Using market data to manage risk The potential of financial modernization
Gary H. Stern
President
Federal Reserve Bank of Minneapolis
This issue of The Region is devoted to financial modernization
and, in particular, to the legislation passed and signed into law late
last year—the Gramm-Leach-Bliley (GLB) Act. The commentaries included
here cover a wide range—from consumer and competitive matters to
concerns about the regulatory and supervisory framework, the insurance
fund and ultimately the ongoing role of commercial banks in the financial
system. As you will see, they represent different perspectives and, in
some instances, view GLB in very different lights.
My principal interest is the constraint placed by GLB on excessive
risk taking by large banks. This is a subject in which I and others
at the Minneapolis Fed have long had an interest, and our interest
is heightened since GLB is likely to lead to more large institutions
engaged in an increasingly expansive array of financial activities.
Examination of GLB from this perspective—the perspective
of risk taking by large banking and financial holding company organizations—yields
divergent implications. In particular, it appears that the recent
legislation unintentionally could contribute to excessive risk taking
by large banks and thus could increase the exposure of taxpayers
to prospective losses. Moreover, as detailed in the article by Minneapolis
Fed supervision staff
, it is not clear that much of the regulatory structure created
by the bill to address this risk taking will prove effective. At
the same time, GLB creates an incipient framework for enhanced management
of risk through increased use of market data in banking regulation.
The legislation does not go as far as it could, but policymakers
have an opportunity to better contain risk taking by fully exploiting
the GLB framework in the future.
Some concerns about GLB
GLB repeals limits on banking organizations' ability to underwrite securities
and insurance and on other financial activities. This was a worthy goal
given the weak foundation for these restrictions. Policymakers must now
be tempted to call a time out, yet the very act of passing modernization
legislation makes it even more important that the regulatory and supervisory
apparatus governing bank risk taking be enhanced.
Reform is needed because the federal government assumes risks
that banks would otherwise have to face. Some of this risk assumption
is explicit (for example, federal insurance on deposits under $100,000).
Risk assumption also occurs via the practice of too-big-to-fail
(TBTF), whereby uninsured creditors of the largest failing banks
have from time to time been protected. TBTF is particularly relevant
post-GLB. Banking organizations can more readily control large insurance
and securities firms, and the failure of these nonbanking firms
could jeopardize the health of the entire banking organization.
This at least raises the possibility that federal bailouts will
protect creditors of nonbanking firms, to say nothing of the possibility
that there may be more large banks whose creditors are deemed worthy
of support.
Policymakers have understood and warned against the increased
exposure that liberalizing financial activities could create. It
appears that the primary GLB response to increased exposure focuses
on organizational rules. For example, firms exercising the expanded
powers allowed under GLB must be legally distinct from insured banks,
and GLB restricts transactions between these legal entities. For
reasons discussed in detail in the aforementioned article, we think
the regulatory and organizational structure created by GLB will
face significant challenges.
In contrast to many of the regulatory steps taken in the legislation,
we believe effective policy responses must go further and address
the underlying risk-taking proclivities of federally protected banking
organizations. We believe increased market discipline and greater
use of market signals by banking regulators hold considerable promise
in addressing effectively excessive risk taking by banks ("A Response
to Critics of Market Discipline," The Region, September 1999). GLB
certainly did not move as aggressively on this front as we might
have liked, legislating what amounts to modest enhancements in the
use of market assessments in bank regulation. But these actions
set the stage for increased future use of market data and discipline.
Market discipline in GLB
There are two actions in the legislation that could create additional
market discipline. First, GLB grants new powers to the subsidiaries
of national banks. But these new powers are conditional on several
factors, one of which is market-oriented and aimed at the 100 largest
national banks. Specifically, these large national bank parents
of the subsidiary must have at least one issue of outstanding long-term
unsecured debt rated in the top three categories by a credit rating
firm (for example, rated at least "A" by a firm like Moody's). The
bank would not be able to make new equity investments in its existing
subsidiaries nor establish new subsidiaries if the bond rating fell
out of the top three categories.
An association representing some of the largest banking organizations
has expressed a desire to escape this provision, arguing that banks
should not be required to issue a particular kind of debt. While
the Comptroller has rejected this suggestion as a matter of law,
such evasion should also be stopped as a matter of sound public
policy. The requirement that banks must issue some public debt compels
them to face market forces, potentially at a time when they would
prefer to raise funds via insured and sheltered money. Indeed, the
interim decision by the Treasury and the Federal Reserve to allow
national banks that fall between the 50th and 100th largest in the
country to escape this issuance requirement could be debated on
these grounds (although there were certainly other strong reasons
for the agencies to take this step).
Second, GLB requires the Board of Governors of the Federal Reserve
System and the Treasury to study the "feasibility and appropriateness"
of requiring large banks to issue subordinated debt. Holders of
this debt are unlikely to receive full and timely payment if the
banking organization issuing the debt fails. Thus, these creditors
have reason to require higher interest rates when the issuer takes
on more risk (that is, apply market discipline). In fact, empirical
research suggests that subordinated debt holders already act on
this incentive.
Regardless of its necessity, the Treasury and Federal Reserve
review of subordinated debt could provide impetus for actually implementing
a regime that increases market discipline. To exert a positive influence,
the report, and Federal Reserve and Treasury policymakers, would
have to recommend a plan for creating credible market prices. This
could be accomplished by mandating subordinated debt issuance by
large banks and by explicitly putting creditors of large banks at
risk of loss. We believe available evidence supports such recommendations.
We would encourage such a plan, which could first call for monitoring
prices on voluntarily issued subordinated debt, or other bank liabilities
such as jumbo CDs, before mandates ensue.
Exploiting the GLB framework
The upside of GLB does not come from the immediate steps it takes.
Rather, the benefit is more likely to come in the future when policymakers
more fully exploit the framework that the legislation created. In
particular, policymakers and regulators have increasingly relied
on "regulatory triggers" that require supervisory action against
a bank when its condition changes (for example, its capital falls
below a prescribed level). Regulators also use triggers to grant
banks regulatory relief (for example, banks with strong management
assessments and capital receive expedited supervisory review). GLB
creates an additional regulatory trigger. A high credit rating allows
the exercise of new powers by national bank subsidiaries and a drop
in the rating could limit such activity. Should policymakers stop
at the use of credit rating or look to other triggers in the future?
It is perhaps faint praise to argue that GLB's reliance on market-based
data is an advance over traditional, lagging indicators of a bank's
financial condition such as capital. Credit ratings do provide regulators
with an assessment of repayment prospects for bond holders, and
the new GLB trigger could very well limit risk taking. Only one
bond issuer has defaulted since 1937 while holding a Moody's rating
of "A" or higher. However, credit ratings are inferior to market
prices as a tool for regulators for several reasons. In particular,
market prices:
incorporate credit ratings plus the host of other factors that
market participants find valuable,
result from tens of thousands of decisions by a huge array
of market participants,
have an almost infinite degree of gradation, and change in
real time, while ratings are much more likely to be static measures
with a limited number of outcomes (for example, "AAA," "AA"),
and
reflect the views of those with money to lose (although credit
rating firms have a profit motive to accurately gauge risk).
Accordingly, we suggest that over time policymakers exploit the trigger
system they have developed by substituting market prices for credit ratings.
This combination of triggers and market prices would provide the nexus
between the credit rating regime established by GLB and its call for a
study of subordinated debt plans. Under one possible scenario, the exercise
of new powers could occur only if the difference, or spread, between the
yield on a bank's subordinated debt and the yields on comparable risk-free
Treasury securities was no greater than the average spread for peer banks.
Clearly, there are a large number of permutations that this price trigger
could take, and the experience gained with the credit rating trigger may
help in the calibration process.
In conclusion, despite some of the reservations expressed above, overall
we think that with work GLB could set the stage for increased market discipline
and, hence, for more effective limitation of excessively risky banking.
At the least, it suggests an outline and opportunity for doing so.
Annual Report
