Testimony of Chairman Alan Greenspan|
Modernization of the financial system
Before the Subcommittee on Financial Institutions and Consumer Credit of the Committee on Banking and Financial Services, U.S. House of Representatives
February 13, 1997
Madam Chairwoman, members of the Subcommittee on Financial Institutions and Consumer Credit, it is a pleasure to appear here today to present the views of the Federal Reserve Board on some broad issues associated with financial modernization. The unremitting pressures of technology and the market are drastically changing the financial landscape and eroding traditional positions of competitors, inducing new competitive strategies and participants, forcing new regulatory responses, and building pressures on the Congress to shape developments in the public interest.
Madam Chairwoman, I know that you have been an active sponsor and supporter of legislation to modernize the financial system. The Board also has been a strong proponent both of expanded financial activities for banking organizations and enhanced opportunities for nonbank financial institutions to enter banking. We continue to support financial modernization because we believe it would provide improved financial services for our citizens. Moreover, both our experience and analysis suggest that the additional risks of new financial products are modest and manageable. Indeed, technology already has resulted in a blurring of product and service-defining lines so dramatic as to make many financial products virtually indistinguishable from each other and the old rules inapplicable. In the process, we have already seen the public benefits, benefits that removal of old barriers could only enhance.
But, as we proceed down the path of reform, reforms both desired for their benefits to the public and required by global markets and new technologies, the Board urges that any modifications be tested against certain standards. In particular, the Board believes that the changes we adopt should be consistent with (1) continuing the safety and soundness of the banking system; (2) limiting systemic risks; (3) contributing to macroeconomic stability; and (4) limiting the spread of both the moral hazard and the subsidy implicit in the safety net.
Thus, if my comments today sound cautious, I want the subcommittee to understand that my observations do not reflect opposition to further freeing of constraints on financial competition. To the contrary. We strongly urge an extensive increase in the activities permitted to banking organizations and other financial institutions, provided these activities are financed at nonsubsidized market rates and do not pose unacceptable risks to our financial system. While a level playing field requires broader powers, it does not require subsidized ones.
Safety Net Implications
Return to top
| In the process, it has profoundly altered the risks and returns in banking. Sovereign credit guarantees have significantly reduced the amount of capital that banks and other depositories need to hold, since creditors demand less of a buffer to protect themselves from the failure of institutions that are the beneficiaries of such guarantees. In different language, these entities have been able to operate with a much higher degree of leverage--that is, to obtain more of their funds from other than the owners of the organization--than virtually all other financial institutions. At the same time, depositories have been able to take greater risk in their portfolios than would otherwise be the case, because private creditors--depositors and others--are less affected by the illiquidity of, or losses on, the banks' portfolios. The end result has been a higher risk-adjusted rate of return on depository institution equity.
Moreover, the enhanced ability to take risk has contributed to economic growth, while the discount window and deposit insurance have contributed to our macroeconomic stability. But all good things have their price. The use of the sovereign credit in banking--even its potential use--creates a moral hazard that distorts the incentives for banks: the banks determine the level of risk-taking and receive the gains therefrom, but do not bear the full costs of that risk. The remainder of the risk is transferred to the government. This then creates the necessity for the government to limit the degree of risk it absorbs by writing rules under which banks operate, and imposing on these entities supervision by its agents--the banking regulators--to assure adherence to these rules. The experience in the 1980s with many insured thrift institutions showed just how dangerous lax enforcement of supervisory rules can be. In the end, some hard lessons were learned, many of which were legislated into the FDIC Improvement Act of 1991.
The subsidy to the banking and other depositories created by the use of the unsurpassable sovereign credit rating of the United States government is an undesirable but unavoidable consequence of creating a safety net. Indeed, one measure of the effectiveness of a safety net is our ability to minimize the subsidy and limit its incidence outside of the area to which it was directed. Some of the value of the subsidy has been passed to depositors of, and borrowers from, banks, for example, as well as to the original bank shareholders. But, the United States government has been remarkably successful in containing the value of most of the subsidy within depository institutions. The bank holding company organizational structure has, on balance, provided an effective means of limiting the use of the sovereign credit subsidy by other parts of the banking organization. To be sure, bank holding companies have indirectly benefited from the subsidy because their major assets are subsidiary banks. The value of the subsidy given to the subsidiary banks has no doubt been capitalized in part into the share prices of holding companies and has improved their debt ratings, lowering their cost of capital. But, holding companies also own nonsubsidized entities that have no direct access to the safety net. Accordingly, both bank holding companies and their nonbank subsidiaries have a higher cost of capital than banks.
This is clear in the debt ratings of bank subsidiaries of bank holding companies, which are virtually always higher than those of their parent holding companies. Moreover, existing law and regulation under Sections 23A and 23B of the Federal Reserve Act require that any credit extended by a bank to its parent or affiliate not only be totally collateralized and subject to quantitative limits, but also be extended at arms-length and at market rates, making a direct transfer of the safety net subsidy difficult.
It is true that a bank could pay dividends from its earnings, earnings which have been enhanced by the safety net subsidy, to fund its parent's nonbank affiliates. However, the evidence appears to be that such transfers generally do not occur. Existing holding company powers are limited and do not offer a broad spectrum of profitable opportunities. Accordingly, it is not surprising that data for the top 50 bank holding companies indicate that transfers from bank subsidiaries to their parents which, like dividends, embody the subsidy, appear to have approximately equaled holding company net transfers to their own shareholders and long-term creditors. This indicates that few subsidized dollars in the aggregate found their way into the equity accounts of holding company nonbank affiliates from the upstreaming of bank funds.
We must, I think, be continually on guard that the subsidy provided by the safety net does not leak outside the institutions for which it was intended and provide a broad subsidy to other kinds of activities. Put another way, we must remain especially vigilant in maintaining a proper balance between a safety net that fosters economic and financial stabilization and one that benefits the competitive position of private businesses for no particular public purpose. As I noted, safety net subsidies have costs in terms of distorted incentives and misallocated resources. That is why the Congress must be cautious in how the sovereign credit is used.
It has been suggested that the bank holding company structure imposes inefficiencies on banking organizations, and that these organizations should thus be given the option of conducting expanded financial activities in a direct subsidiary of the bank. The bank subsidiary may be a marginally more efficient way of delivering such services, but we believe it cannot avoid being a funnel for transferring the sovereign credit subsidy directly from the bank to finance the new powers, thereby imparting a subsidized competitive advantage to the subsidiary of the bank. One can devise rules--such as 23A and 23B--to assure that loans from the bank to its own subsidiaries are limited and at market rates. One can even devise rules to limit the aggregate equity investment made by banks in their subsidiaries. But one cannot eliminate the fact that the equity invested in subsidiaries is funded by the sum of insured deposits and other bank borrowings that directly benefit from the subsidy of the safety net. Thus, inevitably, a bank subsidiary must have lower costs of capital than an independent entity and even a subsidiary of the bank's parent. Indeed, one would expect that a rational banking organization would, as much as possible, shift its nonbank activity from the bank holding company structure to the bank subsidiary structure. Such a shift from affiliates to bank subsidiaries would increase the subsidy and the competitive advantage of the entire banking organization relative to its nonbank competitors.
Return to top
| I am aware that these are often viewed as only highly technical issues, and hence ones that are in the end, of little significance. I do not think so. The issue of the use of the sovereign credit is central to how our financial system will allocate credit, and hence real resources, the kinds of risk it takes, and the degree of supervision it requires. If the Congress wants to extend the use of the sovereign credit further, to achieve a wider range over which the benefits of doing so can accrue, it ought to make that decision explicitly, and accept the consequences of the subsidy on the financial system that come with it. But, it should not, in the name of some technical change, or in search of some minor efficiency, inadvertently expand the use of the sovereign credit. This issue would not be so important were we not in the process of addressing what must surely be a watershed in the revamping of our regulatory structure. We must avoid inadvertently extending the safety net and its associated subsidy without a thorough understanding of the implications of such an extension to the competitive balance and systemic risks of our financial system.
Central to the Board's choice of a financial structure is its desire for one that will be most effective in fostering both a viable financial system and a vibrant economy. These objectives, in our view, would be thwarted if the safety net subsidy directly benefited new activities. With the safety net comes the moral hazard of which I spoke earlier, and its attendant misallocation of resources, and uneven competitive playing field. If the government subsidies directed to banks were channeled to bank subsidiaries, in my judgement, both the benefits and enumerated costs to the financial system and the public would occur.
If banks were permitted to engage in new activities in their own subsidiaries, inevitably virtually all holding companies would shift those activities now conducted in holding company affiliates to bank subsidiaries, eviscerating the holding company structure. Were such shifts to happen solely as the result of operational efficiencies, no one, including the Board, should mourn the demise of the holding company. But if, as I suspect, such shifts occurred because of the attraction of a government subsidy, we should be concerned because the insidious effects of such subsidies would have spread. The evidence from flows between banks and their parents, relative bond ratings, and the administration of Sections 23A and B of the Federal Reserve Act, all strongly suggest that the holding company structure is far more capable of containing the sovereign credit subsidy whose purpose is support of the safety net, not providing expanded competitive advantage.
As new activities hopefully expand for banking organizations, we believe that it is essential that we assure they are financed at market rates, not subsidized ones. This will not always be easy. Containment of subsidies is often implemented through firewalls and other devices which could also inhibit the very synergies which the expansion of activities is meant to achieve. But we have dealt with these trade-offs before and should be able to in the future as well.
The bank holding company organization is increasingly being managed so as to take advantage of the synergies between its component parts in order to deliver better products to the market and higher returns to stockholders. Such synergies cannot occur if the model of the holding company is one in which the parent is just, in effect, a portfolio investor in its subsidiary. Indeed, virtually all of the large holding companies now operate as integrated units and are managed as such.
As bank holding companies began to widen their activities, and as new technologies permitted not only the development of new products but also the systems for controlling them, the banking organization was impelled to develop centralized risk control techniques that crossed legal entities. Today, risk management for the entire company is increasingly centralized not only at the larger and more sophisticated banking organizations, but at other large financial services providers as well. This development reflects the demands of the marketplace, which views banks and their affiliates and other financial businesses and their affiliates as integrated organizations in terms of financial condition, management, and reputation.
Return to top
| To understand the risk controls of the bank, we have first to come to grips with the fact that the organization is interested in risk and its control, not by instrument or legal entity, but for the entire business. This type of control is being adopted by more and more organizations each year, and can only increase as more activities are authorized by the regulators and the Congress. Regulatory policies and operating procedures have had to respond to these realities, to focus on the process of decisionmaking for the total organization. Thus, the Federal Reserve--the historical umbrella supervisor--also has found it necessary to concentrate more on the process that banking organizations use to manage market, credit, operating and exchange rate risk, and less on the traditional after-the-fact evaluation of balance sheets that can and often do change dramatically the day after they have been reviewed by the supervisors. In such a world, process, if not everything, is critical, and that process is determined increasingly at the parent holding company for all of the units of the organization on a consolidated basis.
One could argue--as several witnesses appearing before this subcommittee did on Tuesday--that regulators should only be interested in the entities they regulate and, hence, review the risk evaluation process only as it relates to their regulated entity. Presumably each regulator of each entity--the bank regulators, the SEC, the state insurance and finance company authorities--would look only at how the risk management process affected their units. It is our belief that this simply will not be adequate. Risks managed on a consolidated basis cannot be reviewed on an individual legal entity basis by different supervisors.
Indeed, our experience has been that a problem in one legal entity can have a contagion effect in other entities. If a bank affiliate begins to have difficulty, the market evaluates the problem as the consolidated entity's problem and can bring pressure on all the units. These pressures usually take the form of funding or liquidity difficulties, as creditors seek to reduce their exposure to all units of an organization that seems to be having trouble. Better safe than sorry. Indeed, it is in the cauldron of the payments and settlement system, where decisions involving large sums must be made in short periods, that this contagion effect might be first seen as participants understandably seek to protect themselves from the uncertainty that accompanies this contagion effect. And that is how crises often begin.
These concerns were part of the motivation for the congressional decision just five years ago to require that foreign banks could enter the United States if, and only if, they were subject to consolidated supervision. This decision, which is consistent with the international standards for consolidated supervision of banking organizations, was a good decision then. It is a good decision today, especially for those banking organizations whose disruption could cause major financial disturbances in United States and foreign markets. For foreign and for U.S. banking organizations, retreat from consolidated supervision would, the Board believes, be a significant step backward.
We have to be careful, however, that consolidated umbrella supervision does not inadvertently so hamper the decisionmaking process of banking organizations as to render them ineffectual. The Federal Reserve Board is accordingly in the process of reviewing its supervisory structure and other procedures in order to reflect the aforementioned market-directed shift from conventional balance sheet auditing to evaluation of the internal risk management process. Although focused on the key risk management processes, it would sharply reduce routine supervisory umbrella presence in holding companies. As the committee knows, the Board has recently published for comment proposals to expedite the applications process, and the legislation Congress enacted last year eased such procedures as well. Nonetheless, the Board requests even greater modification to its existing statutory mandate so that the required applications process could be sharply cut back, particularly in the area of nonbank financial services.
We would hope that should the Congress authorize wider activities for financial services holding companies that it recognize that a bank which is a minor part of such an organization (and its associated safety net) can be protected through adequate bank capital requirements and the application of Sections 23A and 23B of the Federal Reserve Act. The case is weak, in our judgment, for umbrella supervision of a holding company which, because it owns only a small bank, does not have material access to the safety net.
Return to top
| As I noted when discussing the safety net and bank subsidiaries, attached to all uses of the sovereign credit come efforts by the government to protect the taxpayer. Those entities interested in banks are really interested in access to the safety net, since it is far easier to engage in the nonsafety net activities of banks without acquiring a bank. If an organization chooses to deliver some of its services with the aid of the sovereign credit by acquiring a bank, it should not be excused from efforts of the government to look out for the stability of the overall financial system. For bank holding companies that own more than a small bank, this implies umbrella supervision. Although that process will increasingly be designed to reduce supervisory presence and be as nonintrusive as possible, umbrella supervision should not be eliminated, but recognized for what it is: the cost of obtaining a subsidy.
Banking and Commerce
As I indicated earlier, it is clear that rapidly changing technologies are altering the nature of what constitutes finance. Indeed, just as the lines between banking and other financial institutions are often already difficult to discern, the boundaries between finance and nonfinance are likely to become increasingly indistinct as we move into the 21st century. For example, computer and software firms will certainly be offering ever more sophisticated financial products. And doubtless financial firms will be offering an increasingly sophisticated array of nonfinancial services. In addition, some of the financial firms who mainly produce products and services that many observers believe should be permissible to banks are also engaged in, or affiliated with, nonfinancial businesses.
Newer technologies will make it highly unlikely that the walling off of any ownership of financial institutions by nonfinancial businesses and vice-versa can be continued very far into the 21st century. Nonetheless, the Board has concluded that it would be wise to move with caution in addressing the removal of the current legal barriers between commerce and banking. The free and open legal association of banking and commerce would be a profound and surely irreversible structural change in the American economy. Hence, we must be careful to assure ourselves that whatever changes are made in our supervisory structure, that it not distort our evolution to the most efficient financial structure as we move into the next century.
Were we fully confident of how the structure would evolve, we could presumably construct today the regulations which would foster that evolution. But we cannot be certain. We thus run the risk of locking in a set of inappropriate regulations that could adversely alter the development of market structures. We cannot be confident we know what the true synergies between finance and nonfinance will be in ten or even five years. Our ability to foresee accurately the future implications of technologies and market developments in banking, as in other industries, has not been particularly impressive. As Professor Rosenberg of Stanford University has pointed out, ". . . mistaken forecasts of future structure litter our financial landscape." Consider the view of the 1960s that the "cashless society" was imminent. Nonetheless, the public preference for paper has declined only gradually. Similarly, just a few years ago conventional wisdom argued that banks were dinosaurs that were becoming extinct. The reality today is far from it. Even more recently, it was argued that banks and nonfinancial firms had to merge in order to save the capital-starved banking system. Today, as you know, virtually all of our banks are very well capitalized.
All these examples suggest that if we change the rules now about banking and commerce under circumstances of uncertainty about future synergies between finance and nonfinance, we might end up doing more harm than good. And, as with all rule changes by government, we are likely to find it impossible to correct our errors promptly. Modifications of such a fundamental structural rule as the separation of banking and commerce should accordingly proceed at a deliberate pace, testing the response of markets and technological innovation to the altered rules in the years ahead. The public needs to have confidence in the regulatory structure, implying that we proceed slowly and cautiously.
Excessive delay, however, would doubtless produce some inequities. Expanded financial activities for banking organizations requires, the Board believes, that those firms operating in markets that banks can enter should, in turn, be authorized to engage in banking. However, some of these nonbanking financial firms already own--or are owned by--nonfinancial entities. A complete commerce and banking prohibition would thus require the divestiture of all nonfinancial activities by those organizations that wanted to acquire or establish banks. The principle of caution suggests an approach which may prove useful. Perhaps those organizations that either have or establish well-capitalized and well-managed bank subsidiaries should be permitted a small basket of nonfinancial assets--a certain percentage of either consolidated assets or capital. A small permissible basket would establish, in effect, a pilot program to evaluate the efficacy of further breaching of the banking and commerce wall. We found that such a slow and deliberate policy worked well with Section 20 affiliates.
Of course, some nonbanking firms would find that their nonfinancial activities would exceed a small basket exemption. Such excess nonconforming assets might be addressed on a case-by-case basis with a scheduled longer-term divestiture to avoid the worst short-term inequities. A basket clause plus case-by-case review of individual situations might also provide a way to make available a common bank and thrift charter to those unitary thrifts that are affiliated with nonfinancial businesses. The Board has no firm opinion on just exactly how such trade-offs might be made, constrained only by the general concerns I summarized earlier.
Return to top