November 12, 2010

Next Steps in Financial Regulatory Reform

Governor Daniel K. Tarullo

At the George Washington University Center for Law, Economics, and Finance Conference on the Dodd-Frank Act, Washington, D.C.

Thank you very much for your invitation to speak at this second annual symposium on financial regulatory reform. Before addressing that topic, I want to say a word about the mortgage foreclosure documentation imbroglio, the latest chapter in the recent sad history of mortgage finance in this country.

The Need to Increase Mortgage Modifications
As you know, inquiries into the extent of, and culpability for, these problems are currently being conducted by banking regulators, other federal agencies, and state attorneys general. Regardless of the findings that emerge, and the steps that servicers and others may take to correct their mistakes, this episode has again drawn attention to what can only be described as a perverse set of incentives for homeowners with underwater mortgages. Homeowners who try to get a modification of the terms of their mortgages are all too frequently subject to delay and disappointment. On the other hand, those who simply stop paying their mortgages have found that they can often stay in their homes for a year or more, rent free, before the foreclosure process moves ahead.

This simply is not a good outcome from any broad perspective--not for the revival of housing markets, not for the banks and investors that hold the delinquent mortgages, and in the longer run, not even for the homeowners themselves, who will ultimately have to move out, taking with them a dark cloud over their creditworthiness.

Several possible explanations have been suggested for this untoward state of affairs--the lack of servicer capacity to execute modifications, purported financial incentives for servicers to foreclose rather than modify, what until recently appeared to be easier execution of foreclosures relative to modifications, limits on the authority of securitization trustees, and conflicts between primary and secondary lien holders. Whatever the merits and relative weights of these various explanations, the social costs of this situation are huge. It just cannot be the case that foreclosure is preferable to modification--including reductions of principal--for a significant proportion of mortgages where the deadweight costs of foreclosure, including a distressed sale discount, are so high. While some banks and other industry participants have stepped forward to increase the rate of modifications relative to foreclosures, many have not done enough. I would hope that both servicers and ultimate holders of the mortgages will take this occasion not just to correct documentation flaws and to contest who should bear the losses of mortgages gone bad, but to invigorate the modification process.

Next Steps Down the Reform Road
I note that while last year's conference was called "Regulatory Reform at the Crossroads," this year's event is entitled "The Dodd-Frank Act and the Road Ahead for Financial Regulatory Reform." The metaphor of a long road ahead following key decisions in Dodd-Frank is an apt one for the Federal Reserve and other regulatory agencies that will, over the next 15 to 20 months, complete implementation of that bill through scores of regulations. The metaphor also applies to elaboration and domestic implementation of the framework for Basel III that was agreed to internationally in September.

Though we may no longer face a major crossroads, the federal banking agencies will certainly encounter numerous forks in the road we are travelling. Choices will be presented during implementation of certain Dodd-Frank provisions that set a general direction for change, but do not mandate a precise route. Others will be encountered as we continue the Basel III exercise, to which I will return in a moment. Still other choices will doubtless be required as all the member agencies of the new Financial Stability Oversight Council evaluate and, potentially, respond to developments in financial markets. Finally, at least one big crossroad still lies ahead--a decision on the future of the government-sponsored housing finance agencies.

We are, of course, in the middle of the Dodd-Frank implementation process. Thus, while I recognize there is enormous interest in where the Federal Reserve and other rule-writing agencies may be headed, I cannot say much about the substance of the regulations that will eventually be proposed and adopted. Indeed, I think it would be inconsistent with the very purpose of administrative law requirements for me to come to my own conclusions, much less opine publicly on them, before we have made our proposals and evaluated all the comments.

What I can say is that, in implementing Dodd-Frank, the Board of Governors will be guided by the same norms of statutory construction that a court would apply. Most importantly, where Congressional intentions are clear from the language of the statute, we must faithfully execute those intentions. Of course, there are a good many provisions that do not admit of a single interpretation, and the implementation of those provisions will require the exercise of discretion by the Federal Reserve or other regulatory agencies.

I can also say that we are following a transparent and inclusive process that goes well beyond the classic notice and comment requirements for agencies adopting regulations. As to transparency: At the Federal Reserve, we are entering into the public record a summary of all communications with non-government groups or individuals regarding matters subject to a potential or proposed rulemaking under Dodd-Frank. As to inclusiveness: We have joined with the other banking agencies in sponsoring a series of joint forums to solicit views from industry, academics, and others on some key issues relevant to Dodd-Frank implementation. We are also hearing views on regulatory implementation at meetings that cover a broad range of topics, such as at our last Consumer Advisory Council session.

Reforming Minimum Capital Requirements
Although they had long used bank capital ratios as a supervisory tool, U.S. bank regulators did not impose explicit minimum capital requirements until the 1980s. The proximate reason for this change was regulatory concern over the decline in capital ratios of the largest banks--a concern reinforced by Congress, as it saw some of those large banks facing enormous losses on their loans to foreign sovereigns. This U.S. regulatory innovation was effectively internationalized a few years later in the original Basel Accord.

At the same time, regulators came to regard capital requirements as a supple prudential tool. As activity and affiliation restrictions were loosened in the United States, capital requirements seemed a promising way to protect the public's interest in the stability of financial institutions that had access to the Federal Reserve's discount window and Federal Deposit Insurance Corporation insurance. Capital requirements promised to provide a buffer against bank losses from any activities in which the bank or its affiliates might engage, a consideration of equal or greater relevance in countries with universal banking models. Some support also developed for the proposition that minimum capital levels could, by maintaining a material equity value for the bank, serve as a disincentive for excessive risk-taking by management and shareholders.

In the ensuing quarter century, the attention of banking regulators around the world had been heavily oriented toward elaborating capital requirements to reflect more precisely the particular risks faced by a financial institution. Capital requirements had, to a considerable extent, become the dominant prudential regulatory tool.

The financial crisis showed that the concentrated, almost all-consuming regulatory focus on refining bank capital requirements in Basel II had come at the expense of attention to other risks in the financial system. In particular, there was insufficient appreciation of the implications of the growth in size, leverage, and maturity transformation levels of the shadow banking system for the balance sheets of commercial banks and for overall financial stability. The limitations of capital requirements as a regulatory tool, such as the frequent lag between declines in asset values and reductions in bank capital, were also confirmed by experience during the crisis.

But it was also evident that capital requirements had simply been set too low in general, and with respect to particular assets. One of the most obvious examples was the capital requirement for asset-backed securities in the trading books of banks. The requirement was based on returns over a 10-day holding period, used a one-year observation period that had been characterized by unusually low price volatility, and neglected the credit risks inherent in these traded instruments. It was also apparent that at least some of the instruments that qualified as "Tier 1 capital" for regulatory purposes were not reliable buffers against losses, at least not on a going concern basis.

It is instructive that during the height of the crisis, counterparties and other market actors looked almost exclusively to the amount of tangible common equity held by financial institutions in evaluating the creditworthiness and overall stability of those institutions--they essentially ignored the Tier 1 and total risk-based capital ratios in regulatory requirements. In the fall of 2008, there was widespread doubt in markets that the common equity of some of our largest institutions was sufficient to withstand the losses that those firms appeared to be facing. This doubt made investors and counterparties increasingly reluctant to deal with these firms, contributing to the severe liquidity strains that characterized financial markets at the time.

It is obvious that the post-crisis regulatory system will not be as dependent on capital requirements as the pre-crisis regime. Dodd-Frank itself is testimony to this fact, as are a number of changes already made by banking agencies. There will be increased emphasis on market discipline, liquidity regulation, activities restrictions, and more effective supervision. But the crisis reinforces the point that robust capital requirements should continue to be a central component of the financial regulatory system. The U.S. banking agencies, and most of our counterparts from countries represented in the Basel Committee on Banking Supervision, made strengthening the capital regime a high priority in the latest financial reform agenda.

Basel III makes a number of important changes to address deficiencies in the pre-crisis capital rules:

First, reflecting both intuitive good sense and market realities during the crisis, Basel III creates a new minimum common equity capital requirement. Moreover, the agreement provides a definition of common equity that will prevent firms or national regulators from including in the calculation of common equity certain assets that could dilute its loss-absorbing character.

Second, the minimum common equity ratio will be set at 4.5% of risk-weighted assets, with an additional requirement for a 2.5% "conservation buffer." The minimum ratio should be understood as defining the amount of common equity needed for a firm to be regarded as a viable financial intermediary. The conservation buffer is a new feature of capital regulation, intended specifically to reflect the losses that a firm may suffer during periods of financial stress.

Thus the concept behind the two-level requirement is that a banking organization should be able to withstand losses associated with systemic stress and still be a viable financial intermediary. This concept is comparable to the approach we adopted during the Supervisory Capital Assessment Program (SCAP) in early 2009. There, you may recall, we used a special stress test to provide a rough estimate of losses that the large banking organizations could face in an adverse scenario and asked that they hold capital sufficient to absorb those losses and still be above common equity levels that would maintain the firms as viable intermediaries.

There is no direct way to calculate how much equity is needed to assure markets that a banking organization is viable. In our internal analysis at the Federal Reserve in preparation for the Basel Committee deliberations, we analyzed distributions of actual losses suffered by larger institutions over the last several decades, on the assumption that an institution that could withstand such losses at a high confidence level would be regarded as a viable going concern. For the conservation buffer, we looked at actual pre-SCAP losses incurred by large banking firms during the recent stress period and SCAP estimates of additional losses associated with the recent stress period. Both these determinations required considerable judgment, and thus we developed ranges, rather than point estimates, for the levels we thought reasonable. In particular, government capital injections and debt guarantees in the fall of 2008 complicated the estimation of losses that might have been incurred in the absence of too-big-to-fail support. The ratios agreed to in Basel were at the lower end of, though still within, the ranges we had calculated.

The practical effect of the two-level approach is that banks under stress may let their common equity ratio drop below the 7% level that is the sum of the minimum and buffer requirements. However, restrictions on capital distributions will result, which will become progressively more stringent as the common equity ratio drops closer to the 4.5% minimum. The buffer is thus designed to forestall banks from continuing to pay dividends even as they come under stress, a practice observed in some institutions during the financial crisis. Realistically, both regulators and markets will expect firms generally to maintain their common equity ratios above 7%.

Third, Basel III makes extensive changes to the risk weights assigned to a financial institution's traded assets and counterparty exposures. As I mentioned earlier, the market-risk requirements of the pre-crisis capital regime were woefully inadequate. In many instances, they simply did not reflect the actual risk assumed by an institution. They also created an invitation to arbitrage credit risks by turning them into traded assets with lower risk weights. It is also noteworthy that the changes in risk weights incorporate some elements of a macroprudential perspective as, for example, in higher capital requirements on equity investments in other financial firms and credit exposures to large financial firms

Fourth, Basel III provides for a minimum leverage ratio, roughly similar to requirements already applicable under national law in the United States and Canada. While the terms of this leverage requirement have been agreed to, there will be a supervisory monitoring period and then a parallel run to assess its impact, particularly in countries with no history of such a requirement, and to provide for adjustments if warranted.

Finally, Basel III has a rather lengthy and complicated transition period, with the new requirements to be phased in between January 1, 2013, and January 1, 2019. We favored a significant transition period, so as to allow firms flexibility in adjusting to the new regime through such means as running off higher risk-weighted assets, adjusting their business models gradually, or using retained earnings to add any new capital that might be required under the new rules. To be honest, however, we did not think the transition period needed to stretch over eight years. In fact, it appears that most U.S. banking entities expect to meet the new requirements considerably sooner. However, the lengthy transition period was an important inducement for some countries to agree to the new, much stronger standards.

With the agreements reached in July and September at meetings of the Governors and Heads of Supervision (GHOS) in Basel, the structure and basic elements of Basel III are clear. Details are still being worked out by the Basel Committee. Then, of course, the U.S. banking agencies will need to implement Basel III through domestic capital regulations.

Basel III is not a perfect agreement, of course. There are things we would have done differently if we were writing a capital regulation on our own. There will surely be some technical challenges in implementing it. It does not really address some pre-crisis problems in capital regulation such as pro-cyclicality. But it is a major step forward for capital regulation. It will raise minimum requirements substantially, ensure that regulatory capital is truly loss absorbing, and discourage some of the risky activities for which the pre-crisis regime required far too little capital.

Basel III was also a major step forward in international cooperation. In all candor, as recently as this past spring I was concerned that we might be unable to agree in Basel on such key issues as a distinct common equity requirement. However, under the strong leadership of Nout Wellink in the Basel Committee and Jean-Claude Trichet in the GHOS, we concluded a good agreement in timely fashion. I believe that another factor in turning things around was that, unlike at some times in the past, the U.S. banking agencies spoke with a single, unified voice in Basel.

Obviously, the benefits of Basel III for financial stability will be realized only if they are implemented rigorously. In this regard, it is important to draw a distinction between, on the one hand, implementation in the sense of enacting national regulations that incorporate the Basel standards and, on the other, implementation in the sense that firms are actually holding the amounts of capital called for by the internationally agreed rules.

The Basel Committee must be able to monitor effectively implementation of, and compliance with, these new capital standards. A number of market analysts have noted that, even under current market risk capital rules, there is considerable apparent variation in the risk-weightings apparently applied by different banks. We are urging the Committee to explore mechanisms for ensuring that these strengthened capital standards lead to a consistency in application, as well as in the provisions of relevant domestic regulations.

Along these lines, we have heard complaints from a few other countries that Basel II is not yet operative for our large, internationally active banking organizations in the United States. As we have explained, despite the substantial resources devoted by both banking organizations and supervisors to the tasks of developing and validating the Advanced Internal Ratings-Based Approach in those institutions, we continue to encounter significant difficulties. The suggestion that U.S. banking organizations have thereby gained a competitive advantage is misplaced, however. For one thing, we required significant capital increases as part of the SCAP and the Troubled Asset Relief Program repayment processes last year. Also, we note that the required capital levels for some foreign banks adopting Basel II apparently declined from Basel I levels.

In-depth oversight by the Basel Committee of implementation and compliance would allow supervisors from all member countries better to understand issues such as these. I suspect it would also result in supervisors learning from one another and thus improving the quality of large institution capital regulation globally. Although, fortunately, Basel III does not present nearly the degree of technical challenge posed by the advanced approach of Basel II, there will still be a good bit of opaqueness in how some of its components are implemented and thus a continuing need for significant monitoring by the Basel Committee.

One piece of unfinished business on the international capital regulatory agenda arises from the agreement by the GHOS in September that systemically important financial institutions should have loss absorbing capacity beyond the Basel III requirements. This international position parallels the Dodd-Frank requirement that the Federal Reserve apply capital requirements to large, interconnected financial institutions that are more stringent than those applied to other banks. We think it serves U.S. interests to develop our plans for implementing our domestic statutory obligation in tandem with our participation in this international process, so as to maximize the chances of convergence of international standards and our own practice. Work on this issue in the Basel Committee and the Financial Stability Board will continue well into next year.

Dividend Policy
Before closing, I want to address requests for renewed or increased dividends by our large bank holding companies, an issue in which there has been substantial recent interest. During the crisis, dividends were eventually suspended or reduced to minimal levels by all the banking organizations covered by the SCAP. As the financial system has stabilized, some firms have indicated an interest in resuming or increasing dividends, or repurchasing shares. We have been concerned with the safety and soundness implications of resuming or increasing capital distributions in the absence of a strong, forward-looking demonstration that the capital position of a firm would be protected even under stressed conditions. Until Basel III was completed and Dodd-Frank enacted, it was obviously difficult for any firm to make that kind of demonstration, since its future capital needs and potential business model changes were obviously unknown.

While there continues to be a relatively high degree of uncertainty about near- to medium-term economic prospects, the basic questions surrounding capital and regulatory reform have now been answered. We anticipate that some firms with high capital levels that have been retaining solid earnings for several quarters will be interested in increasing or resuming dividends. In response to these anticipated requests, we will soon be issuing supervisory guidelines applicable to such requests from the largest holding companies for the first quarter of next year.

Although the details of these guidelines are still being finalized, I can say that our approach to considering such requests will be a conservative one. We will expect firms to submit convincing capital plans that demonstrate their ability to absorb losses over the next two years under an adverse economic scenario that we will specify, and still remain amply capitalized. We also expect that firms will have a sound estimate of any significant risks that may not be captured by the stress testing, such as potential mortgage putback exposures, and the capacity to absorb any consequent losses. The firms will also be asked to show how, even with their proposed capital distributions, they will readily and comfortably meet the Basel III requirements as they come into effect, as well as to accommodate any business model changes that might be necessitated by Dodd-Frank.

The final set of major Dodd-Frank regulations will not be completed until early 2012. A year from now we will be in the midst of a regulatory process implementing Basel III, and there will likely be an active debate over the future of the government-sponsored enterprises. So I see little risk that the third annual conference on financial regulatory reform at George Washington Law School will be entitled "The End of the Road." Maybe, just maybe, I have given you a title for the fourth. Then again, maybe not.

Last Update: November 12, 2010