The Role of Capital in Optimal
Banking Supervision and Regulation
It is my pleasure to join President McDonough and our col- Of course, the focus of this conference is on the
leagues from the Bank of Japan and the Bank of England in future of prudential capital standards. In our deliberations,
hosting this timely conference. Capital, of course, is a topic we must therefore take note that observers both within the
of never-ending importance to bankers and their counter- regulatory agencies and in the banking industry itself are
parties, not to mention the regulators and central bankers raising warning flags about the current standard. These
whose job it is to oversee the stability of the financial sys- concerns pertain to the rapid technological, financial, and
tem. Moreover, this conference comes at a most critical and institutional changes that are rendering the regulatory
opportune time. As you are aware, the current structure of capital framework less effectual, if it is not on the verge of
regulatory bank capital standards is under the most intense becoming outmoded, with respect to our largest, most
scrutiny since the deliberations leading to the watershed complex banking organizations. In particular, it is argued
Basle Accord of 1988 and the Federal Deposit Insurance that the heightened complexity of these large banks’ risk-
Corporation Improvement Act of 1991. taking activities, along with the expanding scope of
In this tenth anniversary year of the Accord, its regulatory capital arbitrage, may cause capital ratios as
architects can look back with pride at the role played by calculated under the existing rules to become increasingly
the regulation in reversing the decades-long decline in bank misleading.
capital cushions. At the time that the Accord was drafted, I, too, share these concerns. In my remarks this
the use of differential risk weights to distinguish among evening, however, I would like to step back from the tech-
broad asset categories represented a truly innovative and, nical discourse of the conference’s sessions and place these
I believe, effective approach to formulating prudential concerns within their broad historical and policy contexts.
regulations. The risk-based capital rules also set the stage Specifically, I would like to highlight the evolutionary
for the emergence of more general risk-based policies nature of capital regulation and then discuss the policy
within the supervisory process. concerns that have arisen with respect to the current capital
structure. I will end with some suggestions regarding basic
Alan Greenspan is the chairman of the Board of Governors of the Federal principles for assessing possible future changes to our
Reserve System. system of prudential supervision and regulation.
FRBNY ECONOMIC POLICY REVIEW / OCTOBER 1998 163
To begin, financial innovation is nothing new, and suited to the times for which they were developed or will
the rapidity of financial evolution is itself a relative con- be ill suited for those banking systems that are at an earlier
cept—what is “rapid” must be judged in the context of the stage of development.
degree of development of the economic and banking struc- Indeed, so long as we adhere in principle to a com-
ture. Prior to World War II, banks in this country did not mon prudential standard, it is appropriate that differing
make commercial real estate mortgages or auto loans. Prior regulatory regimes may exist side by side at any point in
to the 1960s, securitization, as an alternative to the tradi- time, responding to differing conditions between banking
tional “buy and hold” strategy of commercial banks, did systems or across individual banks within a single system.
not exist. Now banks have expanded their securitization Perhaps the appropriate analogy is to computer-chip manu-
activities well beyond the mortgage programs of the 1970s facturers. Even as the next generation of chip is being
and 1980s to include almost all asset types, including cor- planned, two or three generations of chip—for example,
porate loans. And most recently, credit derivatives have Pentium IIs, Pentium Pros, and Pentium MMXs—are
been added to the growing list of financial products. Many being marketed, and at the same time, older generations of
of these products, which would have been perceived as too chip continue to perform yeoman duty within specific
risky for banks in earlier periods, are now judged to be safe applications. Given evolving financial markets, the ques-
owing to today’s more sophisticated risk measurement and tion is not whether the Basle standard will be changed but
containment systems. Both banking and regulation are how and why each new round of change will occur and to
continuously evolving disciplines, with the latter, of which market segment it will apply.
course, continuously adjusting to the former. As it oversees the necessary evolution of the Accord
Technological advances in computers and in tele- for the more advanced banking systems, the regulatory
communications, together with theoretical advances— community would do well to address some of the basic
principally in option-pricing models—have contributed to issues that, in my view, it has not adequately addressed to
this proliferation of ever more complex financial products. date. In so doing, perhaps we can shed some light on the
The increased product complexity, in turn, is often cited as source of our present concerns with the existing capital
the primary reason that the Basle standard is in need of standard. There really are only two questions here: First,
periodic restructuring. Indeed, the Basle standard, like the How should bank “soundness” be defined and measured?
industry for which it is intended, has not stood still over Second, What should be the minimum level of soundness
the past ten years. Since its inception, significant changes set by regulators?
have been made on a regular basis to the Accord, includ- When the Accord was being crafted, many super-
ing, most visibly, the use of banks’ internal models to assess visors may have had an implicit notion of what they meant
capital charges for market risk within trading accounts. All by soundness—they probably meant the likelihood of a
of these changes have been incorporated within a document bank becoming insolvent. Although by no means the only
that is now quite lengthy—and written in appropriately one, this definition of soundness is perfectly reasonable.
dense, regulatory style. Indeed, insolvency probability is the standard explicitly
While no one is in favor of regulatory complexity, used within the internal risk measurement and capital allo-
we should be aware that capital regulation will necessarily cation systems of our major banks. That is, many of the
evolve over time as the banking and financial sectors them- large banks explicitly calculate the amount of capital they
selves evolve. Thus, it should not be surprising that we need in order to reduce to a targeted percentage the proba-
constantly need to assess possible new approaches to old bility, over a given period, that losses would exceed the
problems, even as new problems become apparent. Nor allocated capital and drive the bank into insolvency.
should the continual search for new regulatory procedures But whereas our largest banks have explicitly set
be construed as suggesting that existing policies were ill their own internal soundness standards, regulators really
164 FRBNY ECONOMIC POLICY REVIEW / OCTOBER 1998
have not. Rather, the Basle Accord set a minimum capital information systems and internal systems for quantifying,
ratio, not a maximum insolvency probability. Capital, being pricing, and managing risk.
the difference between assets and liabilities, is of course an It is appropriate that supervisory procedures evolve
abstraction. Thus, it was well understood at the time that to encompass the changes in industry practices, but we
the likelihood of insolvency is determined by the level of must also be sure that improvements in both the form
capital a bank holds, the maturities of its assets and liabili- and the content of the formal capital regulations keep
ties, and the riskiness of its portfolio. In an attempt to pace. Inappropriate regulatory capital standards, whether
relate capital requirements to risk, the Accord divided too low or too high in specific circumstances, can entail sig-
assets into four risk “buckets,” corresponding to minimum nificant economic costs. This resource allocation effect of
total capital requirements of 0 percent, 1.6 percent, capital regulations is seen most clearly by comparing the
4.0 percent, and 8.0 percent, respectively. Indeed, much of Basle standard with the internal “economic capital” alloca-
the complexity of the formal capital requirements arises tion processes of some of our largest banking companies.
from rules stipulating which risk positions fit into which For internal purposes, these large institutions attempt
of the four capital buckets. explicitly to quantify their credit, market, and operating
Despite the attempt to make capital requirements risks by estimating loss probability distributions for various
at least somewhat risk-based, the main criticisms of the risk positions. Enough economic, as distinct from regula-
Accord—at least as applied to the activities of our largest, tory, capital is then allocated to each risk position to satisfy
most complex banking organizations—appear to be war- the institution’s own standard for insolvency probability.
ranted. In particular, I would note three: First, the formal Within credit risk models, for example, capital for internal
capital ratio requirements, because they do not flow from purposes often is allocated so as to hypothetically “cover”
any particular insolvency probability standard, are for the 99.9 percent or more of the estimated loss probability
most part arbitrary. All corporate loans, for example, are distribution.
placed into a single, 8 percent bucket. Second, the require- These internal capital allocation models have
ments account for credit risk and market risk but not much to teach the supervisor and are critical to under-
explicitly for operating and other forms of risk that may standing the possible misallocative effects of inappropriate
also be important. Third, except for trading account capital rules. For example, the Basle standard lumps all
activities, the capital standards do not take account of corporate loans into the 8 percent capital bucket, but the
hedging, diversification, and differences in risk manage- banks’ internal capital allocations for individual loans vary
ment techniques, especially portfolio management. considerably—from less than 1 percent to well over 30 per-
These deficiencies were understood even as the cent—depending on the estimated riskiness of the position
Accord was being crafted. Indeed, it was in response to in question. In the case in which a group of loans attracts
these concerns that, for much of the 1990s, regulatory an internal capital charge that is very low compared with
agencies focused on improving supervisory oversight of the Basle 8 percent standard, the bank has a strong incentive
capital adequacy on a bank-by-bank basis. In recent years, to undertake regulatory capital arbitrage to structure the
the focus of supervisory efforts in the United States has risk position in a manner that allows it to be reclassified
been on the internal risk measurement and management into a lower regulatory risk category. At present, securitiza-
processes of banks. This emphasis on internal processes has tion is, without a doubt, the major tool used by large U.S.
been driven partly by the need to make supervisory policies banks to engage in such arbitrage.
more risk-focused in light of the increasing complexity of Regulatory capital arbitrage, I should emphasize,
banking activities. In addition, this approach reinforces is not necessarily undesirable. In many cases, regulatory
market incentives that have prompted banks themselves to capital arbitrage acts as a safety valve for attenuating the
invest heavily in recent years to improve their management adverse effects of those regulatory capital requirements that
FRBNY ECONOMIC POLICY REVIEW / OCTOBER 1998 165
are well in excess of the levels warranted by a specific activ- requirements on the assets remaining on the book. In the
ity’s underlying economic risk. Absent such arbitrage, a extreme, such “cherry picking” would leave on the balance
regulatory capital requirement that is inappropriately high sheet only those assets for which economic capital allocations
for the economic risk of a particular activity could cause a are greater than the 8 percent regulatory standard.
bank to exit that relatively low-risk business by preventing Given these difficulties with the one-size-fits-all
the bank from earning an acceptable rate of return on its nature of our current capital regulations, it is understand-
capital. That is, arbitrage may appropriately lower the able that calls have arisen for reform of the Basle standard.
effective capital requirements against some safe activities It is, however, premature to try to predict exactly how the
that banks would otherwise be forced to drop by the effects next generation of prudential standards will evolve. One
of regulation. set of possibilities revolves around market-based tools and
It is clear that our major banks have become quite incentives. Indeed, as banks’ internal risk measurement
efficient at engaging in such desirable forms of regulatory and management technologies improve, and as the depth
capital arbitrage, through securitization and other devices. and sophistication of financial markets increase, bank
However, such arbitrage is not costless and therefore not supervisors should continually find ways to incorporate
without implications for resource allocation. Interestingly, market advances into their prudential policies, when
one reason that the formal capital standards do not include appropriate. Two potentially promising applications of this
very many risk buckets is that regulators did not want to principle have been discussed at this conference. One is the
influence how banks make resource allocation decisions. use of internal credit risk models as a possible substitute
Ironically, the “one-size-fits-all” standard does just that, by for, or complement to, the current structure of ratio-based
forcing the bank into expending effort to negate the capital capital regulations. Another approach goes one step further
standard, or to exploit it, whenever there is a significant and uses market-like incentives to reward and encourage
disparity between the relatively arbitrary standard and improvements in internal risk measurement and manage-
internal, economic capital requirements. ment practices. A primary example is the proposed pre-
The inconsistencies between internally required commitment approach to setting capital requirements for
economic capital and the regulatory capital standard create bank trading activities. I might add that precommitment
another type of problem: Nominally high regulatory capi- of capital is designed to work for only the trading account,
tal ratios can be used to mask the true level of insolvency not the banking book, and then for only strong, well-
probability. For example, consider the case in which the managed organizations.
bank’s own risk analysis calls for a 15 percent internal Proponents of an internal-models-based approach
economic capital assessment against its portfolio. If the to capital regulations may be on the right track, but at
bank actually holds 12 percent capital, it would, in all this moment of regulatory development, it would seem
likelihood, be deemed to be well capitalized in a regulatory that a full-fledged, bankwide, internal models approach
sense, even though it might be undercapitalized in the could require a very substantial amount of time and
economic sense. effort to develop. In a paper given earlier today, Federal
The possibility that regulatory capital ratios may Reserve Board economists David Jones and John Mingo
mask true insolvency probability becomes more acute as enumerate their concerns about the reliability of the
banks arbitrage away inappropriately high capital require- current generation of credit risk models. They suggest,
ments on their safest assets by removing these assets from however, that these models may, over time, provide a
the balance sheet via securitization. The issue is not solely basis for setting future regulatory capital requirements.
whether capital requirements on the bank’s residual risk Even in the shorter term, they argue, elements of internal
in the securitized assets are appropriate. We should also credit risk models may prove useful within the super-
be concerned with the sufficiency of regulatory capital visory process.
166 FRBNY ECONOMIC POLICY REVIEW / OCTOBER 1998
Still other approaches are of course possible, Conversely, permitting regulated institutions that
including some combination of market-based and tradi- benefit from the safety net to take risky positions that, in
tional ratio-based approaches to prudential regulation. But the absence of the net, would earn them junk bond ratings
regardless of what happens in this next stage, as I noted for their liabilities is clearly inappropriate. In such a world,
earlier, any new capital standard is itself likely to be super- our goals of protecting taxpayers and reducing the mis-
ceded within a continuing process of evolving prudential allocative effects of the safety net would simply not be
regulations. Just as manufacturing companies follow a realized. Ultimately, the setting of soundness standards
product-planning cycle, bank regulators can expect to should achieve a complex balance—remembering that the
begin working on still another generation of prudential goals of prudential regulation should be weighed against
policies even as proposed modifications to the current the need to permit banks to perform their essential risk-
standard are being released for public comment. taking activities. Thus, capital standards should be struc-
In looking ahead, supervisors should, at a mini- tured to reflect the lines of business and the degree of risk
mum, be aware of the increasing sophistication with which taking chosen by the individual bank.
banks are responding to the existing regulatory framework A second principle should be to continue linking
and should now begin active discussions on the necessary strong supervisory analysis and judgment with rational
modifications. In anticipation of such discussions, I would regulatory standards. In a banking environment charac-
like to conclude by focusing on what I believe should be terized by continuing technological advances, this means
several core principles underlying any proposed changes to placing an emphasis on constantly improving our super-
our current system of prudential regulation and supervision. visory techniques. In the context of bank capital adequacy,
First, a reasonable principle for setting regulatory supervisors increasingly must be able to assess sophisti-
soundness standards is to act much as the market would if cated internal credit risk measurement systems and to
there were no safety net and all market participants were gauge the impact of the continued development in securi-
fully informed. For example, requiring all of our regulated tization and credit derivative markets. It is critical that
financial institutions to maintain insolvency probabilities supervisors incorporate, where practical, the risk analysis
that are equivalent to a triple-A rating standard would be tools being developed and used on a daily basis within the
demonstrably too stringent because there are very few such banking industry itself. If we do not use the best analytical
entities among unregulated financial institutions not subject tools available and place these tools in the hands of highly
to the safety net. That is, the markets are telling us that the trained and motivated supervisory personnel, then we
value of the financial firm is not, in general, maximized at cannot hope to supervise under our basic principle—
default probabilities reflected in triple-A ratings. This sug- supervision as if there were no safety net.
gests, in turn, that regulated financial intermediaries cannot Third, we have no choice but to continue to plan
maximize their value to the overall economy if they are for a successor to the simple risk-weighting approach to
forced to operate at unreasonably high levels of soundness. capital requirements embodied within the current regula-
Nor should we require individual banks to hold tory standard. While it is unclear at present exactly what
capital in amounts sufficient to protect fully against rare that successor might be, it seems clear that adding more
systemic events, which, in any event, may render standard and more layers of arbitrary regulation would be counter-
probability evaluation moot. The management of systemic productive. We should, rather, look for ways to harness
risk is properly the job of the central banks. Individual market tools and market-like incentives whenever possible,
banks should not be required to hold capital against the by using banks’ own policies, behaviors, and technologies
possibility of overall financial breakdown. Indeed, central in improving the supervisory process.
banks, by their existence, appropriately offer banks a form of Finally, we should always remind ourselves that
catastrophe insurance against such events. supervision and regulation are neither infallible nor likely
FRBNY ECONOMIC POLICY REVIEW / OCTOBER 1998 167
to prove sufficient to meet all our intended goals. Put efforts to contain the scope of the safety net or to press for
another way, the Basle standard and the bank examination increases in the quantity and quality of financial disclosures
process, even if structured in optimal fashion, are a second by regulated institutions.
line of support for bank soundness. Supervision and regula- If we follow these basic prescriptions, I suspect
tion can never be a substitute for a bank’s own internal that history will look favorably on our attempts at crafting
scrutiny of its counterparties and for the market’s scrutiny regulatory policy.
of the bank. Therefore, we should not, for example, abandon
The views expressed in this article are those of the author and do not necessarily reflect the position of the Federal Reserve
Bank of New York or the Federal Reserve System. The Federal Reserve Bank of New York provides no warranty, express or
implied, as to the accuracy, timeliness, completeness, merchantability, or fitness for any particular purpose of any information
contained in documents produced and provided by the Federal Reserve Bank of New York in any form or manner whatsoever.
168 FRBNY ECONOMIC POLICY REVIEW / OCTOBER 1998