Chairman Shelby, Senator Sarbanes, and members of the committee, I am pleased to be here this morning to discuss the condition of the U.S. banking system and various related matters. They include improved risk-management practices of banks, the current status and direction of our regulatory efforts to revise capital standards for internationally active banks, deposit insurance, and the ongoing consolidation process within our domestic banking industry.
Growth in the size and complexity of the largest U.S. and foreign banking organizations, in particular, has substantially affected financial markets and the supervisory and regulatory practices of the Federal Reserve and other bank regulatory agencies around the world. It has, in part, required authorities to focus more than before on the internal processes and controls of these institutions and on their ability to manage risk. Only through steady and continued progress in measuring and understanding risk will our banking institutions remain vibrant, healthy, and competitive in meeting the growing financial demands of the nation while keeping systemic risk at acceptable levels. Therefore, the regulatory authorities must provide the industry with proper incentives to invest in risk-management systems that are necessary to compete successfully in an increasingly competitive and efficient global market.
When I last discussed the condition of the banking industry with this committee in June 2001, the industry's asset quality had begun to decline, but from a relatively high level, and banks were generally well positioned to deal with the emerging problems. Moreover, as early as the late 1990s, both the industry and bank supervisors had begun to address the slippage in credit standards that was one of the causes of the drop in asset quality. By most measures, this was an unusually early stage in the economic cycle to begin addressing such deterioration.
Today, with the benefit of hindsight, we can see that the weaknesses I cited then have indeed been mild for the banking system as a whole and that the system remains strong and well positioned to meet customer needs for credit and other financial services. During the past two years, in particular, the industry extended its string of high and often record quarterly earnings. For the full year 2003, commercial banks reported net profits of more than $100 billion while maintaining historically high equity and risk-based capital ratios and enjoying brisk asset growth. Although the demand for business loans and the underwriting of equity securities have been weak over the past few years, banking organizations have continued to benefit from strong demand for household credit, not least for residential mortgage products as interest rates declined substantially.
Moreover, the volume of problem assets in commercial banks declined each quarter last year, including a drop in the fourth quarter of nearly 10 percent, which brought the ratio of problem assets to total loans and foreclosed assets to less than 1 percent--its lowest level since year-end 2000. As a result of this favorable performance, both the size and the number of bank failures in recent years have been exceptionally small. Last year, for example, only two banks, with combined assets of just $1.5 billion, failed.
The results of last year's interagency review of large syndicated loans and internal reports about the level and distribution of their criticized and classified credits lead us to expect still further improvement in the industry's asset quality this year. Notably, the pool of "special mention" credits that are weak but still performing (and which tend to produce the more serious problem assets) has shrunk both in the annual Shared National Credit review and in the quarterly bank reports.
Risk measures derived from prices of publicly traded bank securities--stocks, debt securities, and credit default swaps--also signal that market participants are taking an increasingly positive view of the future of banks. Indeed, these measures suggest the lowest level of market concern about these companies that we have seen during the five-years in which we have tracked them.
The banking industry's relatively benign experience with loan losses these past few years may not be surprising given that the recession was mild by most measures. The experience is more notable, however, when one considers the broader range of shocks and developments that have occurred during this period, including the September 11 attacks, Argentina's credit default, the continuing shift by large and not-so-large firms in this country from bank to capital market financing, and the concentration of recent economic pressures on specific industries and business sectors. These events tended to reduce the overall quality of corporate loan portfolios at banks and contributed significantly to banks' efforts to improve their measurement and management of risks, especially after the substantial credit losses they suffered in the late 1980s and early 1990s. These efforts, aided by the continued trend toward industry consolidation, helped moderate previous concentrations of credit exposures in bank portfolios and fueled greater use of new methods of hedging and managing risk.
At present, credit risk-management practices are perhaps least developed in measuring risk associated with exposures related to construction projects and to the financing of commercial real estate, which have grown rapidly, particularly among regional and community banks. At all banks, such lending represented nearly 19 percent of all bank loans at year-end 2003--the highest level thus far recorded--and accounted for essentially all the loan growth last year at banks with less than $1 billion in assets.
Despite the limited development of formal risk-management practices, credit standards applied to these loans have apparently been quite high. At least, we see as yet no signs of rising credit losses from such lending, and supervisory and market sources indicate that the poor lending practices of the late 1980s and early 1990s have been largely avoided. Nonetheless, the historical record provides ample evidence of the risks associated with this form of lending and of accumulating large credit concentrations in any form of exposure. Supervisors continue to monitor these concentrations and the lending practices and market conditions that will ultimately determine their effects on the banking system.
These and other gradual changes in the balance sheets of banks, along with the sustained decline in market rates, helped compress net interest margins at many banks, as they chose not to reflect the full effect of lower market rates into rates paid on deposits without a specified maturity. As a percentage of earning assets, net interest income of all insured commercial banks declined 27 basis points last year, to 3.80 percent, the lowest level in more than a decade. Although this compression eased slightly during the fourth quarter, we cannot yet tell whether margins have begun to rebound.
This compression of margins needs to be understood in the fuller context of the banks' sensitivity to changes in interest rates and, in particular, the effect of historically low rates on banks' financial performance and condition. At the same time that declining rates were adversely affecting the industry's interest margins, they were also spurring growth in mortgage-related assets and associated loan-origination fees and were producing significant capital gains in bank investment portfolios. Lower interest rates, along with the decline in equity valuations experienced during 2000-2002, also contributed to a substantial inflow of liquid deposits by lessening their opportunity cost.
Under these circumstances, and with a steep yield curve, a banker's natural inclination might be to shift the credit mix and extend the maturity of assets in an attempt to bolster asset yields. To some extent such actions have been taken. Residential mortgage loans and pass-through securities have increased from 17.5 percent of assets in 2000 to 20 percent in 2003. But the manner in which this growth has occurred suggests a balanced assessment of risk. Call Report data indicate that a substantial portion of the increase in mortgage assets has been in adjustable-rate or shorter-term mortgages, particularly at smaller banks. For their part, large institutions also have significant capacity to offset on-balance-sheet exposures through off-balance-sheet transactions.
All told, the available data, industry and supervisory judgments, and the long and successful experience of the U.S. commercial banking system in dealing with changing rates suggest that, in general, the industry is adequately managing its interest rate exposure. Many banks indicate that they now either are interest-rate neutral or are positioned to benefit from rising rates. These views are based partly on specific steps that they have taken to adjust portfolios and partly on judgments about the effects that rising interest rates would have in easing pressure on interest margins. That is, many banks seem to believe that as rates rise--presumably along with greater economic growth--they can increase lending rates more than they will need to increase rates paid on deposits. Certainly, there are always outliers, and some banks would undoubtedly be hurt by rising rates. However, the industry appears to have been sufficiently mindful of interest rate cycles and not to have exposed itself to undue risk.
In other areas, earlier concerns about the effect of the century date change on computer systems, the destruction of infrastructure in the September 11 attacks, and the increased volume and scope of banking transactions generally have also required financial institutions, particularly large institutions, to devote more effort and resources to contingency planning in order to ensure the continuity of their operations. Last fall's power outage and Hurricane Isabel may have offered only limited tests of the industry's improved procedures, but financial firms handled those challenges extremely well.
As the nation's central bank and as a bank supervisor, the Federal Reserve has a strong interest in the continued operation of the U.S. financial system after a disruptive event. To that point, last year, the Federal Reserve Board, the Securities and Exchange Commission, and the Office of the Comptroller of the Currency jointly issued an interagency paper, "Sound Practices to Strengthen the Resilience of the U.S. Financial System." That paper provides guidance that supplements long-standing principles of business continuity planning and disaster recovery and is directed at the entities that pose systemic risk to the financial system, particularly in the context of their clearing and settlement activities. Through the Federal Financial Institutions Examination Council, we also issued revised examination guidance on business continuity planning. This guidance covers a variety of threats to business operations, including terrorism, and will be used in future examinations.
Improved Risk Management in Banks
Independent of continuity planning for unusual events, the basic thrust of recent efforts to improve the management of risk has been better quantification and the creation of a formal and more-disciplined process for recognizing, pricing, and managing risk of all types. In the area of credit risk, by providing those involved with a stronger, more-informed basis for making judgments, this development has enhanced the interaction between lending and
risk-control officers. Operating with better information does not mean that banks will necessarily reduce credit availability for riskier borrowers. It does mean that banks can more knowingly choose their risk profiles and price risk accordingly. Better, more-informed lending practices should also lead to a more-efficient allocation of scarce financial capital to the benefit of the economy at large.
Greater internal transparency and quantification of risk have helped bank managers monitor portfolio performance and identify aspects of the risk-measurement and credit-granting process that begin to move off track. As risk-measurement and disclosure practices evolve, investors and uninsured creditors will also become more motivated and better positioned to understand the risk profile of banks and convey their own views of banking risks. Indeed, accommodating greater and more-informed market discipline is an important goal of bank supervisors.
Perhaps most important, better risk management has already begun to show real potential for reducing the wide swings in bank credit availability that historically have been associated with the economic cycle. Sound procedures for risk quantification generally lead to tighter controls and assigned responsibilities and to less unintended acceptance of risk during both the strengthening and weakening phases of the business cycle. Earlier detection of deviations from expectations leads to earlier corrective actions by bank managers and, as necessary, by bank supervisors.
Better methods for measuring credit risk have also spurred growth in secondary markets for weak or problem assets, which have provided banks with a firmer, sounder basis for valuing these credits and an outlet for selling them and limiting future loss. Insurance companies, hedge funds, and other investors acquire these assets at discounts that they judge are sufficient to meet their expected returns and balance their portfolio risks. The result is greater liquidity for this segment of bank loan portfolios and the earlier removal of weakening credits from bank balance sheets. Portfolio risks have also been increasingly hedged by transactions that do not require asset sales, such as derivatives that transfer credit risk.
With greater use, more-thorough review, and more-extensive historical data, risk modeling has improved in accuracy and will continue to do so. Supervisors are also learning these techniques and are pressing banks to improve their own methods and systems to keep up with the latest developments. In the United States, our leading banking organizations began the process years ago and, in many respects, were in the vanguard of the effort worldwide. Nevertheless, they and the risk-measurement process itself have much further to go.
Recent initiatives of the Basel Committee on Banking Supervision to revise international capital standards have helped focus attention on risk-measurement practices and have encouraged further investment in this area. Moreover, the very improvements in technology that facilitated better bank risk measurement and management have undermined the current regulatory capital regime by creating transactions and instruments that were not conceived when the current regulatory standard was developed.
Although these developments have sometimes helped banks circumvent existing rules, they have also enabled banks to hedge portfolio risk in ways that the current accord does not address well. As a result, the current regulatory capital standard is increasingly unable to establish capital requirements for our largest and most-complex banking organizations that reflect their true underlying risks. We need a more accurate, more risk-sensitive measure of capital adequacy to provide these institutions with appropriate risk-management incentives and to provide ourselves with a more reliable basis for supervising them in a way that focuses on true risks. In the process, such a measure should also enhance our efforts in taking prompt corrective action. For all these reasons, I believe the U.S. banking agencies must remain committed to the process of developing and applying a revised regulatory capital standard for the world's international banks.
Proposed Capital Standards
Last summer, the U.S. banking agencies took another step toward adopting the new capital standard by issuing for public comment an advance notice of proposed rulemaking (ANPR). The conception and design of the proposed standard, referred to as Basel II, are based on techniques developed in recent years by the largest banks, especially those, as I noted, in this country. As the scale and complexity of their activities grew, the banks needed to find better and more efficient ways to understand, manage, and control their risk-taking activities; to promote and respond to the emergence of new markets, such as those for securitized assets; and to make greater use of available technology and financial theory in measuring and managing their risks.
Before the agencies issued the ANPR, numerous changes in the proposed Basel II Accord had already been made in light of earlier comments. Reflecting the comments received on the ANPR, the Basel Committee agreed to extend the period for reaching an agreement in principle until mid-2004 to permit more time for revisions of the proposal to be formulated. Indeed, we have already negotiated some major changes in the international proposal to reflect U.S. public comments. These changes include the adoption of a framework based on unexpected loss and a revised set of rules on securitization. We have also modified the implementation process to ease the burden on banking organizations that operate across borders. These technical changes were high on the list of modifications suggested by commenters.
The shift from a combined "expected" and "unexpected" loss framework to one that focuses on unexpected loss only is crucial to ensuring that the regulatory capital framework is consistent with standard internal banking practices, both here and abroad. That change will also simplify other parts of the proposal. The modification on securitization was imperative to permit U.S. banks to continue participating in important funding markets that they pioneered and to ensure a prudent risk-sensitive capital treatment for securitization exposures. Beyond these achievements, working groups in Basel are considering other U.S. proposals related to refining measures of expected loss, an issue that a number of commenters raised. The U.S. agencies are still trying to reach a consensus on a revised proposal for capital charges on retail credit to put before our colleagues in Basel. The Federal Reserve, for its part, will continue to make every effort to reach consensus on this issue that is both risk-sensitive and workable.
I believe that all the federal banking agencies are committed to achieving a revised accord that reflects the realities of the twenty-first century; that meets our needs for a safe, sound, and competitive banking system; and that addresses the legitimate concerns of the industry. The Federal Deposit Insurance Corporation has raised important issues about capital adequacy, and the Office of the Comptroller of the Currency has expressed significant concerns about a capital structure that may inadvertently disrupt retail credit operations of banks. All the agencies are addressing these concerns by jointly developing proposals to bring to Basel. In working to reach full agreement among ourselves, and ultimately with our colleagues abroad, we all seek a solution that promotes sound banking practices and that we can adequately implement and enforce. I hope that in the days ahead the agencies can close the gap on credit cards within such an overarching framework.
If we can do so, the Basel Committee should be able to reach agreement in principle on a new proposal around midyear, and the U.S. banking agencies expect to evaluate that proposal through another "quantitative impact study" that we plan to conduct at large U.S. banks this fall. Committee members are aware that this survey and public comments on a forthcoming Notice of Proposed Rulemaking may raise still further issues that will need to be addressed before we can implement Basel II in the United States. Of course, other countries have their own national and European Union-wide review processes to conclude, and those consultations too, may raise issues that will require additional attention.
As this committee knows, the U.S. agencies have proposed that in this country the most-advanced version of Basel II is to be required only of the largest, most-complex banking organizations, although we anticipate that some of the other larger banks also will choose to adopt that version. Non-adopters in the United States will continue to operate under the current capital rules. The current regulatory capital regime, as I noted, has become less effective for the largest organizations while consolidation has sharply increased the scale and scope of their activities. In this country, the Basel II proposal focuses on them. The current rules remain appropriate and prudent for other banking organizations in the United States, and the agencies have decided that imposing the cost of new rules on these banking organizations does not pass a cost-benefit test.
Nonetheless, change in the procedures for calculating regulatory capital for larger banks creates uncertainty among those entities to which the new rules would not apply. The comments we received on the ANPR and from the Congress last year indicate that some smaller banks are concerned that their competitive environment will change. More specifically, these fears include the possibility that Basel II will induce adopters, who are likely to have reductions in regulatory capital requirements, to redeploy their capital by acquiring non-adopters or to gain a competitive advantage, particularly in the markets for small business and residential mortgage loans.
To judge the merits of these concerns, the Federal Reserve conducted two technical and empirical analyses of the underlying issues and made the papers available to the public last month; congressional staff members were also briefed. A third study will be completed shortly, and a fourth will commence soon.
The first of these papers, dealing with mergers and acquisitions, found virtually no statistical support for the view that either the level of, or changes in, excess regulatory capital have played a role in past merger and acquisition decisions, which suggests that any future effect of Basel II on such decisions is also likely to be quite small. Moreover, reductions in regulatory capital requirements for adopters relative to the requirements for non-adopters are unlikely to lead to an acceleration in the pace of consolidation.
The second study evaluated the likely effect of Basel II on the competition between adopters and non-adopters in the market for small- and medium-sized business loans. It estimated that the marginal cost of such loans at adopting banks would decline no more than about 16 basis points, on average, and is likely to decline by less than that in most cases.
Importantly, the study also found that most small business loans made by community banks are sufficiently different from those made by either required or likely adopters of Basel II as to make any marginal cost differences virtually irrelevant. Moreover, being riskier, the small business loans made by most community banks are priced so much above the loans made by the large banks that the marginal cost benefit to adopters would not be a material competitive factor. The study did find, however, that the types of small- and medium-sized business loans made by adopters and other large banks are, indeed, similar and similarly priced, so that adopting institutions may have a competitive advantage in many cases over other large banks that choose not to adopt Basel II. I will return to the implication of this finding in a moment.
A paper analyzing competitive effects in the residential mortgage market will be available later this spring, and once the U.S. agencies agree on a proposal regarding the treatment of credit cards, staff members can begin analyzing potential competitive effects of the proposal in that market, as well. All four papers will then be re-evaluated early next year when new data become available from the agencies' next quantitative impact study.
If the evidence following these reviews and a public comment process suggests that implementation of bifurcated capital standards in this country may affect competition in certain markets, the proposals for Basel II may need to be reconsidered. We may need, for example, to modify the application of Basel II in the United States, where permissible under the Basel agreement; negotiate further changes in the international agreement itself; or change the way the current capital rules are applied to institutions that do not adopt the new standard.
In short, if we have sufficient indications that implementation of a new capital standard will distort the balance of competition, we can and will apply policies to mitigate this effect consistent with the risk profile of individual institutions. We cannot, however, respond to an unsubstantiated and generalized fear of change. Such concerns should not halt the evolution of regulatory capital standards for the large, complex banking organizations that play such an important role in our banking system and in global financial markets.
Legislation designed to deregulate U.S. banking markets, technology, and other factors have contributed to significant structural change in the banking industry and to a decline of nearly 40 percent in the number of banking organizations since the mid-1980s, when industry consolidation began. Consolidation activity has slowed sharply in the past five years, but a recent uptick in merger announcements, including a couple of very large transactions, may signal a return to a more rapid pace of bank merger activity. Since 1995, the ten largest U.S. banking organizations have increased their share of domestic banking assets from 29 percent to 46 percent at year-end 2003. Yet, over the past decade, roughly 90 percent of bank mergers have involved a target with less than $1 billion in assets, and three-quarters have involved an acquiree with assets of less than $250 million.
This ongoing consolidation of the U.S. banking system has not, in my judgment, harmed the overall competitiveness of our banking and financial markets. Although they have facilitated consolidation, the reduced barriers to entry--such as were provided by the Riegle-Neal Act's relaxation of interstate banking laws--have provided net competitive benefits to U.S. consumers of financial services.
Other economic forces, such as technological change and globalization, have stimulated competition among depository institutions and between depositories and nonbank providers of financial services. In addition to other credit-extending businesses, our system of depository institutions alone continues to be characterized by many thousands of commercial banks, savings institutions, and credit unions. Measures of concentration in local banking markets, both urban and rural, have actually declined modestly not just since 2000 but since the mid-1990s. Significantly, most households and small and medium-sized businesses obtain the vast majority of their banking services in such local markets.
I would like to turn now to the issue of deposit insurance reform and to the need for some legislative change in this area. As the committee knows, most depository institutions have not paid any deposit insurance premiums since 1996, and in fact, some large institutions that have been chartered in the past eight years have never paid them at all. Under current conditions, not only is a government guarantee being provided free, but also depositories having similar or identical risks are exposed to potentially disparate treatment should one, but not the other, of the deposit insurance funds fall below its funding target. In that situation, the FDIC would be required to impose a charge on one set of depository institutions while continuing to provide free deposit insurance to those in the other fund. Because some depository institutions today have commingled BIF- and SAIF-insured deposits as a result of bank and thrift mergers, this disparate treatment could apply even to different deposit accounts within the same depository institution.
At this time, the Congress has the opportunity to provide the FDIC with greater flexibility to charge risk-based premiums, possibly using market data (for example, rates on uninsured deposits) for the largest banks, to allow such premiums to increase or decrease in a gradual manner over a wider range of fund reserve ratios, and to treat all depositories with similar risk ratings equally and equitably. Such reforms should be implemented in a manner that does not unnecessarily create additional moral hazard and that strengthens, rather than erodes, market discipline.
Higher coverage limits, for example, would exacerbate moral hazard problems without apparent and offsetting benefits. The current level of coverage seems adequate to meet the needs of an overwhelming majority of depositors. First, depositors have certain flexibility in distributing large balances among multiple accounts and depository institutions to obtain higher insurance coverage. Second, the Federal Reserve's latest survey of consumer finances indicates that at year-end 2001 less than 4 percent of U.S. depositor households had any uninsured deposits. Moreover, the median bank IRA/Keogh account balance was only $15,000, well below the existing insurance limit. Finally, community banks have shown themselves just as adept as the largest banks in attracting uninsured deposits when necessary to fund customer loan demand.
Return to top
In closing, Mr. Chairman, let me reiterate that the past decade has been one in which the banking industry has recorded persistent record profits while providing an ever-wider range of products and services to much more diverse groups. The industry's experience during the past several years in dealing with clear weakness in key economic sectors demonstrates the importance of strong capital positions and sound risk-management practices. Bank supervisors worldwide are working to encourage further progress in these areas, through more-accurate and more-effective regulatory capital standards based on even better internal risk-management procedures.