Federal Deposit Insurance Corporation
Federal Reserve Bank of Chicago
Conference on Bank Structure and Competition
May 4, 2000
I am delighted to speak at this 36th Conference on Bank Structure and Competition.
Part of the long and proud history of this Conference has been an ongoing dialogue about deposit insurance pricing. Indeed, the first papers discussing risk-based deposit insurance were presented at the Structure Conference in 1980. From 1980 until the FDIC implemented risk-based premiums on January 1, 1993, more than 20 papers that had deposit insurance pricing either as their main subject or as an important subtext were presented. And doubtless, during that time, many more Structure Conference papers made at least a passing mention of what -- by the early 1990s -- had become the generally accepted academic wisdom: that is to say, the importance of risk-based premiums in mitigating the moral hazard problems that can attend any deposit insurance system.
In 1991, that accepted academic wisdom enjoyed a moment in the sun, being given the force of law when FDICIA required the FDIC to implement a risk-based premium system. Since the FDIC implemented the new premium system in 1993, the number of Structure Conference papers that have had deposit insurance pricing as their main subject has dropped to zero. The dialogue has moved on to other topics. For those who advocate more fundamental changes-such as the complete privatization of the deposit-insurance system, the use of cross-guarantee arrangements, a more formal use of co-insurance to promote market discipline, or the use of bank subordinated debt or FDIC Capital Notes to price risks-the debate is now less about deposit insurance premiums than it is about deposit insurance paradigms. And by adding momentum to the trend toward larger and more complex financial institutions, the Gramm-Leach-Bliley Act should lead us to consider new approaches to regulation, to supervision-and to deposit insurance.
And yet, if depositors are protected fully or partially from loss, the appropriate pricing of that protection will always be an important subject under any financial regulatory paradigm that emerges.
Today, I want to talk about deposit insurance pricing. First, I want to ask how well we are meeting the goals laid down by FDICIA, and the challenges introduced by Gramm-Leach-Bliley. Second, I want to point to some areas where we may be able to do better. And, third, I will talk about how the FDIC intends to move forward on deposit insurance reform.
It is important to address deposit insurance reform now, before the long economic expansion we have all enjoyed turns - as it surely will. An increase in some banks' appetite for risk and a decrease in attention by some to sound banking practices could combine with a potentially more volatile economic environment to increase our insurance losses beyond what the financial statements, the examination ratings, and the premiums being paid by insured institutions might suggest.
How will history judge the effectiveness of the risk-based premium system in achieving the goals for which it was established? In particular, did the premium structure price risk-taking appropriately? Did it promote the sharing of the costs of the deposit insurance system in a fair and equitable manner? Did it adjust to fundamental changes in banking industry structure? These are questions by which future presenters at this Conference may someday judge our deposit insurance pricing arrangements.
It is true that the FDIC's premium system does not price nor differentiate risk as aggressively as a private insurer would. Nevertheless, it does provide a financial stimulus for banks to achieve a well-rated, well-capitalized status. And although this stimulus is only one among many regulatory sanctions that 3-, 4-, and 5-rated banks face, we should not be too quick to dismiss the importance of bank management's desire to report to the Boards of Directors that they are paying the lowest FDIC insurance premium. And because the system uses well-established risk measures, there have been relatively few complaints over the years that the system is arbitrary or unfair.
It is my hope that history will count these features among the strengths of the current risk-based pricing structure - but let's look at some of the weaknesses and where we can do better, as well.
Most banks today pay nothing for deposit insurance. One of the striking consequences of our current price structure is that both new and existing institutions can grow rapidly, thereby lowering their fund's reserve ratio and increasing the need for premium income from all other institutions, at no premium cost to themselves. And as institutions grow larger, the potential movements in the fund reserve ratio resulting from deposit injections are becoming more dramatic as well. Further, and ironically, another example of the effects of a zero price for insurance is that we may see the banking industry paying -- through the insurance funds -- for the failures of banks that never paid an FDIC insurance premium.
This ability of some banks to impose costs on the system at no cost to themselves - as well as the lack of a financial benefit to those banks that choose to reduce the exposure they impose on the rest of the industry - do raise questions about the equity and fairness of our deposit insurance pricing structure, and about its ability to influence bank risk-taking. We at the FDIC should - and we will -- develop options to address these issues.
Another question that future presenters at this Conference may ask is whether our pricing system was sufficiently forward-looking. That is, did we price risk on an ex ante basis, or did we simply penalize banks after the fact, when events proved that the risks they selected were ill-advised? The truth is somewhere between those two extremes, but there is clearly room to be more forward-looking in our deposit insurance pricing. In gearing premiums to the CAMELS ratings, we have set ourselves an easy target. Banks rated 3, 4 and 5 are subject to heightened supervision and generally are in some stage of formal or informal enforcement action. The extra premiums, therefore, simply add to the sanctions already imposed on them.
These comments should not be construed as a suggestion that we need to systematically second-guess the risk-management decisions of safe-and-sound banks through the premium system. But where there are objective indicators of a significantly higher risk profile, we need to consider whether those indicators should be factored into the deposit insurance premium, even where the bank may presently be a strong performer.
In this regard, the 9,500-plus banks in the 1A category-well-capitalized banks rated 1 or 2-encompass a wide range of risk appetites and exposures. Yet, they all pay the same zero price for their insurance. Some of these institutions have a high appetite for, or exposure to risk, but because of their strong financial performance or condition they may retain a favorable examination rating and not pay premiums. I say this, not to rattle my sabers about the need for higher premiums in general, but out of a desire I think we all share, that the premium cost of financing the system should fall to the extent possible on banks that elect to assume the greatest risks.
In this regard, there has always been a small subset of well-rated institutions that had atypically high risk-profiles and deteriorated rapidly, some because of adverse changes in the external economy, and others because of internal problems. We saw those in the 1980s in great numbers. More recently, in 1999, as a result of an uptick in the incidence of unanticipated or high-cost failures, the FDIC experienced its first annual loss in the Bank Insurance Fund since 1991. The banks that accounted for most of that loss were not paying premiums two years or less prior to their failures.
It is too soon to call a trend based on this recent, limited experience. Nevertheless, coming as they have during good economic times, recent failures are a reminder that we need to take a good, hard look at the premiums paid by some FDIC-insured institutions with 1A-ratings, especially those whose performance looks too good to be true - and we will.
As we consider refinements to our pricing structure, we should guard against creating a profusion of hair-splitting, and potentially subjective, distinctions among banks. One of the strengths of the current approach is its simplicity. Ideally, any change to the pricing system would achieve a more forward-looking approach in a simple and straightforward manner. It is too early to say how that ideal can best be achieved, but for purposes of discussion, we could consider ways to put three sources of information to better use. Those three sources of information are the financial reports that banks routinely produce, the onsite information generated by supervisors, and the signals generated by market participants.
First, we need to make the best use we can of the financial reports banks produce already. For each of the last two premium cycles, for example, the FDIC has flagged some 200 1A-rated institutions for further supervisory review. For some of these banks, the result may be an insurance downgrade. The screens used include financial ratios that can be indicative of a high-risk appetite, such as concentrations, growth, high asset yields, and the like. This screening and review process is an interagency effort that will evolve, and will benefit from the input of all interested persons.
Second, there may be ways to enhance our use of supervisory information. Examiners have a wealth of information about risk-management practices of insured institutions. Finding a way to synthesize some of this information in one or more risk ratings-ratings that would be geared more to risk appetite, risk exposure, and risk-management expertise than to financial performance and condition-could provide extremely useful information from the standpoint of insurance pricing. To some extent this is already being done, through the various risk matrices and other risk-focused supervisory approaches used by individual agencies. The development of a widely accepted supervisory framework for rating risks in the great majority of institutions that are in strong financial condition may be difficult to achieve, but would be of extraordinary value to the FDIC.
Refining the use of financial and supervisory information would be a way of building on the underlying pricing philosophy and procedures we already have in place. We should also be willing to consider enhancing existing procedures with new conceptual approaches, specifically a more systematic use of information generated by the market. Here, I would like to return to one of the questions I raised earlier. Namely, is our deposit insurance pricing adjusting appropriately to the evolution of industry structure? Many in this room have observed that the evolution of the financial services industry toward larger and more complex institutions will force supervisors to rely more on transparency and market discipline as a pillar of safety-and-soundness. The announcement last week of the newly created Working Group on Public Disclosure is an exciting development. High-quality disclosures have the potential to be a pillar not only of safety-and-soundness regulation, but also of efforts to use market signals in deposit insurance pricing.
Gramm-Leach-Bliley has two provisions that recognize the potential importance of market discipline in assisting safety-and-soundness regulators. The Act requires the Board of Governors and the Treasury to conduct a study of the feasibility and appropriateness of requiring the largest banks to maintain a minimum level of subordinated debt. It also requires that the granting of new powers to the subsidiaries of the 50 largest national banks be conditional on the parent bank having at least one issue of outstanding long-term unsecured debt rated in one of the top three categories by a credit rating agency.
As we consider how to harness market signals to enhance safety-and-soundness, we should also consider how deposit insurance pricing could benefit from this effort. A subordinated debt requirement -- were it to be implemented -- could result in useful information about the pricing of those instruments, information that could serve as an input to the insurance pricing decision. This idea has, of course, been discussed in many forums, including last week at the Stigler Center Conference and, not surprisingly, in a paper presented here at the Structure Conference in 1991.
Another way to get a market view of the risks the FDIC faces would be for the agency to enter into risk-sharing arrangements with market participants. The private reinsurance study mandated by FDICIA, and completed by the FDIC in 1993, considered one such approach. Since that time, techniques for pricing and transferring financial risk have evolved considerably. It is not unreasonable to suppose that contracts could be developed that allow the FDIC to transfer -- and price in a competitive market -- a portion of risks faced by the insurance funds. The pricing of such financial instruments could provide the FDIC with useful information about the relative risks of banks as perceived by market participants.
Market-based approaches could be especially useful in enhancing deposit insurance pricing for the largest institutions. And we should bear in mind that there is no hard-and-fast constraint that requires the FDIC to maintain a one-size-fits all approach to pricing deposit insurance. FDICIA specifically provided authority for the FDIC to establish different premium systems for large and small institutions insured by the same fund.
What do we need to do about deposit insurance reform?
Last week in Washington, the FDIC sponsored a Deposit Insurance Roundtable to discuss issues regarding insurance pricing, fund adequacy and coverage. The candid exchange of views at the Roundtable, and upcoming sessions scheduled for bank chief executive officers, will serve as a useful starting point for an FDIC options paper on deposit insurance issues that will be released later this summer for public comment.
That options paper will have to recognize and address a significant constraint facing the FDIC's pricing structure for deposit insurance. FDICIA gave the FDIC virtually unlimited flexibility to consider a wide range of information when assigning insurance categories. In 1996, however, the Deposit Insurance Funds Act included a provision that was intended to reduce the FDIC's flexibility to charge a well-capitalized bank for deposit insurance unless the bank is rated 3, 4 or 5. I feel safe in saying that one of the options that will be presented would be to restore the FDIC's discretion in this area.
Now I am sure that when some of you hear the FDIC speaking the words "discretion" and "premiums" in the same breath, you immediately suspect a thinly veiled ploy to ask bankers to open their wallets. That is not the case. The FDIC already has the flexibility to raise premiums for all institutions in order to maintain the designated reserve ratio. We may not be at that point yet, but we're closer than we were one year ago. The probability that the Bank Insurance Fund reserve ratio would fall below 1.25 this year might appear small, but it is rising, and is by no means as remote as one might think.
Why? Because of the factors we are seeing in a small subset of the banks we insure: an aggressive appetite for revenue growth; weaker loan underwriting; deficiencies in internal controls; in short, factors that in an ideal world already would be reflected in the insurance premiums, so that the institutions furthest out on the risk curve pay their share of the costs they impose on others.
If and when we need more assessment revenue to maintain the designated reserve ratio, we would like to be able to raise it in a fair and equitable manner. We would like the verdict of history to be that our deposit insurance pricing was equitable, that it was forward-looking, and that it responded to a changing industry structure.