TESTIMONY OF JOHN D. HAWKE, JR. COMPTROLLER OF THE CURRENCY BEFORE THE COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS UNITED STATES SENATE APRIL 23, 2002 Statement required by 12 U.S. C. § 250 The views expressed herein are those of the Office of the Comptroller of the Currency and do not necessarily represent the views of the President. Introduction Chairman Sarbanes, Senator Gramm and Members of the Committee, I am pleased to have this opportunity to present the views of the Office of the Comptroller of the Currency (OCC) on deposit insurance reform. As the current and most recent past chairmen of the Federal Deposit Insurance Corporation (FDIC) have noted--and as I strongly agree--the system of Federal deposit insurance adopted by the Congress in the early 1930s has served this nation well for the greater part of a century. No massive overhaul of the system is required to ensure that it will continue to contribute to financial confidence and stability in the 21st century. Nonetheless, the efforts so far undertaken to address the weaknesses in the system uncovered during the banking and thrift crises of the late 1980s and early 1990s have not been entirely adequate to the task. Indeed, the legislation adopted in response to those crises has actually constrained the FDIC from taking sensible and necessary actions. This is particularly the case with respect to the FDIC’s ability to price deposit insurance in a way that reflects the risks posed by different depository institutions, and to the funds’ ability to absorb material losses over the business cycle without causing sharp increases in premiums. Failure to address these issues in the current financial environment poses the danger that the next major domestic financial crisis will be exacerbated rather than ameliorated by the Federal deposit insurance system. Current legislative proposals in the House and Senate to reform deposit insurance address most, albeit not all, of the issues raised by the FDIC staff in its excellent and wide- ranging "Options Paper" released in April 2001. Among these issues are (1) how much discretion the FDIC should have to set premiums reflecting the risks posed by individual institutions to the insurance funds; (2) whether strict limits on the size of the insurance funds result in excessive volatility of deposit insurance premiums; (3) whether the deposit insurance coverage limit should be increased and/or indexed to changes in the price level; and, (4) whether the Bank Insurance Fund (BIF) should be merged with the Savings Association Insurance Fund (SAIF). In summary, the OCC recommends that (1) the FDIC be provided with the authority to implement a risk-based premium system for all banks; (2) the current fixed DRR be replaced with a range to allow the FDIC more flexibility in administering the deposit insurance premium structure; (3) coverage limits on deposits should not be increased; and (4) the BIF and SAIF should be merged. We believe that deposit insurance reform also provides an opportunity to strengthen our supervisory structure by eliminating a distortion and unfairness in the current system of funding bank supervision. Currently, a portion of the earnings on the insurance funds, which state and national banks paid into, is diverted to fund the federal supervision of only one class of institutions, state banks supervised by the FDIC. The FDIC has elected not to pass those costs on to the banks they supervise. As a consequence, state nonmember banks pay only a small percentage of the costs of their supervision. In contrast, national banks pay over $400 million each year to cover the full costs of their supervision by the OCC. Ending this anomaly is not just a matter of fairness to national banks. It is a necessary component of allocating the costs and benefits of deposit insurance in an equitable and efficient manner among insured banks. For that reason, in addition to our views on the issues addressed by the legislative proposals to reform deposit insurance, my testimony today will include our suggestion for remedying the inequity that exists in the funding of supervision. Eliminating Constraints on Risk-Based Pricing The ability of the FDIC to set premiums for deposit insurance that reflect the risks posed by individual institutions to the insurance funds is one of the most important issues in the deposit insurance reform debate. The banking and thrift crises of the 1980s revealed the weaknesses of a flat-rate deposit insurance system in which the great majority of sound, prudently managed institutions subsidize the risks assumed by a few institutions. The Congress responded to this glaring deficiency by enacting the Federal Deposit Insurance Corporation Improvement Act (FDICIA) of 1991, which required the FDIC to establish a risk- based system of deposit insurance premiums, thereby bringing the pricing of deposit insurance more in line with the practices of private insurance companies. The FDIC’s initial efforts to implement such a system made meaningful, actuarially-based distinctions among institutions based on the risk each institution posed to the insurance funds, but fell short of creating a well-differentiated structure.Unfortunately, the Deposit Insurance Fund Act (DIFA) of 1996 diminished the FDIC’s discretion to maintain, let alone improve, the risk- based structure of deposit insurance premiums. DIFA effectively prohibited the FDIC from charging a positive premium to any institution in the 1A category--that is, well- capitalized institutions with composite CAMELS ratings of 1 or 2--whenever the reserves of the deposit insurance funds are at or above the designated reserve ratio (DRR) of 1.25 percent of insured deposits. As a result, at December 31, 2001, 92.5 percent of all insured banks fell into that category, and therefore pay nothing for their deposit insurance--even though their risk of loss may be far above zero. Thus, today many institutions--some of which have never paid any deposit insurance premiums--receive a valuable government service free, and very well-managed institutions in effect subsidize riskier, less well-managed institutions. Moreover, quite apart from the risk that a specific bank might present, banks are not required to pay even a minimum "user" fee for the governmentally provided benefit represented by the deposit insurance system-- a benefit without which, as a practical matter, no bank could engage in the business of taking deposits from the public. Aside from the obvious inequity to institutions that contributed heavily to recapitalize the funds after the losses of the 1980s and 1990s, a system in which the vast majority of institutions pays no insurance premium forgoes one of the major benefits of a risk-based pricing system--creating an incentive for good management by rewarding institutions that pose a low risk to the insurance funds. A mandated zero premium precludes the FDIC from charging different premiums to banks with different risks within the 1A category, despite the fact that within the 1A category there are banks that pose very different risks to the funds. Whenever the reserve ratio of the BIF falls below 1.25 percent, however, FDICIA requires the FDIC to charge an assessment rate to all banks high enough to bring it back to the DRR within one year. If that is not feasible, the FDIC must impose an assessment rate of at least 23 basis points. This sharp rise in premiums, or "cliff effect," would hit banks the hardest when they are most vulnerable to earnings pressure. To avoid creating this procyclical volatility in deposit insurance premiums, it would be preferable to offset losses to the funds through more gradual changes in premiums based on the level of the insurance fund relative to the FDIC’s assessment of current risk in the banking system. In short, we believe that as risks in the banking system change relative to the level of the insurance funds, the FDIC should have the authority to adjust premiums on all banks. Increasing Coverage Limits The question of deposit insurance coverage limits is a challenging one, in part because it is extremely easy for depositors to obtain full insurance of deposits in virtually unlimited amounts through multiple accounts. Along with most academic economists and other bank regulators, we are convinced that the sharp increase in the deposit insurance limit from $40,000 to $100,000 in 1980--at a time when the thrift industry was virtually insolvent-- was a serious public policy mistake that increased moral hazard and contributed to the weakening of market discipline that exacerbated the banking and thrift crises of the 1980s and 1990s. By encouraging speculative behavior, it ultimately increased losses to the deposit insurance funds and taxpayers. Proponents of an increase in coverage assert that it would ease liquidity pressures on small community banks and better enable small banks to compete with large institutions for deposits. None of these assertions, however, is supported by substantial evidence. First, we see no compelling evidence that increased coverage levels would offer depositors substantial benefits. Anyone who wants to use insured bank deposits as a means of holding their wealth can do so today virtually without limits, subject only to the minor inconvenience of having to open accounts at multiple banks. Despite the ability of depositors to achieve almost unlimited coverage at banks, money market mutual funds, which have some of the same features as bank transactions accounts and generally offer higher returns than bank deposits, today hold over $2 trillion. Because these funds could easily be placed in insured accounts, these facts suggest that many depositors are not concerned about the additional risk involved in holding their liquid funds in uninsured form and that households are comfortable with the status quo. Second, it is not at all clear that increasing deposit insurance coverage would result in an increase in the deposits of the banking system. One effect could be to cause a shift in deposits among banks. It is far from clear that any such redistribution of existing deposits would favor community banks. Depositors who multiply insurance coverage today by using multiple banks might consolidate their deposits in a single institution if coverage were raised, but there is no way of determining which institutions would be the ultimate beneficiaries when the switching process ended. Moreover, it is quite possible that larger, more aggressive institutions might use the expanded coverage to offer even more extensive governmentally protected investment vehicles to wealthy customers. That could cause an even greater shift of deposits away from community banks and increase the liquidity pressures felt by some. For many of the same reasons that we object to an increase in the general insurance limit, we are also concerned about proposals to use the federal deposit insurance system to favor particular classes of depositors such as municipal depositors. For instance, at year- end 2001, commercial banks had $162 billion in municipal deposits. The FDIC estimated in 1999 that less than one-third of municipal deposits was insured. Applying that 1999 ratio to the 2001 total suggests that nearly $115 billion of municipal deposits at banks are uninsured. A significant increase in the insurance limit for municipal deposits, therefore, would undoubtedly raise the level of insured deposits and put pressure on the DRR. In addition, an increase in insured coverage could spur riskier lending because banks would no longer be required to collateralize the municipal deposits with low-risk securities. Merger of the BIF and the SAIF One of the least controversial issues of deposit insurance reform is the merger of the BIF and the SAIF. The financial conditions of thrifts and banks have converged in recent years, as have the reserve ratios of the two funds, removing one of the primary objections to a merger of the funds. As of the fourth quarter of 2001, the reserve ratio of the BIF was 1.26 percent, while that of the SAIF was 1.37 percent. The reserve ratio of a combined fund would have been 1.29 percent as of the same date. As is described in greater detail below, many institutions now hold some deposits insured by each fund. But under the current structure, BIF and SAIF deposit insurance premiums could differ significantly depending on the relative performance of the two funds, raising the possibility that institutions with similar risks could pay very different insurance premiums. This would unfairly penalize low-risk institutions insured by the fund charging the higher premiums. Despite the tendency for the activities of the banking and thrift industries to converge in recent years, substantial differences remain in their portfolio composition. For example, residential mortgage loans constitute 51 percent of the assets of insured savings institutions but only 15 percent of the assets of insured commercial banks. Largely because of these differences, merger of the two funds would result in significant diversification of risks. A related development affecting the potential for diversification is industry consolidation, which has led to an increased concentration of insured deposits in a relatively few institutions and increased the risks to the deposit insurance funds. According to the FDIC staff, the three largest SAIF-insured institutions held over 15 percent of SAIF-insured deposits in 2001, while the corresponding share of the top three BIF-insured banks was over 13 percent. Merging the funds would reduce these concentrations, and thereby the risk that the failure of a few large institutions could seriously impair the insurance fund. Further, there is significant overlap in the types of institutions insured by the two funds. As of March 2001, 874 banks and thrifts were members of one fund but also held deposits insured by the other fund, and BIF-member institutions held 41 percent of SAIF-insured deposits. Finally, merger of the BIF and the SAIF would undoubtedly result in operational savings as the two funds were combined into one. Increased Flexibility for the Deposit Insurance Funds The OCC supports giving the FDIC the authority to establish a range for the DRR to replace the present arbitrary fixed DRR of 1.25 percent. The FDIC should have the authority to set the range based on its assessment of the overall level of risk in the banking system. We also believe that in establishing the range, the FDIC should provide notice and an opportunity for the public to comment on the proposed range. Adoption of a range and elimination of the 23 basis point "cliff effect" would allow the FDIC more flexibility in administering the premium structure and would minimize the likelihood of sharp increases in premiums during economic downturns when banks can least afford them. When the funds exceed the upper boundary of the DRR range set by the FDIC, the FDIC should be authorized to pay rebates or grant credits against future premiums. To ensure that rebates or credits to insured institutions are equitable, the FDIC should have the authority to assess the nature of the institutions’ claims on the funds. Institutions that have paid little or no insurance premiums to the funds have far less of a claim on rebates or credits than those that contributed to building up the funds. While such rebates or credits seem reasonable on their face, there are two obvious principles that should be observed in determining their size and allocation. First, a system of rebates and credits should not undermine the risk-based premium system. Institutions that paid high insurance premiums because they posed a higher risk to the funds should not receive larger rebates than less risky institutions of the same size. The fact that these high-risk institutions did not fail during that period does not alter the fact that they subjected the funds to greater than average risks. The second principle is that the payment of rebates and credits should not have the unintended consequence of exacerbating the disparity in supervisory fees that now exists between state and nationally chartered banks. Today, the FDIC charges the insurance funds for its costs of supervising state-chartered institutions. National banks, in contrast, pay the full cost of their supervision despite the fact that they have contributed almost 55 percent of the amount in the BIF. For example, in 2001, in addition to $400 million in assessments that national banks paid to the OCC for their own supervision, national banks can be viewed as contributing 55 percent, or about $273 million, of the $525 million that the FDIC spent on state nonmember bank supervision. Failure to take this into account in fashioning a rebate program would be unconscionable. Fee Disparity State banks, on average, pay only modest assessments to state regulators, which represent about 20 percent of the total costs of state bank supervision. Far and away the largest component of state bank supervision is that provided by their federal regulators--the Federal Reserve, in the case of state banks that are members of the Federal Reserve, and the FDIC, in the case of nonmember state banks. In 2001, the Federal Reserve and FDIC together spent over $900 million on state bank supervision. None of this was recovered directly from the banks they supervise. The FDIC absorbs the cost of its supervisory and regulatory activities through charging the BIF and SAIF, while the Federal Reserve uses its interest earnings to absorb its supervisory and regulatory costs. Neither the Federal Reserve nor the FDIC assesses state banks for their costs in providing exactly the same supervisory functions as the OCC provides for--and assesses--national banks. As a result of this subsidy provided by the Federal Reserve and the FDIC, there is a continuing incentive for national banks to convert to state charters. Indeed, state supervisors aggressively proselytize for such conversions, heavily exploiting fee disparity as a major part of their sales pitch. It should be emphasized that fee disparity has no relationship to the relative efficiency of national and state bank supervision. It is entirely a consequence of the fact that state banks are not charged for the major portion of their supervision costs--that provided by their federal regulators. Indeed, the OCC has a strong externally imposed incentive to run its operations efficiently, for if it fails to do so, and must turn to its banks to pick up additional costs, it runs the risk of causing increased conversions of banks from national charters to state charters. Still, the effectiveness of supervision can suffer, and serious inequities can result, when unavoidable pressures on supervisors’ budgets are created. For example, during the wave of large bank failures in the late 1980s and 1990s--a period of stress in the banking system that had not been seen since the Great Depression--significant resource demands were placed on bank supervisors in responding to severe problems in the banking system. Yet just as these demands were being felt, the banking system was under severe earnings pressure. At the OCC this meant significant increases in direct assessments on national banks--14 percent in 1989, another 11 percent in 1991, and 30 percent in 1992. While there were reductions in assessments in subsequent years, one conclusion is inescapable: the OCC assessment mechanism works procyclically in times of stress in the banking system. At the Federal Reserve and the FDIC, similar cost increases were easily absorbed--at the FDIC out of insurance funds and at the Federal Reserve out of revenues that otherwise would have been paid over to the Treasury Department. In other words, the OCC faces the threat of reduced supervisory resources at the very time they are most likely to be needed. National banks face a higher burden of supervisory costs at the very time they are facing a troubled economy. Just as the need to address the 23 basis point "cliff effect" has gained attention, so also should the procyclical distortions raised by the present system of funding supervision. The question, of course, is what to do about this disparity. Proposals to level the playing field by requiring the Federal Reserve and the FDIC to impose new fees on state banks have been dead on arrival in Congress. We believe it is necessary to come up with a new method of funding bank supervision--a method that will strengthen both the state and the federal supervisory processes and ensure that all supervisors have adequate, predictable resources available to carry out effective supervisory programs without imposing additional fees on state banks. Solution There are a number of alternative approaches to solving this problem that one might consider, and we believe that now is the ideal time to do so, as the whole topic of the role of deposit insurance is being reexamined. An idea that we think has considerable appeal would draw on the earnings of the FDIC’s insurance funds to cover the costs of both state and national bank supervision. Today, with the level of the combined funds at about $42 billion and generating earnings of around $2.5 billion per year, there are considerably more funds available to defray the costs of FDIC, OCC, and state supervision than those agencies today spend in total. Working together, and using the present costs of supervision as a baseline, state and federal supervisors could develop a nondiscretionary allocation formula that would reflect not only the breadth of responsibilities of the agencies, but the condition, risk profile, size, and operating environment of the banks they supervise. All agencies would remain free to impose supplemental assessments if they chose, but competitive pressures would presumably work to keep these charges at a minimum. This arrangement would offer some meaningful advantages. First, it would remedy the inequity to national banks that exists today, resulting from the fact that the FDIC funds the supervision of only one class of banks, state nonmember banks, out of the earnings of the deposit insurance funds, to which all banks have contributed. As I mentioned earlier, we estimate that national banks have accounted for more than half of the contributions to the Bank Insurance Fund. Another major advantage to a system under which the OCC and the state supervisory agencies would be funded out of the earnings on the insurance funds is that it would reinvigorate the dual banking system. It would create a regulatory system under which banks choose their charters on the basis of factors such as regulatory philosophy, access, and the perceived quality of supervision. The result would be competition based on characteristics of supervisors that are relevant to maintaining a safe and sound banking system. Conclusion The OCC supports a merger of the BIF and the SAIF and proposals to eliminate the current constraints on deposit insurance premiums. We favor elimination of the current fixed DRR and its replacement with a range that would allow the FDIC more flexibility in administering the deposit insurance premium structure. We oppose an increase in deposit insurance coverage limits at this time. Finally, as the entire role of deposit insurance is being subjected to scrutiny by policymakers and legislators, it is an opportune time to address the distortions and unfairness in the current system of funding bank supervision that I have highlighted in my testimony today. In its April 2001 Deposit Insurance Reform Options Paper, the FDIC reported that "the 5-year failure rate for CAMELS 2-rated institutions since 1984 was more than two-and-a-half times the failure rate for 1-rated institutions." (p. 13) As shown in Chart 1 on page 12, the 5-year failure rate for CAMELS 1-rated institutions (commercial and savings banks) was 0.7 percent, while that for CAMELS 2-rated institutions was 1.8 percent.