The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
For release on delivery
1:00 p.m. EDT
October 6, 1995
Supervisory Challenges Related To New
Financial Products
Remarks by
Susan M. Phillips
Board of Governors of Federal Reserve System
to the
Annual Southern Banking Law and Policy Conference
Sponsored by
College of Law, Georgia State University
and
The Federal Reserve Bank of Atlanta
Atlanta, Georgia
October 6, 1995
I.
Introduction
It is a pleasure to be here and participate in your discussions of current
business, and policy issues facing the banking industry.
In particular, I'm especially happy to
have the opportunity to address the supervisory challenges posed by new financial
products.
As you know, technological and financial innovation is spawning new and
increasingly complex ways for banks and other institutions both to take and to manage risk.
Moreover, securities, derivative contracts, loans and other financial instruments are becoming
increasingly interchangeable and harder to differentiate using traditional benchmarks. In this
environment, supervisors face an important challenge in adapting their existing supervisory
regimes to recognize and take advantage of advances in risk measurement and management.
This challenge applies to supervisors of all types of financial institutions, not just bank
supervisors. Indeed, the blurring of products, business lines and other traditional institutional
distinctions is placing increasing pressure on all financial regulators to achieve some form of
harmonization or convergence in supervisory regimes. Failure to do so can lead to regulatory
arbitrage and its associated market inefficiencies.
Today I would like to discuss areas where supervisors are making significant
progress to meet these challenges. I also hope to identify areas where there is still work to be
done.
2
n.
Role of Supervisors
Supervisors of financial institutions share the common objective of ensuring that
the institutions they supervise are not a source of systemic risk. Additional supervisory
objectives vary depending on particular statutory and regulatory mandates. For example, U.S.
bank and thrift regulators have the important objective of protecting the safety net. Bank
supervisory programs are built around that objective.
In general, all financial institution supervisors pursue their objectives by
ensuring that the institutions they supervise: 1) practice sound risk management; 2) have
adequate capital, and, 3) conduct their activities in a reasonably transparent manner. While
different supervisors may place different emphasis on each of these basic elements, they,
nonetheless, represent a basis upon which to seek possible harmonization of supervisory
regimes. Over the past several years both industry groups and supervisors have been
aggressively addressing each of these elements as they relate to new products and risk
management techniques. These efforts were first in the context of derivative instruments but
more recently have been applied to all types of instruments and portfolios. It is useful to
review this progress.
m.
Sound Risk Management Practices
First, with respect to sound risk management practices, the 1993 Group of
Thirty study on derivatives was indeed a watershed. Supervisors followed suit - U . S . banking
agencies each issued their own guidance on sound risk management practices. Last year an
international communique on sound practices for derivatives was issued jointly by the Basle
3
Committee on Bank Supervision and the International Organization of Securities Commissions
(IOSCO). This joint issuance is notable because it demonstrates the enhanced international
coordination of the supervisors of banks and securities firms.
All of the various sound practice statements released thus far, ranging from the
G-30 study to the BIS/IOSCO releases, emphasize the same risk management fundamentals.
They point to the importance of: 1) active oversight by an institution's board of directors and
senior management; 2) clear policies, procedures, and lines of authority; 3) independent risk
management units; 4) comprehensive risk measurement and reporting systems; and finally,
5) other comprehensive internal controls and audit procedures. It is important to note that
these fundamentals are basic sound practices that well-run institutions have applied to
traditional banking and securities activities for years. To a large extent, therefore, the current
focus on risk management represents an application of tried and true fundamentals to new
activities and financial instruments, but using new risk measurement capabilities.
The Barings, Orange County, and now Daiwa Bank incidents are providing
valuable lessons of the importance of applying even the most basic of these fundamentals to
new products and activities. Too often, the focus of both management and supervisors has
been on the "high-tech" aspects of risk management-the modeling and measurement aspects of
complex instruments or their interrelationships. However, while this is, undoubtedly, an
important element of risk management in this time of rapid product innovation, the postmortums conducted to date on Barings and Orange County clearly point the finger at
fundamental breakdowns in several relatively simple, "low-tech" elements of risk management.
Common sense internal controls involving senior level oversight, segmentation of duties, and
4
independent risk assessment appear to have been violated in both cases. While supervisors
face the challenges of incoiporating new risk management techniques into their supervisory
regimes - including high-tech elements -- it appears that the industry still faces significant
challenges in applying the "basics" to their new products and activities. One positive aspect
of these unfortunate events is that they have served as vivid wake-up calls to the industry and
end user community.
Supervisors in different countries and of different industries use various
channels to ensure that the institutions they supervise follow sound risk management practices.
U.S. banking supervisors rely heavily on the annual full-scope examination. A number of
foreign banking supervisors depend on outside auditors to evaluate the adequacy of controls.
Securities and commodities regulators place significant oversight responsibilities with selfregulatory organizations and external auditors.
All of these players have looked to the various supervisory pronouncements on
sound practices as guides in structuring and evaluating the adequacy of risk management as
applied to new products and activities. Along these lines, one such initiative is the Federal
Reserve's Trading Activities Examination Manual published in 1994. It identifies specific
polices and procedures for assessing the risk management of trading operations for both cash
and derivative instruments and has been well-received by bank and non-bank supervisors.
Management and auditors of many banks and securities firms, as well as consultants, have also
found this document useful in defining the steps they use to evaluate the adequacy of risk
management techniques.
5
IV.
Capital Adequacy
Supervisors are also currently assessing.their approaches to capital adequacy in
light of recent technological and financial innovations. The increasing volume and complexity
of financial transactions and the rapid advances in risk management techniques are making it
increasingly difficult to address prudential adequacy concerns using traditional standardized,
"rule-of-thumb" capital regimes. Accordingly, a common thread in many recent capital
adequacy initiatives is the use of an institution's own internal model to measure the risks
against which regulatory capital would be required.
Basle Market Risk Proposal - The most notable of these efforts is the Basle
market risk proposal.
In April, the Basle Committee recommended a two-pronged approach
for determining minimum capital requirements for the trading activities of internationally
active banks.
Such banks would be permitted to use either: 1) a "standardized" risk measure
that applies uniform measurement procedures and assumptions for all banks; or 2) their own
internal value-at-risk (VAR) models, subject to specified constraints and "qualitative"
standards of risk management. Use of internal models would also have to be supplemented
with a rigorous program of model validation and stress testing.
The Federal Reserve has been the principal advocate internationally of using
internal models. We believe this route provides incentives to promote sound risk management
while minimizing supervisory intrusion that might impede innovation in financial risk
measurement.
However, using daily VAR models alone to determine a capital charge may not
be sufficient. While useful as a day-to-day management tool, daily VAR calculations do not
6
incorporate the stress tests which would assess capital adequacy under a variety of economic
and market conditions. Indeed, most large institutions do not, themselves, rely solely on their
daily VAR estimates for allocating capital or evaluating its adequacy. While supervisors want
to build on internal models, ultimately adjustments may be required to produce a credible,
consistent, and meaningful capital charge for market risk.
At present, the Basle proposal
involves constraints on certain model parameters in order to produce sufficient comparability
among models of different banks and to obtain a sufficiently conservative measure of risk.
Changes to these constraints are being considered as a result of the public comment process.
E. U. 's Capital Adequacy Directive - A driving influence on the convergence in
the capital rules of banks and securities firms is the existence of large universal banks in
Europe. This harmonization is underway by means of the implementation of the European
Union's Capital Adequacy Directive or CAD. Due to be implemented in 1996, the CAD
poses a capital calculation system similar to the standard approach proposed by Basle.
A
limited internal model option is also available in the CAD, but the standard approach is also
more consistent with capital structures used by securities broker-dealers or investment bankers.
Nevertheless, it is expected that, ultimately, the CAD will be "adjusted" to conform with the
Basle proposal when that initiative is finalized.
SEC Capital Rules - The use of internal models in determining U.S. securities
firms' capital requirements is on a somewhat slower track. Currently, the SEC is reviewing
its capital rules with an eye toward establishing a unified "hair-cut" methodology for fixed
income products that, while not implementing an internal model measurement system, would
recognize the risk reducing properties of some hedge positions. Cognizant of other
7
supervisors' efforts regarding internal models, SEC staff is monitoring securities firms'
internal model calculations.
Proposals to use internal models for measuring capital adequacy have also
surfaced with regard to the unregulated derivative product subsidiaries of securities firms.
In
its recent report, the securities industry's Derivatives Products Group (DPG) recommends the
use of institutions' own VAR calculations to assess capital adequacy.
The Group endorses
the use of some of the same modeling parameters identified in the Basle proposal. Similarly,
it also identifies the need for supplemental stress testing and a regime of model validation.
The recommendation, along with other initiatives, suggests that internal model approaches to
capital adequacy may offer an avenue to capital harmonization between U.S. bank and security
regulators.
Pre-commitment approach - While the immediate focus of bank supervisors has
been on the use of internal models in the context of the Basle proposals, let me point out that
the Board has issued a concept paper on a possible future approach to capital adequacy that
could be viewed, in my judgement, as "the ultimate internal model approach". This so-called
"pre-commitment" approach would require a bank to specify the amount of capital allocated
to support market risks. The bank would be expected to manage its trading portfolio to limit
cumulative trading losses over some interval to the capital allocated. This capital commitment
would be publicly disclosed. To assure that adequate capital is committed for the risks
involved, the regulator would assess penalties on institutions failing to limit losses to their
capital commitment. The assessment of any penalties would also be publicly disclosed.
8
This proposal is an intriguing concept, ripe for discussion and additional
research.
The specification of the penalty function is problematic since it would need to be
sufficiently strong, certain, and public, to have a credible impact. Yet its imposition could
come at the worst possible time for a bank and could have safety and soundness implications.
The element of public disclosure of capital commitments, VAR estimates, and trading results,
and the incentives that such disclosure would entail, are particularly interesting. The
competitive issues involved in such disclosures clearly must be assessed and there are
undoubtedly other issues to be considered. I urge all of you to give this concept serious
thought and comment accordingly.
V.
Disclosure, Regulatory Reporting and Accounting Initiatives
Just as with sound practices and capital adequacy, there are a number of
initiatives underway to advance the transparency of new products and activities in an
institution's risk profile. In both 1994 and 1995, U.S. bank supervisors dramatically increased
the amount of information collected on derivatives activities in bank Call reports. The
voluntary framework advanced by the DPG recommended similar enhancements to the
reporting of securities firms' derivatives activities to supervisors.
Internationally, in May, the Basle Committee on Banking Supervision and
IOSCO's Technical Committee issued a joint document advancing a "framework" to guide
banking and securities regulators in determining the kinds of information that could assist in
the supervision of institutions' derivatives activities. An important part of that framework
identifies minimum information needed to assess the effects of derivatives on an institution's
9
risk profile. Both international bank and securities firm supervisors are actively pursuing
implementation of the common minimum framework advanced in that joint paper. The two
Committees plan to update the joint supervisory information framework periodically. All in
all, this framework should help promote more consistent methods of supervisory evaluation of
banks' and securities firms' exposures arising from new products and activities, including
derivatives.
Other international initiatives are also underway with regard to disclosure.
In
1994, the BIS published the Fisher Report which calls for firms to disclose more quantitative
information drawn from their own internal risk management processes. In addition, the Basle
Supervisors Committee is currently conducting a comprehensive survey and analysis of the
1994 derivatives disclosures in the annual reports of the top dealer banks in the Basle member
countries. A report by the Basle Supervisors Committee is expected later this year on ways of
improving these disclosures. IOSCO has also indicated an interest in joining in this effort.
Any joint document arising from such a common effort should significantly advance the
interest of enhanced disclosure of the effects of new products on the risk profiles of both banks
and securities firm, and should result in greater consistency and comparability in these
disclosures over time.
Furthermore, there are opportunities for greater convergence in accounting rules
for banks and securities firms - particularly for their financial products ~ by aligning U.S.
bank regulatory accounting rules with generally accepted accounting principles ("GAAP").
Currently, securities firms prepare their regulatory financial reports on a basis consistent with
GAAP, and the same holds true for regulatory financial reports and general purpose financial
10
statements of bank holding companies. On the other hand, the regulatory accounting
principles underlying bank Call Reports differ in certain respects from GAAP. For example,
generally, for bank Call Report purposes, futures and forward contracts used as hedges must
be marked to market, whereas GAAP permits the changes in the market values of these
instruments to be reflected in earnings on a basis consistent with the income or expense of the
items being hedged. Also, certain asset securitization transactions involving recourse that are
reported as sales under GAAP are treated as financings for Call Report puiposes. Thus, the
assets, liabilities, equity, and earnings reported in bank Call Reports may differ from related
items reported in financial statements and regulatory reports prepared in accordance with
GAAP.
Eliminating these differences between regulatory reporting standards and GAAP
will result in greater consistency between bank Call Reports and the financial statements and
regulatory reports of bank holding companies and securities firms. Since bank Call Reports
and bank holding companies' regulatory financial reports are generally available to the public,
the elimination of so-called "GAAP/RAP differences" will make it easier for analysts,
investors, and other users to understand Call Report information and compare it with related
bank holding company information. At the same time, regulatory burden will be significantly
reduced since banking organizations will not need to keep two basic sets of books. Any
concerns about GAAP can be addressed by other means, such as by collecting supervisory
information in memoranda items and by making adjustments to regulatory capital ratios, as
appropriate. This is the approach that the Federal Reserve has used for the regulatory reports
11
of bank holding companies for many years. Accordingly, I am hopeful that we will soon see
the adoption of GAAP for bank Call Reports.
VI.
Concluding Remarks.
To conclude and summarize, I hope this review of supervisory initiatives and
accomplishments illustrates that supervisors are making concerted efforts to keep pace with
market practices and financial innovations. Just as innovation poses new challenges to the
industry, it also poses challenges to supervisors. Supervisors are responding and are making
significant progress in adapting their existing supervisory regimes.
In several important
areas, supervisory initiatives seem to be running along parallel tracks and showing some
headway in the difficult task of harmonizing supervisory regimes among different industries.
However, significant challenges still remain. Disclosure is one area where
greater progress and harmonization could be beneficial. While the Fisher report significantly
advanced consideration of the issues regarding disclosure, the report itself recognizes a lack of
consensus on the appropriate disclosures of new products and activities and their effect on the
institution's risk profile. This point is illustrated in last month's Federal Reserve Bulletin
article that reviews the 1994 derivatives disclosures made by major U.S. banks. Although
1994 disclosures were found to be dramatically improved over 1993 disclosures, there was
significant diversity of methods used by the top ten dealer banks in presenting information
about their derivatives activities. While diversity in disclosure may be appropriate given the
wide range of possible risk profiles, some harmonization may be necessary to facilitate
12
analysis and public understanding. Hopefully, continued industry and supervisory efforts in
this area will advance harmonization.
Accounting -
The dearth of definitive accounting guidance for complex and
sophisticated instruments and strategies is another area ripe for further private and public
initiatives.
The lack of definitive accounting rules for derivatives is particularly troubling ~
we need to press forward to enhance our accounting framework for derivatives and achieve
greater consistency in accounting treatment across various types of instruments.
Internal models - We also need to make progress on the use of internal models
for supervisory purposes. In some ways internal model development is still in its infancy,
particularly when stretched to bank supervisory and capital adequacy purposes. Moreover, the
harmonization of supervisory regimes for bank and securities firms has even further to go.
In concluding, let me point out that while there are clear areas of supervisory
harmonization underway, supervisory convergence in all areas need not be the overriding goal.
Although supervisors may share objectives, the specific techniques used to promote sound
practices, capital adequacy, and transparency need not be identical. Indeed, in some areas,
more harmonization is not always the answer. Supervisory regimes must reflect the
differences in the regulated industries' business practices and needs. One size regulation or
supervision may not fit all. Moreover, some competition and supervisory differences may be
healthy and provide for constructive experimentation. Harmonization does appear to be
occurring where clear economic incentives are driving supervisors to recognize the effects of
dramatic technological and financial innovation in the financial services industries. But
supervision, like business development is, and I believe should be, an evolutionary process.
@FedFRASER