way, will yield better regulatory outcomes by reducing uncertainty. Moreover, we must be credible and follow through once the systematic policy is articulated. Otherwise, the benefits will not be realized.
One systematic approach would be to extend the philosophy behind the prompt corrective action (PCA) provisions of FDICIA for commercial banks to other kinds of identifiable, systemically important financial firms. These firms should be subjected to greater regulatory oversight in general to reduce the probability of insolvency. But given the limits to regulators' ability to foresee and control events, we must also design regulations and mechanisms that seek to enhance the effectiveness of market discipline. We should also ensure that our regulatory interventions are as transparent and as systematic as possible, to help align market expectations with our actions and outcomes.
As before, the first step is to define, identify, and quantify systemic risk. Economists in academia and government have been working on practical methods for measuring systemic risk.3 Our goal should not be to find one all-encompassing measure of systemic risk but to develop a menu of useful indicators to guide regulators' attention to evolving problems. Information from securities markets, such as correlations among spreads on credit default swaps, can be useful. Regulators might expand the range of available market indicators by introducing new securities designed to aggregate market estimates of systemic risks. For example, academics have proposed using contingent capital securities4 or a market for insurance against capital impairment5 as possible supplements to regular capital requirements. The market prices of these instruments might provide regulators with useful signals of systemic stress.
Armed with such signals, regulators would be able to react — indeed, should be required to react — in a more timely way to increased stress in markets or institutions, following guidelines similar to the PCA rules in FDICIA. Elevated indicators of systemic stress could first trigger enhanced information collection and regulatory scrutiny. Signs of further stress could lead to further regulatory actions such as increased premiums, increased regulatory capital, or perhaps requirements to better insulate systemically important segments. In these ways, firms generating systemic risk would be taxed for the externalities generated by their activities. As indicators of systemic risk rose further, they might trigger recapitalizations, as in recent proposals in which banks would be required to hold a certain amount of convertible debt that would be converted into equity under well-specified conditions. This convertible debt could play an important role in financial firm resolutions by providing a quick, transparent method for recapitalization. The potential for conversion, which would dilute shareholder and creditor claims, should improve market discipline. Finally, serious danger signals would trigger planning for closure or some other resolution procedure.
While the precise details require more study, I believe that these interventions should be as transparent and public as possible. Interventions should be graduated based on observable indicators. Regulators might publicize the indicators they are monitoring. They might also consider publishing a list of firms they view as containing systemically significant segments. There is some risk to publicizing such a list, since conditions can change quickly. However, I think the benefits might outweigh the costs, since market participants will form expectations about which firms or portions of firms are systemically significant in any case.
Finally, I would like to turn to the role of the Federal Reserve in supporting financial stability. Fed Chairman Ben Bernanke has suggested that the Federal Reserve have a formal mandate to regulate systemically important payments and settlement systems.6 This aim is consistent with the Fed's existing mandate under the Federal Reserve Act to ensure the integrity, efficiency, and accessibility of the payment system. It is also essential to reducing systemic risk. Of course, as I have already mentioned, determining precisely which systems are systemically important and how to regulate them requires careful consideration.
Others have suggested that the Fed become the grand overseer of the stability of the entire financial system. Here, I think we should proceed with great care. We must avoid giving the Fed a mandate for financial or systemic stability that is too vague or too sweeping or that lacks clearly defined principles. We must set objectives that are both feasible and clearly defined. Otherwise, over-promising puts at risk the credibility of the central bank and jeopardizes the Fed's ability to meet its other important objectives: price stability and sustainable economic growth. Instability in the general level of prices — whether inflation or deflation — is itself a significant source of financial instability. Consequently, we must make sure that in trying to cure one source of financial instability, we do not sow the seeds of another.
The financial crisis has underscored the need for reforming the regulatory and supervisory architecture of our financial system. Today I've outlined three considerations as we pursue reform. First, we must address the too-big-to-fail and too-interconnected-to-fail issue by developing a resolution mechanism that reduces the systemic costs of failure. Second, we must develop ways to measure systemic risk and impose higher regulatory burdens, in a systematic and transparent way, on firms that generate increasing levels of systemic risk. Third, the Fed should play a central role in reducing systemic risk through an explicit mandate to regulate systemically important payments and settlements systems. However, we should not ask the Fed to take on responsibilities that would undercut its ability to achieve its monetary policy objectives.
We have the opportunity to design a robust and dynamic regulatory structure that balances the need to reduce financial system instability with the need to preserve financial system innovation. The importance of this undertaking cannot be overstated. The Federal Reserve has several working groups looking carefully at financial system architecture and reform. The scholarship exhibited by the New York University researchers, others in academia, and the regulatory agencies provides useful information as we work toward meeting that goal.