Description
The aim of this paper is to examine the question of the specialness of banks by addressing
concerns raised in the recent studies and deriving policy implications for the future of banking. The
specialness of banks has been well documented in the finance literature. More recent research, however,
calls into question the special nature of banks.
Journal of Financial Economic Policy
Depositor discipline and the bank ' s incentive to monitor
Michael F. Ferguson Bradley A. Stevenson
Article information:
To cite this document:
Michael F. Ferguson Bradley A. Stevenson , (2014),"Depositor discipline and the bank ' s incentive to
monitor", J ournal of Financial Economic Policy, Vol. 6 Iss 2 pp. 98 - 111
Permanent link to this document:
http://dx.doi.org/10.1108/J FEP-06-2013-0022
Downloaded on: 24 January 2016, At: 21:48 (PT)
References: this document contains references to 37 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 110 times since 2014*
Users who downloaded this article also downloaded:
Sheilla Nyasha, Nicholas M Odhiambo, (2014),"Bank-based financial development and economic growth:
A review of international literature", J ournal of Financial Economic Policy, Vol. 6 Iss 2 pp. 112-132 http://
dx.doi.org/10.1108/J FEP-07-2013-0031
Galina Hale, J oão A.C. Santos, (2014),"Do banks propagate debt market shocks?", J ournal of Financial
Economic Policy, Vol. 6 Iss 3 pp. 270-310 http://dx.doi.org/10.1108/J FEP-03-2014-0023
J ill M. Hendrickson, Mark W. Nichols, Daniel R. Fairchild, (2014),"Bank branch location and stability
during distress", J ournal of Financial Economic Policy, Vol. 6 Iss 2 pp. 133-151 http://dx.doi.org/10.1108/
J FEP-07-2013-0026
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Depositor discipline and the
bank’s incentive to monitor
Michael F. Ferguson
Finance and Real Estate Department, University of Cincinnati, Cincinnati,
Ohio, USA, and
Bradley A. Stevenson
Department of Economics and Finance, Bellarmine University, Louisville,
Kentucky, USA
Abstract
Purpose – The aim of this paper is to examine the question of the specialness of banks by addressing
concerns raised in the recent studies and deriving policy implications for the future of banking. The
specialness of banks has been well documented in the fnance literature. More recent research, however,
calls into question the special nature of banks.
Design/methodology/approach – We use event study methodology to study 423 bank loan
announcements from1988 to 1996 and examine the returns relative to proxies for the bank’s monitoring
incentives and skill using ordinary least squares (OLS) regressions.
Findings – Our results indicate borrower abnormal announcement returns are positively related to
proxies for the bank’s monitoring incentives and skill as measured by: the ratio of uninsured deposits to
total loans; a risk-adjusted measure of recovered charge-offs; and the relative bank-to-borrower capital
ratio.
Research limitations/implications – The results reveal how the fragile nature of the bank’s
structure improves the bank’s incentives to monitor borrowers.
Practical implications – Our results can inform the current debates in the Fed and in Congress
surrounding reapplying the Glass-Steagall Act and limiting the size of banks. We showthat banks were
special before the Gramm-Leach-Bliley Act and when fewer banks belonged to the too-big-to-fail
category. This suggests that reregulating banks to re-establish their fragile nature will re-establish
them as information-generating intermediaries instead of just transactional institutions.
Originality/value – Our fndings have not previously been documented but are broadly consistent
with models developed by Calomiris and Kahn (1991) and especially Diamond and Rajan (2001).
Keywords Demand deposits, Bank monitoring, Loan announcement returns
Paper type Research paper
1. Introduction
The mix of private and public oversights of US fnancial institutions, and the effcacy of
private oversight has varied widely over time. Kane (1977) coined the term “regulatory
dialectic” to describe the back and forth between regulators and those trying to evade
their oversight. As part of this ebb and fow, periodic changes are made to the public
oversight component. Two major recent changes in public oversight are the
Gramm-Leach-Bliley repeal of Depression-era restrictions on combining commercial
JEL classifcation – G20, G21, G30, G32
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JFEP
6,2
98
Journal of Financial Economic Policy
Vol. 6 No. 2, 2014
pp. 98-111
© Emerald Group Publishing Limited
1757-6385
DOI 10.1108/JFEP-06-2013-0022
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and investment banking. This was intended to be, and was viewed as, a “deregulatory”
change. After the 2007-2008 fnancial crisis, the Dodd-Frank reforms were passed. This
act was intended to be, and was widely viewed as, a movement toward stricter
regulation.
One way to view regulatory changes is to view them as a change in the relative
importance of public and private oversight of banks. Specifcally, at different times
depositors and capital providers have been relied on to be the primary monitors of banks
(often referred to as “market discipline”). Implicit in a movement toward greater public
oversight of bank safety and soundness is that market oversight is inadequate.
This paper explores the role of private oversight of banks from a unique angle – we
revisit the literature on bank loan “specialness”, which suggests banks that provide the
most value added are those that provide the best monitoring services. Because
monitoring services are provided after making a loan, the provision of monitoring
services is subject to a classic agency problem. Who monitors the bank management to
ensure it actually provides these value-added services?
Calomiris and Kahn (1991) argue uninsured depositors play a monitoring role,
standing ready to “run” the bank in case of mismanagement. Diamond and Rajan (2000,
2001) develop equilibrium models in which the bank’s capital structure is central to the
provision of specialized monitoring services. Specifcally, uninsured depositors
withdraw capital whenever the return from loan monitoring is too low. This ensures
managers expend resources to monitor loans. Thus, a “fragile” bank capital structure
(i.e. relying on very short-term, very liquid debt) subject to runs resolves the lender’s
agency problem.
We document market reaction to bank loan announcements consistent with the
notion that uninsured bank depositors play an important monitoring role. In a sample of
bank loans that have positive average announcement returns (i.e. they appear to be
“special”), the announcement returns are increasing in the proportion of uninsured
deposits. This suggests that the provision of the lender’s special services is dependent
upon the availability of market discipline. This has interesting implications for
regulatory reform. If bank liabilities (i.e. deposits) are too safe, the specialness of banks
may be compromised. To the extent policy makers view credit analysis and loan
origination as welfare-enhancing activities, they should consider costs associated with
reduced depositor discipline as they craft legislation and new regulations.
We have chosen to take a moderately “historical” look at bank loan markets by
examining data from prior to the passage of the Gramm-Leach-Bliley Act (GLBA). Our
reason is there have been a number of changes made recently to the regulatory
environment that have potentially signifcantly reduced the specialness of banks. Also,
a number of current proposals imply or explicitly propose undoing many of the changes
made by GLBA. So to better inform policy makers, it is useful to improve our
understanding of bank lending prior to the passage of GLBA.
What changes affected the nature of banks? First, with the passage of GLBAin 1999,
also called the Financial Services Modernization Act, banks were allowed to act as both
commercial and investment banks. On one hand, this change might allow for better
screening and monitoring due to information synergies between the two banking
functions. Alternatively, combining commercial and investment banking allowed banks
to become bigger and more interconnected which exacerbates the too-big-to-fail (TBTF)
problem. Increasing the TBTF problem impairs the special nature of banks (if their
99
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
specialness is due to their fragile nature, as described by Diamond and Rajan (2001))
because they are no longer “fragile”.
Second, even prior to GLBA, banks were growing larger and spreading
geographically due to deregulation of intrastate and interstate expansion. Banks that
serve borrowers farther away gain greater geographic diversity in their portfolios.
While this may improve effciency and diversify portfolios, it may result in banks
lending to more distant, less familiar (i.e. more opaque) borrowers. Petersen and Rajan
(2002) note that banks tend to lend to closer more opaque borrowers than non-banks.
Furthermore, DeYoung et al. (2008) show that increased distance between bank lenders
and borrowers and the increased use of credit scoring has led to increased default rates
for small business borrowers. Thus, GLBAhas facilitated the recent trend toward banks
behaving more like non-bank lenders.
Finally, Basel II and the introduction of small business scoring have increased the use
of more analytical, less relationship-based methods to judge credit risk. Berger et al.
(2005) suggest that small business credit scoring systems (SBCSS) reduce information
asymmetry because loans made with SBCSS tend to have longer maturities indicating
less information asymmetry.
Our hypothesis is that the market discipline provided by depositors is crucial to the
provision of bank monitoring. What is bank monitoring? The theoretical literature
draws a vivid picture. A common theme dating back to Kane and Malkiel (1965) and
Black (1975) is banks monitor some aspects of their borrowers’ activities. Subsequent
research suggests, relative to other lenders, banks have superior monitoring
technologies and more detailed knowledge of a borrower’s investment opportunities,
infuence over project choice and, importantly, infuence over effort choice[1]. Besanko
and Kanatas (1993) describe bank monitoring as an active process of tracking the effort
exerted by the entrepreneur via long-term relationships where the bank can monitor the
frm’s collateral balances, fnancial transactions and agent effort. In this environment,
banks have a direct infuence on borrower decisions, which is consistent with the view
of banks as relationship lenders. Boot (2000) defnes relationship lending as repeated
interactions over time with the same borrower that allows the bank to develop
frm-specifc, private information. As Rajan (1992) observes, this intimate knowledge of
a borrower’s projects can be so extensive as to give the lender a decided bargaining
advantage over the borrower. In short, monitoring is an active, ongoing, intrusive and,
ultimately, expensive process that requires skill and proper incentives.
The proxies for monitoring skill and incentive we identify are:
• the ratio of the bank’s uninsured deposits to total loans;
• the relative bank-to-borrower capital ratio; and
• a risk-adjusted measure of recovered charge-offs.
These proxies for a bank’s skill and incentive to monitor borrowers are associated with
higher excess returns surrounding bank loan announcements. This implies the excess
returns associated with bank loan announcements are a refection of the monitoring
benefts generated by the bank. That “monitoring creates value” is the key insight
gleaned from the strand of the empirical “banks-are-special” literature that began with
Fama (1985) and James (1987). Evidence that uninsured deposits have an important
impact on the “specialness” of banks is the primary contribution of our paper. This
JFEP
6,2
100
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
evidence implies that policy makers and regulators should consider the potential costs
of reducing market discipline of banks. Specifcally, if banks become too safe they may
no longer be particularly valuable as monitoring lenders.
The rest of the paper is as follows. Section 2 reviews the literature on the special
nature of banks. Section 3 describes the data used for the loan announcement and excess
return analysis. Section 4 details empirical proxies for monitoring. Section 5 examines
the relationship between those proxies and bank loan announcement abnormal returns.
Section 6 concludes.
2. Bank specialness
James and Smith (2000) summarize the literature on the “specialness” of bank loans:
In general, these studies suggest that private lenders add the greatest value for so-called
‘informationally-intensive’ borrowers – borrowers who face the greatest potential information
problems when issuing public securities […] The fact that bank loans are collateralized,
contain strict covenants, and are typically short term improves a bank’s skill to monitor
informationally-intensive loans.
Thus, the existing literature tells us about differences in borrowers or about differences
in loan contracts. This literature confrms Fama’s initial observation that bank loans are
“special” and banks provide value to their borrowers they cannot get elsewhere from
specialized “monitoring” services. Notably, previous research is relatively silent on
lender characteristics[2].
James (1987) and others provide evidence that banks are special, but this specialness
has been contested in recent years. One argument is that banks are no longer special.
Another, stronger argument is banks never were special.
The frst argument comes from Fields et al. (2006). They fnd loan announcement
effects diminish over time, disappearing toward the end of their sample (from 1980 to
2003). They attribute this to changes in the information environment (Petersen and
Rajan, 2002), increased use of risk management techniques such as credit scoring
(Schuermann, 2004) and increased capital standards. Also, they note the small impact of
the 2001 recession on lending activity. Several recent papers argue bank loans continue
to be special[3]. Nonetheless, we are concerned about the potentially changing nature of
the special role of banks. Therefore, we conduct our analysis on a pre-GLBA sample of
bank loan announcements.
More fundamentally, Maskara and Mullineaux (2011) claim the loan announcement
effects in earlier studies never existed because the sample of loans in studies like Billett
et al. (1995) do not represent the universe of borrowers because the announced loans are
subject to self-selection bias. The fear is that the sample is biased by announcements
characterized by disproportionately positive excess returns because borrowers only
announce “good news”. They fnd the abnormal positive returns for the period examined
by Billet et al. (1995) are actually zero in a data set including announced and
unannounced loans. They acknowledge for the smallest frms there are signifcant
abnormal returns.
Because we identify loan announcements via media reports, we are concerned about
the self-selection bias described above. However, our sample of announcements from
1988 to 1996 exhibits rich cross-sectional variation and is not dominated by positive,
“good news” announcements. The maximumand minimumexcess returns are 47.63 per
cent and ?15.44 per cent for a range of 63.07 per cent with a mean excess return of 0.89
101
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
per cent; 54.85 per cent of the announcements are positive. The inter-quartile range is
4.46 per cent (high ? 2.93 per cent; low ? ?1.53 per cent). Figure 1 plots the
standardized excess returns.
As our goal is to relate variation in bank loan announcement returns to lender
characteristics, the cross-sectional variation is more important than the mean. Figure 1
and the range statistics suggest substantial cross-sectional variation; moreover, the
sample is balanced between positive and negative returns. We feel confdent our sample
is not signifcantly affected by a sample selection bias.
3. Data
3.1. Excess returns
We identify 1,065 loan announcements by commercial banks to non-fnancial frms
appearing on the Reuters wire, DowJones Newswires or in The Wall Street Journal from
1988 to 1996[4]. Articles were selected by searching for the terms “line of credit”, “credit
line”, “credit facility”, “credit agreement”, “credit extension”, “new loan”, “loan
agreement”, “loan renewal”, “loan revision”, “loan extension”, “fnance company loan”,
“termloan”, “commercial loan”, “bank loan”, “loan facility” or “working capital facility”.
We collected the borrowing frm’s name, announcement date, the lead lender, lenders
in the syndicate, initiation or renewal, facility amount, previous agreement amount,
maturity, purpose, changes to interest rates and covenants and loan type. Loans were
dropped due to unidentifed lender, confounding news events, lack of data from center
for research in securities prices (CRSP) (returns) and COMPUSTAT (lender and
borrower data) and identifcation as a real estate investment trust (REIT) or fnancial
frm[5]. Removing these observations leaves a clean sample of 423 loan announcements.
The descriptive statistics appear in Table I..
0
10
20
30
40
50
60
70
80
90
<
-
2
0
.
5
%
-
2
0
.
5
%
t
o
-
1
9
.
5
%
-
1
9
.
5
%
t
o
-
1
8
.
5
%
-
1
8
.
5
%
t
o
-
1
7
.
5
%
-
1
7
.
5
%
t
o
-
1
6
.
5
%
-
1
6
.
5
%
t
o
-
1
5
.
5
%
-
1
5
.
5
%
t
o
-
1
4
.
5
%
-
1
4
.
5
%
t
o
-
1
3
.
5
%
-
1
3
.
5
%
t
o
-
1
2
.
5
%
-
1
2
.
5
%
t
o
-
1
1
.
5
%
-
1
1
.
5
%
t
o
-
1
0
.
5
%
-
1
0
.
5
%
t
o
-
9
.
5
%
-
9
.
5
%
t
o
-
8
.
5
%
-
8
.
5
%
t
o
-
7
.
5
%
-
7
.
5
%
t
o
-
6
.
5
%
-
6
.
5
%
t
o
-
5
.
5
%
-
5
.
5
%
t
o
-
4
.
5
%
-
4
.
5
%
t
o
-
3
.
5
%
-
3
.
5
%
t
o
-
2
.
5
%
-
2
.
5
%
t
o
-
1
.
5
%
-
1
.
5
%
t
o
-
0
.
5
%
-
0
.
5
%
t
o
0
.
5
%
0
.
5
%
t
o
1
.
5
%
1
.
5
%
t
o
2
.
5
%
2
.
5
%
t
o
3
.
5
%
3
.
5
%
t
o
4
.
5
%
4
.
5
%
t
o
5
.
5
%
5
.
5
%
t
o
6
.
5
%
6
.
5
%
t
o
7
.
5
%
7
.
5
%
t
o
8
.
5
%
8
.
5
%
t
o
9
.
5
%
9
.
5
%
t
o
1
0
.
5
%
1
0
.
5
%
t
o
1
1
.
5
%
1
1
.
5
%
t
o
1
2
.
5
%
1
2
.
5
%
t
o
1
3
.
5
%
1
3
.
5
%
t
o
1
4
.
5
%
1
4
.
5
%
t
o
1
5
.
5
%
1
5
.
5
%
t
o
1
6
.
5
%
1
6
.
5
%
t
o
1
7
.
5
%
.
1
7
5
t
o
.
1
8
5
1
8
.
5
%
t
o
1
9
.
5
%
1
9
.
5
%
t
o
2
0
.
5
%
>
2
0
.
5
%
N
u
m
b
e
r
o
f
P
r
e
d
i
c
?
o
n
E
r
r
o
r
s
p
e
r
C
a
t
e
g
o
r
y
Two-Day Predic?on Errors
Figure 1.
Prediction error
distribution
JFEP
6,2
102
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.2. Bank monitoring proxies
Lender data collected fromcall reports included total loans, total assets, total C&I loans,
fees and interest on C&I loans, owner’s equity, total deposits, deposits over $100,000,
deposits under $100,000, amount of loan commitments, amount of non-accrual and past
due loans and the amount of predicted loan losses. Table II contains proxy
characteristics from the call report data.
If a commercial bank and a foreign or investment bank are mentioned as the lead
lender or arranger, the domestic commercial bank is treated as the lead lender. If one
bank is identifed as the lead lender, data are collected for it only. If multiple banks are
mentioned but there is no indication of rank, data are collected for all lenders.
Table III reports two-day standardized prediction errors for the full sample and
subsamples and illustrates loan announcements are associated with positive excess
returns. The average abnormal return (prediction error (PE) ? 0.89 per cent) is larger
than reported in most previous studies and it is signifcantly different from zero. In all,
54.85 per cent of the sample has a positive announcement effect[6].
The univariate analysis shows, except for loan yield and capital ratio, the abnormal
returns for bank loan announcements are greater when the lender characteristic is above
the median value. However, the only signifcant difference is for uninsured deposits.
4. Monitoring proxies and control variables
4.1 Proxies
4.1.1 Uninsured deposits. Diamond and Rajan (2001) and Calomiris and Kahn (1991)
emphasize the monitoring role uninsured depositors play in disciplining management.
To account for bank size, we divide total uninsured deposits by total loans. Banks with
lowratios face less discipline fromdepositors, while high ratios imply greater discipline
Table I.
Characteristics of bank
loan announcements
Loan and bank characteristics
N ?423; N ?99 with public debt
Mean Maximum Minimum
Loan amount ($ millions) 98.7 (40.0) 2050.0 0.5
Per cent with multiple lead banks 47.75 per cent
Borrower size (USD millions) 460.9 (131.2) 8567.3 1.6
Loan/MV of equity 0.78 (0.50) 8.2 0.02
Loan maturity (in years) 3.2 (3.0) 10.0 0.1
10-day cumulative stock return 1.09 per cent (0.0 per cent) 69.697 per cent ?41.67 per cent
Stock return volatility 7.84 per cent (3.59 per cent) 811.08 per cent 0.99 per cent
Notes: The Loan amount includes the amount drawn down immediately as well as the unused portion
of the line. Borrower size is defned as the sum of the book value of total liabilities, liquidating value of
preferred stock and the market value of common stock as of end of the year prior to the announcement.
Loan/MV equity is the dollar amount of the loan divided by the market value of the borrower’s assets.
In all, 217 of 423 announcements did not provide the Loan maturity. Firms are deemed to have Public
debt outstanding if COMPUSTATreports a debt rating at the end of the year prior to the announcement
or if COMPUSTAT reports a non-zero balance for public debt for the prior year. The 10-day cumulative
stock return is measured in the ten days preceding the announcement, and the stock return volatility is
measured from day ?200 to day ?51 before the announcement. Means are reported with medians in
parentheses
103
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and more intense monitoring. Thus, higher ratios should translate into more positive
announcement effects.
4.1.2 Capital ratio. As reported by Besanko and Kanatas (1993), entrepreneurs exert
effort related to their stake in the frm. Outside funding decreases effort and decreases
the probability of project success. Bank monitoring can substitute for the incentive
provided by owner’s equity. In their model, moral hazard arises because banks cannot
commit to a level of monitoring. Holmstrom and Tirole (1997) develop a related model
where banks substitute their own capital (through the incentive to monitor) for the
borrower’s capital. Our proxy is the ratio of total bank equity to total bank assets
divided by the ratio of total borrower equity to total borrower assets. Higher values
indicate increased incentive for the bank to monitor relative to the borrower’s incentive.
Therefore, larger values should be associated with greater excess returns.
4.1.3 Recovered charge-off per cent. Diamond and Rajan (2001) argue the bank’s
advantage in relationship lending is, via monitoring, acquiring specialized knowledge of
the borrower’s assets, creating an advantage in liquidating those assets. Thus, better
monitors should recover more of charged-off loans. We proxy this with the ratio of
recovered loans to the total amount of charged-off loans. This measures the bank’s
effectiveness in collecting on defaulted loans, where the higher percentage collected, the
greater the bank’s specialized knowledge.
However, better monitors may attract riskier loans. A bank that is better in an
absolute sense at liquidating assets may have a lower recovered charge-off per cent. To
control for this, we adjust for the bank’s loan portfolio risk, interacting the per cent
recovered with a proxy for risk measured by the bank’s past due loans[7]. After
controlling for bank risk, recovered charge-offs will be positively related to abnormal
returns such that, for a given risk level, better monitors will recover more.
4.1.4 Loan yield. Bernanke and Gertler (1989) develop an optimal auditing model of
bank monitoring where higher repayment (i.e. yield) for riskier investments lowers
borrower consumption in the good state (where repayment occurs). This reduces the
potential opportunity cost of falsely claiming the bad state has occurred which increases
the moral hazard of the borrower and the need for costly bank monitoring. Controlling
for bank loan portfolio risk, as in recovered charge-off per cent, the average yield earned
by the lender should proxy for the monitoring performed in its lending relationships.
The coeffcient should be positive. For a given level of risk, higher yields equate to more
Table II.
Proxies for bank
monitoring
Bank monitoring proxies
Full sample N ?423
Mean Maximum Minimum
Uninsured deposits 0.38801 (0.34350) 3.53115 0.017486
Capital ratio 0.37555 (0.1975) 49.8102 ?25.1106
Loan yield 0.053693 (0.052055) 0.21053 0.026946
Recovered charge-off per cent 0.63642 (0.32895) 14.7731 0
Past due loans/assets 0.032721 (0.011005) 2.00868 0
LC/bank assets 0.000706 (0.0000007) 0.0244832 0
Loan loss provision 0.025992 (0.022947) 0.11799 0
Note: Median values are reported in parentheses
JFEP
6,2
104
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
monitoring. The proxy is the ratio of fees and interest on C&I loans to the total amount
of C&I loans.
4.1.5 Loan commitments/total bank assets. As stated by Rajan (1998), banks lend to
clients to maintain their reputation as liquidity providers. By monitoring, they know
Table III.
Two-day standardized
prediction errors
Proxy variables
Number of
observations
Two-day
excess returns
(per cent) t-stats
Percent of
positive excess
returns
Whole sample 423 0.89 3.03*** 54.85
By lender characteristic
Defaulted loan recovery
Recovered charge-Off per cent
Above median 211 0.94 2.01** 56.40
Below median 212 0.83 2.36** 53.30
Difference 0.11 0.19
Past due loan/assets (PDLA)
Above median 210 1.30 2.73*** 55.14
Below median 213 0.48 1.40 55.24
Difference 0.82 1.40
Recovered charge-off per cent ?PDLA
Above median 209 1.11 2.36** 54.55
Below median 214 0.67 1.90* 55.14
Difference 0.44 0.76
Capital ratio
Above median 212 0.81 2.37** 58.02
Below median 211 0.97 2.02** 51.66
Difference ?0.16 ?.27
Uninsured deposits
Above median 211 1.45 2.91*** 57.82
Below median 212 0.33 1.08 51.87
Difference 1.12 1.91*
Loan yield
Loan yield
Above median 212 0.88 2.32** 55.19
Below median 211 0.90 2.00** 54.50
Difference ?0.02 ?.04
Yields ?PDLA
Above median 210 1.29 2.71*** 55.71
Below median 213 0.49 1.43 53.99
Difference 0.80 1.36
LC/Bank assets
Above median 212 0.92 2.28** 56.60
Below median 211 0.85 2.00** 53.08
Difference 0.07 0.12
Notes: The t-statistic is showing signifcance from zero, except for the t-stats for the rows labeled
“Difference”. The t-stats in the rows labeled “Difference” are tests for signifcance between the mean for
the portion of the sample above the mean less the mean of the sample below the median in a given
sample; *signifcant at the 10 per cent level; **signifcant at the 5 per cent level; ***signifcant at the
1 per cent level
105
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
when to not lend to the frm. Thus, banks know when the loss of reputation from not
lending will be less than the loss incurred from a bad loan. Superior monitoring banks
can provide more liquidity through standby letters of credit and loan commitments
while maintaining their reputations. The proxy is the ratio of standby letters and loan
commitments to total bank assets. High ratios provide greater incentive to monitor, so
excess returns on loan announcements should be positively associated with this proxy.
4.2 Control variables
4.2.1 Borrower characteristics. Slovin et al. (1992) report smaller borrowers have greater
abnormal returns, so we include the log of total assets. Billett et al. (1995) and Johnson
(1997) fnd run-ups are negatively associated with abnormal returns. We include
run-ups using the cumulative return over the previous 10 days prior to the loan
announcement. Billett et al. (1995) and Hadlock and James (2002) fnd volatility of the
borrower’s equity returns (stock volatility) is positively associated with excess returns.
We calculate daily volatility for the period from 200 days to 50 day prior to the
announcement. Finally, Hadlock and James (2002) use a public debt dummy (public
debt) as a proxy for the relative contracting costs of bank debt.
4.2.2 Bank size and loan terms. Larger banks suffer from a variety of governance
problems and may have lower incentives to monitor. The coeffcient for bank size (log of
total bank assets) should be negative. We also control for the maturity of the loan and for
the size of the loan relative to the borrower’s equity to capture any effects due to material
changes in the borrower’s leverage.
4.2.3 SIC codes. Recovering 80 per cent on loans in an industry with non-specifc
collateral, i.e. lumber, may not be impressive but the same recovery rate where collateral
is more specifc to the borrower, i.e. stamping equipment for autos, may be
extraordinary. Due to the size of our data set, we control for industry using the frst digit
of the standard industrial classifcation (SIC) code.
5. Results
Table IVreports the results of regressions of the two-day standardized prediction errors on
various subsets of the monitoringproxies andcontrol variables describedinSections 4.1and
4.2. Three strong proxies for bank monitoring and skill emerge fromthe empirical analysis:
uninsured deposits, capital ratio, and recovered charge-off per cent ?past due.
Model (1) examines lender characteristics without any of the control variables.
Capital ratio, recovered charge-off per cent, and recovered ?past due are all signifcant.
The negative coeffcient on recovered charge-off per cent refects the intuition that better
monitors attract riskier loans and greater recovery rates indicate inferior monitoring
skill. However, results for recovered ? past due show a positive coeffcient indicating
that, for a given level of risk, higher recovery rates indicate superior active monitoring
skill, as expected. The positive coeffcient for uninsured deposits is expected but not
signifcant with a p-value of 0.1025. The positive, signifcant sign on capital ratio
indicates the relative value of bank and borrower capital which leads to a greater
incentive to monitor the borrower is valuable to markets.
Loan yield ?past due is signifcant but of the opposite sign of our expectation. loan
commitments (LC)/Bank assets is insignifcant, but its sign is in the expected direction.
For loan yield ? past due, the negative coeffcient indicates, for a given level of risk,
banks with lower yields are associated with higher excess returns and is counter to our
JFEP
6,2
106
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table IV.
Effects of bank
monitoring on borrower
returns
I
n
d
e
p
e
n
d
e
n
t
v
a
r
i
a
b
l
e
s
(
1
)
(
1
a
)
(
2
)
(
2
a
)
(
3
)
B
i
g
(
4
)
S
m
a
l
l
I
n
t
e
r
c
e
p
t
0
.
3
7
8
[
0
.
6
2
]
0
.
2
5
7
[
0
.
3
9
]
0
.
3
1
0
[
0
.
5
1
]
0
.
1
3
7
[
0
.
2
1
]
0
.
3
9
4
[
0
.
2
0
]
0
.
5
6
6
[
0
.
7
9
]
B
a
n
k
s
i
z
e
(
L
o
g
o
f
T
A
)
0
.
0
0
0
[
0
.
0
0
]
0
.
0
0
2
[
0
.
0
9
]
0
.
0
0
9
[
0
.
3
3
]
0
.
0
1
3
[
0
.
4
7
]
0
.
0
0
3
[
0
.
0
7
]
?
0
.
0
0
4
[
?
0
.
1
1
]
M
o
n
i
t
o
r
i
n
g
p
r
o
x
i
e
s
U
n
i
n
s
u
r
e
d
d
e
p
o
s
i
t
s
0
.
2
6
7
*
[
1
.
7
2
]
0
.
2
6
4
*
[
1
.
6
8
]
0
.
2
9
2
*
[
1
.
8
6
]
0
.
2
8
9
*
[
1
.
8
2
]
?
0
.
3
1
3
[
?
0
.
3
9
]
0
.
3
5
9
*
*
[
2
.
1
3
]
C
a
p
i
t
a
l
r
a
t
i
o
0
.
0
4
0
*
*
*
[
2
.
6
2
]
0
.
0
4
0
*
*
*
[
2
.
6
3
]
0
.
0
4
8
*
*
[
2
.
6
3
]
0
.
0
0
1
[
0
.
0
2
]
B
a
n
k
c
a
p
i
t
a
l
r
a
t
i
o
1
.
4
5
4
[
1
.
0
8
]
1
.
6
4
2
[
1
.
1
7
]
B
o
r
r
o
w
e
r
c
a
p
i
t
a
l
r
a
t
i
o
?
0
.
0
5
2
[
?
0
.
2
9
]
0
.
0
0
1
[
0
.
0
0
]
R
e
c
o
v
e
r
e
d
c
h
a
r
g
e
-
o
f
f
p
e
r
c
e
n
t
?
0
.
0
6
3
*
[
?
1
.
7
0
]
?
0
.
0
6
4
*
[
?
1
.
6
8
]
?
0
.
0
5
9
[
?
1
.
5
9
]
?
0
.
0
6
1
[
?
1
.
5
9
]
?
0
.
0
6
7
[
?
0
.
4
1
]
?
0
.
0
6
6
[
?
1
.
5
5
]
R
e
c
o
v
e
r
e
d
?
p
a
s
t
d
u
e
0
.
4
2
5
*
*
[
2
.
2
7
]
0
.
4
2
3
*
*
[
2
.
2
4
]
0
.
4
2
4
*
*
[
2
.
2
7
]
0
.
4
1
8
*
*
[
2
.
2
1
]
3
.
6
0
2
[
0
.
3
6
]
0
.
4
9
7
[
1
.
1
4
]
L
o
a
n
y
i
e
l
d
?
5
.
5
9
4
[
?
1
.
4
1
]
?
5
.
3
9
0
[
?
1
.
3
1
]
?
5
.
1
9
5
[
?
1
.
3
0
]
?
4
.
8
4
2
[
?
1
.
1
8
]
?
1
6
.
0
3
2
[
?
1
.
2
3
]
?
4
.
9
8
1
[
?
1
.
1
2
]
L
o
a
n
y
i
e
l
d
?
p
a
s
t
d
u
e
?
3
7
.
9
0
6
*
*
*
[
?
2
.
8
4
]
?
3
7
.
2
2
6
*
*
*
[
?
2
.
7
5
]
?
3
6
.
8
8
7
*
*
*
[
?
2
.
7
5
]
?
3
5
.
7
0
8
*
*
*
[
?
2
.
6
3
]
?
6
5
.
5
4
2
[
?
0
.
6
1
]
?
4
5
.
2
1
2
[
?
1
.
1
5
]
L
C
/
b
a
n
k
a
s
s
e
t
s
1
3
.
5
2
5
[
0
.
6
5
]
1
5
.
0
8
6
[
0
.
7
2
]
1
4
.
4
1
5
[
0
.
7
0
]
1
6
.
4
7
3
[
0
.
7
9
]
1
1
8
.
9
1
2
[
0
.
7
2
]
1
0
.
7
9
0
[
0
.
5
0
]
B
o
r
r
o
w
e
r
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
L
o
g
(
M
V
e
q
u
i
t
y
)
?
0
.
0
3
8
[
?
1
.
1
6
]
?
0
.
0
4
2
[
?
1
.
2
2
]
?
0
.
0
8
1
[
?
0
.
5
4
]
?
0
.
0
0
3
[
?
0
.
0
5
]
R
u
n
-
u
p
?
1
.
1
0
4
*
*
*
[
?
2
.
7
0
]
?
1
.
1
0
2
*
*
*
[
?
2
.
6
7
]
?
3
.
3
2
9
*
*
*
[
?
2
.
7
4
]
?
0
.
9
3
2
*
*
[
?
2
.
0
8
]
S
t
o
c
k
v
o
l
a
t
i
l
i
t
y
0
.
0
4
2
[
0
.
5
2
]
0
.
0
3
7
[
0
.
4
4
]
1
3
.
7
8
9
[
0
.
9
3
]
0
.
0
4
2
[
0
.
5
0
]
P
u
b
l
i
c
d
e
b
t
0
.
0
4
4
[
0
.
4
1
]
0
.
0
3
3
[
0
.
3
0
]
0
.
3
6
1
[
1
.
3
4
]
?
0
.
0
3
9
[
?
0
.
3
1
]
C
o
n
t
r
a
c
t
t
e
r
m
s
L
o
a
n
s
i
z
e
/
M
V
e
q
u
i
t
y
0
.
0
3
9
[
0
.
7
5
]
0
.
0
4
1
[
0
.
7
5
]
0
.
2
9
6
[
1
.
0
6
]
0
.
0
3
8
[
0
.
6
7
]
M
a
t
u
r
i
t
y
0
.
0
2
3
[
1
.
0
3
]
0
.
0
2
4
[
1
.
0
6
]
0
.
0
0
7
[
0
.
1
3
]
0
.
0
4
0
[
1
.
4
6
]
N
u
m
b
e
r
o
f
a
n
n
o
u
n
c
e
m
e
n
t
s
4
2
3
4
2
3
4
2
3
4
2
3
8
2
3
4
1
R
2
0
.
0
8
6
0
0
.
0
7
3
4
0
.
1
1
0
1
0
.
0
9
7
7
0
.
4
2
9
3
0
.
1
1
1
2
N
o
t
e
s
:
E
a
c
h
r
e
g
r
e
s
s
i
o
n
c
o
n
t
r
o
l
s
f
o
r
t
h
e
b
o
r
r
o
w
e
r
’
s
S
I
C
c
o
d
e
a
t
t
h
e
o
n
e
-
d
i
g
i
t
l
e
v
e
l
b
e
c
a
u
s
e
i
n
d
i
c
a
t
i
n
g
i
n
d
u
s
t
r
y
a
t
t
h
e
t
w
o
-
d
i
g
i
t
S
I
C
l
e
v
e
l
w
o
u
l
d
h
a
v
e
c
r
e
a
t
e
d
m
a
n
y
g
r
o
u
p
s
w
i
t
h
o
n
l
y
o
n
e
,
t
w
o
o
r
t
h
r
e
e
f
r
m
s
.
C
o
n
t
r
o
l
s
a
r
e
a
l
s
o
i
n
c
l
u
d
e
d
f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
t
h
e
l
o
a
n
.
F
o
r
b
r
e
v
i
t
y
,
c
o
e
f
f
c
i
e
n
t
s
f
o
r
S
I
C
a
n
d
l
o
a
n
p
u
r
p
o
s
e
a
r
e
n
o
t
i
n
c
l
u
d
e
d
.
T
h
e
n
u
m
b
e
r
s
i
n
b
r
a
c
k
e
t
s
a
r
e
t
h
e
t
-
s
t
a
t
i
s
t
i
c
s
f
o
r
e
a
c
h
c
o
e
f
f
c
i
e
n
t
;
*
*
*
s
i
g
n
i
f
c
a
n
t
a
t
?
0
.
0
1
;
*
*
s
i
g
n
i
f
c
a
n
t
a
t
?
0
.
0
5
;
*
s
i
g
n
i
f
c
a
n
t
a
t
?
0
.
1
0
107
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
expectation. However, it would be consistent with a relationship lender capturing its
borrower and, thus, capturing a greater share of the surplus[8]. The insignifcant sign on
LC/Bank assets may mean the provision of liquidity is not perceived as an incentive to
monitor, or providing liquidity is not as important as the other, more signifcant factors.
Control variables are added to generate Model (2). The control variables have little
effect, except on uninsured deposits and recovered charge-off per cent. Uninsured
deposits are signifcant and positive. This supports our hypothesis that uninsured
demand depositors improve the bank’s incentive to monitor its borrowers. Recovered
charge-off per cent is no longer signifcant with a t-stat of ?1.63. Among the controls,
run-up is signifcant (z ? 2.67). Consistent with Billett et al. (1995) and Johnson (1997),
run-up is negatively related to excess returns on loan announcements.
Models (1a) and (2a) break apart the capital ratio, so the amount of bank equity is not
scaled by the amount of borrower equity. Neither the bank nor borrower capital ratio is
signifcant, indicating that bank equity capital matters most when the amount of frm
capital is lowand the bank is needed to monitor a borrower with lowincentive for effort.
Models (3) and (4) estimate Model (2) separately for large and small borrowers,
respectively. While the model does not ft the small borrowers as well, uninsured
deposits are signifcant. This is consistent with Slovin et al.’s (1992) notion that smaller
frms are the primary benefciaries of bank monitoring.
Overall, three proxies are strongly associated with greater positive bank loan
announcement returns. The results indicate the specialness of bank loans identifed in
previous literature is, at least in part, due to monitoring skill and incentive. Both
uninsured depositors and bank capital discipline banks by giving them greater
incentives to monitor. Also, the signifcance of recovery rates indicate that through their
ongoing relationships with borrowers, banks can closely monitor them, learn valuable
private information about a borrower’s business and alternative uses for its assets and
provide added value to their borrowers.
6. Conclusion
In this paper, we explore the role of private oversight of banks (i.e. market discipline)
from a unique angle – we revisit the literature on the “specialness” of bank loans. That
“monitoring creates value” is the key insight gleaned from this literature. The
hypothesis tested in this paper is that the market discipline provided by depositors is
crucial to the provision of monitoring which makes banks special.
Motivated by previous research on bank specialness, our approach is to link salient
lender characteristics that refect the lender’s monitoring skill and incentives to bank
loan announcement returns. The specifc proxies for monitoring skill and incentive we
identify are:
• the ratio of the bank’s uninsured deposits to total loans;
• the relative bank-to-borrower capital ratio; and
• a risk-adjusted measure of recovered charge-offs.
We fnd these proxies for a bank’s skill and incentive to monitor borrowers are
associated with higher excess returns surrounding bank loan announcements. This
implies that the excess returns associated with bank loan announcements are in fact a
refection of the monitoring benefts generated by the bank.
JFEP
6,2
108
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The most intriguing relation identifed by our analysis is that there is a link between
the bank’s capital structure – specifcally the proportion of uninsured depositors – and
positive borrower equity returns around bank loan announcements. This has not
previously been documented but it is broadly consistent with models developed by
Calomiris and Kahn (1991) and, especially, Diamond and Rajan (2001). More generally,
we fnd that the market rewards (via excess announcement returns) borrowers who
borrow from lenders who are more likely to provide monitoring benefts.
The evidence that uninsured deposits have an important impact on the “specialness”
of banks is the primary contribution of our paper. This evidence implies that policy
makers and regulators should consider the potential costs of reducing market discipline
of banks. We document that banks were special before the passage of the GLBA when
fewer banks were treated as too-big-to-fail. This suggests that bank regulation needs to
maintain/re-establish some elements of the fragile capital structure that exists at banks
that are not too-big-to-fail. Specifcally, if banks become too safe, they may no longer be
particularly valuable as monitoring lenders. Rather than information-generating
intermediaries, they will merely be transactional institutions.
Notes
1. See, for example, Chan (1983), Diamond (1984, 1991), Ramakrishnan and Thakor (1984), Berlin
and Loeys (1988), Rajan (1992, 1998), Besanko and Kanatas (1993), Rajan and Winton (1995),
Holmstrom and Tirole (1997) and Diamond and Rajan (2001).
2. Exceptions include Billett et al. (1995), Byers et al. (1998) and Marsh (2006). Each utilized the
lender’s credit rating as a proxy for lender quality. While credit ratings are a function of lender
quality and borrower quality, we are focused on the inputs to the credit rating equation, not
the overall level of bank creditworthiness.
3. Gande and Saunders (2012) show banks are special in the production of information and
monitoring, while the special nature of banks is changing in that the sales of loans and
secondary market trading of loans also generate positive borrower returns. Guner (2006) fnds
borrowers whose loans are sold pay lower rates, suggesting that banks that retain and
monitor loans are provide a higher-value service. Ross (2010) fnds more favorable loan
announcements when one of three “dominant” US banks is the lead underwriter due to strong
monitoring reputations.
4. Loans are often unfunded agreements such as revolvers, term loans, working capital
agreements, etc.
5. “Contaminated” means a story appeared within one day on either side of the announcement
and included information about items such as dividends, earnings announcements,
managerial changes, joint ventures, assets sales/purchases or new product announcements.
6. A two-day announcement window is used because wire service announcements may arrive
after the close of trading on day zero. The market model is estimated using 100 trading days
before, t ??120 to ?21, and 100 days after, t ?21 to 120, the announcement date.
7. Financial regulators look to past due loans as an indicator of the quality of a bank’s loan
portfolio. Additionally, Berger et al. (1991) document that the ratio of past due loans to total
assets predicts future bank problems. Therefore, we use this as our measure of bank portfolio
risk. Specifcally, we use a three-year average of this ratio.
109
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
8. An alternative interpretation is that some lenders provide “commodity” funding and compete
on price rather than monitoring services. Ross (2010) fnds that the dominant banks in the
early 2000s capture market share by offering lower-cost funds.
References
Berger, A., Espinosa-Vega, M., Frame, W. and Miller, N. (2005), “Debt maturity, risk and
asymmetric information”, The Journal of Finance, Vol. 60 No. 6, pp. 2895-2923.
Berger, A., King, K. and O’Brien, J. (1991), “The limitations of market value accounting and a more
realistic alternative”, Journal of Banking and Finance, Vol. 15 No. 4, pp. 753-783.
Berlin, M. and Loeys, J. (1988), “Bond covenants and delegated monitoring”, The Journal of
Finance, Vol. 43 No. 2, pp. 397-412.
Bernanke, B. and Gertler, M. (1989), “Agency costs, net worth, and business fuctuations”, The
American Economic Review, Vol. 79 No. 1, pp. 14-31.
Besanko, D. and Kanatas, G. (1993), “Credit market equilibrium with bank monitoring and moral
hazard”, The Review of Financial Studies, Vol. 6 No. 1, pp. 213-232.
Billett, M., Flannery, M. and Garfnkel, J. (1995), “The effect of lender identity on a borrowing
frm’s equity return”, The Journal of Finance, Vol. 50 No. 2, pp. 699-718.
Black, F. (1975), “Bank funds management in an effcient market”, Journal of Financial Economics,
Vol. 2 No. 4, pp. 323-339.
Boot, A. (2000), “Relationship banking: what do we know?”, Journal of Financial Intermediation,
Vol. 9 No. 1, pp. 7-25.
Byers, S., Fraser, D. and Shockley, R. (1998), “Lender identity and borrower returns: the evidence
fromforeign bank loans to U.S. corporations”, Global Finance Journal, Vol. 9 No. 1, pp. 81-94.
Calomiris, C. and Kahn, C. (1991), “The role of demandable debt in structuring optimal banking
arrangements”, The American Economic Review, Vol. 81, pp. 497-513.
Chan, Y. (1983), “On the positive role of fnancial intermediation in allocation of venture capital in
a market with imperfect information”, The Journal of Finance, Vol. 38 No. 5, pp. 1543-1568.
DeYoung, R., Glennon, D. and Nigro, P. (2008), “Borrower-lender distance, credit scoring, and loan
performance: evidence from informational-opaque small business borrowers”, Journal of
Financial Intermediation, Vol. 17 No. 1, pp. 113-143.
Diamond, D. (1984), “Financial intermediation and delegated monitoring”, Review of Economic
Studies, Vol. 51 No. 3, pp. 393-414.
Diamond, D. (1991), “Monitoring and reputation: the choice between bank loans and directly
placed debt”, Journal of Political Economy, Vol. 99, pp. 689-721.
Diamond, D. and Rajan, R. (2000), “Atheory of bank capital”, The Journal of Finance, Vol. 55 No. 6,
pp. 2431-2465.
Diamond, D. and Rajan, R. (2001), “Liquidity risk, liquidity creation and fnancial fragility: a
theory of banking”, Journal of Political Economy, Vol. 109 No. 2, pp. 287-327.
Fama, E. (1985), “What’s different about banks?”, Journal of Monetary Economics, Vol. 15 No. 1,
pp. 29-39.
Fields, L., Fraser, D., Berry, T. and Byers, S. (2006), “Do bank loan relationships still matter?”,
Journal of Money, Credit, and Banking, Vol. 38 No. 5, pp. 1195-1209.
Gande, A. and Saunders, A. (2012), “Are banks still special when there is a secondary market for
loans?”, The Journal of Finance, Vol. 67 No. 5, pp. 1649-1684.
Guner, A. (2006), “Loan sales and the cost of corporate borrowing”, Review of Financial Studies,
Vol. 19 No. 2, pp. 687-716.
JFEP
6,2
110
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hadlock, C. and James, C. (2002), “Do banks provide fnancial slack?”, The Journal of Finance,
Vol. 57 No. 3, pp. 1383-1419.
Holmstrom, B. and Tirole, J. (1997), “Financial intermediation, loanable funds, and the real sector”,
The Quarterly Journal of Economics, Vol. 112 No. 3, pp. 663-691.
James, C. (1987), “Some evidence on the uniqueness of bank loans”, Journal of Financial
Economics, Vol. 19 No. 2, pp. 217-235.
James, C. and Smith, D. (2000), “Are banks still special? newevidence on their role in the corporate
capital-raising process”, Journal of Applied Corporate Finance, Vol. 13 No. 1, pp. 52-63.
Johnson, S. (1997), “The effect of bank reputation on the value of bank loan agreements”, Journal
of Accounting, Auditing, and Finance, Vol. 12 No. 1, pp. 83-100.
Kane, E. (1977), “Good intentions and unintended evil, the case against selective credit allocation”,
Journal of Money, Credit and Banking, Vol. 9, pp. 55-69.
Kane, E. and Malkiel, B. (1965), “Bank portfolio allocation, deposit variability, and the availability
doctrine”, Quarterly Journal of Economics, Vol. 79, pp. 257-261.
Marsh, I. (2006), “The effect of lenders’ credit risk transfer activities on borrowing frms’ equity
returns”, Bank of Finland Research Discussion Papers.
Maskara, P. and Mullineaux, D. (2011), “Information asymmetry and self-selection bias in bank
loan announcement studies”, Journal of Financial Economics, Vol. 101 No. 3, pp. 684-694.
Petersen, M. and Rajan, R. (2002), “Does distance still matter? the information revolution in small
business lending”, The Journal of Finance, Vol. 57 No. 6, pp. 2533-2570.
Rajan, R. (1992), “Insiders and outsiders: the choice between informed and arm’s length debt”, The
Journal of Finance, Vol. 47 No. 4, pp. 1367-1400.
Rajan, R. (1998), “The past and future of commercial banking viewed through an incomplete
contracts lens”, Journal of Money, Credit, and Banking, Vol. 30 No. 3, pp. 524-550.
Rajan, R. and Winton, A. (1995), “Covenants and collateral as incentives to monitor”, The Journal
of Finance, Vol. 50 No. 4, pp. 1113-1146.
Ramakrishnan, S. and Thakor, A. (1984), “Information reliability and a theory of fnancial
intermediation”, Review of Economic Studies, Vol. 51 No. 3, pp. 415-432.
Ross, D. (2010), “The ‘dominant bank effect’: how high lender reputation affects the information
content and terms of bank loans”, Review of Financial Studies, Vol. 23 No. 7, pp. 2730-2756.
Schuermann, T. (2004), “Why were banks better off in the 2001 recession?”, Federal Reserve Bank
of New York, Current Issues in Economics and Finance, Vol. 10 No. 1, pp. 1-10.
Slovin, M., Johnson, S. and Glascock, J. (1992), “Firmsize and the information content of bank loan
announcements”, Journal of Banking and Finance, Vol. 16 No. 6, pp. 1057-1071.
Corresponding author
Bradley A. Stevenson can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
111
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_336556074.pdf
The aim of this paper is to examine the question of the specialness of banks by addressing
concerns raised in the recent studies and deriving policy implications for the future of banking. The
specialness of banks has been well documented in the finance literature. More recent research, however,
calls into question the special nature of banks.
Journal of Financial Economic Policy
Depositor discipline and the bank ' s incentive to monitor
Michael F. Ferguson Bradley A. Stevenson
Article information:
To cite this document:
Michael F. Ferguson Bradley A. Stevenson , (2014),"Depositor discipline and the bank ' s incentive to
monitor", J ournal of Financial Economic Policy, Vol. 6 Iss 2 pp. 98 - 111
Permanent link to this document:
http://dx.doi.org/10.1108/J FEP-06-2013-0022
Downloaded on: 24 January 2016, At: 21:48 (PT)
References: this document contains references to 37 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 110 times since 2014*
Users who downloaded this article also downloaded:
Sheilla Nyasha, Nicholas M Odhiambo, (2014),"Bank-based financial development and economic growth:
A review of international literature", J ournal of Financial Economic Policy, Vol. 6 Iss 2 pp. 112-132 http://
dx.doi.org/10.1108/J FEP-07-2013-0031
Galina Hale, J oão A.C. Santos, (2014),"Do banks propagate debt market shocks?", J ournal of Financial
Economic Policy, Vol. 6 Iss 3 pp. 270-310 http://dx.doi.org/10.1108/J FEP-03-2014-0023
J ill M. Hendrickson, Mark W. Nichols, Daniel R. Fairchild, (2014),"Bank branch location and stability
during distress", J ournal of Financial Economic Policy, Vol. 6 Iss 2 pp. 133-151 http://dx.doi.org/10.1108/
J FEP-07-2013-0026
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Depositor discipline and the
bank’s incentive to monitor
Michael F. Ferguson
Finance and Real Estate Department, University of Cincinnati, Cincinnati,
Ohio, USA, and
Bradley A. Stevenson
Department of Economics and Finance, Bellarmine University, Louisville,
Kentucky, USA
Abstract
Purpose – The aim of this paper is to examine the question of the specialness of banks by addressing
concerns raised in the recent studies and deriving policy implications for the future of banking. The
specialness of banks has been well documented in the fnance literature. More recent research, however,
calls into question the special nature of banks.
Design/methodology/approach – We use event study methodology to study 423 bank loan
announcements from1988 to 1996 and examine the returns relative to proxies for the bank’s monitoring
incentives and skill using ordinary least squares (OLS) regressions.
Findings – Our results indicate borrower abnormal announcement returns are positively related to
proxies for the bank’s monitoring incentives and skill as measured by: the ratio of uninsured deposits to
total loans; a risk-adjusted measure of recovered charge-offs; and the relative bank-to-borrower capital
ratio.
Research limitations/implications – The results reveal how the fragile nature of the bank’s
structure improves the bank’s incentives to monitor borrowers.
Practical implications – Our results can inform the current debates in the Fed and in Congress
surrounding reapplying the Glass-Steagall Act and limiting the size of banks. We showthat banks were
special before the Gramm-Leach-Bliley Act and when fewer banks belonged to the too-big-to-fail
category. This suggests that reregulating banks to re-establish their fragile nature will re-establish
them as information-generating intermediaries instead of just transactional institutions.
Originality/value – Our fndings have not previously been documented but are broadly consistent
with models developed by Calomiris and Kahn (1991) and especially Diamond and Rajan (2001).
Keywords Demand deposits, Bank monitoring, Loan announcement returns
Paper type Research paper
1. Introduction
The mix of private and public oversights of US fnancial institutions, and the effcacy of
private oversight has varied widely over time. Kane (1977) coined the term “regulatory
dialectic” to describe the back and forth between regulators and those trying to evade
their oversight. As part of this ebb and fow, periodic changes are made to the public
oversight component. Two major recent changes in public oversight are the
Gramm-Leach-Bliley repeal of Depression-era restrictions on combining commercial
JEL classifcation – G20, G21, G30, G32
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JFEP
6,2
98
Journal of Financial Economic Policy
Vol. 6 No. 2, 2014
pp. 98-111
© Emerald Group Publishing Limited
1757-6385
DOI 10.1108/JFEP-06-2013-0022
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and investment banking. This was intended to be, and was viewed as, a “deregulatory”
change. After the 2007-2008 fnancial crisis, the Dodd-Frank reforms were passed. This
act was intended to be, and was widely viewed as, a movement toward stricter
regulation.
One way to view regulatory changes is to view them as a change in the relative
importance of public and private oversight of banks. Specifcally, at different times
depositors and capital providers have been relied on to be the primary monitors of banks
(often referred to as “market discipline”). Implicit in a movement toward greater public
oversight of bank safety and soundness is that market oversight is inadequate.
This paper explores the role of private oversight of banks from a unique angle – we
revisit the literature on bank loan “specialness”, which suggests banks that provide the
most value added are those that provide the best monitoring services. Because
monitoring services are provided after making a loan, the provision of monitoring
services is subject to a classic agency problem. Who monitors the bank management to
ensure it actually provides these value-added services?
Calomiris and Kahn (1991) argue uninsured depositors play a monitoring role,
standing ready to “run” the bank in case of mismanagement. Diamond and Rajan (2000,
2001) develop equilibrium models in which the bank’s capital structure is central to the
provision of specialized monitoring services. Specifcally, uninsured depositors
withdraw capital whenever the return from loan monitoring is too low. This ensures
managers expend resources to monitor loans. Thus, a “fragile” bank capital structure
(i.e. relying on very short-term, very liquid debt) subject to runs resolves the lender’s
agency problem.
We document market reaction to bank loan announcements consistent with the
notion that uninsured bank depositors play an important monitoring role. In a sample of
bank loans that have positive average announcement returns (i.e. they appear to be
“special”), the announcement returns are increasing in the proportion of uninsured
deposits. This suggests that the provision of the lender’s special services is dependent
upon the availability of market discipline. This has interesting implications for
regulatory reform. If bank liabilities (i.e. deposits) are too safe, the specialness of banks
may be compromised. To the extent policy makers view credit analysis and loan
origination as welfare-enhancing activities, they should consider costs associated with
reduced depositor discipline as they craft legislation and new regulations.
We have chosen to take a moderately “historical” look at bank loan markets by
examining data from prior to the passage of the Gramm-Leach-Bliley Act (GLBA). Our
reason is there have been a number of changes made recently to the regulatory
environment that have potentially signifcantly reduced the specialness of banks. Also,
a number of current proposals imply or explicitly propose undoing many of the changes
made by GLBA. So to better inform policy makers, it is useful to improve our
understanding of bank lending prior to the passage of GLBA.
What changes affected the nature of banks? First, with the passage of GLBAin 1999,
also called the Financial Services Modernization Act, banks were allowed to act as both
commercial and investment banks. On one hand, this change might allow for better
screening and monitoring due to information synergies between the two banking
functions. Alternatively, combining commercial and investment banking allowed banks
to become bigger and more interconnected which exacerbates the too-big-to-fail (TBTF)
problem. Increasing the TBTF problem impairs the special nature of banks (if their
99
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
specialness is due to their fragile nature, as described by Diamond and Rajan (2001))
because they are no longer “fragile”.
Second, even prior to GLBA, banks were growing larger and spreading
geographically due to deregulation of intrastate and interstate expansion. Banks that
serve borrowers farther away gain greater geographic diversity in their portfolios.
While this may improve effciency and diversify portfolios, it may result in banks
lending to more distant, less familiar (i.e. more opaque) borrowers. Petersen and Rajan
(2002) note that banks tend to lend to closer more opaque borrowers than non-banks.
Furthermore, DeYoung et al. (2008) show that increased distance between bank lenders
and borrowers and the increased use of credit scoring has led to increased default rates
for small business borrowers. Thus, GLBAhas facilitated the recent trend toward banks
behaving more like non-bank lenders.
Finally, Basel II and the introduction of small business scoring have increased the use
of more analytical, less relationship-based methods to judge credit risk. Berger et al.
(2005) suggest that small business credit scoring systems (SBCSS) reduce information
asymmetry because loans made with SBCSS tend to have longer maturities indicating
less information asymmetry.
Our hypothesis is that the market discipline provided by depositors is crucial to the
provision of bank monitoring. What is bank monitoring? The theoretical literature
draws a vivid picture. A common theme dating back to Kane and Malkiel (1965) and
Black (1975) is banks monitor some aspects of their borrowers’ activities. Subsequent
research suggests, relative to other lenders, banks have superior monitoring
technologies and more detailed knowledge of a borrower’s investment opportunities,
infuence over project choice and, importantly, infuence over effort choice[1]. Besanko
and Kanatas (1993) describe bank monitoring as an active process of tracking the effort
exerted by the entrepreneur via long-term relationships where the bank can monitor the
frm’s collateral balances, fnancial transactions and agent effort. In this environment,
banks have a direct infuence on borrower decisions, which is consistent with the view
of banks as relationship lenders. Boot (2000) defnes relationship lending as repeated
interactions over time with the same borrower that allows the bank to develop
frm-specifc, private information. As Rajan (1992) observes, this intimate knowledge of
a borrower’s projects can be so extensive as to give the lender a decided bargaining
advantage over the borrower. In short, monitoring is an active, ongoing, intrusive and,
ultimately, expensive process that requires skill and proper incentives.
The proxies for monitoring skill and incentive we identify are:
• the ratio of the bank’s uninsured deposits to total loans;
• the relative bank-to-borrower capital ratio; and
• a risk-adjusted measure of recovered charge-offs.
These proxies for a bank’s skill and incentive to monitor borrowers are associated with
higher excess returns surrounding bank loan announcements. This implies the excess
returns associated with bank loan announcements are a refection of the monitoring
benefts generated by the bank. That “monitoring creates value” is the key insight
gleaned from the strand of the empirical “banks-are-special” literature that began with
Fama (1985) and James (1987). Evidence that uninsured deposits have an important
impact on the “specialness” of banks is the primary contribution of our paper. This
JFEP
6,2
100
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
evidence implies that policy makers and regulators should consider the potential costs
of reducing market discipline of banks. Specifcally, if banks become too safe they may
no longer be particularly valuable as monitoring lenders.
The rest of the paper is as follows. Section 2 reviews the literature on the special
nature of banks. Section 3 describes the data used for the loan announcement and excess
return analysis. Section 4 details empirical proxies for monitoring. Section 5 examines
the relationship between those proxies and bank loan announcement abnormal returns.
Section 6 concludes.
2. Bank specialness
James and Smith (2000) summarize the literature on the “specialness” of bank loans:
In general, these studies suggest that private lenders add the greatest value for so-called
‘informationally-intensive’ borrowers – borrowers who face the greatest potential information
problems when issuing public securities […] The fact that bank loans are collateralized,
contain strict covenants, and are typically short term improves a bank’s skill to monitor
informationally-intensive loans.
Thus, the existing literature tells us about differences in borrowers or about differences
in loan contracts. This literature confrms Fama’s initial observation that bank loans are
“special” and banks provide value to their borrowers they cannot get elsewhere from
specialized “monitoring” services. Notably, previous research is relatively silent on
lender characteristics[2].
James (1987) and others provide evidence that banks are special, but this specialness
has been contested in recent years. One argument is that banks are no longer special.
Another, stronger argument is banks never were special.
The frst argument comes from Fields et al. (2006). They fnd loan announcement
effects diminish over time, disappearing toward the end of their sample (from 1980 to
2003). They attribute this to changes in the information environment (Petersen and
Rajan, 2002), increased use of risk management techniques such as credit scoring
(Schuermann, 2004) and increased capital standards. Also, they note the small impact of
the 2001 recession on lending activity. Several recent papers argue bank loans continue
to be special[3]. Nonetheless, we are concerned about the potentially changing nature of
the special role of banks. Therefore, we conduct our analysis on a pre-GLBA sample of
bank loan announcements.
More fundamentally, Maskara and Mullineaux (2011) claim the loan announcement
effects in earlier studies never existed because the sample of loans in studies like Billett
et al. (1995) do not represent the universe of borrowers because the announced loans are
subject to self-selection bias. The fear is that the sample is biased by announcements
characterized by disproportionately positive excess returns because borrowers only
announce “good news”. They fnd the abnormal positive returns for the period examined
by Billet et al. (1995) are actually zero in a data set including announced and
unannounced loans. They acknowledge for the smallest frms there are signifcant
abnormal returns.
Because we identify loan announcements via media reports, we are concerned about
the self-selection bias described above. However, our sample of announcements from
1988 to 1996 exhibits rich cross-sectional variation and is not dominated by positive,
“good news” announcements. The maximumand minimumexcess returns are 47.63 per
cent and ?15.44 per cent for a range of 63.07 per cent with a mean excess return of 0.89
101
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
per cent; 54.85 per cent of the announcements are positive. The inter-quartile range is
4.46 per cent (high ? 2.93 per cent; low ? ?1.53 per cent). Figure 1 plots the
standardized excess returns.
As our goal is to relate variation in bank loan announcement returns to lender
characteristics, the cross-sectional variation is more important than the mean. Figure 1
and the range statistics suggest substantial cross-sectional variation; moreover, the
sample is balanced between positive and negative returns. We feel confdent our sample
is not signifcantly affected by a sample selection bias.
3. Data
3.1. Excess returns
We identify 1,065 loan announcements by commercial banks to non-fnancial frms
appearing on the Reuters wire, DowJones Newswires or in The Wall Street Journal from
1988 to 1996[4]. Articles were selected by searching for the terms “line of credit”, “credit
line”, “credit facility”, “credit agreement”, “credit extension”, “new loan”, “loan
agreement”, “loan renewal”, “loan revision”, “loan extension”, “fnance company loan”,
“termloan”, “commercial loan”, “bank loan”, “loan facility” or “working capital facility”.
We collected the borrowing frm’s name, announcement date, the lead lender, lenders
in the syndicate, initiation or renewal, facility amount, previous agreement amount,
maturity, purpose, changes to interest rates and covenants and loan type. Loans were
dropped due to unidentifed lender, confounding news events, lack of data from center
for research in securities prices (CRSP) (returns) and COMPUSTAT (lender and
borrower data) and identifcation as a real estate investment trust (REIT) or fnancial
frm[5]. Removing these observations leaves a clean sample of 423 loan announcements.
The descriptive statistics appear in Table I..
0
10
20
30
40
50
60
70
80
90
<
-
2
0
.
5
%
-
2
0
.
5
%
t
o
-
1
9
.
5
%
-
1
9
.
5
%
t
o
-
1
8
.
5
%
-
1
8
.
5
%
t
o
-
1
7
.
5
%
-
1
7
.
5
%
t
o
-
1
6
.
5
%
-
1
6
.
5
%
t
o
-
1
5
.
5
%
-
1
5
.
5
%
t
o
-
1
4
.
5
%
-
1
4
.
5
%
t
o
-
1
3
.
5
%
-
1
3
.
5
%
t
o
-
1
2
.
5
%
-
1
2
.
5
%
t
o
-
1
1
.
5
%
-
1
1
.
5
%
t
o
-
1
0
.
5
%
-
1
0
.
5
%
t
o
-
9
.
5
%
-
9
.
5
%
t
o
-
8
.
5
%
-
8
.
5
%
t
o
-
7
.
5
%
-
7
.
5
%
t
o
-
6
.
5
%
-
6
.
5
%
t
o
-
5
.
5
%
-
5
.
5
%
t
o
-
4
.
5
%
-
4
.
5
%
t
o
-
3
.
5
%
-
3
.
5
%
t
o
-
2
.
5
%
-
2
.
5
%
t
o
-
1
.
5
%
-
1
.
5
%
t
o
-
0
.
5
%
-
0
.
5
%
t
o
0
.
5
%
0
.
5
%
t
o
1
.
5
%
1
.
5
%
t
o
2
.
5
%
2
.
5
%
t
o
3
.
5
%
3
.
5
%
t
o
4
.
5
%
4
.
5
%
t
o
5
.
5
%
5
.
5
%
t
o
6
.
5
%
6
.
5
%
t
o
7
.
5
%
7
.
5
%
t
o
8
.
5
%
8
.
5
%
t
o
9
.
5
%
9
.
5
%
t
o
1
0
.
5
%
1
0
.
5
%
t
o
1
1
.
5
%
1
1
.
5
%
t
o
1
2
.
5
%
1
2
.
5
%
t
o
1
3
.
5
%
1
3
.
5
%
t
o
1
4
.
5
%
1
4
.
5
%
t
o
1
5
.
5
%
1
5
.
5
%
t
o
1
6
.
5
%
1
6
.
5
%
t
o
1
7
.
5
%
.
1
7
5
t
o
.
1
8
5
1
8
.
5
%
t
o
1
9
.
5
%
1
9
.
5
%
t
o
2
0
.
5
%
>
2
0
.
5
%
N
u
m
b
e
r
o
f
P
r
e
d
i
c
?
o
n
E
r
r
o
r
s
p
e
r
C
a
t
e
g
o
r
y
Two-Day Predic?on Errors
Figure 1.
Prediction error
distribution
JFEP
6,2
102
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.2. Bank monitoring proxies
Lender data collected fromcall reports included total loans, total assets, total C&I loans,
fees and interest on C&I loans, owner’s equity, total deposits, deposits over $100,000,
deposits under $100,000, amount of loan commitments, amount of non-accrual and past
due loans and the amount of predicted loan losses. Table II contains proxy
characteristics from the call report data.
If a commercial bank and a foreign or investment bank are mentioned as the lead
lender or arranger, the domestic commercial bank is treated as the lead lender. If one
bank is identifed as the lead lender, data are collected for it only. If multiple banks are
mentioned but there is no indication of rank, data are collected for all lenders.
Table III reports two-day standardized prediction errors for the full sample and
subsamples and illustrates loan announcements are associated with positive excess
returns. The average abnormal return (prediction error (PE) ? 0.89 per cent) is larger
than reported in most previous studies and it is signifcantly different from zero. In all,
54.85 per cent of the sample has a positive announcement effect[6].
The univariate analysis shows, except for loan yield and capital ratio, the abnormal
returns for bank loan announcements are greater when the lender characteristic is above
the median value. However, the only signifcant difference is for uninsured deposits.
4. Monitoring proxies and control variables
4.1 Proxies
4.1.1 Uninsured deposits. Diamond and Rajan (2001) and Calomiris and Kahn (1991)
emphasize the monitoring role uninsured depositors play in disciplining management.
To account for bank size, we divide total uninsured deposits by total loans. Banks with
lowratios face less discipline fromdepositors, while high ratios imply greater discipline
Table I.
Characteristics of bank
loan announcements
Loan and bank characteristics
N ?423; N ?99 with public debt
Mean Maximum Minimum
Loan amount ($ millions) 98.7 (40.0) 2050.0 0.5
Per cent with multiple lead banks 47.75 per cent
Borrower size (USD millions) 460.9 (131.2) 8567.3 1.6
Loan/MV of equity 0.78 (0.50) 8.2 0.02
Loan maturity (in years) 3.2 (3.0) 10.0 0.1
10-day cumulative stock return 1.09 per cent (0.0 per cent) 69.697 per cent ?41.67 per cent
Stock return volatility 7.84 per cent (3.59 per cent) 811.08 per cent 0.99 per cent
Notes: The Loan amount includes the amount drawn down immediately as well as the unused portion
of the line. Borrower size is defned as the sum of the book value of total liabilities, liquidating value of
preferred stock and the market value of common stock as of end of the year prior to the announcement.
Loan/MV equity is the dollar amount of the loan divided by the market value of the borrower’s assets.
In all, 217 of 423 announcements did not provide the Loan maturity. Firms are deemed to have Public
debt outstanding if COMPUSTATreports a debt rating at the end of the year prior to the announcement
or if COMPUSTAT reports a non-zero balance for public debt for the prior year. The 10-day cumulative
stock return is measured in the ten days preceding the announcement, and the stock return volatility is
measured from day ?200 to day ?51 before the announcement. Means are reported with medians in
parentheses
103
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and more intense monitoring. Thus, higher ratios should translate into more positive
announcement effects.
4.1.2 Capital ratio. As reported by Besanko and Kanatas (1993), entrepreneurs exert
effort related to their stake in the frm. Outside funding decreases effort and decreases
the probability of project success. Bank monitoring can substitute for the incentive
provided by owner’s equity. In their model, moral hazard arises because banks cannot
commit to a level of monitoring. Holmstrom and Tirole (1997) develop a related model
where banks substitute their own capital (through the incentive to monitor) for the
borrower’s capital. Our proxy is the ratio of total bank equity to total bank assets
divided by the ratio of total borrower equity to total borrower assets. Higher values
indicate increased incentive for the bank to monitor relative to the borrower’s incentive.
Therefore, larger values should be associated with greater excess returns.
4.1.3 Recovered charge-off per cent. Diamond and Rajan (2001) argue the bank’s
advantage in relationship lending is, via monitoring, acquiring specialized knowledge of
the borrower’s assets, creating an advantage in liquidating those assets. Thus, better
monitors should recover more of charged-off loans. We proxy this with the ratio of
recovered loans to the total amount of charged-off loans. This measures the bank’s
effectiveness in collecting on defaulted loans, where the higher percentage collected, the
greater the bank’s specialized knowledge.
However, better monitors may attract riskier loans. A bank that is better in an
absolute sense at liquidating assets may have a lower recovered charge-off per cent. To
control for this, we adjust for the bank’s loan portfolio risk, interacting the per cent
recovered with a proxy for risk measured by the bank’s past due loans[7]. After
controlling for bank risk, recovered charge-offs will be positively related to abnormal
returns such that, for a given risk level, better monitors will recover more.
4.1.4 Loan yield. Bernanke and Gertler (1989) develop an optimal auditing model of
bank monitoring where higher repayment (i.e. yield) for riskier investments lowers
borrower consumption in the good state (where repayment occurs). This reduces the
potential opportunity cost of falsely claiming the bad state has occurred which increases
the moral hazard of the borrower and the need for costly bank monitoring. Controlling
for bank loan portfolio risk, as in recovered charge-off per cent, the average yield earned
by the lender should proxy for the monitoring performed in its lending relationships.
The coeffcient should be positive. For a given level of risk, higher yields equate to more
Table II.
Proxies for bank
monitoring
Bank monitoring proxies
Full sample N ?423
Mean Maximum Minimum
Uninsured deposits 0.38801 (0.34350) 3.53115 0.017486
Capital ratio 0.37555 (0.1975) 49.8102 ?25.1106
Loan yield 0.053693 (0.052055) 0.21053 0.026946
Recovered charge-off per cent 0.63642 (0.32895) 14.7731 0
Past due loans/assets 0.032721 (0.011005) 2.00868 0
LC/bank assets 0.000706 (0.0000007) 0.0244832 0
Loan loss provision 0.025992 (0.022947) 0.11799 0
Note: Median values are reported in parentheses
JFEP
6,2
104
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
monitoring. The proxy is the ratio of fees and interest on C&I loans to the total amount
of C&I loans.
4.1.5 Loan commitments/total bank assets. As stated by Rajan (1998), banks lend to
clients to maintain their reputation as liquidity providers. By monitoring, they know
Table III.
Two-day standardized
prediction errors
Proxy variables
Number of
observations
Two-day
excess returns
(per cent) t-stats
Percent of
positive excess
returns
Whole sample 423 0.89 3.03*** 54.85
By lender characteristic
Defaulted loan recovery
Recovered charge-Off per cent
Above median 211 0.94 2.01** 56.40
Below median 212 0.83 2.36** 53.30
Difference 0.11 0.19
Past due loan/assets (PDLA)
Above median 210 1.30 2.73*** 55.14
Below median 213 0.48 1.40 55.24
Difference 0.82 1.40
Recovered charge-off per cent ?PDLA
Above median 209 1.11 2.36** 54.55
Below median 214 0.67 1.90* 55.14
Difference 0.44 0.76
Capital ratio
Above median 212 0.81 2.37** 58.02
Below median 211 0.97 2.02** 51.66
Difference ?0.16 ?.27
Uninsured deposits
Above median 211 1.45 2.91*** 57.82
Below median 212 0.33 1.08 51.87
Difference 1.12 1.91*
Loan yield
Loan yield
Above median 212 0.88 2.32** 55.19
Below median 211 0.90 2.00** 54.50
Difference ?0.02 ?.04
Yields ?PDLA
Above median 210 1.29 2.71*** 55.71
Below median 213 0.49 1.43 53.99
Difference 0.80 1.36
LC/Bank assets
Above median 212 0.92 2.28** 56.60
Below median 211 0.85 2.00** 53.08
Difference 0.07 0.12
Notes: The t-statistic is showing signifcance from zero, except for the t-stats for the rows labeled
“Difference”. The t-stats in the rows labeled “Difference” are tests for signifcance between the mean for
the portion of the sample above the mean less the mean of the sample below the median in a given
sample; *signifcant at the 10 per cent level; **signifcant at the 5 per cent level; ***signifcant at the
1 per cent level
105
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
when to not lend to the frm. Thus, banks know when the loss of reputation from not
lending will be less than the loss incurred from a bad loan. Superior monitoring banks
can provide more liquidity through standby letters of credit and loan commitments
while maintaining their reputations. The proxy is the ratio of standby letters and loan
commitments to total bank assets. High ratios provide greater incentive to monitor, so
excess returns on loan announcements should be positively associated with this proxy.
4.2 Control variables
4.2.1 Borrower characteristics. Slovin et al. (1992) report smaller borrowers have greater
abnormal returns, so we include the log of total assets. Billett et al. (1995) and Johnson
(1997) fnd run-ups are negatively associated with abnormal returns. We include
run-ups using the cumulative return over the previous 10 days prior to the loan
announcement. Billett et al. (1995) and Hadlock and James (2002) fnd volatility of the
borrower’s equity returns (stock volatility) is positively associated with excess returns.
We calculate daily volatility for the period from 200 days to 50 day prior to the
announcement. Finally, Hadlock and James (2002) use a public debt dummy (public
debt) as a proxy for the relative contracting costs of bank debt.
4.2.2 Bank size and loan terms. Larger banks suffer from a variety of governance
problems and may have lower incentives to monitor. The coeffcient for bank size (log of
total bank assets) should be negative. We also control for the maturity of the loan and for
the size of the loan relative to the borrower’s equity to capture any effects due to material
changes in the borrower’s leverage.
4.2.3 SIC codes. Recovering 80 per cent on loans in an industry with non-specifc
collateral, i.e. lumber, may not be impressive but the same recovery rate where collateral
is more specifc to the borrower, i.e. stamping equipment for autos, may be
extraordinary. Due to the size of our data set, we control for industry using the frst digit
of the standard industrial classifcation (SIC) code.
5. Results
Table IVreports the results of regressions of the two-day standardized prediction errors on
various subsets of the monitoringproxies andcontrol variables describedinSections 4.1and
4.2. Three strong proxies for bank monitoring and skill emerge fromthe empirical analysis:
uninsured deposits, capital ratio, and recovered charge-off per cent ?past due.
Model (1) examines lender characteristics without any of the control variables.
Capital ratio, recovered charge-off per cent, and recovered ?past due are all signifcant.
The negative coeffcient on recovered charge-off per cent refects the intuition that better
monitors attract riskier loans and greater recovery rates indicate inferior monitoring
skill. However, results for recovered ? past due show a positive coeffcient indicating
that, for a given level of risk, higher recovery rates indicate superior active monitoring
skill, as expected. The positive coeffcient for uninsured deposits is expected but not
signifcant with a p-value of 0.1025. The positive, signifcant sign on capital ratio
indicates the relative value of bank and borrower capital which leads to a greater
incentive to monitor the borrower is valuable to markets.
Loan yield ?past due is signifcant but of the opposite sign of our expectation. loan
commitments (LC)/Bank assets is insignifcant, but its sign is in the expected direction.
For loan yield ? past due, the negative coeffcient indicates, for a given level of risk,
banks with lower yields are associated with higher excess returns and is counter to our
JFEP
6,2
106
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table IV.
Effects of bank
monitoring on borrower
returns
I
n
d
e
p
e
n
d
e
n
t
v
a
r
i
a
b
l
e
s
(
1
)
(
1
a
)
(
2
)
(
2
a
)
(
3
)
B
i
g
(
4
)
S
m
a
l
l
I
n
t
e
r
c
e
p
t
0
.
3
7
8
[
0
.
6
2
]
0
.
2
5
7
[
0
.
3
9
]
0
.
3
1
0
[
0
.
5
1
]
0
.
1
3
7
[
0
.
2
1
]
0
.
3
9
4
[
0
.
2
0
]
0
.
5
6
6
[
0
.
7
9
]
B
a
n
k
s
i
z
e
(
L
o
g
o
f
T
A
)
0
.
0
0
0
[
0
.
0
0
]
0
.
0
0
2
[
0
.
0
9
]
0
.
0
0
9
[
0
.
3
3
]
0
.
0
1
3
[
0
.
4
7
]
0
.
0
0
3
[
0
.
0
7
]
?
0
.
0
0
4
[
?
0
.
1
1
]
M
o
n
i
t
o
r
i
n
g
p
r
o
x
i
e
s
U
n
i
n
s
u
r
e
d
d
e
p
o
s
i
t
s
0
.
2
6
7
*
[
1
.
7
2
]
0
.
2
6
4
*
[
1
.
6
8
]
0
.
2
9
2
*
[
1
.
8
6
]
0
.
2
8
9
*
[
1
.
8
2
]
?
0
.
3
1
3
[
?
0
.
3
9
]
0
.
3
5
9
*
*
[
2
.
1
3
]
C
a
p
i
t
a
l
r
a
t
i
o
0
.
0
4
0
*
*
*
[
2
.
6
2
]
0
.
0
4
0
*
*
*
[
2
.
6
3
]
0
.
0
4
8
*
*
[
2
.
6
3
]
0
.
0
0
1
[
0
.
0
2
]
B
a
n
k
c
a
p
i
t
a
l
r
a
t
i
o
1
.
4
5
4
[
1
.
0
8
]
1
.
6
4
2
[
1
.
1
7
]
B
o
r
r
o
w
e
r
c
a
p
i
t
a
l
r
a
t
i
o
?
0
.
0
5
2
[
?
0
.
2
9
]
0
.
0
0
1
[
0
.
0
0
]
R
e
c
o
v
e
r
e
d
c
h
a
r
g
e
-
o
f
f
p
e
r
c
e
n
t
?
0
.
0
6
3
*
[
?
1
.
7
0
]
?
0
.
0
6
4
*
[
?
1
.
6
8
]
?
0
.
0
5
9
[
?
1
.
5
9
]
?
0
.
0
6
1
[
?
1
.
5
9
]
?
0
.
0
6
7
[
?
0
.
4
1
]
?
0
.
0
6
6
[
?
1
.
5
5
]
R
e
c
o
v
e
r
e
d
?
p
a
s
t
d
u
e
0
.
4
2
5
*
*
[
2
.
2
7
]
0
.
4
2
3
*
*
[
2
.
2
4
]
0
.
4
2
4
*
*
[
2
.
2
7
]
0
.
4
1
8
*
*
[
2
.
2
1
]
3
.
6
0
2
[
0
.
3
6
]
0
.
4
9
7
[
1
.
1
4
]
L
o
a
n
y
i
e
l
d
?
5
.
5
9
4
[
?
1
.
4
1
]
?
5
.
3
9
0
[
?
1
.
3
1
]
?
5
.
1
9
5
[
?
1
.
3
0
]
?
4
.
8
4
2
[
?
1
.
1
8
]
?
1
6
.
0
3
2
[
?
1
.
2
3
]
?
4
.
9
8
1
[
?
1
.
1
2
]
L
o
a
n
y
i
e
l
d
?
p
a
s
t
d
u
e
?
3
7
.
9
0
6
*
*
*
[
?
2
.
8
4
]
?
3
7
.
2
2
6
*
*
*
[
?
2
.
7
5
]
?
3
6
.
8
8
7
*
*
*
[
?
2
.
7
5
]
?
3
5
.
7
0
8
*
*
*
[
?
2
.
6
3
]
?
6
5
.
5
4
2
[
?
0
.
6
1
]
?
4
5
.
2
1
2
[
?
1
.
1
5
]
L
C
/
b
a
n
k
a
s
s
e
t
s
1
3
.
5
2
5
[
0
.
6
5
]
1
5
.
0
8
6
[
0
.
7
2
]
1
4
.
4
1
5
[
0
.
7
0
]
1
6
.
4
7
3
[
0
.
7
9
]
1
1
8
.
9
1
2
[
0
.
7
2
]
1
0
.
7
9
0
[
0
.
5
0
]
B
o
r
r
o
w
e
r
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
L
o
g
(
M
V
e
q
u
i
t
y
)
?
0
.
0
3
8
[
?
1
.
1
6
]
?
0
.
0
4
2
[
?
1
.
2
2
]
?
0
.
0
8
1
[
?
0
.
5
4
]
?
0
.
0
0
3
[
?
0
.
0
5
]
R
u
n
-
u
p
?
1
.
1
0
4
*
*
*
[
?
2
.
7
0
]
?
1
.
1
0
2
*
*
*
[
?
2
.
6
7
]
?
3
.
3
2
9
*
*
*
[
?
2
.
7
4
]
?
0
.
9
3
2
*
*
[
?
2
.
0
8
]
S
t
o
c
k
v
o
l
a
t
i
l
i
t
y
0
.
0
4
2
[
0
.
5
2
]
0
.
0
3
7
[
0
.
4
4
]
1
3
.
7
8
9
[
0
.
9
3
]
0
.
0
4
2
[
0
.
5
0
]
P
u
b
l
i
c
d
e
b
t
0
.
0
4
4
[
0
.
4
1
]
0
.
0
3
3
[
0
.
3
0
]
0
.
3
6
1
[
1
.
3
4
]
?
0
.
0
3
9
[
?
0
.
3
1
]
C
o
n
t
r
a
c
t
t
e
r
m
s
L
o
a
n
s
i
z
e
/
M
V
e
q
u
i
t
y
0
.
0
3
9
[
0
.
7
5
]
0
.
0
4
1
[
0
.
7
5
]
0
.
2
9
6
[
1
.
0
6
]
0
.
0
3
8
[
0
.
6
7
]
M
a
t
u
r
i
t
y
0
.
0
2
3
[
1
.
0
3
]
0
.
0
2
4
[
1
.
0
6
]
0
.
0
0
7
[
0
.
1
3
]
0
.
0
4
0
[
1
.
4
6
]
N
u
m
b
e
r
o
f
a
n
n
o
u
n
c
e
m
e
n
t
s
4
2
3
4
2
3
4
2
3
4
2
3
8
2
3
4
1
R
2
0
.
0
8
6
0
0
.
0
7
3
4
0
.
1
1
0
1
0
.
0
9
7
7
0
.
4
2
9
3
0
.
1
1
1
2
N
o
t
e
s
:
E
a
c
h
r
e
g
r
e
s
s
i
o
n
c
o
n
t
r
o
l
s
f
o
r
t
h
e
b
o
r
r
o
w
e
r
’
s
S
I
C
c
o
d
e
a
t
t
h
e
o
n
e
-
d
i
g
i
t
l
e
v
e
l
b
e
c
a
u
s
e
i
n
d
i
c
a
t
i
n
g
i
n
d
u
s
t
r
y
a
t
t
h
e
t
w
o
-
d
i
g
i
t
S
I
C
l
e
v
e
l
w
o
u
l
d
h
a
v
e
c
r
e
a
t
e
d
m
a
n
y
g
r
o
u
p
s
w
i
t
h
o
n
l
y
o
n
e
,
t
w
o
o
r
t
h
r
e
e
f
r
m
s
.
C
o
n
t
r
o
l
s
a
r
e
a
l
s
o
i
n
c
l
u
d
e
d
f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
t
h
e
l
o
a
n
.
F
o
r
b
r
e
v
i
t
y
,
c
o
e
f
f
c
i
e
n
t
s
f
o
r
S
I
C
a
n
d
l
o
a
n
p
u
r
p
o
s
e
a
r
e
n
o
t
i
n
c
l
u
d
e
d
.
T
h
e
n
u
m
b
e
r
s
i
n
b
r
a
c
k
e
t
s
a
r
e
t
h
e
t
-
s
t
a
t
i
s
t
i
c
s
f
o
r
e
a
c
h
c
o
e
f
f
c
i
e
n
t
;
*
*
*
s
i
g
n
i
f
c
a
n
t
a
t
?
0
.
0
1
;
*
*
s
i
g
n
i
f
c
a
n
t
a
t
?
0
.
0
5
;
*
s
i
g
n
i
f
c
a
n
t
a
t
?
0
.
1
0
107
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
expectation. However, it would be consistent with a relationship lender capturing its
borrower and, thus, capturing a greater share of the surplus[8]. The insignifcant sign on
LC/Bank assets may mean the provision of liquidity is not perceived as an incentive to
monitor, or providing liquidity is not as important as the other, more signifcant factors.
Control variables are added to generate Model (2). The control variables have little
effect, except on uninsured deposits and recovered charge-off per cent. Uninsured
deposits are signifcant and positive. This supports our hypothesis that uninsured
demand depositors improve the bank’s incentive to monitor its borrowers. Recovered
charge-off per cent is no longer signifcant with a t-stat of ?1.63. Among the controls,
run-up is signifcant (z ? 2.67). Consistent with Billett et al. (1995) and Johnson (1997),
run-up is negatively related to excess returns on loan announcements.
Models (1a) and (2a) break apart the capital ratio, so the amount of bank equity is not
scaled by the amount of borrower equity. Neither the bank nor borrower capital ratio is
signifcant, indicating that bank equity capital matters most when the amount of frm
capital is lowand the bank is needed to monitor a borrower with lowincentive for effort.
Models (3) and (4) estimate Model (2) separately for large and small borrowers,
respectively. While the model does not ft the small borrowers as well, uninsured
deposits are signifcant. This is consistent with Slovin et al.’s (1992) notion that smaller
frms are the primary benefciaries of bank monitoring.
Overall, three proxies are strongly associated with greater positive bank loan
announcement returns. The results indicate the specialness of bank loans identifed in
previous literature is, at least in part, due to monitoring skill and incentive. Both
uninsured depositors and bank capital discipline banks by giving them greater
incentives to monitor. Also, the signifcance of recovery rates indicate that through their
ongoing relationships with borrowers, banks can closely monitor them, learn valuable
private information about a borrower’s business and alternative uses for its assets and
provide added value to their borrowers.
6. Conclusion
In this paper, we explore the role of private oversight of banks (i.e. market discipline)
from a unique angle – we revisit the literature on the “specialness” of bank loans. That
“monitoring creates value” is the key insight gleaned from this literature. The
hypothesis tested in this paper is that the market discipline provided by depositors is
crucial to the provision of monitoring which makes banks special.
Motivated by previous research on bank specialness, our approach is to link salient
lender characteristics that refect the lender’s monitoring skill and incentives to bank
loan announcement returns. The specifc proxies for monitoring skill and incentive we
identify are:
• the ratio of the bank’s uninsured deposits to total loans;
• the relative bank-to-borrower capital ratio; and
• a risk-adjusted measure of recovered charge-offs.
We fnd these proxies for a bank’s skill and incentive to monitor borrowers are
associated with higher excess returns surrounding bank loan announcements. This
implies that the excess returns associated with bank loan announcements are in fact a
refection of the monitoring benefts generated by the bank.
JFEP
6,2
108
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The most intriguing relation identifed by our analysis is that there is a link between
the bank’s capital structure – specifcally the proportion of uninsured depositors – and
positive borrower equity returns around bank loan announcements. This has not
previously been documented but it is broadly consistent with models developed by
Calomiris and Kahn (1991) and, especially, Diamond and Rajan (2001). More generally,
we fnd that the market rewards (via excess announcement returns) borrowers who
borrow from lenders who are more likely to provide monitoring benefts.
The evidence that uninsured deposits have an important impact on the “specialness”
of banks is the primary contribution of our paper. This evidence implies that policy
makers and regulators should consider the potential costs of reducing market discipline
of banks. We document that banks were special before the passage of the GLBA when
fewer banks were treated as too-big-to-fail. This suggests that bank regulation needs to
maintain/re-establish some elements of the fragile capital structure that exists at banks
that are not too-big-to-fail. Specifcally, if banks become too safe, they may no longer be
particularly valuable as monitoring lenders. Rather than information-generating
intermediaries, they will merely be transactional institutions.
Notes
1. See, for example, Chan (1983), Diamond (1984, 1991), Ramakrishnan and Thakor (1984), Berlin
and Loeys (1988), Rajan (1992, 1998), Besanko and Kanatas (1993), Rajan and Winton (1995),
Holmstrom and Tirole (1997) and Diamond and Rajan (2001).
2. Exceptions include Billett et al. (1995), Byers et al. (1998) and Marsh (2006). Each utilized the
lender’s credit rating as a proxy for lender quality. While credit ratings are a function of lender
quality and borrower quality, we are focused on the inputs to the credit rating equation, not
the overall level of bank creditworthiness.
3. Gande and Saunders (2012) show banks are special in the production of information and
monitoring, while the special nature of banks is changing in that the sales of loans and
secondary market trading of loans also generate positive borrower returns. Guner (2006) fnds
borrowers whose loans are sold pay lower rates, suggesting that banks that retain and
monitor loans are provide a higher-value service. Ross (2010) fnds more favorable loan
announcements when one of three “dominant” US banks is the lead underwriter due to strong
monitoring reputations.
4. Loans are often unfunded agreements such as revolvers, term loans, working capital
agreements, etc.
5. “Contaminated” means a story appeared within one day on either side of the announcement
and included information about items such as dividends, earnings announcements,
managerial changes, joint ventures, assets sales/purchases or new product announcements.
6. A two-day announcement window is used because wire service announcements may arrive
after the close of trading on day zero. The market model is estimated using 100 trading days
before, t ??120 to ?21, and 100 days after, t ?21 to 120, the announcement date.
7. Financial regulators look to past due loans as an indicator of the quality of a bank’s loan
portfolio. Additionally, Berger et al. (1991) document that the ratio of past due loans to total
assets predicts future bank problems. Therefore, we use this as our measure of bank portfolio
risk. Specifcally, we use a three-year average of this ratio.
109
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
8. An alternative interpretation is that some lenders provide “commodity” funding and compete
on price rather than monitoring services. Ross (2010) fnds that the dominant banks in the
early 2000s capture market share by offering lower-cost funds.
References
Berger, A., Espinosa-Vega, M., Frame, W. and Miller, N. (2005), “Debt maturity, risk and
asymmetric information”, The Journal of Finance, Vol. 60 No. 6, pp. 2895-2923.
Berger, A., King, K. and O’Brien, J. (1991), “The limitations of market value accounting and a more
realistic alternative”, Journal of Banking and Finance, Vol. 15 No. 4, pp. 753-783.
Berlin, M. and Loeys, J. (1988), “Bond covenants and delegated monitoring”, The Journal of
Finance, Vol. 43 No. 2, pp. 397-412.
Bernanke, B. and Gertler, M. (1989), “Agency costs, net worth, and business fuctuations”, The
American Economic Review, Vol. 79 No. 1, pp. 14-31.
Besanko, D. and Kanatas, G. (1993), “Credit market equilibrium with bank monitoring and moral
hazard”, The Review of Financial Studies, Vol. 6 No. 1, pp. 213-232.
Billett, M., Flannery, M. and Garfnkel, J. (1995), “The effect of lender identity on a borrowing
frm’s equity return”, The Journal of Finance, Vol. 50 No. 2, pp. 699-718.
Black, F. (1975), “Bank funds management in an effcient market”, Journal of Financial Economics,
Vol. 2 No. 4, pp. 323-339.
Boot, A. (2000), “Relationship banking: what do we know?”, Journal of Financial Intermediation,
Vol. 9 No. 1, pp. 7-25.
Byers, S., Fraser, D. and Shockley, R. (1998), “Lender identity and borrower returns: the evidence
fromforeign bank loans to U.S. corporations”, Global Finance Journal, Vol. 9 No. 1, pp. 81-94.
Calomiris, C. and Kahn, C. (1991), “The role of demandable debt in structuring optimal banking
arrangements”, The American Economic Review, Vol. 81, pp. 497-513.
Chan, Y. (1983), “On the positive role of fnancial intermediation in allocation of venture capital in
a market with imperfect information”, The Journal of Finance, Vol. 38 No. 5, pp. 1543-1568.
DeYoung, R., Glennon, D. and Nigro, P. (2008), “Borrower-lender distance, credit scoring, and loan
performance: evidence from informational-opaque small business borrowers”, Journal of
Financial Intermediation, Vol. 17 No. 1, pp. 113-143.
Diamond, D. (1984), “Financial intermediation and delegated monitoring”, Review of Economic
Studies, Vol. 51 No. 3, pp. 393-414.
Diamond, D. (1991), “Monitoring and reputation: the choice between bank loans and directly
placed debt”, Journal of Political Economy, Vol. 99, pp. 689-721.
Diamond, D. and Rajan, R. (2000), “Atheory of bank capital”, The Journal of Finance, Vol. 55 No. 6,
pp. 2431-2465.
Diamond, D. and Rajan, R. (2001), “Liquidity risk, liquidity creation and fnancial fragility: a
theory of banking”, Journal of Political Economy, Vol. 109 No. 2, pp. 287-327.
Fama, E. (1985), “What’s different about banks?”, Journal of Monetary Economics, Vol. 15 No. 1,
pp. 29-39.
Fields, L., Fraser, D., Berry, T. and Byers, S. (2006), “Do bank loan relationships still matter?”,
Journal of Money, Credit, and Banking, Vol. 38 No. 5, pp. 1195-1209.
Gande, A. and Saunders, A. (2012), “Are banks still special when there is a secondary market for
loans?”, The Journal of Finance, Vol. 67 No. 5, pp. 1649-1684.
Guner, A. (2006), “Loan sales and the cost of corporate borrowing”, Review of Financial Studies,
Vol. 19 No. 2, pp. 687-716.
JFEP
6,2
110
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hadlock, C. and James, C. (2002), “Do banks provide fnancial slack?”, The Journal of Finance,
Vol. 57 No. 3, pp. 1383-1419.
Holmstrom, B. and Tirole, J. (1997), “Financial intermediation, loanable funds, and the real sector”,
The Quarterly Journal of Economics, Vol. 112 No. 3, pp. 663-691.
James, C. (1987), “Some evidence on the uniqueness of bank loans”, Journal of Financial
Economics, Vol. 19 No. 2, pp. 217-235.
James, C. and Smith, D. (2000), “Are banks still special? newevidence on their role in the corporate
capital-raising process”, Journal of Applied Corporate Finance, Vol. 13 No. 1, pp. 52-63.
Johnson, S. (1997), “The effect of bank reputation on the value of bank loan agreements”, Journal
of Accounting, Auditing, and Finance, Vol. 12 No. 1, pp. 83-100.
Kane, E. (1977), “Good intentions and unintended evil, the case against selective credit allocation”,
Journal of Money, Credit and Banking, Vol. 9, pp. 55-69.
Kane, E. and Malkiel, B. (1965), “Bank portfolio allocation, deposit variability, and the availability
doctrine”, Quarterly Journal of Economics, Vol. 79, pp. 257-261.
Marsh, I. (2006), “The effect of lenders’ credit risk transfer activities on borrowing frms’ equity
returns”, Bank of Finland Research Discussion Papers.
Maskara, P. and Mullineaux, D. (2011), “Information asymmetry and self-selection bias in bank
loan announcement studies”, Journal of Financial Economics, Vol. 101 No. 3, pp. 684-694.
Petersen, M. and Rajan, R. (2002), “Does distance still matter? the information revolution in small
business lending”, The Journal of Finance, Vol. 57 No. 6, pp. 2533-2570.
Rajan, R. (1992), “Insiders and outsiders: the choice between informed and arm’s length debt”, The
Journal of Finance, Vol. 47 No. 4, pp. 1367-1400.
Rajan, R. (1998), “The past and future of commercial banking viewed through an incomplete
contracts lens”, Journal of Money, Credit, and Banking, Vol. 30 No. 3, pp. 524-550.
Rajan, R. and Winton, A. (1995), “Covenants and collateral as incentives to monitor”, The Journal
of Finance, Vol. 50 No. 4, pp. 1113-1146.
Ramakrishnan, S. and Thakor, A. (1984), “Information reliability and a theory of fnancial
intermediation”, Review of Economic Studies, Vol. 51 No. 3, pp. 415-432.
Ross, D. (2010), “The ‘dominant bank effect’: how high lender reputation affects the information
content and terms of bank loans”, Review of Financial Studies, Vol. 23 No. 7, pp. 2730-2756.
Schuermann, T. (2004), “Why were banks better off in the 2001 recession?”, Federal Reserve Bank
of New York, Current Issues in Economics and Finance, Vol. 10 No. 1, pp. 1-10.
Slovin, M., Johnson, S. and Glascock, J. (1992), “Firmsize and the information content of bank loan
announcements”, Journal of Banking and Finance, Vol. 16 No. 6, pp. 1057-1071.
Corresponding author
Bradley A. Stevenson can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
111
Depositor
discipline
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_336556074.pdf