Gary H. Stern - President, 1985-2009
Published June 1, 1991 | June 1991 issue
Reform of the financial services industry is currently at the forefront of America's domestic agenda. The administration, members of Congress, banking associations and regulatory agencies have all advanced ideas that, they believe, are essential to the industry's future.
While they differ in details, most of these reform proposals generally address three issues: supervisory and regulatory reform, including capital-based supervision and changes in deposit insurance coverage; geographic and activities expansion for banks, including nationwide interstate banking and broader securities powers; and regulatory agency restructuring and consolidation.
Conceptually, there is little to object to in these ideas, but in practice these measures are inadequate on two distinct levels. First, the proposed reforms of the deposit insurance system do not address the fundamental flaw in the system, namely the moral hazard problem. As matters now stand, risk taking in banking is underpriced and, as a consequence of this mispricing, depository institutions whose liabilities are insured take on more risk than they should, especially since insurance coverage is virtually without limit. In fact, as matters now stand, it is possible for even insolvent institutions to increase their coverage and the size of their operations. The results of current deposit insurance policy are evident (in the extreme) in the savings and loan industry and in the string of earnings "disappointments" at many large commercial banks over time.
The second level on which many of the supervisory and regulatory reform proposals are inadequate is in their practical implementation. We may agree on the desirability of higher capital standards but, so far as I know, no one is seriously proposing a return to the capital levels prevailing before federal deposit insurance, although presumably those are roughly the levels required to contain moral hazard. As it is, some institutions have difficulty meeting the modest tangible leverage ratio now in effect.
Moreover, how do you rigorously maintain capital standards and impose prompt supervisory discipline without market value accounting? It seems that much, perhaps too much, is left to the discretion of the bank supervisors, a problem also encountered by suggestions for risk-based deposit insurance premia administered by supervisors. As the problems of the industry suggest, we may already be asking the supervisors to do too much.
Another example of excessive reliance on supervisors may be found in the so-called "credit crunch" controversy. Personally, I think this is one of the more exaggerated issues of the day. But to the extent that there is a credit crunch, it has resulted in part from the inherent difficulties bank examiners have in distinguishing between "good" and "bad" real estate loans. Do we really want to introduce more supervisory discretion in banking?
In thinking about the banking and financial system of the future, and the principles which, from a public policy perspective, might guide its evolution, several considerations come to the fore.
How do we implement these principles? I doubt that government officials can or should provide a detailed road map. Preferably, we call on the market or, more precisely, make it possible for market forces to play an increasing role in determining the future shape and scope of the banking and financial services industry.
To accomplish this, we must start by correcting the incentives and redressing the misallocation of resources resulting from the current deposit insurance system and the "too big to fail" doctrine. As matters now stand, we have a system that relies on extensive regulation and supervision to offset the moral hazard inherent in deposit insurance. When this fails, as in the savings and loan industry, the taxpayer is at risk.
To help straighten out incentives and increase market discipline on insured depository institutions, we at the Federal Reserve Bank of Minneapolis have previously proposed a form of coinsurance. The basic elements are as follows. Over and above the nominal $100,000 insurance limit (and only one insured account, in the system, per customer), depositors would explicitly be at risk for, say, 10 percent of their deposit, so that insurance coverage over $100,000 would be 90 percent, in contrast to the de facto 100 percent coverage prevailing today. This 10 percent exposure, we believe, would give (large) depositors ample incentive to pay attention to the calibre of the institutions with which they do business, again in contrast to current practice where such incentive is largely lacking.
Other changes would follow from this one. For example, increased disclosure of banks' financial condition is likely, especially as strong institutions find it in their interest to make their virtues known. A much more fully developed market for information about the condition of banks would emerge over time. Depositors might also become more interested in diversifying the list of institutions with which they do business.
Several caveats about this proposal are, of course, in order. Such a change in deposit insurance coverage cannot be introduced overnight. A period of preparation and phase-in should be allowed, so that both banks and depositors can adjust. Moreover, coinsurance is not, in my view, a substitute for supervision. Rather, it is a complement to it that relies on market discipline to a greater extent than is the case today.
Further, coinsurance is likely to increase the frequency of bank runs. Before recoiling at the thought, we should recognize that this is not all bad. On the one hand, runs that occur at open but insolvent institutions represent a market version of prompt intervention and are desirable, especially if one believes there is excess capacity in banking. On the other hand, runs that occur at solvent institutions or threaten to become systemic are problematic, but there is the Federal Reserve discount window to provide liquidity. That is why the window was established in the first place. I recognize, in contemplating a higher incidence of bank runs, that there may be an understandable reluctance to embrace coinsurance. But it is not as if the current system has worked all that well; witness the cost of resolving the savings and loan problem borne by the taxpayer. Think about the current debate about how to "recapitalize" the Bank Insurance Fund.
There are at least two other significant advantages of coinsurance beyond the market discipline it directly imposes. First, it is the only reform proposal of which I am aware that, at least potentially, provides a vehicle for eliminating the pernicious doctrine of too big to fail. While many, perhaps most, analysts, policy-makers and bankers agree that too big to fail must be discarded as a policy, such agreement in itself is insufficient. A mechanism, a method for terminating the policy must be identified.
The reason that coinsurance provides a vehicle for dealing with too big to fail is that it caps the size of depositor exposure and, if desired, the exposure of other creditors as well when a bank fails. Remember that under the coinsurance proposal a depositor is at risk for only 10 percent of his balance over $100,000. Thus, spillovers are held to modest proportions, potentially allowing supervisors to treat institutions of any size identically. In the Continental Illinois case, perhaps the precedent setter for the too big to fail policy, concern for the large number of small banks which had correspondent relations with, and considerable exposure to, Continental was a major impediment to closing and liquidating the organization.
Coinsurance could also quickly end the debate about the advisability of imposing market value accounting on banks. Assuming that coinsurance would contribute to effective depositor discipline on insured institutions, insisting on marking a bank's balance sheet to market would be unnecessary. Depositors would use all available tools and make whatever calculations they felt necessary to assess the safety and soundness of the institutions with which they conduct business.
With increased market discipline, regulators should be far less concerned than they are today about the range of activities in which insured institutions engage, whether such activities are in the bank or in the holding company, or the geographic scope and size of the operation. So long as large depositors understand they are truly at risk and act accordingly, risk taking will be more accurately priced than it is today and bank management will have an incentive to act more prudently. With the proper incentives in place, we can leave it to bank management to pursue profit opportunities by determining the services they want to provide and where they want to provide them.
In leaving these decisions to bank management, I suspect we will end up with a financial services industry at least as diverse as, and more competitive, domestically and internationally, than the one we have today. Some firms will operate internationally, some will confine themselves to a local market. Some large firms will carve out profitable niches and stick to them; other large firms will aggressively expand both service lines and geographically. There will be mergers and there will be consolidation, and over time there will probably be fewer firms doing a traditional banking business than there are today. But this hardly need concern us, given the large number of firms in the business.
In my view, the essential elements of reform of the banking system are clear. We should increase market discipline on banks by introducing coinsurance, a step that could also facilitate termination of the policy of too big to fail. Once coinsurance is in place, banks should be permitted to move ahead as they see fit with geographic and activities expansion. Legislation will of course be required, so none of this is simple. Nevertheless, I am convinced it is the best of the alternatives.
Special study: Too Big To Fail