ESSAYS ON ISSUES

THE FEDERAL RESERVE BANK JULY 2010

OF CHICAGO NUMBER 276a

Chicag­o Fed Letter

Searching for the New Normal: The Rebuilding Process for Risk
Management—A conference summary
by Carl R. Tannenbaum, senior vice president, Supervision and Regulation, and Steven VanBever, lead supervision analyst,
Supervision and Regulation

The Chicago Fed’s Supervision and Regulation Department, in conjunction with DePaul
University’s Center for Financial Services, sponsored its third annual Financial Institution
Risk Management Conference on April 6–7, 2010. The conference concentrated on
comprehensive risk management, lessons learned, and headline issues.

In addition to overviews of the risk- For example, enhancements to micro-

management landscape, this year’s con- prudential regulations, which focus on
ference focused on commercial real individual institutions, need to be supple-
estate (CRE), financial modeling, capital mented by macroprudential supervision,
planning, and risk and compensation. which considers risks to the financial
This Chicago Fed Letter provides a summary system as a whole. However, such macro-
of the relevant research presented and prudential approaches also face obstacles,
discussions held by the bankers, academ- such as knowing exactly when to inter-
ics, and supervisors in attendance. vene in a potential asset bubble. Evans
advocated a multipronged approach fea-
Opening the conference, Ali Fatemi,
Materials presented at the DePaul University, described how risk
turing strengthened capital requirements,
conference are available at a comprehensive approach to risk man-
aversion on the part of both bankers and
agement, a macroprudential supervisor,
www.chicagofed.org/ regulators had reduced the availability
and a process for effectively resolving
webpages/events/2010/ and increased the cost of credit. Establish-
insolvencies at large institutions.
ing the “new normal” requires efficient
risk_conference.cfm. allocation of credit at the lowest cost; According to Evans, central banks should
therefore, he called for the proper bal- play a key role in financial stability and
ance of private and regulatory incentives in the supervision and regulation of
to achieve this goal. Carl R. Tannenbaum, financial institutions. For one thing, a
Federal Reserve Bank of Chicago, con- central bank without supervisory respon-
tinued by contrasting the turbulent cir- sibilities would have to confront any fi-
cumstances of the previous conference nancial crisis using only monetary policy.
(held in the immediate aftermath of the In such a case, the central bank might
worst of the financial crisis) with the have to act against exuberance in finan-
relative calm surrounding the current cial markets by tightening monetary
conference, reflecting improvements policy more than would be indicated by
in the economy and financial system. macroeconomic considerations alone.

Policymaker perspectives A former Governor of the Federal Reserve

System, Randall S. Kroszner, currently
Charles L. Evans, president and CEO,
of the University of Chicago, surveyed
Federal Reserve Bank of Chicago, offered
some of the lessons about risk manage-
his thoughts on some of the challenges
ment learned from the global financial
banks and regulators are likely to face.
crisis. He first provided an overview of
some of the fragilities of the current Richard C. Cahill, Federal Reserve Bank such as anti-money laundering, business
financial system. Then he listed “seven of New York, moderated a panel of three continuity, and information security.
deadly sins of risk management” that risk chief risk officers—Terry J. Bulger, BMO Kallembach, the company’s chief infor-
managers should strive to avoid. These Financial Group; John S. Fleshood, mation officer, also serves as its enterprise
included allowing accounting values to Wintrust Financial Corporation; and risk officer, responsible for overseeing
obscure economic realities, failing to Larry J. Kallembach, MB Financial the management-level risk committee.
fully capture risk concentrations, ignor- Corporation. In addition to identifying Other important pieces of the enterprise-
ing risks faced by funding counterparties, what was distinctive about their firms, wide view of risk are provided by a risk-
being overconfident during periods of panelists shared the lessons they had management department (under the
high market liquidity, and failing to ad- learned from the recent financial crisis, administration area), a board-level credit
equately model and manage tail risk.1 including the need to better aggregate committee, internal loan review (which
reports to the audit committee), and
the chief credit officer.
The “new normal” will include more effective chief risk officers,
Commercial real estate outlook
better use of financial models, improved capital planning, and
Weaknesses in managing CRE concentra-
more alignment of compensation with risk. tions were responsible for much of the
current deterioration in banking condi-
tions affecting many small and mid-sized
Kroszner concluded by suggesting re- risks and to strengthen the culture and
banks. Furthermore, policymakers are
forms for risk managers and policy- stature of risk management.
concerned that continuing weaknesses in
makers to consider. Risk-management
Two years ago, BMO embarked on a CRE could impede the economic recov-
functions should be independent, have
structured risk improvement plan, said ery. James D. Shilling, DePaul University,
sufficient stature in the organization,
Bulger. One goal of the plan was to in- moderated a panel that provided diverse
and consider the full range of risks.
crease risk transparency: BMO enhanced perspectives on CRE. The panelists were
Central clearing of derivatives should be
its risk reporting to its board and senior Robert Bach, Grubb and Ellis Company;
encouraged to enhance market resil-
management and completed in-depth Timothy Riddiough, University of
iency and mitigate interconnectedness
risk assessments for all its trading desks. Wisconsin–Madison; and Brian D. ­
problems. The resolution regime for
Bulger stated that his organization’s Gordon, Federal Reserve Bank of Chicago.
large financial institutions should also
priorities for 2010 are managing risk
be improved. Finally, regulators should Bach provided data and analysis on cur-
(including the problem loan portfolio)
improve the monitoring of liquidity rent trends in CRE markets, including
and simultaneously building capabili-
risks and reform capital requirements. rising vacancy rates, falling rents, sharply
ties for the future. The latter includes
reduced investment volumes, and a grow-
CEO and chief risk officer perspectives strengthening core risk-management
ing pool of distressed assets. Riddiough
practices, expanding a sound risk culture
The conference featured perspectives analyzed how weaknesses in residential
throughout the firm, and more effec-
on risk management from a bank CEO lending had combined with macroeco-
tively managing capital.
and a panel of bank chief risk officers. nomic distress to undermine CRE mar-
Christopher J. Murphy III, CEO, 1st According to Fleshood, Wintrust is some- kets. According to Riddiough, in the
Source Corporation, said that his orga- what unusual in that its $12 billion in total short term CRE lending faces a con-
nization had avoided much of the recent assets is distributed among 15 individual- tinuing lack of liquidity, and its longer-
credit meltdown. However, the company ly chartered community banks. Under term recovery depends on the return
had experienced virtually every imagin- this structure, Wintrust’s board and sub- of securitization markets, albeit under
able type of failure, ranging from basic committees concentrate on enterprise- a stricter regulatory regime.
credit problems to an accounting dispute wide risks. The subsidiary banks’ boards
Gordon offered suggestions on how to
and an information technology security and their subcommittees, in turn, focus
estimate CRE losses. Loss estimation is
breach. He detailed the lessons learned more on individual banks’ performance
critical for stress testing, analysis of capital
from these various experiences. In and adherence to corporate policies.
adequacy, and evaluation of the adequacy
Murphy’s opinion, the greatest failure Key challenges with such a decentralized
of loan-loss provisioning (funds set aside
of all was allowing human hubris to structure are effectively aggregating
as an allowance for bad loans). Gordon
creep into the company. Murphy empha- risks and tailoring risk reporting to the
cautioned that broad loss rates should
sized building a culture based on con- needs of diverse audiences.
not be applied to individual portfolios,
stantly reinforced values of integrity and
MB Financial also has a somewhat atypi- which can be highly diverse. He also
honesty. He also stressed that manage-
cal structure in that its enterprise-wide emphasized that capital is meant for
ment’s incentives should not be overly
risk function evolved out of the opera- unexpected losses, not expected losses,
influenced by the short-term stock
tions/technology area, which focuses on which should be reflected in loan-loss
price performance of the organization.
traditionally underemphasized risks, reserves. Finally, Gordon highlighted
the key points in the banking regulatory crisis. However, there were clearly short- important firms for the implied govern-
agencies’ guidance on prudent CRE loan comings in the implementation and use ment support they receive. He concluded
workouts.2 This guidance emphasizes of models. It will be relatively easy to cor- that a premium is superior because it is
that excessive foreclosures are in no one’s rect some of the technical problems of likely to 1) more efficiently and effectively
best interest and that loans should not models and to use data that cover at least discourage excessive risk-taking by finan-
be adversely classified solely because the an entire deep business cycle. But it will cial institutions and 2) more transparently
value of collateral has declined. Overall, be harder to model tail events, to effec- address the TBTF problem.
better estimation of CRE losses would tively use the information that models
Finally, Mason applied a circular five-step
have improved the adequacy of loan-loss provide, and to change the culture of how
risk-management cycle (develop goals,
reserves and capital at banks with high models are used. Specifically, decision-
identify/quantify exposures, define
CRE concentrations and reduced the makers need to better understand models,
philosophy, implement program, and
severity of the current banking crisis. to integrate analysis and judgment more,
evaluate and control) to a wide range of
and to use models to inform decision-
Financial modeling historical risk scenarios, both financial
making (not only justify decisions already
and nonfinancial. Failure to adhere to
Many have identified failures of (and in- made), said Lucas.
the cycle can lead to large unexpected
appropriate use of) financial models as
Capital planning losses, as in the case of Barings Bank in
among the key causes of the financial
1995 or Société Générale in 2008.
crisis. Tannenbaum examined the In light of the large number of financial
strengths and weaknesses of modeling institutions whose capital proved inad- Risk and compensation
with a panel composed of William H. equate during the financial crisis, inter-
Alteration of incentive compensation prac-
Schomburg III, State Street Corporation; nal capital planning has become a key
tices is also high on the reform agenda.
Michael Alix, Federal Reserve Bank ­ focus of bank supervision. It was the
Keith M. Howe, DePaul University, led
of New York; and Deborah J. Lucas, subject of a panel moderated by Andre
a panel on risk and compensation that
Congressional Budget Office. Reynolds, Federal Reserve Bank of
featured Steven N. Kaplan, University of
Chicago. This panel featured Tanya K.
Schomburg said that models provide an Chicago; Kevin J. Murphy, University of
Smith, Office of the Comptroller of the
analytic framework to assess risks and a Southern California; and James W. ­
Currency; Ron Feldman, Federal Reserve
common language to communicate these Nelson, Federal Reserve Bank of Chicago.
Bank of Minneapolis; and Joseph R.
risks to others. To provide the necessary
Mason, Louisiana State University. Kaplan sought to determine whether
controls over models, State Street uses
poorly designed top executive compen-
a highly structured four-level approach, Smith outlined the “new normal” for cap-
sation at financial firms had fueled the
comprising the model owners, validation ital planning. Capital planning encom-
financial crisis. If this had been the case,
and assessment groups, and a high-level passes identification of risks and risk
we would have expected to find that top
risk-management committee with final tolerance, risk measurement, goal set-
bank executives were rewarded for
responsibility. Schomburg emphasized ting (for risks and capital), analysis of
that modeling needs to be supplemented capital supply and demand, assessment
with stress tests and expert judgment. of a range of operating requirements, Charles L. Evans, President; Daniel G. Sullivan, Senior
Vice President and Director of Research; Douglas D. Evanoff,
However, a common weakness of model- and development of capital contingency Vice President, financial studies; Jonas D. M. Fisher,
ing, stress tests, and expert judgment is plans. All of these must be incorporated Vice President, macroeconomic policy research; Daniel
Aaronson, Vice President, microeconomic policy research;
a bias toward recent data.3 into a sound governance framework. The William A. Testa, Vice President, regional programs, and
end result of capital planning should not Economics Editor; Helen O’D. Koshy and Han Y. Choi,
Alix provided a supervisory perspective. Editors; Rita Molloy and Julia Baker, Production
be just a single number. Instead, such plan-
He stressed the need to understand the Editors; Sheila A. Mangler, Editorial Assistant.
ning should constitute a well-articulated,
purpose of a particular model before try- Chicago Fed Letter is published by the Economic
well-supported, and well-understood Research Department of the Federal Reserve Bank
ing to determine its effectiveness. In his
process surrounding the many facets of Chicago. The views expressed are the authors’
view, one commonly perceived problem and do not necessarily reflect the views of the
of capital and risk.
with risk models—that they don’t produce Federal Reserve Bank of Chicago or the Federal
Reserve System.
sufficiently distressed results—is attrib- Feldman considered capital in relation
© 2010 Federal Reserve Bank of Chicago ­
utable not to technical shortcomings but to the too-big-to-fail (TBTF) problem. Chicago Fed Letter articles may be reproduced in
rather to the inability of users to “think TBTF refers to the provision of discre- whole or in part, provided the articles are not ­
outside the box” (i.e., to consider plausi- tionary government support to the un- reproduced or distributed for commercial gain
and provided the source is appropriately credited.
ble, but never experienced, shock sce- insured creditors of financial institutions Prior written permission must be obtained for
narios). Other problems are the lack perceived to pose systemic risk. He any other reproduction, distribution, republica-
tion, or creation of derivative works of Chicago Fed
of reliability of input data and the un- compared two options for addressing Letter articles. To request permission, please contact
critical use of models’ output. TBTF—a capital surcharge for systemically Helen Koshy, senior editor, at 312-322-5830 or
email Helen.Koshy@chi.frb.org. Chicago Fed
important institutions and a premium
Lucas was not convinced that models Letter and other Bank publications are available
(e.g., incorporated into deposit-insurance at www.chicagofed.org.
themselves had failed during the recent
premiums) that charges systemically ISSN 0895-0164
short-term results with large amounts of “clawback” provisions (where rewards indicated that the “new normal” will in-
upfront cash pay; bank executives did are recovered if critical indicators on corporate stronger, fortress-like balance
not hold sufficiently large amounts of which bonuses were based are revised sheets and more effective chief risk offi-
stock to align their interests with those in the future). He also recommended cers and boards. While many of the firms
of shareholders; and firms utilizing more basing bonuses on value creation rather that became troubled did have chief risk
short-term pay and less stock ownership than on sheer volume of transactions. officers that were formally independent
as compensation performed worse in the of business line management, often
Nelson presented the Federal Reserve’s
crisis. Recent research that Kaplan cited these individuals lacked sufficient in-
proposed guidance on sound incentive
did not support these propositions. fluence to restrain excessive risk-taking.
compensation policies.4 This guidance
Therefore, poorly designed top execu- For chief risk officers to gain more in-
is based on three fundamental principles.
tive compensation does not appear to fluence, Ludwig supported producing
First, incentive compensation arrange-
have played a significant role in the targeted, high-quality risk reporting to
ments should not provide employees
financial crisis, especially compared with boards rather than generating a massive
with incentives to take risks beyond an
other factors. Kaplan argued that more tome (known in one firm as “the brick”)
organization’s ability to effectively iden-
regulation of top bank-executive pay is that nobody reads. In addition, risk-
tify and manage those risks. Second, such
unnecessary and would have negative management models should place more
arrangements should be compatible with
unintended consequences, such as driv- emphasis on tail events, become more
effective risk management and controls.
ing the most talented employees to un- forward-looking, and draw on real-time,
Third, these policies should be sup-
regulated sectors, such as hedge funds, enterprise-wide data.
ported by strong corporate governance,
private equity funds, and boutique firms.
including active and effective board For Ludwig, concentration levels in
Murphy was also pessimistic about in- oversight. This proposed guidance is CRE at small and mid-sized banks were
creased government regulation of com- being supplemented by two supervisory less of a problem than their business
pensation. He said that regulation is initiatives—one for large, complex models that ignored other types of
often designed to be punitive and ad- banking organizations and another for lending opportunities, such as business
vance political agendas rather than to be the remaining organizations. lending. Ludwig also recommended
constructive and foster creation of share- that banks improve the quality and di-
holder value. Nevertheless, Murphy Summing up versity of their earnings by emphasizing
argued that compensation practices in Eugene A. Ludwig, Promontory Financial deposit accounts, developing sustain-
financial services could be improved. Group, was the U.S. Comptroller of the able fee income, and looking hard for
He suggested bonus deferrals and Currency over the period 1993–98. He cost-saving opportunities.

1 Technically, tail risk is a form of portfolio 2 For further details, see the Federal Reserve the preceding few years would not have
risk that arises when the possibility that an press release on prudent CRE loan workouts included any instances of sharp declines in
investment will move more than three stan- at www.federalreserve.gov/newsevents/ asset values and thus would have produced
dard deviations from the mean is greater press/bcreg/20091030a.htm. overly optimistic modeling results.
than what is shown by a normal distribution. 3 Financial models often use recent data that 4
See the Federal Reserve press release at
More broadly, the term is used to refer to do not cover past business cycles. For ex- www.federalreserve.gov/newsevents/
the risk of large unexpected losses for the ample, during the late 2000s, data from only press/bcreg/20091022a.htm.
financial sector as a whole.