Speech

For release on delivery 9:00 a.m. CDT (10:00 a.m. EDT) May 10, 1990 Remarks by Alan Greenspan Chairman, Board of Governors of the Federal Reserve System before the Annual Conference on Bank Structure and Competition Federal Reserve Bank of Chicago Chicago, Illinois May 10, 1990 SUBSIDIES AND POWERS IN COMMERCIAL BANKING It is a pleasure to be invited again to speak to the Chicago Federal Reserve Bank Conference on Banking Structure and Competition. Two years ago I spoke to this Conference on the need to repeal the Glass-Steagall Act. This morning I want to renew that call, but also to explore in some depth one issue inherent in the expansion of bank powers -- implications for the reach of the safety net. There are, I believe, two major strands in the evolving debate about the U.S. banking system that events now require us to confront. Over the last decade or so, the debate has been mainly about the first strand, deregulation and the responses to technological change and global competition. A second strand concerns the special relationship of the government to depository institutions and how that relationship might evolve as those organizations take advantage of the greater operating scope allowed by deregulation. The issue, of course, is the safety net and the associated supervision and regulation that comes with access to that safety net. Most recently, the savings and loan failures have focused our attention on deposit insurance reform, but I would like to suggest this morning that the deposit insurance issue is simply a subset, albeit an important one, of this broader second strand. In addition, there are questions about the degree of subsidy in banking -2- unplied by the safety net, about the possibilities of extending that subsidy if -- or should I say as — deregulation is expanded, and about the implications of any such extension. In a textbook model of a commercial bank in a market economy, banks raise funds from stockholders and depositors and lend those funds to businesses, households, and government. In the process, they intermediate between their depositors and borrowers, attracting the borrowers with lower rates than would be required on securities sold in the open market and the depositors by proffering diversification, convenience, liquidity, and payments services. Intermediation benefits lenders and borrowers because the intermediary service adds economic value by applying specialized knowledge, informed credit judgment, scale, diversification, and technology to the process of transforming saving into investment. Indeed, if the intermediation process does not create economic value -- that is, if the operating costs of banking exceed the return required by owners — the bank will fail. Even when profitable, textbook banks are not without risk. Banks can make errors in selecting assets or balancing maturities, and the associated credit, interest rate, and liquidity risks can cause them to fail. Pure market-economy banks are thus forced to maintain substantial capital to attract funds. While the historical data are admittedly -3- distorted by a number of factors, it is still instructive to note that in 1840, the average U.S. banks' equity-capital-to- total-asset ratio was around 50 percent. While generally declining over the next 75 years, equity ratios were still around 12 percent in the late 1920s. Such high equity capital ratios were not the choice of bankers, but rather the result of market pressures to provide comfort to depositors that banks could, in fact, live up to their side of the agreement. As with many market solutions, its secondary incentive reinforced the primary objective: with so much of the owners' money funding the bank, risk appetites were constrained, strengthening the likely ability of banks to fulfill their obligations. By 1974, the equity capital ratio of the largest banks had fallen to less than 4 percent. Last year, after over a decade of strenuous supervisory effort, offset by sizable loan loss reserving, the equity-capital-to-total-asset ratio was about 5 percent for the 25 largest banks. In 1989, equity capital ratios at all U.S. banks were about half of those in the late 1920s. The driving force that has permitted equity capital- to-asset ratios in banking to go from a little less than 25 percent in 1890 to a little over 6 percent in 1990 is the set of statutory and regulatory changes that have drastically lowered the risk to depositors. A complete list might be rather lengthy, but this morning I would like to focus on what -4- I think are the major factors reducing depositor risk premiums and permitting banks in the United States to operate with considerably less capital than free market models would imply, namely deposit insurance and access to the discount window and to Fedwire. Deposit insurance and the discount window -- the major elements of the safety net -- were designed to afford depositors an extra measure of protection from loss and by doing so to shield the aggregate real economy from some of the worst effects of instability in banking markets. A loss of confidence in the soundness of one or more banks by depositors can engender a contagious withdrawal of deposits — a "run" on banks in general. The associated forced liquidation of bank assets, in turn, can not only put still other banks under pressure, but can also cause difficulties among enterprises that lose access to credit because they rely on banks for funds. Moreover, disruptions in the payments mechanism -- particularly the large dollar payments system -- can have devastating effects on the flow of trade and commerce. The framers of the Federal Reserve Act were particularly sensitive to disruptions spreading if individual banks were unable to honor requests for deposit withdrawals. The discount window was designed to cushion shocks to the banking system by giving solvent institutions an opportunity to liquify their good, though illiquid, assets so as to be able to meet withdrawal requests. The ability of a depository -5- institution to hold a less liquid, and presumably more profitable, portfolio with "windows" access than without it, is a measure of the subsidy accorded by the access. The discount window was not suited to meeting massive liquidity withdrawals when there were widespread bank insolvencies and deposit insurance helped to fill that gap by bolstering the confidence of at least small depositors under such circumstances. The combination of deposit insurance and a central bank providing discount window credit has made the currency drains that dominated the 19th and early 20th century banking literature an anachronism. The potential for massive bank-to- bank shifts of deposits from one set of banks to another remains, but the United States has not suffered a financial panic or systemic bank run in the last 50 years. In large part this reflects the safety net, whose existence, as much as its use, has helped to sustain confidence, although general macro policies, and perhaps luck, also have played a role. Increased macro-economic stability is a real benefit, and should not be taken lightly. However, it should be emphasized that these benefits are not purchased without cost to our economy. The safety net lowers the risk premium on bank liabilities, permitting banks to operate with lower capital or with higher-risk asset portfolios. Both the lower deposit rates and the smaller capital base reduce banks' costs, permitting them to profit more from the same portfolio. -6- The capitalized value of this benefit is captured by stockholders so long as entry is restricted. To be sure, the total subsidy may not ultimately show through in substantially higher bank profits. Through competition, lower costs are shared with customers in lower risk-adjusted loan rates and more attractive deposit instruments. In addition, to gain access to the safety net banks assume a substantial and potentially costly regulatory burden, including such things as reserve requirements, deposit insurance premiums, Community Reinvestment Act obligations, and general supervision of their business decisions. But even if the safety net has little effect on bank profits, it still distorts resource allocation in our economy. By giving governmental assurances to bank depositors of the availability of their funds, the safety net enables banks to have larger, riskier, asset portfolios than would be possible in a market-driven intermediation process. Without the safety net, additional lending risks would have to be reflected in some combination of higher deposit costs, greater liquid asset holdings, or a larger capital base, and these in turn would constrain risk-taking. In theory, one might be able to price safety net access at something approximating its free market value and thereby remove the subsidy to depository institutions, depositors, and borrowers from banks. Such a step could reduce risk-taking and the associated credit-allocation -7- distortions of the safety net without reliance on heavy-handed regulation and supervision. But the safety net has been constructed by government because private market decisions cannot adequately incorporate the perceived costs to the economy of systemic risk. Thus the price of the safety net as offered by government should be lower than its market value to individual private participants, necessitating some prudential regulation. This governmental propensity to curb excesses resulting from distorted incentives, partly in order to limit taxpayer exposure, has been afforded inadequate attention, at least until recently. Ideally, one would like banks to be managed as if there were no safety net, to see their profits reflect solely the value added from intermediation, and not supplemented by, or perhaps even dominated by, the subsidy inherent in the safety net. Put differently, we want to avoid banks' benefiting from risk premiums in their assets that are not reflected fully — if at all — in their liabilities and capital costs. And, similarly, we ought to be at least aware that some of the pricing distortions of the safety net are reflected in lower loan rates or higher credit availability for riskier borrowers than would be the case in the textbook market economy model described earlier. The safety net has a tendency to benefit speculative and riskier ventures at the expense of sounder ones. Indeed, the safety net tends, other things equal, to increase the nation's overall real rate of -8- interest by facilitating the ability of riskier borrowers to translate their potential credit demands to effective control over resources, crowding out projects that would be economic at lower real rates. Rules, regulations, and supervision cannot substitute for market signals; they can only attempt to filter the worst missignals that seem to suggest to bank management that unusual risk taking is permissible, if not desirable. Still, to individual bankers, such regulation can seem quite onerous and overly constraining. In an environment of global competition, rapid financial innovation, and technological change, bankers understandably feel that the old portfolio and affiliate rules, and constraints on permissible activities of affiliates, are no longer meaningful and likely to result in a shrinking banking system. This has led some bankers to argue that perhaps they should turn in their bank charter in order to be able to do the business denied to them by statute and regulation. But, if one places any weight on the theory of revealed preference, it is perhaps significant that no matter what the rhetoric, no commercial bank has given up its charter in order to become a financial institution without deposit insurance or access to the discount window. Indeed, many nonbanking institutions are trying very hard to obtain a bank charter. I suggest that the subsidy of the safety net is not irrelevant in explaining these events. Some banks find deposit insurance the key benefit. Others -- -9- relying importantly on the wholesale money market for finance -- find the discount window the primary element in the safety net. Virtually all banks would require larger capital bases to operate as financial intermediaries without one or both of these elements. Banks also consider access to Fedwire an important factor in their operations as an intermediary in the payments or securities business. Access to Fedwire is often overlooked as a component of the safety net. The benefit of Fedwire to banks and their customers is the ability it gives to make sizable transfers of funds during the day without necessarily having first received funds to cover the transfer, and to receive funds with absolute assurance that such transfers will not subsequently be revoked by the failure of the sending party. Meaningful access to Fedwire in turn requires daylight credit extended by Reserve Banks. The degree of subsidy associated with Fedwire will decline as pricing for daylight overdraft credit is implemented. But exclusive access, even if priced, is an aspect of the safety net, not unlike the discount window itself, permitting banks to engage in activities that institutions without access cannot develop. The parallel with the discount window is relevant in another way. When granting daylight credit on Fedwire, the Federal Reserve must consider the implications of the institution incurring the overdraft being unable, for operational or other -10- reasons, to extinguish its debit position by the end of the day. In such circumstances, the Reserve Bank could be forced to provide a discount window loan. It is for this reason that access to Fedwire is -- and should continue to be -- limited to those to whom the Federal Reserve can provide discount window credit. I should note that some insured depositors with access to the discount window are prohibited daylight credit on Fedwire — or are required to post collateral for such access — because they are undercapitalized. The safety net, in sum, has provided measurable benefits to the U.S. economy: it has cushioned disturbances, provided flexibility, protected depositors, insulated banks from the contagion of deposit runs, and has virtually eliminated financial panics. But it also has had, and will continue to have, real costs. Over and above the real taxpayer costs when supervision fails to constrain the worst excesses, the safety net distorts market signals, induces managers to take on risk that does not offer the possibility of commensurate economic benefits, requires supervisors and regulators to monitor and modify behavior induced by distorted market signals, increases borrowing by riskier firms, and probably increases the real interest rate. In the context of these trade-offs, how should we consider the interactions of the safety net and wider powers for banking organizations? The concern, of course, is that permitting banking organizations to engage in a larger number -11- of activities will spread the safety net ever wider and wider through a network of bank affiliates. Such an extension would threaten to intensify the adverse effects of suppressing market signals, and raise questions about what the safety net is designed to protect. By design, firewalls should avoid the extension of the safety net, keeping it under the commercial banks alone. But the more effective the firewall, the more it reduces synergies and undercuts the reason for granting any additional powers to banking organizations in the first place. If firewalls are modest by design, the safety net will of course be widened as new powers are granted. Let us be quite clear how the safety net is transferred to affiliates of banks, because it is not direct access to the safety net by affiliates that is being discussed. Rather, transference occurs by (1) bank use of insured deposits to make loans to, or purchase assets from, affiliates and/or (2) the lowering of financing costs of bank affiliates because the market believes that the risk of difficulties spilling over from affiliates to banks will induce the authorities to support the affiliate in times of duress in order to protect the bank. The effects of the first channel can be, and have been, limited by law and regulation. Continued strenuous effort by the authorities may be able to mitigate the reputational spillover effects of the second channel. -12- But if the supervisory efforts to enforce firewalls are considered too burdensome, or the spread of the safety net were of sufficient concern because of its implications for distorting risk-taking, one option would be to grant no new powers to U.S. banking. It seems reasonably clear -- but not certain -- that the result of such an approach would be that the banking system would contract over time in relative, if not absolute, terms. Such a development might have little consequence for customers — both depositors and borrowers — who would be served by other institutions as new channels evolve. Nevertheless, there may be reasons to be concerned should this occur. Capital and specialized personnel would have to migrate from banks to unregulated institutions and this shift of resources has at least short-run economic cost. This is not a simple issue and requires a balancing of benefits and risks before we acquiesce in the existing regulations and limits, coupled with waves of technological change and global market innovations, to divert banking resources to other institutional structures. In addition, in such an environment, we must also address the question of whether a safety net stretched under a shrinking proportion of the financial system would be adequate to accomplish one of its goals — assuring financial stability. An expanding noninsured nonbank financial system may develop the capital base and the market discipline from -13- its creditors sufficient to reduce significantly any systemic risk that might otherwise occur from such a structural shift. On the other hand, particularly as financial institutions engage in more and more similar activities, disruptions and pressures in nonbanking financial markets may create systemic risk similar to that faced in earlier years in a narrower banking system. If nonbank financial institutions affiliated with banks are truly subject to market discipline, it may not be necessary to face the implications of a wider safety net in order to constrain macro risk in the real economy. But if events lead to wider bank holding company powers and more limited firewalls, requiring a wider safety net on stability grounds, we face the distinct possibility of distorted market signals over a wider range of markets, excessive risk taking by financial institutions, and misallocation of resources. In such circumstances, we would be well advised to consider ways to limit the costs of the safety net. Almost surely access to the safety net would bring with it political pressures for enhanced supervisory oversight. As a result, the package that transfers the benefits of the safety net to bank affiliates would probably require that the supervisors look over the shoulder not only of the bank but also of the affiliates with whom the bank can deal. Particularly with a wider safety net, we should probably develop additional market-simulating ways of limiting moral hazard, choosing from among risk-based -14- insurance premiums, more frequent supervisory evaluation of assets, prompt resolution for undercapitalized banks, and priced daylight overdrafts, to name just a few. We might also consider the desirability of returning to capital requirements that more closely simulate those that the market dictated in a world without a safety net. Such an approach would increase capital costs for the banking organization with wider powers. Indeed, there are those who argue that a high-capital entity would not be able to compete in domestic and international markets. Whatever the relative competitive balance might be, however, the stockholder and management of such an organization would tend to have a less aggressive risk appetite and would certainly be more able to absorb risk on its own. What is becoming increasingly clear, if it was not before, is that deposit insurance, the Fedwire, and the liquification services of central banking are not free lunches. They provide more macro stability, but they misprice risk. Our body politic appears to have chosen macro stability. Nonetheless, the costs of the associated safety net have not always been sufficiently considered and reform in the safety net should be high on our agenda. There is no optimal solution, I fear, and, as in all policy reviews, a series of trade-offs will be necessary.

Alan Greenspan (1990, May 9). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_19900510_greenspan

@misc{wtfs_speech_19900510_greenspan,
  author = {Alan Greenspan},
  title = {Speech},
  year = {1990},
  month = {May},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_19900510_greenspan},
  note = {Retrieved via When the Fed Speaks corpus}
}