DONALD E. POWELL
FEDERAL DEPOSIT INSURANCE CORPORATION
DEPOSIT INSURANCE REFORM
COMMITTEE ON BANKING, HOUSING AND URBAN AFFAIRS
April 23, 2002
Room 538, Dirksen Senate Office Building
Chairman Sarbanes, Senator Gramm, and members of the Committee, it is a great pleasure to appear before you this morning to discuss deposit insurance reform. This is one of the key priorities of the Federal Deposit Insurance Corporation and I appreciate this Committee's interest in promoting the discussion.
Deposit insurance has been a significant element of financial stability in this country for nearly 70 years and helped us through two major banking crises. During the crisis of the 1980s and early 1990s, the FDIC and the Resolution Trust Corporation resolved 2,362 failures of insured institutions involving more than $700 billion in assets. The last time we saw a banking crisis of that magnitude was during the early 1930s. Yet, the outcomes were very different. I believe this is due in large part to the presence of the FDIC and its stabilizing presence in the marketplace. Deposit insurance played a significant role in ending the banking crisis of the Great Depression by re-establishing financial stability. During the more recent crisis, there were no bank panics, no disruptions to financial markets and no debilitating effect on overall economic activity.
We sometimes fail to appreciate the truly remarkable value and accomplishments of federal deposit insurance, and we tend to undervalue or disregard its importance when times are good. The FDIC has played a key role in maintaining public confidence in our financial sector through good times and bad, and we are proud of this record.
The deposit insurance system protects depositors and helps the economy by preventing bank panics and stabilizing the financial system. It should accomplish this without causing other problems. Specifically, it should not increase banks' incentive to engage in riskier behavior than would be possible in the absence of insurance — the moral hazard problem. And, most of all, deposit insurance should never again cost the taxpayers a dime.
Many of the rules put in place by Congress in the Federal Deposit Insurance Corporation Improvement Act of 1991 (FDICIA) are designed to ensure that the deposit insurance funds are adequate, that the deposit insurance program is operated in a manner that is fiscally and economically responsible, and that banks and thrifts — rather than the taxpayers — fund the system. We understand that many features of the current system exist for good reason, and we have not lost sight of this in developing our proposals.
I have been at the FDIC about eight months now. I arrived with a banker's natural skepticism about the need for deposit insurance reform, but I quickly became convinced. While the current system is not in need of a radical overhaul, flaws in the system could actually prolong an economic downturn, rather than promote the conditions necessary for recovery. The current system also distorts incentives in ways that exacerbate the moral hazard problem. These flaws can be corrected only by legislation, and I appreciate this Committee's attention to this issue.
The work that the FDIC staff did in coming up with its recommendations for reform a year ago is a model for how government agencies should create public policy proposals. The staff did its homework and kept all of the players — Congress, the banking industry, scholars and experts, and the public — involved every step of the way. They prepared an excellent report on deposit insurance reform with very important recommendations and I have full confidence in that product.
Since I came to the FDIC, I have had a chance to add my own thoughts to the FDIC's recommendations and together we have had a chance to refine the proposals. This morning, I would like to give you our view on the best course for reform.
Specifically, I will address several areas: merging the funds, deposit insurance fund management and pricing of deposit insurance premiums, a one-time assessment credit, and deposit insurance coverage.
While I understand that much of the debate has centered on coverage, I want to first emphasize what we regard as even more critical — merging the funds, and improving the FDIC's ability to manage the fund and price premiums properly to reflect risk. These changes are needed to provide the right incentives to insured institutions and to improve upon the insurance system's role as a stabilizing economic factor, while also preserving the obligation of banks and thrifts to fund the system. I will discuss each of the issues in turn.
MERGING THE BIF AND THE SAIF
We should merge the Bank Insurance Fund (BIF) and the Savings Association Insurance Fund (SAIF). There is a strong consensus on this point within the industry, among regulators and within the Congress.
Originally, the two funds were intended to insure bank and savings association deposits separately. From the point of view of the insured depositor, there is virtually no difference between banks and thrifts. Moreover, many institutions currently hold both BIF- and SAIF-insured deposits. More than 40 percent of SAIF-insured deposits now are held by commercial banks.
A merged fund would be stronger and better diversified than either fund standing alone. In addition, a merged fund would eliminate the possibility of a premium disparity between the BIF and the SAIF. As long as there are two deposit insurance funds, with independently determined assessment rates, the prospect of a premium differential exists. When such a price disparity exists, banks and thrifts naturally gravitate to the lower price, wasting time and money trying to circumvent restrictions that prohibit them from purchasing deposit insurance at the lowest price — an undesirable result. A merged fund would have a single assessment rate schedule and the prospect of different prices for identical deposit insurance coverage would be eliminated.
The potential for differing rates is not merely theoretical. The BIF reserve ratio at the end of 2001 stood at 1.26 percent ($1.26 in reserves for every $100 of insured deposits), barely above the Designated Reserve Ratio (DRR) of 1.25 percent, while the SAIF reserve ratio stood at 1.37 percent. The FDIC Board will decide within the next few weeks whether BIF rates must be raised for the second half of 2002 to maintain the reserve ratio at the DRR.
For all of these reasons, the FDIC has advocated merging the BIF and the SAIF for a number of years, and I wholeheartedly agree. Any reform plan must include merging the funds.
FUND MANAGEMENT AND PREMIUM PRICING
Two statutory mandates currently govern the FDIC's management of the deposit insurance funds. One of these mandates can put undue pressure on the industry during an economic downturn. The other prevents the FDIC from charging appropriately for risk during good economic times. Together, they lead to volatile premiums.
When a deposit insurance fund's reserve ratio falls below the 1.25 percent DRR, the FDIC is required by law to raise premiums by an amount sufficient to bring the reserve ratio back to the DRR within one year, or charge at least 23 basis points until the reserve ratio meets the DRR. Thus, if a fund's reserve ratio falls slightly below the DRR, premiums need not necessarily increase much. On the other hand, if a fund's reserve ratio falls sufficiently below the DRR, average premiums will increase to 23 basis points, at a minimum.
The potential for 23-basis point rates is problematic because, during a period of heightened insurance losses, both the economy in general and depository institutions in particular are more likely to be distressed. A 23-basis point premium at such a point in the business cycle would be pro-cyclical and result in a significant drain on the net income of depository institutions, thereby impeding credit availability and economic recovery.
When a fund's reserve ratio is at or above the 1.25 percent DRR (and is expected to remain above 1.25 percent), current law prohibits the FDIC from charging premiums to institutions that are both well-capitalized, as defined by regulation, and well-managed (generally defined as those with the two best CAMELS examination ratings). Right now, 92 percent of banks and thrifts are well-capitalized and well-managed and pay the same rate for deposit insurance - zero. Significant and identifiable differences in risk exposure exist among these 92 percent of insured institutions. To take just one example, since the mid-1980s, institutions rated CAMELS 2 have failed at more than two-and-one-half times the rate of those rated CAMELS 1.
This provision of law produces results that are contrary to the principle of risk-based premiums, a principle that applies to all insurance. Because the current system does not charge appropriately for risk, this increases the potential for moral hazard. This also means that safer banks unnecessarily subsidize riskier banks. Both as an actuarial matter and as a matter of fairness, riskier banks should shoulder more of the industry's deposit insurance assessment burden.
In addition, the current statute also permits banks and thrifts to bring new deposits into the system without paying any premiums. Essentially, the banks that were in existence before 1997 endowed the funds, and newcomers are not required to contribute to the ongoing costs of the deposit insurance system. Since 1996, more than 900 new banks and thrifts have joined the system and never paid for the insurance. I know this firsthand because I chartered a bank in Texas in the late 1990s and we never paid a dime in deposit insurance premiums. Other institutions have grown significantly without paying additional premiums.
These problems can be solved by eliminating the existing inflexible statutory requirements and by giving the FDIC Board of Directors the discretion and flexibility to set appropriate targets for the fund ratio, determine the speed of adjustment toward the target using surcharges or assessment credits as necessary, and charge premiums based on risk at all times, regardless of the reserve ratio.
The FDIC recognizes that accumulating money in the insurance fund to protect depositors and taxpayers means less money in the banking system for providing credit. The current system strikes a balance by establishing a reserve ratio target of 1.25 percent. The existing target appears to be a reasonable starting point for the new system — with a modification to allow the reserve ratio to move within a range to ensure that banks are charged steadier premiums. The point of the reforms is neither to increase assessment revenue from the industry nor to relieve the industry of its obligation to fund the deposit insurance system; rather, it is to distribute the assessment burden more evenly over time and more fairly across insured institutions.
In my view, the reserve ratio target should remain relatively steady over the longer run and move only in response to fundamental changes that are expected to alter the risk exposure of the fund for the foreseeable future. The target should not be viewed as a short-run instrument that should rise and fall continuously with the business cycle. The key to fund management would be to bring the fund ratio back toward the target in an appropriate timeframe when it moves away in either direction. Presumably, the farther the movement away from the target, the larger would be the expected credits, rebates, or surcharges, other things equal, in order to slow the momentum and pull the ratio back toward the target. However, the greater the range over which the FDIC has discretion to manage the fund, the more flexibility we will have to eliminate the system's current pro-cyclical bias.
The FDIC would prefer to steer clear of hard triggers, caps and mandatory credits or rebates. Automatic triggers that "hard-wire" or mandate specific Board actions are likely to produce unintended adverse effects, not unlike the triggers in the current law. They would add unnecessary rigidity to the system and could prevent the FDIC from responding effectively to unforeseen circumstances. For a new deposit insurance fund management plan to work effectively, the Board must have the flexibility to respond appropriately to differing economic and industry conditions. For example, a given reserve ratio may warrant a credit rather than a rebate, and a smaller rather than a larger credit, depending upon economic conditions, industry performance, possible failures and other circumstances. The legislation could contain an expectation of a rebate in certain circumstances with a requirement that the Board justify any alternative decision.
While I believe that the FDIC Board needs greater discretion to manage the fund, I am not suggesting that we need unfettered discretion and I recognize the need for accountability. The FDIC will work with the Congress to develop guidelines on an appropriate range for the fund ratio around the target — as well as some direction for the FDIC Board's management of the fund ratio levels — and to develop reporting requirements for the FDIC's actions to manage the funds. For example, the Board could be expected to adopt rules that operate to pull the reserve ratio back, at an appropriate speed, whenever it moves away from the target in either direction and to publish a schedule for recapitalizing the fund whenever the ratio falls significantly below the target, as current law requires.
Charging Premiums Based upon Risk
How would premiums work if the FDIC could set them according to the risks in the institutions we insure? First, and foremost, the FDIC would attempt to make them fair and understandable. We would strive to make the pricing mechanism simple and straightforward. I believe that we can accomplish our goals on risk-based premiums with relatively minor adjustments to the FDIC's current assessment system.
I am aware of the concern about using subjective indicators to determine bank premiums. We will be sensitive to that issue and work to ensure that we use objective indicators to the extent possible to measure risk in institutions. Any system we adopt will be transparent and open. The industry and the public at large will have the opportunity to weigh in on any changes we propose through the notice-and-comment rulemaking process.
Using the current system as a starting point, the FDIC should consider additional objective financial indicators, based upon the kinds of information that banks and thrifts already report, to distinguish and price for risk more accurately within the existing least-risky (1A) category. The sample "scorecard" included in the FDIC's April 2001 report represents one approach. In this example, banks currently in the best-rated category were divided into three groups using six financial ratios in addition to capital and CAMELS ratings (net income, nonperforming loans, other real estate owned, non-core funding, liquid assets, and growth). Actuarial analysis showed that different premiums would be warranted for these three groups, based on the FDIC's loss experience since 1984.
We have since had many discussions with bankers and trade-group representatives regarding this sample scorecard, and we are making adjustments. We are willing to listen to the industry and Congress regarding alternative pricing schedules that also may be analytically sound.
The FDIC's report also indicated that for the largest banks and thrifts, it might be possible to augment such financial ratios with other information, including market-based data, so long as the final result is fair and does not discriminate in favor of or against banks merely because they happen to be large or small.
In short, I believe the right approach is to use the FDIC's historical experience with bank failures and with the losses caused by banks that have differing characteristics to create sound and defensible distinctions. However, we will not follow the results of our statistical analysis blindly — we recognize the need for sound judgment in designing the premium system.
ASSESSMENT CREDITS FOR PAST CONTRIBUTIONS
One result of the FDIC's current inability to price risk appropriately is that the deposit insurance system today is almost entirely financed by institutions that paid premiums prior to 1997. More than 900 newly chartered institutions, with more than $70 billion in insured deposits, have never paid premiums for the deposit insurance they receive. Many institutions have greatly increased their deposits since 1996, yet paid nothing more in deposit insurance premiums.
Since they are not paying premiums, new institutions and fast-growing institutions have benefited at the expense of their older and slower-growing competitors. Under the present system, rapid deposit growth lowers a fund's reserve ratio and increases the probability that additional failures will push a fund's reserve ratio below the DRR, resulting in an immediate increase in premiums for all institutions.
An assessment credit could be used as a mechanism to address the fairness issue that has arisen. I am reluctant to mandate a cash payment out of the fund at this time, given the recent fall in reserve ratios. But we can achieve the desired result by giving the banks and thrifts that built up the funds a credit toward their future assessments.
A reasonable way to allocate the initial assessment credit would be according to a snapshot of institutions' relative assessment bases at the end of 1996, the first year that both funds were fully capitalized. Each institution would get a share of the total amount to be credited to the industry based on its share of the combined assessment base at yearend 1996. For example, an institution that held one percent of the industry assessment base in 1996 would get one percent of the industry's total assessment credit. Relative shares of the 1996 assessment base represent a reasonable proxy for relative contributions to fund capitalization, while avoiding the considerable complications that can be introduced by attempting to reconstruct the individual payment histories of all institutions.
Institutions that had low levels of deposits on December 31, 1996, but subsequently experienced significant deposit growth would receive relatively small assessment credits to be applied against future premiums. Institutions that never paid premiums would receive no assessment credit. Institutions that made significant contributions into the deposit insurance funds would pay a lower net premium than institutions that paid little or nothing into the fund. Such an assessment credit would provide a transition period whereby banks that contributed in the past could defer any payment of net premiums.
The combination of risk-based premiums and assessment credits tied to past contributions to the fund would help us fix the remaining problems related to rapid growers and new entrants. Regular risk-based premiums for all institutions would mean that fast-growing institutions would pay increasingly larger premiums as they gather deposits. Fast growth, if it posed greater risk, also could result in additional premiums through the operation of the FDIC's expanded discretion to price risk.
DEPOSIT INSURANCE COVERAGE
As a life-long banker, I can tell you that deposit insurance is important not only to individuals and families, but also to many small businesses, community banks, charities, and some local governments. In fact, as an agent of financial stability, deposit insurance is important to the entire economy. If you believe as I do that deposit insurance is important, the declining value of coverage should be a subject of concern.
Our recommendation is simple: the coverage limit should be indexed from the present level of $100,000 to ensure that the value of deposit insurance in the economy does not wither away over time. In real terms, this does not expand coverage. It simply holds it steady over time. I challenge those who oppose indexation to either provide a number better than $100,000 or defend the principle of eroding deposit insurance. Over the 22 years since Congress last increased the coverage limit, the real value of the $100,000 coverage limit has declined to $47,000 in 1980 dollars. I also believe that indexing the limit on a set basis will prevent any unintended consequences that may result from making large adjustments on an irregular basis, which would seem to address the argument that sudden sharp increases resulted in economic disruptions in the past.
My suggestion would be to index the $100,000 limit to a widely accepted index, such as the Consumer Price Index or the Personal Consumption Expenditures Chain-Type Index, and adjust it every five years. The first adjustment would take place on January 1, 2005. We should make adjustments in round numbers — say, increments of $10,000 — and the coverage limit should not decline if the price level falls. These seem like the right elements of an indexing system, but I am willing to support any reasonable method of indexing that ensures the public understands that the FDIC's deposit insurance protection will retain its value. I look forward to working with the Congress to find a method of indexing that works.
There has been some opposition to the FDIC's indexing proposal on the grounds that it would increase the federal safety net. Frankly, I am puzzled by this. I am not recommending the safety net be increased. I am simply recommending the safety net not be scaled back inadvertently because of inflation.
Some argue against indexing by contending that the current limit already is too high, and they point to the problems of the 1980s in support of their concern. With respect to the problems of the 1980s, Congress undertook a comprehensive review of the deposit insurance system with FDICIA. Significant changes were made to many features of the system. Congress adopted prompt corrective action, the least-cost test and risk-based premiums as means to address moral hazard, but did not lower coverage. Moreover, as noted earlier, the real value of coverage today is far lower than it was even when the limit was increased to $40,000 in 1974, roughly a decade before the problems appeared.
I do believe, however, there is one class of deposits for which Congress should consider raising the insurance limit, and that is retirement accounts. These accounts are uniquely important and protecting them is consistent with existing government policies that encourage long-term saving. When we think about saving for retirement in this day and age, $100,000 is not a lot of money. Middle-income families routinely save well in excess of this amount. The shift from defined benefit plans to defined contribution plans has resulted in significant wealth being accumulated outside traditional pension funds.
Protecting retirement accounts is consistent with government policies encouraging savings and responsible retirement planning. Congress recently raised the annual maximum IRA contributions to $3,000 this year and larger amounts in future years and allowed those over 50 to make catch-up contributions. Some precedent exists for providing retirement accounts with special insurance treatment. In 1978, Congress raised coverage for IRA and Keogh deposit accounts to $100,000, while leaving basic coverage for other deposits at $40,000.
The $220 billion of IRA and Keogh deposits currently at banks and thrifts is not large compared to the volume of overall deposits. Thus, if the coverage limit were raised for IRA and Keogh deposits, the initial effect on the fund reserve ratio would not be dramatic. However, the total volume of IRAs and Keoghs in the economy is more than $2.5 trillion and estimating the influx of retirement account deposits as a result of higher coverage is subject to uncertainty. We would also note that a phasing-in of higher coverage limits for retirement account deposits could allow for some measure of control over the effect on the fund reserve ratio. I urge the Congress to give serious consideration to raising the insurance limit on retirement accounts.
Deposit insurance has been a bulwark of the nation's economy since its creation. It is no less important today. We must ensure that it can continue to stabilize the economy and protect depositors in the future.
Deposit insurance reform is not about increasing assessment revenue from the industry or relieving the industry of its obligation to fund the deposit insurance system. Rather, it is to distribute the assessment burden more evenly over time and more fairly across insured institutions. This is good for the industry, good for depositors and good for the overall economy.
I take very seriously the responsibility of prudently managing the fund and maintaining adequate reserves — it is extremely important to the industry and to the financial stability of our country. We have only to look back at the bank and thrift crises of the 1980s and 1990s to understand this. The existing deposit insurance system has served us well, and we must be mindful of this in contemplating changes.
The recommendations we have made would retain the essential characteristics of the present system and improve upon them. While I am Chairman, I will do all I can to ensure that the FDIC manages the insurance fund responsibly and is properly accountable to the Congress, the public and the industry. Our recommendations will ensure that future Chairmen can do so as well.
The Congress has an excellent opportunity to remedy flaws in the deposit insurance system before those flaws cause actual damage either to the banking industry or our economy as a whole. The banking industry remains strong despite the recent downturn. The FDIC has put forward some important recommendations for improving our deposit insurance system. While I believe we should remain flexible with regard to implementation, as a former banker and, as the FDIC's new Chairman, I believe that we should work together to make these reform proposals a reality, and I commend this Committee for providing us the opportunity to discuss this important issue.