History shows that expanding the powers of banking organizations while
maintaining an adequate supervisory framework to prevent excessive risk
taking is a daunting task. Past expansions of bank powers have, unfortunately,
often been coupled with inadequate policy regimes for addressing new risk-taking
opportunities. Not infrequently, additional legislation has subsequently
been required to establish a more robust response. Although it is difficult
to predict, this pattern may repeat itself with the Gramm-Leach-Bliley
(GLB) Act of 1999. In our view, the provisions of GLB result in four significant
First, it is not clear whether regulators can effectively identify
and respond to inadequate risk management in the largest banking
Second, the objectives of holding company regulation and supervision
may be too vague under the GLB framework, as are the means by
which GLB will be implemented.
Third, the shift to functional regulation (securities regulators
examining securities firms regardless of affiliation, bank regulators
examining banks, etc.) will require difficult-to-achieve coordination
Finally, these challenges may lead creditors of nonbanking financial
firms affiliated with banking organizations to believe they will
receive government protection if these firms fail. Such thinking
could lead to excessive risk taking by these nonbanking affiliates.
Obviously Congress can enact additional safeguards if these
implementation challenges prove too serious to adequately address
within the GLB framework. Potential congressional responses include
(1) consolidating regulators, (2) increasing the command and control
authority of regulators, and (3) expanding the role of the market's
assessment of banking organization risk taking in the regulatory
process. Increasing regulatory authority has been the usual response
in previous legislative efforts to repair the results of expanded
powers. However, increased use of markets may have a greater chance
of passage this time thanks, in part, to some of the provisions
The historical premise:
Expansion of powers but incomplete regulatory responses
The passage of GLB was the culmination of a multidecade effort
to eliminate many of the restrictions on the activities of banking
organizations. Nothing as comprehensive as GLB has been approved
in the past, but over the years Congress enacted a variety of
smaller steps to increase the powers of banks. While this historical
record is complex and somewhat ambiguous it suggests a general
trendCongress expands the powers of banking organizations
only to find that the regulatory structure it created to address
the potential for increased risk taking is deficient.
Two laws passed during the 1980s, the Depository Institutions
Deregulation and Monetary Control Act and the Garn-St. Germain
Depository Institutions Act, illustrate this trend. Both laws
took steps, such as removing interest rate ceilings on deposit
accounts and broadening permissible activities, to deregulate
the banking industry. In adopting these expanding authorities
for banks, legislators and regulators assumed that the existing
regulatory framework adequately addressed the risk taking that
the new powers made possible.
However, thrift and bank failures in the mid-1980s and early
1990s demonstrated that risk taking was not adequately under control.
The Financial Institutions Reform, Recovery, and Enforcement Act
of 1989 (FIRREA) was the first step to redress the deficiencies
of the earlier legislation. This law increased the powers of regulators
to take action against banks deemed to be high risk. FIRREA was
followed by the FDIC Improvement Act of 1991, which created a
supervisory system characterized by more frequent examinations,
additional capital, and immediate and automatic action against
banks with financial problems.
The question facing analysts of GLB, and ultimately the public,
is whether this cycle will occur again. The importance of this
question is underscored by the fact that sandwiched between expansion
legislation of the early 1980s and the regulatory regime of the
late 1980s and early 1990s were billions of dollars of losses
to deposit insurance funds.
How GLB addresses expanded powers
The authors of GLB did not merely assume that the existing regulatory
structure was sufficient to address the new powers available to
banking organizations. Rather, the legislation takes some steps
to mitigate against risk taking associated with new powers.
- Restriction of powers. GLB provides methods for banking
organizations to carry out securities, insurance, merchant banking
and other powers while restricting the powers in at least three
ways. First, the law forbids certain activities outright, such
as commercial operations unrelated to financial activities.
Second, the law requires certain activities to be carried out
by subsidiaries of the financial holding company (FHC) rather
than the bank itself. Finally, regulators have the authority
to determine what new powers will be allowed in the future.
- Institutions must be "fit." GLB restricts the grant
of expanded powers to institutions that meet several regulatory
tests of financial and regulatory health. In general, the insured
banks controlled by organizations that carry out the new powers
have to meet a capital test, be "well managed, and have a satisfactory
or better rating under the Community Reinvestment Act." Firms
have a deadline to get back into compliance if they fail to
meet these standards in the future. Regulators can require banking
organizations to cease new activities and even divest if noncompliance
- The Fed's role in FHCs. The Federal Reserve will supervise
and regulate FHCs. The Federal Reserve will have access to data
on risks across the entire organization, as well as information
on the firm's management of those risks. Regulators will thus
be in position to evaluate and presumably act on risks that
threaten the safety and soundness of the insured banks. The
Federal Reserve can also require FHCs to hold capital to absorb
losses that they face.
- Limiting safety net spillovers. GLB takes a variety
of steps to limit the chance that nonbank creditors will think
they are insulated from financial loss. These creditors might
infer such protection because many creditors of their large
bank affiliates assume government bailouts will shield them
from loss. Moreover, regulators fearing systemic risk or contagion
may have incentives to protect these creditors to prevent other
banking organizations, which may have direct exposure or similar
operations to the failing firm, from suffering losses. To limit
this so-called safety net spillover, GLB takes the following
- Requires nonbanking activities to take place in legal
entities separate from the bank.
- Limits the transactions that can occur between a bank
and nonbank affiliates.
- Creates a system of functional regulation to limit direct
contact between bank examiners and nonbanking affiliates.
The open question is whether the responses of GLB to limit risk taking
will prove effective given the challenges the new policy regime will face.
Challenges facing the GLB framework
There are at least four major areas where the GLB framework will
face challenges: (1) regulatory identification and response to
high-risk activities, (2) holding company supervision and regulation,
(3) functional regulation and (4) limits on safety net spillovers.
Identification and response to high-risk activities
Bank regulation relies primarily on the regulators' ability to determine
how much risk is posed by the banking organization's various operations.
GLB continues this general approach of depending on identification of
risks by regulators. Yet, the growing complexity of the largest banking
organizations has made it more difficult for regulators to master the
operations of the firms they supervise. This task will likely be even
more complicated as banking organizations move into GLB-approved activities
where supervisors have less experience, such as merchant banking.
Even if ex ante identification of potential problems
was not an issue, the response to problems envisioned by GLB might
be. As noted, GLB gives bank regulators the power to prevent out-of-compliance
FHCs from carrying out expanded powers and allows forcible divestiture.
Although these authorities sound reasonable and effective in theory,
we have some doubt about their practical effect. This is especially
true for the more Draconian sanctions, like divestiture. There
is virtually no precedent to support the view that this sort of
cleaving of the going concern can be accomplished without fatally
damaging all of the component parts in the process. In other words,
the costs of divestiture may be so high as to greatly reduce its
use. This same logic applies to regulators' willingness to force
a FHC to stop carrying out expanded activities that generate income.
Regulators may be reluctant to take such actions for fear of making
a banking organization even more vulnerable to failure.
Holding company supervision and regulation
GLB counts on effective supervision of holding companies to protect
against failures of insured depositories. However, the effectiveness
of this holding company supervision and even its objectives are
somewhat unclear in the new GLB environment. Traditionally, bank
holding companies engaged in bank-focused activities and had a
bank-dominated culture. Banking or banking-like services were
the core products for the consolidated entity resulting in a familiar
framework in which banking supervisors could examine and judge
the solvency of the holding company.
More recently holding companies have developed with more nonbanking
assets and cultural influences and GLB will accelerate this trend.
Indeed, one of the first firms to publicly announce its intent
to become a FHC was Charles Schwab Inc., as part of its planned
acquisition of U.S. Trust. For such organizations, it may be less
obvious how the FHC's operations relate to the insured banks it
controls. As a result, it will be unclear how to examine these
FHCs and how to determine the objectives such exams should achieve.
Imposing new regulations on these FHCs will also raise significant
challenges. For example, GLB gives the Federal Reserve the right
to issue consolidated capital requirements for FHCs. Trying to
determine the riskiness of the entire firm and thereby determine
appropriate capital levels could prove nearly impossible. Regulators
may therefore create relatively simple capital standards such
as a leverage ratio (for example, a requirement that capital equal
some percent of total assets). Yet, such simple measures, by definition,
often end up setting capital at levels that are inadequate protection
against failure or so high that they lead to distorted behavior
on the part of banking organizations.
There are at least two reasons to doubt the ultimate effectiveness
of a system of functional regulation. First, the objectives and
skills of the different regulators involved with supervision of
FHCs and their subsidiaries vary considerably. For example, the
primary mission of the securities regulators may lead to a focus
on investor protection and market integrity rather than the "safety
and soundness" of the securities firm. As a result, bank regulators
may not find the reports and information generated by the securities
regulators sufficient to ascertain the threat the securities affiliate
poses to the insured banks. This perceived deficiency might encourage
bank regulators to obtain direct examination access to the securities
Second, functional regulation requires regulators to share information
and operate in a coordinated fashion. At times regulators capture
their conclusions in written reports, but much of their assessments
will reside with the examiners themselves or in more informal
written material. Distributing informal analysis poses a logistical
test that could inhibit full sharing of data and analysis as could
data security concerns. Whatever the reasons, failure to fully
share information among regulators would limit accurate risk assessments.
Safety net spillovers
GLB took several steps to address the potential for spillover
of the bank safety net. But, these steps may not change regulators'
concerns and thus may not change creditors' perceptions. For example,
GLB requires that many nonbanking activities take place outside
the banking organization. This requirement places the creditors
of nonbanking operations one step removed from the insured bank.
But does this legal distinction change the underlying fear of
contagion that motivates ex post protection in the first
place? Likewise, we find it hard to accept that the presence or
absence of functional regulation makes much difference for creditors.
Is it credible to believe that policymakers are less concerned
about the effect of a nonbank failure on an affiliated bank merely
because bank examiners did not examine the nonbank firm? In total,
GLB may not be particularly effective in limiting safety net spillovers.
This limitation would make the regulators' job even more difficult
because significant parts of the FHC will now face lower than
previous levels of market discipline and therefore have an incentive
to take on too much risk.
The existence of challenges does not mean that the GLB framework
is doomed to future amendments. Indeed, the framework created
by the legislation could be the proper response to the new powers
and new world in which banking organizations find themselves.
If the challenges do prove too difficult to address, Congress
has at least three types of additional reforms to which it could
turn. These include consolidation of regulators, increased authority
for bank regulators and increased use of market assessment of
If the coordination between multiple regulators poses a significant
problem post-GLB, then one potential solution is some form of
consolidation. Consolidation could occur between regulators (for
example, combining securities and bank regulation) or within one
type of regulator (for example, creating a national insurance
regulator). These combinations might reduce regulatory burden
and produce more coherent risk assessments. However, this type
of reform has been suggested on many occasions only to raise serious
objections focusing on a lack of regulatory options and the downsides
of consolidating power.
Increase authority for bank regulators
Legislators who believe that banking organizations will take on
too much risk post-GLB can give regulators more authority to limit
the use of expanded powers. Such steps could include making it
more difficult for firms to qualify for FHC status, placing additional
restrictions on the types of new powers that are available to
FHCs, and further constraining the types of exposures banks can
have to nonbanking affiliates.
Legislators could also remove regulatory discretion in addressing
the expanded powers of risky FHCs and provide risky firms with
less due process. This could speed the delivery of supervisory
action against these institutions when it is deemed appropriate.
Congress has taken such steps in the past, including requiring
specific regulatory actions based on a bank's capital levels.
The underlying concern with these reforms is the difficulty
regulators face in setting the appropriate level of risk taking
by banking organizations. Regulators surely identify many cases
where risk taking is, in some sense, "too high" and respond appropriately.
However, augmented regulatory power could lead to too little risk
taking by banks. While due process surely allows additional risk
taking it also protects against abuse of power.
Finding the right regulatory balance was becoming trickier before
GLB became law. The underlying problem is the rising complexity
and scope of operations of banking organizations as well as the
blunt tools that regulators have at their disposal. In fact, these
concerns have led this Reserve bank, among others, to call for
Congress to consider a third optionthe increased use of
market signals and disciplinefor reforming the legislative
Increase market discipline
GLB envisions a slightly increased use of the market's assessment
of risk in regulating banks. For example, the law requires certain
banks carrying out expanded powers to have relatively high credit
ratings. It also calls for a study of a plan that would put a
group of creditors at large banking organizations at clearer risk
of loss. The rationale and potential benefits of these aspects
of GLB, along with a suggestion that the law did not go far enough,
are discussed in detail in President Gary Stern's
column. The gist of the argument is that bank creditors with
money to lose can provide valuable price pressure on banking organizations
and provide measures of banking organizations' riskiness through
the prices they charge. This Reserve bank has long supported plans
to increase such market discipline and ensure that the price signals
are incorporated into the regulatory framework. Both steps should
help manage risk taking, particularly by the new large banking
organizations created by GLB. And while President Stern argues
that the legislation does not go as far as it could in terms of
market discipline or signals, he also notes that policymakers
have an opportunity to better contain risk taking by fully exploiting
the GLB framework in the future. We would only emphasize that
market data and discipline initiatives are particularly well suited
to bolster the potentially weak areas of GLB. Market-based reforms
could provide additional risk assessments on large, complex organizations
and would directly address safety net spillovers by putting creditors
at credible risk of loss.
The recent expansion of powers for banking organizations may be the first
of several legislative steps to adapt to a new banking environment. In
particular, the law may not have created an adequate regime to manage
the risk taking of the new banking organizations it created. If so, Congress
may need to enact additional reforms and those based on market data and
discipline may prove the most attractive.