How Does the Stock Market View Bank Regulatory Capital Forbearance Policies?

AuthorXIAOXIA YE,VAN SON LAI
Date01 December 2020
DOIhttp://doi.org/10.1111/jmcb.12692
Published date01 December 2020
DOI: 10.1111/jmcb.12692
VAN SON LAI
XIAOXIA YE
How Does the Stock Market View Bank Regulatory
Capital Forbearance Policies?
During the subprime crisis, the Federal Deposit Insurance Corporation
(FDIC) has shown, once again, laxity in resolving and closing insolvent in-
stitutions. Ronn and Verma (1986) call the tolerance level below which a
bank closure is triggered the regulatory policy parameter.We derive a model
in which we make this parameter stochastic and bank specic to infer the
stock market view of the regulatory capital forbearance value. For565 U.S.
listed banks during 1990 to 2012, the countercyclical forbearance fraction
in capital, most substantial in recessions, could represent 17%, on average,
of the market valuation of bank equity and could go as high as 100%.
JEL codes: G17, G21, G28
We are grateful to the kindness and insights of the Coeditor Robert DeYoung and we thank the two
anonymous referees for their most valuable comments on a previous version of this paper.The paper has
been seminalist for Best Paper Awardin the Markets and Institutions Category at the 2017 FMA annual
meeting. We acknowledgethe Fonds d’Assurance Conrad Leblanc, the Laval University Faculty of Busi-
ness Administration Research Ofce, the Risk Management Institute (RMI) of the National University of
Singapore Credit Research Initiative, and the Social Sciences and Humanities Research Council of Canada
for their nancial support. We appreciate Linda Allen, William Bazley, Lamont Black, Dion Bongaerts,
Craig Brown, Chun-Hao Chang, Brian Clark, Bart Diris, Michal Dzielinski, Helyoth Hessou, Mia Hin-
nerich, Ai Jun Hou, Erik Kole, Martien Lubberink, Mike Mao, Sabur Mollah, Lars Nordén, Richard Paap,
Wook Sohn, Mohammad Tajik, Wing WahTham, DmitriVinogradov, Dong Zhang, Jin Zhang, Lei Zhao,
Lu Zhao, and participants at the nance faculty seminar of Stockholm Business School, the 2014 Con-
ferences of the Financial Engineering & Banking Society (FEBS), the International Finance and Banking
Society (IFABS), the Victoria University of Wellington (VUW) Finance Workshop, the 2015 FMA Eu-
ropean Conference, the 9th Annual Risk Management Conference, the 4th International Conference on
Credit Analysis and Risk Management, the 2016 MFS Conference, the 2017 FMA Annual Meeting, and
the seminar at Econometric Instituate of Erasmus School of Economics for their helpful comments. We
also thank Professor James Hamilton for sharing with us the data used in Hamilton (2014); Wei-Fang
Niu and Fan Yang for excellent research assistance at the early stage of this project while Lai was vis-
iting RMI; Huanjia Chen for his excellent technical supports. Lai also thanks Professor Jin-Chuan Duan
for his hospitality. The computations in the paper were partially performed with resources provided by
the Swedish National Infrastructure for Computing (SNIC) at the Uppsala Multidisciplinary Center for
Advanced Computational Science (UPPMAX).
VS L is a Faculty of Business Administration, Laval University, Canada, and IPAGBusiness
School, Paris, France (E-mail: vanson.lai@fsa.ulaval.ca). X Y is with University of Liverpool
Management School, United Kingdom (E-mail: xiaoxia.ye@liverpool.ac.uk).
Received April 27, 2019; and accepted in revised form August 19, 2019.
Journal of Money, Credit and Banking, Vol. 52, No. 8 (December 2020)
© 2020 The Ohio State University
1874 :MONEY,CREDIT AND BANKING
Keywords: bank regulatory closure rules or policy parameter, bank
insolvency, regulatory forbearance, market-based closure rules, nancial
crises
U     to improve
the design and implementation of bank regulatory policies, bank risk management,
and deposit insurance has long been an important topic in the banking literature.
Notwithstanding the efcient market hypothesis not necessarily holding and the
emergence of endogenous risk (see, e.g., Danielsson, Shin, and Zigrand 2012) espe-
cially during nancial crises, Flannery (1998, 2001), Gunther, Levonian, and Moore
(2001), Krainer and Lopez (2004), and others, do conrm that market information is
useful for ranking banks and provides incremental information for bank regulators’
supervisory monitoring and assessment.
Taking advantage of the relatively higher liquidity and efciency of listed bank
equity prices, our study focuses on the use of stock market information to infer the
market perception of the regulatory closure rules, known following Ronn and Verma
(1986) as the regulatory policy parameter. This parameter is most often driven at
least in part by politics, especially in the case of systemically important nancial
institutions (SIFIs), which will be subject to enhanced capital requirements according
to the current Basel 3 regulatory framework. This parameter represents a (hypo-
thetical, conjectural, or even real) limit, expressed as a percentage ρ(0ρ1)
of the total debt value Dof the bank at the time of supervisory audit, beyond
which the dissolution of assets by regulatory bodies would be a reasonable alternative.
If the value of the bank falls between ρDand D, the insuring agency forbears (e.g.,
the Savings and Loans crisis in the 1980s), and in the case of extreme market turmoil,
the government intervenes, as witnessed during crises such as the 2007–09 subprime
crisis via the Troubled Asset Relief Program (TARP,see Veronesi and Zingales 2010),
infusing up to (1 ρ)Dand making it equal to D. If the bank value falls below ρD,
the insuring agency steps in to dissolve the assets of the bank. A compelling reason
for not closing an insolvent bank is the loss of signicant franchise value (stemming
from core deposits, customer relationships, and valued personnel) that occurs after
a Federal Deposit Insurance Corporation (FDIC) seizure. If the market insolvency
closure rule is strictly followed, the policy parameter is equal to one, and there is reg-
ulatory capital forbearance when the policy parameter is below one. Along with other
assumptions, Ronn and Verma (1986) x ρat a constant 0.97 to yield an aggregate
weighted average premium of 1/
12% for the U.S. deposit insurance premium. Since,
in Ronn and Verma’s model, the value of forbearance is assumed to be the value of
the capital assistance, many authors call this practice regulatory capital forbearance.
Under the Basel 3 countercyclical capital buffer framework, bank regulators
would try to ensure that banks build up capital levels during good times so that they
can run it down in bad times (see, e.g., Drehmann et al. 2010, Hanson, Kashyap,
and Stein 2011). In effect, to conduct countercyclical capital requirements policies,
regulators are compelled to adopt a time-varying policy parameter. To the best of
VANSON LAI AND XIAOXIA YE :1875
our knowledge, only Lai (1996) treats the policy parameter as a stochastic process
to reect the uncertainty of the bank closure rules.1
In this paper, we extend Lai’s(1996) framework by modeling ρas a more realistic
stochastic process that is mean reverting and bound by zero and one. To justify this,
we argue that the regulatory forbearance policy can be treated as “reduced form” and
described by a state variable. Further, the policy is constrained by economic, legal,
political, regulatory competition, and bureaucratic considerations that are mean
reverting throughout their respective cycles. Two main sources of uncertainty re-
garding the variation of bank regulatory forbearance may be posited: (i) asymmetric
information and (ii) stochastic state variables. In the rst case, because of condential
information obtained from on-site examinations, private information may exist that
is known only to the regulator.2For actively stock-listed banks in efcient markets
with increased disclosure, such a scenario may be considered not as a major source of
uncertainty. Forclosely held banks, it is more plausible but we discard this possibility
by only studying banks with available equity prices. In the second situation, we
suppose that while the policy of the regulator is known to all (e.g., the Too Big to Fail
[TBTF] doctrine), it is a function of some external, hardly predictable state variables,
structural changes, or unforeseeable events. While asymmetric information and
stochastic state variables are two possible sources of regulatory uncertainty, only
the latter is explicitly modeled in this paper.3With this justication, we develop an
enhanced Ronn and Verma (1986) model to infer the market-based, bank-specic,
and more exible policy parameter from the market value of bank equity.4Then,
based on the calibration of our model with U.S. listed banks, by gauging the size
of the capital forbearance value and contributing to the ongoing debate on adequate
bank capital, we seek answers to the following two research questions: (i) How does
the time-varying capital forbearance portion embedded in bank equity depend on
various banks’ own risk and businesscycle variables? and (ii) How do banks’ market-
assessed intrinsic (i.e., devoid of the forbearance subsidy) capital ratios (or inverse
leverage ratios) react to various business cycles and the banks’ own risk variables?
Toward this end, we rst develop a two-factor model, in which we model ρas
an exponential of a negative Cox–Ingersoll–Ross (CIR) process (Cox, Ingersoll Jr,
1. Kane (1986) treats safety-net guarantees as a two-part option: a taxpayer put and a knock-in,
stop-loss call on the rm’s assets; while Allen and Saunders (1993) model forbearance as forfeiture by
the deposit insurer of the value of its call component of the deposit insurance option.
2. DeYoung et al.’s (2001) empirical results indicate that on-site examinations do produce value-
relevant information about the future safety and soundness of banks not reected immediately in their
debenture prices whereas we focus on their stock prices.
3. Not postulating forbearance as the outcome of a random process, political-economy theorists treat
it as a function of insurer resources and workloads and of insolvent rms’ size, complexity, and political
clout. When rms are truly too big or too many to fail, some of the insurer’s deposit insurance calls are
never or almost never going to be exercised.This allows these rms a real option to expand their balance
sheets and “gamble for resurrection.”
4. To compare the current market capital–asset ratio to the current regulatory capital-asset ratio for
a U.S. sample of publicly traded bank holding companies and savings and loans associations in 1990,
Cordell and King (1995) present an approach for extending the forbearance factor to be bank and time
specic. However, Cordell and King (1995) arbitrarily assign values for the critical value of the policy
parameter and use an interrelated approximation of what they call the conditional value of forbearance.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT