Journal of Financial Economics

Journal of Financial Economics 103 (2012) 2040
Contents lists available at SciVerse ScienceDirect
Journal of Financial Economics

journal homepage: www.elsevier.com/locate/jfec
Behavioral consistency in corporate nance:

CEO personal and corporate leverage$
Henrik Cronqvist a, Anil K. Makhija b,, Scott E. Yonker c
a
b
c
Robert Day School of Economics and Finance, Claremont McKenna College, 500 E 9th St, Claremont, CA 91711, USA
Fisher College of Business, The Ohio State University, 2100 Neil Ave, Columbus, OH 43210, USA
Kelley School of Business, Indiana University, 1309 East Tenth Street, Bloomington, IN 47405, USA
a r t i c l e in f o
abstract
Article history:
Received 23 August 2010
Received in revised form
11 February 2011
Accepted 1 March 2011
Available online 16 August 2011
We nd that rms behave consistently with how their CEOs behave personally in the
context of leverage choices. Analyzing data on CEOs leverage in their most recent
primary home purchases, we nd a positive, economically relevant, robust relation
between corporate and personal leverage in the cross-section and when examining CEO
turnovers. The results are consistent with an endogenous matching of CEOs to rms
based on preferences, as well as with CEOs imprinting their personal preferences on the
rms they manage, particularly when governance is weaker. Besides enhancing our
understanding of the determinants of corporate capital structures, the broader contribution of the paper is to show that CEOs personal behavior can, in part, explain
corporate nancial behavior of the rms they manage.
& 2011 Elsevier B.V. All rights reserved.
JEL classication:
G30
G32
G34
Keywords:
Corporate nance
Behavioral consistency theory
CEO personal leverage
Corporate leverage
I just dont like to owe money.

William F. Laporte, CEO of American Home Products
that carried no debt until after his 17-year leadership
(Forbes, September 1, 1968, p. 87)
1. Introduction
Since the start of modern capital structure research
with the seminal work of Modigliani and Miller (1958),
$
We are thankful for comments by Sumit Agarwal, Brent Ambrose, Zahi Ben-David, Tim Burch, Jeff Coles, Harry DeAngelo, Werner DeBondt, Rudi
Fahlenbrach, Mark Flannery, David Funder, Nikolay Halov, Eric Helland, Eric Hughson, Danling Jiang, Kose John, Andrew Karolyi, Jon Karpoff, Sandy Klasa,
Alok Kumar, Tim Loughran, Angie Low, Lisa Meulbroek, Enrico Perotti, Richard Rosen, Tony Sanders, Bill Schwert (the editor), Hersh Shefrin, Janet Smith,
Fatma Sonmez, Meir Statman, Per Stromberg,

Geoff Tate, Dick Thaler, Mike Weisbach, Karen Wruck, David Yermack, Robert Yonker, and seminar
participants at Amsterdam Business School, Baruch College, Behavioral Finance conference at DePaul University, California Corporate Finance Conference,
Chapman University, Claremont McKenna College, Erasmus University, European Summer Symposium in Financial Markets in Gerzensee, Financial
Management Association, Florida State University, KAIST, Mitsui Finance Symposium at the University of Michigan, Nanyang Technological University,
NBER Summer Institute in Corporate Finance, Ohio State University, Queens University, Sungkyunkwan University, University of Miami, and Western
Finance Association. We especially appreciate the detailed feedback by an anonymous referee, David Hirshleifer, Mike Lemmon, and Rene Stulz.
Alexander Bonnett, Rishi Desai, Hannah Gregg, and Andrew Hom provided excellent research assistance, and Rudi Fahlenbrach shared his database of
founder-CEOs. Cronqvist is thankful for research funding from the Financial Economics Institute at Claremont McKenna College.
Corresponding author. Tel.: 1 614 292 1899.
E-mail addresses: hcronqvist@cmc.edu (H. Cronqvist), makhija_1@sher.osu.edu (A.K. Makhija), syonker@indiana.edu (S.E. Yonker).
0304-405X/$ - see front matter & 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jneco.2011.08.005
H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040
21
nancial economists have devoted signicant effort to

studying the determinants of corporate leverage. The focus
of most empirical work has been on market, industry, and
rm characteristics. Yet, rms that are similar in terms
of these fundamentals often choose very different corporate leverage. This has led researchers to recently study
the impact of personal characteristics of the rms top
executive, the Chief Executive Ofcer, CEO (Bertrand and
Schoar, 2003; Frank and Goyal, 2009b; Graham, Harvey,
and Puri, 2009; Malmendier, Tate, and Yan, 2010). Our
paper extends this work, but we focus on personal decisions made by CEOs that are in the same domain as the
analyzed corporate decision. Specically, we attempt to
explain corporate capital structures based on what CEOs
have revealed about themselves and their debt tolerance
through past personal leverage choices. The scientic basis
for this hypothesis is an extensive set of well-cited studies
on behavioral consistency theory, i.e., the notion that
individuals behave consistently across situations. We nd
that this is a promising empirical approach in corporate
nance because rms are found to behave consistently
with how their CEOs behave personally in the context of
leverage decisions. Besides enhancing our understanding
of the determinants of corporate capital structures, the
broader novel contribution of the paper is to show that
CEOs personal behavior can, in part, predict corporate
nancial behavior of the rms they manage.1
Until recently, most prior empirical studies assume, at
least implicitly, that a rms CEO does not impact corporate leverage decisions. If it takes a certain type of
individual to rise to the top of a rm, then CEOs are
homogeneous or close substitutes for one another. Alternatively, there may be signicant differences across CEOs,
but they do not affect rms if governance constrains CEOs
from imprinting their personal preferences on the rms
they manage. In either case, rms in the same industry
with similar fundamentals choose similar capital structures despite being managed by different CEOs. In contrast, several researchers have recently taken the position
that differences in terms of personal preferences/tastes
across CEOs may indeed impact corporate leverage decisions. For example, in a recent and extensive review of
empirical capital structure papers, Parsons and Titman
(2008, p. 24) state that CEOs personal characteristics,
such as managerial preferences, may also affect capital
structures. A similar prediction is provided by Opler and
Titman (1994, p. 1021) who state that [d]ifferences in
management tastesycould also explain differences in
leverage ratios within an industry.
Indeed, nancial economists have recently examined
some observable CEO characteristics as potential determinants of corporate leverage.2 Overall, the empirical
evidence is ambiguous. For example, Bertrand and

Schoar (2003) show that older CEOs choose lower leverage, and having a MBA does not signicantly affect
corporate capital structures. But, Malmendier, Tate, and
Yan (2010) report that older CEOs take on more debt, and
Frank and Goyal (2009b) show that MBAs are associated
with more leverage. Frank and Goyal (2009b,p. 5) conclude that, leverage choices are not all that closely
connected to readily observable managerial traits, suggesting that we are still missing identication of important CEO characteristics. By focusing on CEOs personal
leverage, our approach offers to capture the mix and
interplay of the underlying CEO beliefs and preferences
that are relevant to a debt decision.
In contrast to existing studies of CEO characteristics, our
approach is based on behavioral consistency theory.3 An
individual, in our case a rms CEO, is predicted to behave
consistently across situations. Although we have not previously noted the term behavioral consistency in research in
nancial economics, we are aware of several recent studies
in economics, nance, and accounting which are supportive
of this notion. Perhaps the most important example is
Barsky, Juster, Kimball, and Shapiro (1997), who show a
positive relation across individuals between all the risky
behaviors they study: holding stocks rather than Treasury
bills, risky entrepreneurial activity, and smoking and alcohol
consumption. In a corporate nance context, Malmendier
and Tate (2005) nd that CEOs who show signs of overcondence in their personal portfolios are overcondent
also in corporate investment decisions. Hong and
Kostovetsky (forthcoming) nd that portfolio managers
who make personal campaign contributions to Democrats
invest relatively less of the portfolios they manage in rms
deemed socially irresponsible. Hutton, Jiang, and Kumar
(2010) nd that Republican CEOs pursue more conservative
corporate policies than do Democrats. Chyz (2010) nds
that CEOs who are personally more tax aggressive manage
rms with more tax avoidance activities. In sum, the
personal preferences and choices of decision-makers such
as CEOs and portfolio managers seem to partly explain their
professional decisions.
In this paper, we apply behavioral consistency theory
to corporate nance by studying CEOs personal leverage
(as in their choice of mortgage for their primary residences) and the corporate leverage of the rms they
manage. We choose the nancing of the CEOs primary
homes because it involves the domain of debt decisions,
the home purchase is an important decision, and mortgage debt tends to be the most important source of debt,
even if not a measure of total personal indebtedness.
Notably, behavioral consistency theory only requires us to
identify and use a relevant comparable situation and not
1
Borghans, Duckworth, Heckman, and Weel (2008) is a very
informative overview of the economics of personal characteristics, and
they conclude: There is a lot of room for cooperation and exchange of
ndings and methods between personality psychology and economics.
(p. 84). We view our paper as an attempt to engage in such exchange of
methods and ndings.
2
Later in this paper, we review related empirical studies. But, it is
worth pointing out that there also exists theoretical research which
(footnote continued)
incorporates heterogeneity in CEOs personal characteristics into models
of corporate capital structure decisions. For example, Cadenillas,
Cvitanic, and Zapatero (2004) model the relation between managerial
risk aversion and leverage, while Hackbarth (2008) models the relation
with optimism or overcondence.
3
Seminal references include Allport (1937, 1966), Epstein (1979,
1980), and Funder and Colvin (1991).
22
the overall indebtedness of the CEO. Based on behavioral

consistency theory, we predict that corporate and personal leverage are positively related.4 There is, however, a
competing hypothesis which predicts that CEOs with
more personal leverage prefer lower corporate leverage
to countervail their high personal nancial risk in their
portfolio. That is, the hedging hypothesis predicts an
inverse relation between personal and corporate leverage.
It is not clear a priori as to which effectbehavioral
consistency or hedgingis stronger, and in the end, it is
an empirical question whether CEOs personal leverage
decisions explain the corporate leverage of the rms they
manage.
We start our empirical analysis in the spirit of Liu and
Yermack (2007) and construct a database with detailed
information on CEOs primary homes and mortgages.5 In
the U.S., data on individuals total wealth and personal
indebtedness are not available, but data on home purchase prices and mortgages have recently become available to researchers. Our data are from the Lexis-Nexis
public records database and other public data sources
(e.g., county assessor databases), and we show that our
database is representative of ExecuComp rms, i.e., it is
representative of large U.S. public rms. We use the
mortgage to purchase price ratio, or loan-to-value (LTV)
ratio, as the main measure in our empirical analysis, but
we check the robustness to the use of other measures. We
nd signicant heterogeneity across CEOs in personal
home leverage: the range is from zero to 100% and the
standard deviation is 35%. That is, some CEOs choose
signicantly higher personal leverage, either because of
specic debt preferences, or because of other economic
factors which we also consider.
We rst regress corporate leverage on personal home
leverage, and we nd a positive, statistically signicant,
and robust relation. That is, CEOs who are more conservative in terms of their personal leverage are found to
manage rms that choose more conservative corporate
capital structures.6 The economic magnitude of the estimated effect is comparable with other empirical determinants of corporate leverage. Suppose, for example, that we
compare two CEOs, one with the median personal home
leverage and one with a one-standard-deviation lower
leverage. The estimated effect implies 2.5 percentage
points (20%) lower corporate leverage. Personal home
leverage adds a little less explanatory power (incremental
adjusted R2) than rm size and protability, but more
explanatory power than other determinants of corporate
leverage, e.g., tangibility. Our results are robust to
4
A positive relation may not be supported if behaviors are situation-specic (e.g., Mischel, 1968; Slovic, 1972a,b; Endler and
Magnusson, 1976).
5
Liu and Yermack (2007) nd that rm performance deteriorates
when CEOs acquire very large mansions, but unlike our paper, they do
not examine the relation between personal leverage and corporate
capital structures.
6
We measure corporate leverage in year 2004 while personal home
leverage is measured at the time of the most recent primary home
purchase, which at the median is ve years earlier, reducing concerns
that an omitted contemporaneous variable such as mortgage/interest
rates jointly explains both personal and corporate leverage.
numerous robustness checks, including using personal

leverage measures that are not subject to a concern about
the scaling by purchase price (e.g., we use a debt/no
debt indicator rather than a continuous measure) and
the use of excess personal leverage where we control
for determinants of personal home leverage. CEO personal
leverage is found to have an effect on corporate leverage
even after we control for decade-old corporate leverage,
so we conclude that personal leverage explains corporate
leverage beyond a persistence of capital structure effect.
Our results are also robust to controlling for other CEO
characteristics recently proposed in corporate nance
research, i.e., CEO personal leverage captures variation
in corporate leverage that is not simply subsumed by
other CEO characteristics. We also conrm our ndings
regarding a positive relation between corporate and
personal leverage when we examine changes in capital
structure around CEO turnovers.
What are the economic mechanisms through which the
positive relation between CEO personal and corporate
leverage arises? One mechanism seems to be endogenous
matching of CEOs to rms. CEOs with certain personal
characteristics match more optimally with rms that have
demand for those characteristics. Economic explanations for
such optimal matching include more efcient risk allocation: CEOs who are willing to bear more nancial risk match
more optimally with rms for which more nancial leverage is optimal. Our evidence supports CEOrm matching
because we nd that rms systematically replace a CEO
with one with a similar personal debt preference. That is,
our CEO changes analysis reveals that CEOs who are more
conservative in terms of their personal leverage are replaced
with similar CEOs. Such CEOrm matching can partly
explain the persistence of corporate capital structures, as
reported by Lemmon, Roberts, and Zender (2008). An
alternative mechanism which can explain a positive relation
between personal and corporate leverage is that CEOs
imprint their personal preferences on the capital structures
of the rms they manage, whether optimal or not. Our
evidence supports this mechanism as well, since we nd
that the corporate-personal leverage relation is stronger for
rms with weaker corporate governance. That is, some of
our ndings are consistent with studies which report that
agency problems have an effect on corporate capital structures (e.g., Jung, Kim, and Stulz, 1996; Berger, Ofek, and
Yermack, 1997).
The rest of the paper is organized as follows. In Section
2, we discuss our hypotheses regarding the relation
between CEO personal leverage and the corporate leverage of the rms they manage. In Section 3, we describe
and summarize our data. In Section 4, we study the
relation between CEOs personal and corporate leverage.
In Section 5, we report further empirical evidence,
emphasizing the mechanisms through which the positive
relation between personal and corporate leverage arises.
Section 6 concludes.
2. Empirical predictions
In this section, we propose two empirical predictions for
a relation between CEO personal and corporate leverage.
2.1. Behavioral consistency

One hypothesis, which we refer to as behavioral
consistency, is based on a large number of well-cited
studies (e.g., Allport, 1937, 1966; Epstein, 1979, 1980;
Funder and Colvin, 1991).7 If the extent to which an
individual exhibits a behavior in one situation is predictable from the extent to which the individual exhibits the
behavior in another situation, then there is evidence of
behavioral consistency. In the context of this paper,
behavioral consistency predicts a positive relation
between CEO personal and corporate leverage.
Suppose that CEOs, when making leverage decisions in
both the personal and corporate domain, consider their
own personal debt preferences (in addition to other
factors that should impact leverage decisions). That is, in
addition to considering market, industry, and rm characteristics, CEOs who are particularly averse to debt,
prefer corporate capital structures with relatively less
debt, all else equal. This may potentially result in agency
costs of managers imprinting their personal debt preferences on the rms they manage, but such agency costs
may be reduced. First, rms may select CEOs whose
personal debt preferences closely match the rms optimal corporate leverage. Economic explanations for such
matching of CEOs to rms include more efcient risk
allocation, i.e., CEOs willing to bear more nancial risk
match more optimally with rms for which more nancial leverage is optimal. One way to examine this hypothesis is to test whether personal debt choices by CEOs,
several years before their appointments, are related to
corporate leverage after the appointment. We also test
whether rms exhibit a persistent preference for CEOs
that have similar personal leverage choices, in the sense
that current and former CEOs of a particular rm have
similar personal leverage. Second, governance may reduce
the possibility that CEOs imprint their own personal debt
preferences on the rms they manage. We test whether
CEO personal leverage and corporate leverage are more
strongly positively related when governance is relatively
weaker.
2.2. Hedging
The hedging hypothesis predicts an inverse relation
between personal and corporate leverage. Specically, we
predict that CEOs with more personal leverage prefer
lower corporate leverage to counterbalance the risk in
their overall personal portfolios. This hypothesis assumes
that excessive corporate leverage and nancial distress
are costly for the CEO personally. There exists evidence to
support such a prediction. Gilson (1989) nds increased
CEO turnover if rms are in default, bankrupt, or privately
restructuring their debt,8 dismissed CEOs are commonly
7
The relative impact of persons versus situations on behavior
constitutes a long-lasting debate in psychology research. Those on the
person side believe that there is relatively consistent variation across
individuals in, e.g., their thoughts and behaviors.
8
See Weisbach (1988) and Warner, Watts, and Wruck (1988) for
evidence of past performance and CEO turnover.
23
not employed by another public rm for at least three

years while retained CEOs experience compensation
reductions (Gilson and Vetsuypens, 1993), and CEOs of
nancially distressed rms hold fewer seats on other
boards following their departures (Gilson, 1990).
3. Personal leverage of CEOs in the U.S.

3.1. Database construction
Based on public data sources, we construct a new
database with information on the primary homes (i.e.,
the homes where they reside) and mortgages of CEOs of
large U.S. public rms in 2004.9 We focus on a CEOs
primary home, as it constitutes the vast majority of most
CEOs real estate holdings. We examine this year because
it is recent enough that there is reasonable coverage by
public data sources, but we also check the robustness to
the use of data from other years. A detailed description of
the database construction and summary statistics for
CEOs primary homes are provided in Appendix A.
We compute the leverage which each CEO used in the
purchase of his most recent primary home. Specically,
HomeLev is the sum of the primary and other mortgage
liens, at the time of the home purchase, scaled by the
purchase price.10 In the real estate literature, this measure
is commonly referred to as the loan-to-value (LTV) ratio.
Mortgages and home equity loans/lines are likely the
most important sources of debt for CEOs as the interest
rate is generally lower than for uncollateralized loans
(e.g., credit card debt), and mortgages also come
with interest deductibility and may as a result be
used rst.
It is important to recognize that while we analyze
corporate leverage in 2004, personal home leverage is
generally measured in another year, thus reducing concerns about both leverage measures being jointly determined (e.g., by macroeconomic conditions and interest/
mortgage rates in the same year). In Fig. 1, we report a
timeline and a frequency distribution describing when the
CEOs in our sample purchased their most recent primary
homes. We nd that the median year in the gure is 1999.
That is, the median CEO in our database had owned his
primary home for ve years in 2004, so personal leverage
is measured, on average, ve years earlier than corporate
leverage.
9
In this paper, we focus on CEOs, and not Chief Financial Ofcers
(CFOs), because it is very costly to collect data on all executives. CFOs
report to CEOs, not vice versa, so CEOs sign off on important capital
structure decisions. Chava and Purnanandam (2010) nd that CEOs
matter for capital structure choices, while CFOs may matter more for,
e.g., debt-maturity decisions, which we do not study. Also, Graham,
Harvey, and Puri (2009) report that CEOs believe that capital structure is
one of the central corporate decisions that they have the most control
over. Some 15.1% of the CEOs surveyed indicate that they choose capital
structures with no input from others, compared to only 3.1% for CFOs.
10
One problem with the nonexistence of a mortgage record for a
CEO is that it results in HomeLev 0, although the reason could be: (i) no
mortgage was used; or (ii) missing data. To try to include the former and
exclude the latter, we require the purchase price to be available for an
observation to remain in the sample.
Number of CEO home purchases
24
We nd that in terms of both corporate leverage and rm

characteristics, e.g., rm size and protability, our sample
of rms seems representative of large U.S. public rms.
We recognize, however, that the capital structure literature is not limited to large public rms and that capital
structure can be more relevant for smaller rms, since
they are often the ones with the most severe departures
from the Modigliani and Miller (M&M) assumptions.
50
40
30
20
3.3. Determinants of CEOs personal leverage
10
0
Corporate leverage
measured
1970 1975 1980 1985 1990 1995 2000 2004 2008

Year
Fig. 1. CEO home purchase timing. This gure shows the distribution, by
purchase year, of the most recent home purchase for the 605 CEOs of
non-nancial S&P 1,500 rms in ofce at the end of 2004 for whom we
were able to calculate HomeLev.
3.2. Summary statistics

Table 1 reports summary statistics. Panel A shows that
the median HomeLev is 47%. Conditional on having a
mortgage, we nd that the median CEO home leverage
is 66%. CEOs home leverage is somewhat lower than the
U.S. median, which was 75% in 2005, as can be seen in the
nal column of the table. However, the most important
conclusion from the table is the very wide range of
HomeLev: from zero to 100% leverage (i.e., zero downpayment on the home). The variation, as measured by the
standard deviation, is also signicant at 35%.
Panel B contains alternative measures of personal leverage. As noted, 66.0% of CEOs use a mortgage at the time of
the purchase of their primary home. Some CEOs obtain
mortgages after the time of the home purchase (renancing): 73.8% of the CEOs use a mortgage backed by their
primary home at some point in time. For some CEOs, we nd
forms of home leverage other than mortgages. This debt
includes home equity lines/loans or other forms of shortterm debt nancing. The table shows that 22.0% of CEOs
never lever, i.e., we nd no evidence of any form of personal
home leverage. That is, there is signicant heterogeneity
across CEOs in terms of their choices of personal leverage.
Panels C and D show summary statistics for corporate
leverage measures and rm characteristics of our sample
rms.11 These variables are dened in Appendix B. In the
last two columns of Panels C and D of Table 1, we report
the means and medians for the entire ExecuComp universe in 2004. We also perform t-tests and Wilcoxon
Mann-Whitney tests for differences in sample means
and medians of our sample and the ExecuComp universe.
11
Following the approach employed in other work on capital
structure (e.g., Lemmon, Roberts, and Zender, 2008), we winsorize and
truncate leverage at 1.0.
Why do some CEOs have a higher personal home

leverage than others? We recognize several potentially
important determinants of personal leverage: individual
preferences, and also economic factors such as home
prices in the geographic region of the home, macroeconomic conditions (mortgage rates) at the time of the
home purchase, and personal taxes.12
Table 2 reports results from regressing HomeLev on a set
of potential determinants of CEOs personal leverage. In
column 1, we include the CEOs age at the time of the home
purchase (PurAge). We expect an inverse relation as older
CEOs are likely to have accumulated more wealth and, as a
result, are less capital constrained. In column 2, we provide
an alternative measure of wealth: a dummy variable that is
equal to one if the home was purchased after the purchaser
became CEO (PurAfterCeo). In column 3, we include the log of
the median home price in the geographic region (county) of
the CEOs home (LnMedHmVal). CEOs who reside in regions
where residential real estate is relatively more expensive are
expected to use more debt because they may not compensate completely by reducing their demand for housing.13 In
column 4, we include the 30-year xed mortgage rate at the
time when the CEO purchased the home (MortRate30). In
column 5, we include the ve-year lagged market return
prior to the month when the CEO purchased his home
(MktRet5yr ). In column 6, we include all of these potential
determinants at the same time, forming our baseline regression for determinants of personal home leverage.
We nd that older CEOs seem to be less capital
constrained: ten years reduce personal home leverage
by about 3.2 percentage points, though this effect is not
statistically signicant at conventional levels. CEOs who
purchase their homes after taking ofce use 6.6 percentage points less leverage. We also nd that CEOs in
geographic regions with relatively higher real estate
prices are signicantly more levered in their homes. The
difference between Los Angeles county in California and
Cuyahoga county in Ohio implies 7.5 percentage points
higher leverage. Finally, CEOs who purchased their homes
when mortgage rates were relatively low use more
leverage: a 100 basis points lower 30-year xed rate
implies about 6.1 percentage points more home leverage.
As a robustness check, we add in column 7 purchase year
12
A review of the real estate literature reveals that there is no
standard predictive model for loan-to-value ratio.
13
Where to live is an endogenous choice, but living very far from
the corporate headquarters is associated with signicant diseconomies,
so executives are commonly constrained to live in the region of the
corporate headquarters.
25
Table 1
Summary statistics.
The table reports summary statistics for a sample of 605 S&P 1,500 rms. Panels A and B display statistics relating to the nancing of primary residences by
CEOs and Panels C and D display statistics of rm-level leverage measures and control variables. In Panel A, HomeLev is determined at the time of the CEO
home purchase, and is computed as the mortgage divided by the purchase price of the home. If a mortgage is found but if any one of the mortgage amount,
purchase price, or the improvement cost (if the home is new construction) is unavailable, then HomeLev is set to missing. Statistics for HomeLev are for the
unconditional sample, and for HomeLev9Mort the reported sample statistics are conditional on the CEO using a mortgage to nance the home. Mortgage
amount is the sum of the rst and second mortgages at the time of the CEOs home purchase. Panel B reports the percent of the sample and number of
observations that use mortgage nance in the purchase of their primary residence (Mortgage usage at purchase), that use either a mortgage at the time of the
purchase or debt nancing on their home at some point in time (Home leverage usage), and for which there is no public record that the CEO ever used debt
(Never use leverage). The U.S. median data are tabulated from 2005 data provided by the Federal Housing Finance BoardPeriodic Summary Tables and the
2005 American Community Survey Subject Tables. Mortgage amounts are displayed in 2005 home price dollars. Values are adjusted using the Ofce of Federal
Housing Enterprise Oversights National Home Price Index. In Panel C, the corporate leverage variables are total debt to market value of assets (TDM) and total
debt to book value of assets (TDA). All debt measures are computed as of the end of the calendar year 2004. In Panel D, the control variables are market-to-book
ratio (Mktbk), the log of total assets (Assets), protability (Prot), tangibility of assets (Tang), and median industry leverage (IndusLev). All control variables are
computed as of the end of the calendar year 2003, i.e., with a lag of one year compared to the corporate leverage measures. ExecuComp MEAN and MEDIAN
values are calculated from 1,351 U.S.-based, non-nancial, and non-utility rms covered by ExecuComp in 2004. Stars in the ExecuComp MEAN and MEDIAN
columns denote signicance levels from testing the difference between the sample mean (median) and the ExecuComp sample mean (median) using a t-test
(Wilcoxon Mann-Whitney test). Signicance levels are denoted by n, nn, nnn, which correspond to 10%, 5%, and 1% levels, respectively. All leverage and control
variables are computed from S&Ps Compustat database. Detailed denitions are found in Appendix B.
Panel A
MED
HomeLev
HomeLev9Mort
Mortgage amount ($1000s)
MEAN
STD
MIN
MAX
0.47
0.66
0.40
0.63
0.35
0.21
0.00
0.01
1.00
1.00
1,047.00
1,233.00
973.00
54.00
8,626.00
N
608
385
430.00
U.S. MED
0.75
212.00
Panel B
Mortgage usage at purchase

Home leverage usage
Never use leverage
66.0
73.8
22.0
642
642
642
Panel C
TDM
TDA
MEAN
STD
10th
Percentile
50th
90th
MEAN
0.179
0.205
0.198
0.188
0.000
0.000
0.126
0.185
0.425
0.415
605
605
0.178
0.205
1.769
7.119
0.113
0.247
0.154
1.241
1.606
0.139
0.183
0.153
0.710
5.164
0.009
0.050
0.004
1.428
6.950
0.121
0.201
0.104
3.134
9.349
0.240
0.515
0.369
605
605
605
605
605
1.796
7.140
0.112
0.268nn
0.166n
ExecuComp
MEDIAN
0.130
0.187
Panel D
Mktbk
Assets
Prot
Tang
IndusLev
xed effects to account for any differences across purchase years in legislation and market conditions not
picked up by mortgage rates and market returns.
Can differences in personal taxes explain the variation
in personal home leverage? This does not seem to be the
case for several reasons. First, the tax code in the U.S.
allows married (single) taxpayers to deduct interest on
home mortgages up to $1 million ($500,000) on up to two
homes. Out of the mortgages in our database, only 9.6%
are exactly $1 million. Only 11.7% of the CEOs have 100%
HomeLev if their home purchase price is below $1 million
or a $1 million mortgage if it is above the tax deductability threshold. Second, in column 8 of Table 2 we
control for the ratio of a CEOs total compensation which
is not tax deferrable (TaxIncRatio), i.e., salary and other
cash compensation (e.g., bonus) divided by total compensation. CEOs with a larger proportion of their
1.366
6.961
0.124
0.210
0.134nn
compensation in the form of non-tax deferrable income

may be expected to use more debt to reduce their taxes,
but the estimated coefcient is close to zero ( 0.0053)
and not statistically signicant. In column 9, we control
for the log of the CEOs total cash compensation. However,
the estimated coefcient on this variable is negative and
statistically signicant at the 10%-level, which seems to
be more supportive of a capital constraint than a tax
explanation.14 We conclude that taxes do not seem to
explain the variation in personal home leverage.
14
The number of observations is reduced in columns 8 and 9
because we require data on the CEOs compensation at the time of the
home purchase. In several cases, such data are missing because the CEO
purchased the home at a time when the CEO was not a top-executive
covered by ExecuComp or before the start date of the ExecuComp
database.
26
Table 2
Determinants of personal leverage.
The table reports the coefcients and standard errors from regressing HomeLev on determinants of personal leverage. The sample is non-nancial S&P
1,500 rms during 2004. HomeLev is the ratio of mortgage value to purchase price used by the rms CEO in his most recent primary home purchase.
PurAge is the age of the CEO at the time of his home purchase. PurAfterCeo is a dummy variable that is equal to one if the home was purchased after the
purchaser became CEO. LnMedHmVal is the natural logarithm of the median home value in the county in which the CEOs primary residence is located.
County-level median home value data are obtained from the 2005 American Community Survey. MortRate30 is the prevailing 30-year conventional xed
mortgage rate in the month and year of the CEOs home purchase. Data on monthly mortgage rates are obtained from the Federal Reserve Economic
Database series MORTG. MktRet 5yr is the ve-year annualized return of the value-weighted Center for Research on Security Prices (CRSP) index ending on
the last day of the month prior to the CEOs home purchase. TaxIncRatio is the ratio of CEO compensation for which the CEO cannot defer the tax liability.
LnCashComp is the natural logarithm of the total cash compensation (ExecuComp data item TOTAL CURR) of the CEO in the year of the home purchase
adjusted to 2005 dollars. This compensation includes salary plus bonuses. It is computed as the CEOs salary plus bonus divided by total compensation in
the year of the home purchase (ExecuComp items TOTAL CURR / TDC1). The table reports White (1980) heteroskedasticity-consistent standard errors.
Signicance levels are denoted by n, nn, nnn, which correspond to 10%, 5%, and 1% levels, respectively.
(1)
PurAge
(2)
(3)
(4)
(5)
0.0001
(0.0017)
PurAfterCeo
0.0444
(0.0311)
0.0679nnn
(0.0254)
LnMedHmVal
0.1355
(0.1664)
MktRet 5yr
(7)
(8)
0.0048
(0.0022)
0.0663nn
(0.0337)
0.0671n
(0.0355)
0.0672nn
(0.0261)
6.0716nnn
(0.7285)
4.8663
(4.3941)
0.4475nn
(0.1792)
0.2542
(0.6674)
TaxIncRatio
0.0053
(0.0638)
0.0344n
(0.0193)
LnCashComp
Intercept
AdjR2
N
Fixed effects
(9)
nn
0.0032
(0.0020)
0.0746nnn
(0.0247)
3.9221nnn
(0.6008)
MortRate30
(6)
0.3981nnn
(0.0848)
0.4138nnn
(0.0166)
0.0017
605
No
0.0018
605
No
0.4558
(0.3231)
0.0102
605
No
0.7008nnn
(0.0496)
0.3844nnn
(0.0240)
0.0402
605
No
0.0006
605
No
4. Empirical evidence
In this section, we examine whether personal leverage is
related to corporate leverage, i.e., we test the empirical
predictionsbehavioral consistency versus hedgingfrom
Section 2.
4.1. Regression results

4.1.1. Baseline model
Table 3 reports results from regressing corporate
leverage on personal home leverage using ordinary least
squares (OLS).15 In column 1 of Panel A, we nd that the
estimated coefcient on CEO personal home leverage is
positive (0.0632) and statistically signicant at the
1%-level (t-statistic 2.80). We report White (1980) heteroskedasticity-consistent standard errors in this and all
other model specications. The positive relation between
personal leverage and corporate leverage supports the
behavioral consistency hypothesis.
15
We checked that these results are similar to those from a Tobit
model. Also, because corporate leverage may have outliers, we checked
that our results are robust to truncating leverage at zero and one, or
winsorizing at the 0.5%-level in each tail of the distributions.
0.0463
(0.3441)
0.0785
605
No
0.2323
(0.4758)
0.1010
605
PurYear
0.4248nnnn
(0.0314)
0.6552nnn
(0.1321)
0.0027
364
No
0.0053
392
No
In column 2, we include lagged rm-level characteristics as control variables: the market-to-book ratio
(Mktbk) as a measure of growth opportunities, the log of
total assets (Assets) measuring rm size, protability
(Prot), and the tangibility of the rms assets (Tang) as
a measure of collateral. We choose this set of controls to
follow Frank and Goyal (2009a).16 In column 3, we control
for industry leverage by including IndusLev, the median
total debt to market value of assets ratio in the rms
industry, following Frank and Goyal (2009a). In column 4,
which we label our baseline model for corporate leverage, we include all these controls at the same time. The
rm-level control variables have the expected signs. Most
importantly, we nd that the estimated coefcient on
HomeLev is still positive (0.0718) and statistically signicant at the 1%-level.
16
Frank and Goyal (2009a) explore the relative importance of a very
large set of potential determinants of corporate leverage. We include the
controls that they conclude are the most reliable determinants of
corporate leverage. In untabulated regressions, we also checked that
our results are robust to the inclusion of other controls. For example, we
included Sales instead of Assets, a different collateral measure (inventory
plus net property, plant, and equipment scaled by assets) instead of
Tang, and we included Z-score by Altman (1968).
27
4.1.2. Industry xed effects

An alternative to control for industry in research on
corporate capital structures is to include industry xed
effects. In column 5, we include industry xed effects
dened at the two-digit Standard Industry Classication
(SIC) code level. However, we nd that the estimated
coefcient on personal leverage is still positive (0.0784)
and statistically signicant at the 1%-level.17 This withinindustry result should reduce concerns related to matching of CEOs to industries based on risk aversion and debt
preferences.
CEOs during three years prior to 2004, in total 84

observations. HomeLevPrev is the personal leverage of
these previous CEOs. In column 8 of Table 3, we then
regress corporate leverage in the last full year of a
previous CEOs tenure on HomeLevPrev and the control
variables. The estimated coefcient is 0.1285, signicant
at the 5%-level.19 That is, a positive relation between CEO
personal leverage and corporate leverage is found also if
we consider a completely different set of CEOs and
corporate leverage in a different set of years compared
to our original sample.
4.1.3. Non-linearity
A non-linear relation may mask support for the hedging hypothesis. It may be that only CEOs with the highest
home leverage choose to countervail their personal leverage through corporate capital structure decisions. That is,
we may nd an inverse relation between personal and
corporate debt, but only for the CEOs who are the most
highly levered. We choose an 80% cutoff to dene high
personal leverage because of the standard in the U.S.
mortgage industry related to down payments. Specically, we dene HL80 to be an indicator variable that is
one if HomeLev 40:80, and zero otherwise. However,
column 6 shows that there is no evidence of CEOs offsetting their personal leverage by changing their rms
leverage in a countervailing way, not even among the
CEOs who are the most highly levered.
4.1.6. Robustness checks

We performed a number of robustness checks
(untabulated):
Book leverage: We repeat our analysis with book
leverage rather than market leverage as the dependent
variable. We nd that the estimated coefcient on HomeLev is 0.062 in the baseline specication (column 4 of
Table 3). It is statistically signicant at the 1%-level. Thus,
our results are unchanged by using book leverage instead
of market leverage.
Analyzing other years: Because it may be argued that
2004 is a special year, we re-estimated the baseline
regression model for each year 20002008.20 We nd
that the estimated coefcient on HomeLev is statistically
signicant at least at the 5%-level for each year except
2007. The magnitudes of the coefcients vary from 0.0500
to 0.0958.
Placebo analysis: Using data from 20002008, we
regress corporate leverage on personal leverage for rms
where the CEO in 2004 was not the CEO in that year,
either because the CEO had not been appointed yet or
because there had been a CEO turnover after 2004. In this
case we do not predict a signicant effect, and the
estimated coefcient on HomeLev is indeed not statistically signicant.
Other controls: There is evidence that CEOs purchases
of large homes signal poor future performance (Liu and
Yermack, 2007). However, controlling for the natural log
of the square footage of the home, the number of rooms,
or the natural log of the purchase price in 2005 home
price dollars does not change our results. Another concern
is that HomeLev measures regional effects because of
possible relations between geography, personal, and corporate leverage. There is evidence that rural rms have
more debt in their capital structures than otherwise
similar urban rms (Loughran, 2008). However, for such
a result to explain the CEO home leverage effect, it has to
be that CEOs of rural rms have more home leverage than
CEOs of urban rms, which contradicts our previous
results that CEOs in regions with higher median home
prices have higher home leverage. We still re-estimated
our baseline model specication including state xed
effects, but our result remains unchanged.
4.1.4. Excess personal leverage

It may be argued that our empirical prediction is for
excess personal home leverage, i.e., the residual personal leverage controlling for other determinants for personal leverage. In column 7 of Table 3, we therefore
decompose HomeLev into two components: the portion
predicted by economic factors (HomeLevPredict) and the
unexplained portion (HomeLevRes). We obtain these predicted and residual components from the model estimated in column 6 of Table 2.18 We nd that the
residual component of HomeLev is positive and highly
signicant, while the predicted portion has no explanatory power for corporate leverage. As an example, while
residing in a geographic region with relatively high real
estate prices is found to signicantly increase a CEOs
personal home leverage (as we found in Table 2), this
predictable component is not driving our result; it is the
residual component that is signicantly related to corporate capital structures.
4.1.5. Out-of-sample evidence
We also report out-of-sample evidence. Specically,
we collect personal home leverage data for all previous
17
We also used Fama and French (1997) industry classications and
the results do not change (untabulated).
18
An alternative to our reduced-form approach is a structural
model. The advantage of a structural model is that, if HomeLev is
capturing risk aversion, then we can estimate managers risk aversion
parameters and interpret those parameters. However, if something other
than risk aversion is driving the relationship between personal and
corporate leverage, then the model is mispecied.
19
This result is robust to including year xed effects (untabulated).

It is important to recognize that this analysis only involves those
individuals who were actually CEOs in these years, i.e., when we deviate
from 2004, we have to drop observations as we have only complete
personal leverage data for individuals who were CEOs in 2004.
20
28
Table 3
CEO personal and corporate leverage.
Panel A of the table reports coefcients and standard errors from regressing the total debt to market value of assets of the rm in 2004 (TDM) on
determinants of capital structure, using OLS estimation. Control variables are constructed using 2003 data and dened as in Appendix B. The sample is
non-nancial S&P 1,500 rms. HomeLev is dened as the ratio of mortgage value to purchase price used by the rms CEO in his most recent primary
home purchase. Column 5 includes industry xed effects by two-digit SIC code. HL80 is a dummy variable that equals one if HomeLev4 0:80, and zero
otherwise. HomeLevPredict and HomeLevRes are the predicted and residual series from regression 6 in Table 2. In column 8 we estimate the baseline
regression using home leverage data for all previous CEOs during three years prior to 2004, in total 84 observations. HomeLevPrev is the home leverage of
these previous CEOs. The dependent variable in this regression is TDM during the last year of tenure of the previous CEO and control variables are lagged
by one year. Panel B of the table displays the incremental increase in adjusted-R-square caused by adding each of the determinants of capital structure to
a model which includes only industry xed effects by two-digit SIC code. The dependent variable in Panel B is TDM. The table reports White (1980)
heteroskedasticity-consistent standard errors. Signicance levels are denoted by n, nn, nnn, which correspond to 10%, 5%, and 1% levels, respectively.
Panel A
(1)
HomeLev
0.0632nnn
(0.0226)
(2)
(3)
0.0781nnn
(0.0192)
0.0666nnn
(0.0192)
(4)
0.0718nnn
(0.0179)
(5)
(6)
0.0784nnn
(0.0180)
HL80
(7)
(8)
0.0631nnn
(0.0202)
0.0662
(0.4122)
HL80 HomeLev
0.0963
(0.4388)
HomeLevPredict
0.0419
(0.0576)
0.0828nnn
(0.0190)
HomeLevRes
0.1285nn
(0.0592)
HomeLevPrev
Mktbk
0.0454nnn
(0.0076)
0.0281nnn
(0.0058)
0.0346nnn
(0.0072)
0.0279nnn
(0.0058)
0.0273nnn
(0.0057)
Assets
0.0284nnn
(0.0049)
0.0199nnn
(0.0045)
0.0202nnn
(0.0047)
0.0201nnn
(0.0046)
0.0204nnn
(0.0045)
Prot
0.2616nnn
(0.0846)
0.2691nnn
(0.0683)
0.2011nn
(0.0887)
0.2727nnn
(0.0687)
0.2721nnn
(0.0664)
0.2668nn
(0.1265)
Tang
0.2458nnn
(0.0452)
0.0659
(0.0468)
0.1374nn
(0.0598)
0.0659
(0.0468)
0.0624
(0.0469)
0.0138
(0.1294)
0.5604nnn
(0.0675)
0.5562nnn
(0.0666)
0.5934nnn
(0.1453)
0.7247nnn
(0.0555)
IndusLev
Intercept
AdjR2
N
Fixed effects
0.1535nnn
(0.0102)
0.0106
605
No
0.0409nnn
(0.0109)
0.0054
(0.0409)
0.2890
605
No
0.3230
605
No
0.5607nnn
(0.0670)
0.0136
(0.0352)
0.4160
605
No
0.0388
(0.0975)
0.0134
(0.0354)
0.4220
605
Indus
0.0286
(0.0408)
0.4150
605
No
0.4190
605
No
0.0265n
(0.0134)
0.0425nnn
(0.0153)
0.1375
(0.1004)
0.4460
84
No
Panel B
(1)
(2)
(3)
(4)
(5)
0.0742nnn
(0.0198)
HomeLev
0.0462nnn
(0.0066)
Mktbk
0.0190nnn
(0.0051)
Assets
0.2094nnn
(0.0665)
Prot
0.1624nnn
(0.0627)
Tang
Intercept
AdjR2
0.1401
(0.1036)
0.1123
(0.0835)
0.1709
(0.1069)
0.0051
(0.1057)
0.1501
(0.1112)
0.0946
(0.1078)
0.3000
0.3160
0.0160
0.3760
0.0760
0.3190
0.0190
0.3200
0.0200
0.3100
0.0100
605
Indus
605
Indus
605
Indus
605
Indus
605
Indus
605
Indus
DAdjR2
N
Fixed effects
(6)
4.2. Economic magnitude of estimated effect
4.3. Persistence in corporate leverage

There is evidence that corporate capital structures are
relatively persistent (e.g., Leary and Roberts, 2005;
Lemmon, Roberts, and Zender, 2008). One concern is that
CEO personal home leverage is capturing a persistent
leverage effect. As a result, in Table 4, we report the
baseline model specication, and also control for the
rms own leverage a decade ago (column 1), ve years
ago (column 2), and in the previous year (column 3).
Controlling for a rms own leverage a decade or ve
years ago does not change our result. The estimated
coefcients are 0.0838 and 0.0697, respectively, and
statistically signicant at the 1%-level. The magnitude of
the estimated coefcient on HomeLev decreases to 0.0245,
but remains statistically signicant at the 10%-level, when
146
150
103
105
100
Frequency
The economic magnitude of the estimated effect is

comparable to that of some other determinants of corporate leverage, and thus economically relevant. For example, a rm with a CEO with 100% home leverage has a debt
ratio which is 7.2 percentage points higher than a similar
rm with a CEO with zero debt, based on column 4 of
Table 3. We may also compare two CEOs, one with the
median home leverage in our sample and another with a
one-standard-deviation higher leverage. The estimated
difference in corporate leverage is about 2.5 percentage
points ( 0.072 0.35). Because the median total debt to
market value of assets ratio is 12.6% in our sample, this
implies about 20% higher corporate leverage. As a comparison, a one-standard-deviation change in rm size
corresponds to 25% higher leverage, and the effect of a
corresponding market-to-book or protability change is
similar.
Panel B of Table 3 illustrates the explanatory power of
HomeLev relative to other determinants of capital structure. The model in column 1 shows that 30% of the
variation in rms capital structures is explained by
industry xed effects. When HomeLev is added to this
model, the explanatory power increases by 0.0160.3160
(5.33%). Columns 36 show the explanatory power of
other rm-level determinants of capital structure. Only
Mktbk explains a substantially larger increment in
adjusted R2 compared to HomeLev.
An alternative approach to illustrate the economic
magnitude of the estimated effect is to compute predicted
corporate capital structures for each rm in our sample
based on the baseline model specication in Panel A
column 4 in Table 3, with and without HomeLev as an
explanatory variable. In Fig. 2, we report a histogram of
the absolute value of the difference between the predicted
values from these models as a measure of corporate
capital structure effects directly explained by the CEOs
personal leverage. We nd that the median deviation is
0.0244, with a range from zero to 0.0500. That is, because
of the effect of CEOs personal leverage, the median rms
debt ratio deviates about 2.4 percentage points from the
rms debt ratio as predicted by a standard model (without controlling for home leverage).
29
67
50
48 47
54
18
11
0
0.00
0.01
0.02
0.03
0.04
0.05
0
0.06
Fig. 2. Absolute deviations from predicted corporate leverage. The gure

shows the distribution of the absolute deviations from predicted
corporate leverage due to the CEOs debt preference as measured by
HomeLev, which is dened as the ratio of mortgage value to purchase
price used by the rms CEO in his most recent primary home purchase.
Absolute deviations are computed as the absolute value of the difference
between the tted values from a regression of corporate leverage (TDM)
on Mktbk, Assets, Prot, Tang, and IndusLev, i.e., the baseline model
specication, column 4 of Panel A in Table 3 for the sample of 605 S&P
1,500 rms in 2004.
we control for the previous years leverage.21 The overall

conclusion is that CEO personal leverage explains corporate leverage beyond a persistent capital structure effect.
4.4. Zero personal leverage and other alternative measures
A concern with HomeLev is the effect of the scaling (i.e.,
scaling the mortgage with the purchase price of the most
recent primary home) because it may be argued that the
CEOs total wealth is the preferred scaling. A CEOs total
net worth cannot be measured with U.S. data, but we have
attempted to address this concern using several alternative approaches where the scaling is not relevant. First,
we analyze CEOs with zero personal leverage. These CEOs
are particularly interesting as they seem to deviate the
most from what we would expect from the perspective of
personal taxes and interest deductibility, possibly because
of particularly strong debt preferences. In column 1 of
Table 5, we include an indicator variable that is one if
there is no public record of the CEO ever using any
mortgage, home equity line/loan, or other form of shortterm debt home nancing for any home, and zero otherwise. That is, these CEOs have zero personal home
leverage (ZeroPersLev). We nd that the capital structures
of rms of CEOs who never lever personally are
21
In the specication with previous years leverage, the estimated
coefcients on the control variables change dramatically, often more
dramatically than HomeLev. For example, M/B and rm size are no
longer statistically signicant. The coefcient on corporate leverage in
the previous year is large and statistically signicant at all levels, but it is
difcult to explain exactly what this means as it is absorbing many
effects, including M/B, rm size, and CEO personal leverage.
30
Table 4
Effects of persistence in corporate leverage.
The table reports coefcients and standard errors from regressing the
total debt to market value of assets of the rm in 2004 (TDM) on
determinants of capital structure, using OLS estimation. Control variables are constructed using 2003 data and dened as in Table 1 of the
paper. The sample is non-nancial S&P 1,500 rms. HomeLev is dened
as the ratio of mortgage value to purchase price used by the rms CEO
in his most recent primary home purchase. FirmLeveragei is rm
leverage lagged by i years. The table reports White (1980) heteroskedasticity-consistent standard errors. Signicance levels are denoted by
n nn nnn
, ,
, which correspond to 10%, 5%, and 1% levels, respectively.
(1)
HomeLev
0.0838nnn
(0.0241)
FirmLeverage10
0.3262nnn
(0.0725)
(2)
0.0697nnn
(0.0175)
(3)
0.0245n
(0.0144)
0.4159nnn
(0.0512)
FirmLeverage5
0.7218nnn
(0.0842)
FirmLeverage1
Mktbk
0.0316nnn
(0.0082)
0.0138nnn
(0.0053)
0.0013
(0.0028)
Assets
0.0092
(0.0061)
0.0185nnn
(0.0044)
0.0013
(0.0029)
Prot
0.3074nnn
(0.0912)
0.3213nnn
(0.0702)
0.0781nn
(0.0360)
Tang
0.0872
(0.0568)
0.0590
(0.0449)
0.0089
(0.0230)
IndusLev
0.4012nnn
(0.0861)
0.3522nnn
(0.0669)
0.1307nn
(0.0595)
Intercept
0.0432
(0.0518)
AdjR2
N
0.4920
322
0.0683n
(0.0369)
0.5540
504
0.0111
(0.0212)
0.7880
605
signicantly different than those who are levered at some

point. Specically, rms with zero personal leverage CEOs
have 4.9 percentage points less corporate leverage.22
Second, we re-estimate our baseline model specication using alternative measures of personal leverage that
are also not scaled. In column 2, we include an indicator
variable, Mort, which is one if the CEO uses a mortgage at
the time of the purchase of his primary residence, and
zero otherwise. We nd that the estimated coefcient is
positive (0.0464) and signicant at the 1%-level. In column 3, we include another measure, MortRe, an indicator variable that is one if the CEO uses a mortgage at the
time of purchase or any other time, and zero otherwise.
The result is similar. That is, CEOs who use mortgage
nancing manage rms with more debt in their corporate
capital structures compared to CEOs who do not use any

mortgage nancing.
Finally, if the scaling is driving our results, then HomeLev may simply capture CEO wealth, whereas if it is the
numerator driving the result (the total mortgage), then
HomeLev is more likely capturing CEOs debt preference.
In column 4, we decompose HomeLev and nd that the
natural log of the CEOs total mortgage amount in 2005
dollars (LnMortAmt) is signicantly positively related to
corporate leverage, whereas the coefcient on the natural
log of the total purchase price in 2005 dollars (LnPurPrice)
is not signicant. We conclude that the particular scaling
used for our measure of personal leverage is not critical
for our result.
Another concern is that a CEO has multiple homes, but
HomeLev measures personal leverage based on the primary home only. We have therefore collected data on all
homes (i.e., primary, vacation, and any other home) for a
random subsample of 100 CEOs. Specically, we collected
data on the LTV ratio for each home they owned simultaneously with the primary home already in our database.
We nd that 82% of the CEOs owned two or fewer homes,
the maximum being ve homes. We then compute the
value-weighted LTV ratio across all homes (TotHomeLev)
for each CEO. We are able to compute TotHomeLev for 92
of the 100 random CEOs because some own homes in
states where data are not complete. Several conclusions
can be drawn from this exercise. First, the primary home
captures most of the value of the homes and loans. The
median dollar amount of mortgages on second homes is
about half the mortgages on the primary residence.
Second, the correlation between HomeLev and TotHomeLev is very high (0.87), and statistically signicant at the
1%-level. Third, CEOs exhibit consistent behavior across
their homes. For example, the median leverage in the
second home is 57% when the CEO has some leverage
in his primary home, compared to 0% when the CEO has
no leverage in his primary home. Finally, both measures
are signicantly related to corporate leverage. Column 5
of Table 5 reports that the estimated coefcient on
TotHomeLev is 0.0983 and statistically signicant at the
10%-level.
We also re-estimated our baseline model specication
for CEOs with equity ownership (in 2005 dollars) lower
than the median at the time of their primary home
purchase. We do so because we want to study the CEOs
who are less likely to own a large number of homes and
for which the primary home leverage measure therefore
may be more precise. We nd that the estimated coefcient on HomeLev is 0.1105 and statistically signicant at
the 5%-level (untabulated).
4.5. Timing of personal leverage choices
22
We do not predict a perfect positive correlation ( 1) between

personal and corporate leverage so that zero-leverage CEOs must
necessarily manage zero-leverage rms. Our hypothesis based on the
notion of behavioral consistency is that CEOs who never lever personally
are found in rms with statistically signicantly lower corporate
leverage than CEOs with at least some personal home leverage, ceteris
paribus.
Because we measure corporate leverage in year 2004,

while personal home leverage is measured at the time of
the most recent primary home purchase (which at the
median is ve years earlier), any concerns that an omitted
contemporaneous variable jointly explains both personal
and corporate leverage should be reduced. Nonetheless,
we report additional evidence from a detailed analysis of
31
Table 5
Zero personal leverage and alternative measures.
The table reports coefcients and standard errors from regressing the total debt to market value of assets of the rm in 2004 (TDM) on determinants of
capital structure, using OLS estimation. Control variables are constructed using 2003 data and dened as in Appendix B. The sample is non-nancial S&P
1,500 rms. ZeroPersLev is an indicator variable that equals one if there is no public record that the CEO ever used debt, and zero otherwise. Mort is a
dummy variable that takes a value of one if the CEO uses a mortgage to nance the purchase of his home and takes a value of zero otherwise. MortRe is
an indicator variable that equals one if there is evidence that the CEO uses a mortgage at the time of purchase or some time other than the time of
purchase for his primary residence, and zero otherwise. LnMortAmt is the natural logarithm of the real value of the total mortgage amount used by the
CEO in his most recent home purchase. LnPurPrice is the natural logarithm of the real purchase price of the CEOs most recent primary home purchase.
Real mortgage values and purchase prices are computed in 2005 home price dollars using the Ofce of Federal Housing Enterprise Oversights National
Home Price Index. TotHomeLev is the value-weighted average of the LTV ratios of all homes owned by the CEO simultaneously to his primary residence
(including the primary residence) and was calculated for a random sample of 100 CEOs from the 2004 sample. We are able to compute TotHomeLev for 92
of the 100 random CEOs because some own homes in states where data are not complete. The table reports White (1980) heteroskedasticity-consistent
standard errors. Signicance levels are denoted by n, nn, nnn, which correspond to 10%, 5%, and 1% levels, respectively.
(1)
ZeroPersLev
(2)
(3)
(4)
0.0494nnn
(0.0138)
0.0464nnn
(0.0124)
Mort
0.0512nnn
(0.0131)
MortRe
0.0033nnn
(0.0009)
0.0080
(0.0074)
LnMortAmt
LnPurPrice
0.0281nnn
(0.0059)
0.0177nnn
(0.0045)
0.2712nnn
(0.0690)
0.0713
(0.0475)
0.5729nnn
(0.0672)
0.0381
(0.0334)
0.0288nnn
(0.0059)
0.0184nnn
(0.0045)
0.2656nnn
(0.0694)
0.0671
(0.0472)
0.5657nnn
(0.0666)
0.0050
(0.0355)
0.0283nnn
(0.0059)
0.0180nnn
(0.0044)
0.2708nnn
(0.0690)
0.0700
(0.0474)
0.5711nnn
(0.0669)
0.0115
(0.0356)
0.0282nnn
(0.0059)
0.0194nnn
(0.0045)
0.2683nnn
(0.0691)
0.0643
(0.0476)
0.5641nnn
(0.0669)
0.1035
(0.1043)
0.0983n
(0.0518)
0.0158
(0.0158)
0.0265nnn
(0.0091)
0.1046
(0.0715)
0.1131
(0.0905)
0.6867nnn
(0.1694)
0.1348n
(0.0757)
0.4120
605
0.4130
605
0.4140
605
0.4110
605
0.5220
92
TotHomeLev
Mktbk
Assets
Prot
Tang
IndusLev
Intercept
AdjR2
N
(5)
the timing of CEOs personal home leverage choices to

address remaining concerns that an omitted variable
jointly explains CEO personal and corporate leverage.
First, we examine CEOs who purchased their primary
homes, and thus chose personal home leverage, before
being appointed CEOs of their rms. In column 1 of
Table 6, we nd that the estimated coefcient on HomeLev
is 0.0647 and statistically signicant at the 1%-level. That
is, CEOs show similar debt preferences personally, even
before they became the rms CEO, as they do in their
current role as their rms top executive.
Second, in columns 2 and 3 we compare recent, dened
as within ve years of 2004, versus earlier personal leverage
choices. We nd that both estimated coefcients on HomeLev are positive: 0.0471 versus 0.0882. The coefcient for
more recent leverage choices is signicant at the 1%-level,
while the coefcient for those who chose leverage more
than ve years ago is signicant at the 10%-level. Not
surprisingly, personal leverage measured more recently
seems to be a more precise estimate of a CEOs debt
preferences, but we nd a positive relation also when we
relate personal leverage choices more than ve years ago to
current corporate capital structures.
4.6. CEO characteristics

In this section, we rst review the existing empirical
evidence related to a relation between CEO personal
characteristics and corporate capital structures. We nd
that the literature provides ambiguous evidence for most
characteristics. For example, Bertrand and Schoar (2003)
show that older CEOs choose lower leverage, and having
an MBA does not signicantly affect corporate capital
structures. But, Malmendier, Tate, and Yan (2010) report
that older CEOs take on more debt, and Frank and Goyal
(2009b) show that MBAs are associated with more leverage. Other characteristics that have been proposed and
analyzed recently in the literature include a Depression
Baby indicator, nancial expertise and education, tenure,
and overcondence. Frank and Goyal (2009b,p. 5) conclude that, leverage choices are not all that closely
connected to readily observable managerial traits.
Nevertheless, we next examine these CEO personal
characteristics and compare them to HomeLev in explaining corporate leverage. We collect data on CEO characteristics from Marquis Whos Who and the Notable Names
Database (NNDngdenitions in previously referenced
32
Table 6
Timing of personal leverage choices.
The table reports coefcients and standard errors from regressing
different measures of corporate leverage computed in 2004 on determinants of capital structure, using OLS estimation. Control variables are
constructed using 2003 data and dened as in Appendix B. The sample is
non-nancial S&P 1,500 rms. HomeLev is dened as the ratio of
mortgage value to purchase price used by the rms CEO in his most
recent primary home purchase. Column 1 reports regression results
using observations in which the CEOs most recent home purchase was
made prior to becoming CEO. Column 2 reports regression results using
observations in which the CEOs most recent home was purchased prior
to 1999. Column 3 reports regression results using observations in
which the CEOs most recent home was purchased after 1998. The table
reports White (1980) heteroskedasticity-consistent standard errors.
Signicance levels are denoted by n, nn, nnn, which correspond to 10%,
5%, and 1% levels, respectively.
Prior to
being CEO
(1)
Earlier
leverage choices
(2)
Recent
leverage choices
(3)
0.0647nnn
(0.0226)
0.0471n
(0.0263)
0.0882nnn
(0.0260)
Mktbk
0.0275nnn
(0.0070)
0.0196nnn
(0.0067)
0.0450nnn
(0.0095)
Assets
0.0162nnn
(0.0054)
0.0188nnn
(0.0058)
0.0210nnn
(0.0069)
Prot
0.2623nnn
(0.0796)
0.3217nnn
(0.0771)
0.1311
(0.0817)
Tang
0.0885
(0.0563)
0.0082
(0.0617)
0.1233n
(0.0736)
IndusLev
0.6258nnn
(0.0780)
0.6550nnn
(0.0966)
0.4565nnn
(0.0915)
Intercept
0.0008
(0.0438)
HomeLev
AdjR2
N
0.4240
427
0.0097
(0.0474)
0.4180
296
0.0101
(0.0524)
0.4170
309
papers.23 We also collect data on measures of CEO wealth

because wealthier CEOs may be more willing to lever up,
both personally and in the rms they manage. We use
data from ExecuComp on the natural log of the market
value of the CEOs equity ownership in the rm lagged
one year compared to corporate leverage (EqOwn), and a
founder-CEO indicator (Founder) from Fahlenbrach (2009)
because the wealthiest CEOs are founders.
Table 7 reports the results. In Panel A, we show
descriptive statistics for the CEO personal characteristics.
We also report correlations between HomeLev and these
23
Because we do not have access to Malmendier, Tate, and Yans
(2010) Longholder data, in our analysis we use a classication of CEOs
based on media articles. We review articles on our sample CEOs for the
three years prior to 2004 in The New York Times, Business Week, The
Economist, and The Wall Street Journal. Articles in which the words
condent, condence, optimistic, and optimism were used in
association with the CEO were classied to imply a condent CEO. Along
with articles negating overcondence, articles with cautious, conservative, reliable, practical, frugal, and steady were classied
to imply a cautious CEO. We dene Condent as an indicator variable
with value one if the number of articles implying a condent CEO
exceeds the number implying a cautious CEO, and zero otherwise.
Cautious is one if the number of cautious articles exceed the number
of condent articles, and zero otherwise.
characteristics, and nd several to be signicant: CEOs

with more equity ownership, founders, and older CEOs
have lower personal home leverage, and those characterized as cautious in news articles also have signicantly
less leverage personally. These unconditional correlations
suggest that HomeLev draws up beliefs and preferences in
an expected way. In Panel B, we add each of the CEO
characteristics, with the exception of HomeLev, to our
baseline model. In Panel C, we add HomeLev. Note that the
control variables (rm and industry level) are included,
but not reported. Also note that the adjusted R-square in
Panel C is higher in every case compared to the corresponding regression in Panel B, i.e., adding HomeLev
provides additional explanatory power. Importantly, the
coefcient on personal leverage is positive and statistically signicant at the 1%-level in every case, and has a
largely similar value to that in our baseline model. Finally,
in column 1 of Panel D, we re-estimate the baseline model
specication with all the CEO characteristics, except
HomeLev. Most of the coefcient estimates are not statistically signicant. In column 2, we add HomeLev. We nd
that the estimated coefcient on HomeLev is 0.0602, and
statistically signicant at the 5%-level. Only EqOwn retains
its signicant coefcient among the other variables. We
conclude that personal leverage is signicantly related to
corporate leverage even after controlling for other CEO
characteristics analyzed in the literature.
5. Further evidence and extensions
5.1. CEO turnover and changes in corporate leverage
We examine CEO turnover and corporate leverage
changes. We refer to CEOs as current (i.e., CEOs in 2004)
versus previous CEOs, and identify all CEO turnovers
during the previous three years. We have data on personal
home leverage for 84 previous CEOs.24 HomeLevPrev is the
personal home leverage of the previous CEO. For previous
CEOs, we calculate corporate leverage as of the last full year
of the tenure of the CEO. For example, if the previous CEO
left ofce on June 15, 2002, then we associate the end of the
year 2001 corporate leverage with this CEO, as long as he
was in ofce for all of 2001. For current CEOs, we calculate
corporate leverage for the rst full year that the CEO was in
ofce. Thus, we compute corporate leverage associated with
the two different CEOs two years apart in order to ensure
that the rm capital structure choices we analyze are in fact
attributable to the two different CEOs.
Table 8 shows summary statistics and regression
results for the CEO turnover analysis. We dene HomeLevChg to be HomeLev HomeLevPrev. Panel A shows that
there are 39 observations with HomeLevChg 4 0, i.e., the
new CEO has more personal leverage than the previous
24
We nd 149 CEO turnovers, and are able to nd primary
residences for 108, or 72.5%, of them, i.e., a comparable percentage to
the one for our original sample (75.2%). We are able to calculate home
leverage for 89 of these CEOs (HomeLevPrev) after dropping eight
observations that involve new construction and 11 observations with
missing purchase prices. For ve of them, the previous CEOs tenure was
for less then one full calendar year, so we decided to exclude them.
Panel A
EqOwn
Founder
Age
Tenure
DepBaby
Cohort
Military
MBA
PriorCFO
FinBack
Condent
Cautious
Mean
St. dev.
8.3824
0.0463
55.0000
7.1372
0.0033
1945.0000
0.0615
0.3743
0.1229
0.1453
0.0645
0.0099
2.1632
0.2103
6.8385
6.5432
0.0574
7.2919
0.2405
0.4846
0.3288
0.3528
0.2458
0.0992
535
605
605
605
605
605
358
358
358
358
605
605
Corr. with
HomeLev
0.2006nnn
0.0707n
0.1694nnn
0.1312nnn
0.0667
0.1615nnn
0.0652
0.0153
0.0642
0.0265
0.0445
0.1159nnn
Table 7
CEO characteristics.
The table reports the relationship between HomeLev and various other CEO characteristics. In Panel A, summary statistics and correlations with HomeLev are reported for each of the CEO characteristics. In
Panel B, the coefcients and standard errors are reported for various CEO characteristics as determinants of corporate leverage. The estimates are obtained from regressing the total debt to market value of
assets of the rm in 2004 (TDM) on the baseline model of capital structure reported in Table 3 Panel A column 4 with HomeLev omitted plus the noted characteristic, using OLS estimation. Coefcient estimates
and standard errors for the control variables are not reported, but are included in all regressions in Panels B, C, and D. In Panel C, HomeLev is added as a determinant of capital structure for the regressions in
Panel B. Coefcient estimates and standard errors are reported for both HomeLev and the specic CEO characteristic being tested. In Panel D, we test several characteristics jointly. HomeLev is dened as the
ratio of mortgage value to purchase price used by the rms CEO in his most recent primary home purchase. EqOwn is the log of the market value of the CEOs equity ownership in the rm. Ownership data are
from ExecuComp and price data are from CRSP. Founder is an indicator variable equal to one if the CEO is the founder of the company, and zero otherwise. The data on founder CEOs are from Fahlenbrach
(2009). Age is the age of the CEO in 2004. Tenure is the number of years the CEO held the CEO position as of 2004. DepBaby is a dummy variable that is equal to one if the CEO was born during the period 1920 to
1929. Cohort is the decade in which the CEO was born (i.e., if the CEO was born in 1945, Cohort is 1940). Military is a dummy variable that is equal to one if the CEO has military experience and is zero otherwise.
MBA is a dummy variable that is equal to one if the CEO has an MBA and is zero otherwise. PriorCFO is a dummy variable that is equal to one if the CEO was ever CFO of a company and is zero otherwise. FinBack
is a dummy variable that is equal to one if PriorCFO equals one or the CEO has a degree in the area of nance and is zero otherwise. The data on CEOs career paths, educational background, and military history
are hand-collected from Marquis Whos Who database and the NNDB online database. We are able to identify 358 (59.1%) of the 605 CEOs in the sample using these sources. Condent is a dummy variable that
is equal to one if in the years 2001 through 2003 more news articles use condent adjectives than cautious adjectives and is zero otherwise. Cautious is a dummy variable that is equal to one if in the years 2001
through 2003 more news articles use cautious adjectives than condent adjectives and is zero otherwise. In constructing both the Cautious and Condent variables, we follow the methodology outlined in
Malmendier, Tate, and Yan (2010). The table reports White (1980) heteroskedasticity-consistent standard errors. Signicance levels are denoted by n, nn, nnn, which correspond to 10%, 5%, and 1% levels,
respectively.
Panel B
EqOwn
(1)
CEO characteristic
0.373
535
0.0411n
(0.0234)
0.403
605
Age
(3)
0.0000
(0.0008)
0.401
605
Tenure
(4)
0.0011
(0.0009)
0.402
605
DepBaby
(5)
0.0694nnn
(0.0223)
0.401
605
Cohort
(6)
0.0003
(0.0008)
0.401
605
Military
(7)
0.0604
(0.0512)
0.395
358
MBA
(8)
0.0092
(0.0180)
0.390
358
PriorCFO
(9)
FinBack
(10)
0.0016
(0.0219)
0.0025
(0.0205)
0.389
358
0.389
358
Condent
(11)
0.0023
(0.0306)
0.401
605
Cautious
(12)
0.0277
(0.0498)
0.401
605
33
AdjR2
N
0.0080nnn
(0.0031)
Founder
(2)
34
Table 7 (continued)
Panel C
EqOwn
(1)
CEO characteristic
HomeLev
0.0323
(0.0231)
Age
(3)
0.0006
(0.0008)
Tenure
(4)
DepBaby
(5)
0.0005
(0.0009)
0.0408n
(0.0227)
Cohort
(6)
0.0008
(0.0008)
Military
(7)
MBA
(8)
0.0661
(0.0505)
0.0076
(0.0178)
PriorCFO
(9)
FinBack
(10)
0.0034
(0.0217)
0.0001
(0.0199)
Condent
(11)
0.0001
(0.0300)
0.0667nnn
(0.0188)
0.0703nnn
(0.0179)
0.0737nnn
(0.0180)
0.0702nnn
(0.0181)
0.0713nnn
(0.0180)
0.0745nnn
(0.0180)
0.0728nnn
(0.0240)
0.0699nnn
(0.0236)
0.0704nnn
(0.0238)
0.0702nnn
(0.0237)
0.0718nnn
(0.0179)
0.388
535
0.416
605
0.416
605
0.416
605
0.415
605
0.416
605
0.409
358
0.403
358
0.402
358
0.402
358
0.415
605
Cautious
(12)
0.0017
(0.0509)
0.0717nnn
(0.0180)
0.415
605
Panel D
(1)
(2)
0.0104nn
(0.0041)
0.0255
(0.0345)
0.0015
(0.0019)
0.0989n
(0.0540)
0.0010
(0.0011)
0.0013
(0.0341)
0.0169
(0.0181)
0.0170
(0.0231)
0.0195
(0.0271)
0.0720nn
(0.0318)
Yes
0.364
319
0.0602nn
(0.0252)
0.0092nn
(0.0041)
0.0293
(0.0340)
0.0018
(0.0019)
0.0854
(0.0539)
0.0013
(0.0011)
0.0020
(0.0340)
0.0141
(0.0182)
0.0168
(0.0226)
0.0196
(0.0266)
0.0567
(0.0349)
Yes
0.375
319
HomeLev
EqOwn
Founder
Tenure
DepBaby
Cohort
Military
MBA
FinBack
Condent
Cautious
Baseline controls
AdjR2
N
AdjR2
N
0.0061nn
(0.0030)
Founder
(2)
35
Table 8
CEO turnover and changes in corporate leverage.
Panel A of this table reports summary statistics for changes in CEO HomeLev for 84 S&P 1,500 non-nancial rms. HomeLev is dened as the ratio of
mortgage value to purchase price used by the rms CEO in his most recent primary home purchase and HomeLevPrev is the home leverage of the rms
previous CEO. Panel B reports regression results using the 84 sample rms for which changes in HomeLev are calculated. Column 1 of Panel B reports
coefcients and standard errors from regressing HomeLev on HomeLevPrev. Columns 2 and 3 of Panel B report coefcients and standard errors from
regressing the change in the total debt to market value of assets of the rm (TDMChg) on changes in the determinants of capital structure, using OLS
estimation. Control variables are constructed using one-year lagged data and are dened as in Appendix B. HomeLevChg is dened as HomeLev
HomeLevPrev. HomeLevDecr is a dummy variable that takes a value of one if the HomeLev of the incumbent CEO is less than the HomeLev of the previous
CEO. HomeLevIncr is a dummy variable that takes a value of one if the HomeLev of the incumbent CEO is greater than the HomeLev of the previous CEO.
TDM0 is the year-end TDM of the last full year of the previous CEOs tenure. Columns 2 and 3 include xed effects for the rst year of tenure of the current
CEO. The table reports White (1980) heteroskedasticity-consistent standard errors. Signicance levels are denoted by n, nn, nnn, which correspond to 10%,
Panel A
Increases in HomeLev
No change in HomeLev
Decreases in HomeLev
Changes in HomeLev
MED
MEAN
STD
MIN
MAX
0.35
0.00
0.29
0.00
0.41
0.00
0.36
0.06
0.27
0.00
0.25
0.42
0.03
0.00
0.93
0.93
0.95
0.00
0.06
0.95
N
39
15
30
84
Panel B
Dependent variable
Intercept
HomeLevPrev
HomeLev
(1)
TDMChg
(2)
0.3435nnn
(0.0512)
0.2319nn
(0.1007)
0.0674n
(0.0346)
0.056
84
0.1609nn
(0.0712)
0.0142
(0.0104)
0.0065
(0.0312)
0.3103nn
(0.1292)
0.1166
(0.1755)
0.0139
(0.2132)
0.094
83
0.1152nnn
(0.0418)
0.0302
(0.0345)
0.1599nn
(0.0626)
0.0123
(0.0089)
0.0186
(0.0292)
0.2985nn
(0.1321)
0.0392
(0.1687)
0.0276
(0.1794)
0.168
83
No
Time
Time
HomeLevDecr
HomeLevIncr
TDM0
MktbkChg
AssetsChg
ProtChg
TangChg
IndusLevChg
Fixed effects
0.0145
(0.0487)
0.0622n
(0.0329)
HomeLevChg
AdjR2
N
TDMChg
(3)
CEO, 30 observations with HomeLevChg o 0, and 15 observations with no change (often zero or 80% home leverage).
We construct indicator variables for a leverage increase
(HomeLevIncr) and decrease (HomeLevDecr). As can be
seen in the table, the mean (median) increase in personal
home leverage is 0.41 (0.35), while the mean (median)
decrease is 0.36 (0.29).
We report two results from the CEO turnover analysis.
First, in column 1 of Panel B, we regress the current CEOs
personal home leverage on the previous CEOs leverage. We
nd a positive (0.2319) and statistically signicant, at the
5%-level, relation between the home leverage of the current
and previous CEOs. That is, if the previous CEO of a rm had
relatively low personal leverage, the current CEO also tends
to have low personal leverage. Second, in column 2, we
regress changes in corporate leverage on HomeLevChg,
changes in the control variables, the corporate leverage in

the last full year of the previous CEO (TDM0), and year xed
effects. We nd changes in CEO personal leverage predict
changes in corporate leverage.25 The estimated coefcient
on HomeLevChg is positive (0.0622) and statistically signicant at the 10%-level. The result of this changes analysis is
consistent with the cross-sectional regressions, but here the
identication comes from CEO turnover within rms. Firms
change corporate leverage in a way that is, at least in part,
predicted by the difference in personal leverage between
the new and previous CEOs.
25
We checked the robustness by including a measure of changes in
expected ination using data from the Livingston Survey (www.phila
delphiafed.org/research-and-data/real-time-center/livingston-survey/)
but the results remain unchanged (untabulated).
36
In column 3, we decompose the change in home

leverage associated with the change in CEO by introducing HomeLevDecr and HomeLevIncr in the regression,
which leaves out the cases with no changes in home
leverage. We nd that the positive relation in column 2
arises from decreases in TDM for new CEOs that have
lower home leverage than the previous CEO. The coefcient of HomeLevDecr is negative and signicant at the 1%level, while the coefcient of HomeLevIncr is insignicant.
It would appear that CEOs with preferences for less debt
are more proactive in managing corporate debt, or that
rms seeking lower leverage nd matches more easily
with conservative CEOs.
5.2. Why are CEO personal and corporate leverage related?
5.2.1. Endogenous matching of CEOs and rms
One mechanism through which the positive relation
between personal and corporate leverage arises is endogenous matching of CEOs and rms. Different CEOrm pairs
may differ signicantly in their match quality, so that a
specic CEO matches well with one rm but not another.
Economic explanations for such matching include more
efcient risk allocation such that CEOs who are willing to
bear more nancial risk match most optimally with rms
for which higher corporate leverage is optimal. In equilibrium, CEOs with specic personal characteristics match
with rms that have demand for those characteristics.
Differences in match quality across CEOs can explain why
rms boards spend so much effort on ex ante screening
prior to appointing a specic new CEO.
We found that a rm commonly replaces a CEO with low
personal leverage with a new similar CEO, and vice versa.
This evidence is supportive of an endogenous CEOrm
matching model in which rms persistently select CEOs
with specic preferences for leverage and nancial risktaking. If the rm believes that the CEO will imprint his
personal leverage preference on the rm, then one way to
mitigate value-destroying effects of such debt preference is
simply to choose a manager whose debt preference is
aligned with the optimal capital structure of the rm. In
some situations, the rm prefers a change and chooses a
new CEO who is more conservative than the previous CEO,
and in the years following such a CEO turnover, we indeed
observe a decrease in corporate leverage.
5.2.2. CEOs imprinting personal preferences
If CEOs imprint their personal preferences on the capital
structures of the rms they manage, corporate governance
structures may play an important role. More specically, it
is in rms with relatively weaker governance that we expect
CEOs to imprint their preferences. We examine whether
variation in governance results in different effects of personal leverage on corporate leverage.
Table 9 reports our results.26 First, we study incentivebased compensation. Using data from ExecuComp, we
26
Control variables in this table follow our baseline specication,
but we do not report coefcient estimates and standard errors in the
table.
Table 9
Effects of corporate governance.
The table reports coefcients and standard errors from regressing
different measures of corporate leverage computed in 2004 on determinants of capital structure, using OLS estimation. Unreported control
variables are constructed using 2003 data and include those in the
baseline model dened in column 4 of Table 3, Panel A. The sample is
non-nancial S&P 1,500 rms. HomeLev is dened as the ratio of
mortgage value to purchase price used by the rms CEO in his most
recent primary home purchase. IncentPay is the ratio of CEO incentive
compensation to total compensation in 2003. It is computed as the CEOs
total compensation minus salary and deferred compensation divided by
his total compensation (ExecuComp items (TDC1 SALARY DEFER RPT
AS INC TOT)/TDC1). SmallBoard is an indicator variable that is one if the
number of directors on the rms board is less than or equal to the
median number of board members per rm in the sample in 2004 (nine
directors or less), and is zero otherwise. GoodGov is an indicator variable
that equals one if the 2004 governance index of Gompers, Ishii, and
Metrick (2003) is less than or equal to six, and zero otherwise. The data
on the governance index and board size are from RiskMetrics. The table
reports White (1980) heteroskedasticity-consistent standard errors.
Signicance levels are denoted by n, nn, nnn, which correspond to 10%,
(1)
(2)
nnn
HomeLev
0.1280
(0.0445)
IncentPay
0.0030
(0.0356)
HomeLev IncentPay
(3)
nnn
0.0928
(0.0234)
0.0614nnn
(0.0194)
0.1163
(0.0749)
SmallBoard
0.0172
(0.0173)
0.0711nn
(0.0324)
HomeLev SmallBoard
GoodGov
0.0341n
(0.0187)
HomeLev GoodGov
0.0253
(0.0397)
Baseline controls
AdjR2
N
Yes
0.4010
580
Yes
0.4480
483
Yes
0.4320
535
dene IncentPay as the CEOs total compensation minus

salary and deferred compensation divided by total compensation. In column 1, we interact HomeLev with incentive pay and nd that the effect of personal leverage is
lower, but not signicantly so (t-statistic 1.55) when the
CEOs incentive pay is a larger proportion of total
compensation.
Second, we examine a measure of board governance,
the size of the board.27 See Yermack (1996) for evidence
on board size and governance. We dene SmallBoard to be
an indicator variable that is one if the number of directors
on the rms board is less than or equal to the median,
and zero otherwise. In column 2, we interact HomeLev
with this board governance measure. The interaction
effect is 0.0711 and statistically signicant at the 5%level, so more efcient board governance seems to reduce
27
We do not have sufcient cross-sectional variation in our 2004
sample to study board independence (because of the Sarbanes-Oxley Act
and the new listing rules by NYSE and Nasdaq).
37
Table 10
Corporate valuation effects.
The table reports coefcients and standard errors from regressing rm value (Q) on various determinants of rm value, using OLS estimation. Control
variables are constructed using 2003 data and Q is dened as 2004 Mktbk dened as in Table 1. AbsDev is the absolute value of the difference between the
predicted values from a standard corporate leverage model specication with and without HomeLev included. Fig. 2 displays a histogram of this variable.
AbsDevQ4 is an indicator variable which is one if the rm is in the quartile with the largest absolute deviation. Assets is the natural logarithm of total rm
assets. EBIT is a measure of protability and is dened as EBIT/Sales. CAPEX is a measure of capital expenditures and is dened as CAPEX/Sales. SPDum is a
dummy variable that is one if the rm was a member of the S&P 500 in 2004 and zero otherwise. Lev is 2003 TDM as dened in Appendix B. IndusQ is the
median Q value for the rms four-digit SIC industry for the universe of Compustat rms. Denitions for SmallBoard and GoodGov are found in Table 9. The
sample is non-nancial S&P 1,500 rms. The table reports White (1980) heteroskedasticity-consistent standard errors. Signicance levels are denoted by
n nn nnn
, ,
, which correspond to 10%, 5%, and 1% levels, respectively.
(1)
AbsDev
(2)
(3)
(4)
0.2618nn
(0.1305)
0.4112nnn
(0.1084)
0.0713
(0.1054)
0.3541n
(0.1957)
0.2754nnn
(0.1046)
7.1425
(6.2225)
AbsDevQ4
SmallBoard
AbsDevQ 4 SmallBoard
GoodGov
0.3638nnn
(0.0720)
0.0037
(0.0053)
0.7382
(0.6352)
1.1106nnn
(0.1887)
0.0126nnn
(0.0038)
0.5387nnn
(0.1209)
3.3817nnn
(0.6643)
0.3775nnn
(0.0742)
0.0035
(0.0052)
0.7704
(0.6178)
1.1220nnn
(0.1910)
0.0135nnn
(0.0036)
0.5327nnn
(0.1205)
3.3910nnn
(0.6170)
0.1896nnn
(0.0535)
0.9289nn
(0.3903)
0.7850n
(0.4147)
0.7640nnn
(0.1639)
1.7203nnn
(0.2934)
0.3908nnn
(0.1196)
2.4401nnn
(0.3784)
0.1468
(0.1158)
0.3909n
(0.2135)
0.1756nnn
(0.0462)
0.5119nnn
(0.1735)
1.0423nn
(0.4123)
0.7173nnn
(0.1573)
1.5329nnn
(0.2477)
0.3709nnn
(0.1069)
2.4421nnn
(0.3491)
0.2490
605
0.2520
605
0.2890
483
0.2850
535
AbsDevQ 4 GoodGov
Assets
EBIT
CAPEX
SPDum
Lev
IndusQ
Intercept
AdjR2
N
CEOs ability to imprint their specic preferences on their

rms.28
We also collect data from RiskMetrics on the G-index
by Gompers, Ishii, and Metrick (2003), and construct a
measure of good external governance (GoodGov), i.e., an
indicator variable that is one if the rm has a G-index
smaller than or equal to six, and zero otherwise.29 In
column 3, we interact HomeLev with this measure of
governance and nd that the interaction effect is negative
( 0.0253) as predicted, but not statistically signicant.30
28
It should be noted, however, that smaller boards are not uniformly the more efcient form for all types of rms, as shown in Coles,
Daniel, and Naveen (2008), though by design we have excluded nancial
rms which they suggest may benet from larger boards.
29
We use a slightly different cutoff than Gompers, Ishii, and Metrick
(2003). They use a G-index cutoff of ve in their paper to dene
Democracy rms. We have very few such observations in our sample.
However, 67 rms in our sample have a G-index of six or less. We
believe that our cutoff still captures the most well-governed rms,
according to the G-index measure, in our sample.
30
Interestingly, the coefcient on the GoodGov dummy variable is
estimated at 0.0341 and is signicant at the 10%-level. This nding is
indicative of one of the problems with using the G-index as a measure of
governance: the G-index identies young growth rms as bettergoverned rms, but young growth rms are poor candidates for high
Finally, Table 10 examines valuation effects. In column 1,

we regress Q on AbsDev and a set of control variables.31 We
nd that the estimated coefcient on AbsDev is negative, but
not statistically signicant.32 In column 2 we split AbsDev
into quartiles, and dene AbsDevQ4 to be an indicator
variable that is one if the rm is in the quartile with the
most extreme deviations, and regress Q on AbsDevQ4. The
estimated coefcient is negative and statistically signicant
at the 5%-level. In columns 3 and 4 we interact AbsDevQ4
with two previously dened governance measures, SmallBoard and GoodGov, respectively. In both cases the coefcient on the interaction between AbsDevQ4 and the
governance measure is positive and statistically signicant
at the 10% level, providing evidence that good corporate
(footnote continued)
leverage. If the G-index is not a measure of good governance, but rather
a measure of young growth rms, we nd that our results are robust to
controlling for young growth rms as evidenced by the positive and
statistically signicant coefcient on HomeLev in column 3 of Table 9.
31
Recall that AbsDev is the absolute value of the difference between
the predicted values from a standard corporate leverage model specication with and without HomeLev included.
32
See, e.g., Korteweg (2010) for estimates of the markets valuation
of corporate leverage deviations.
38
governance can reduce the value-destroying effect of CEOs

pushing their rms capital structures toward their preferred debt preferences. This evidence is consistent with
long-standing arguments (Jensen and Meckling, 1976) that
CEOs do not always choose capital structures with a valueenhancing level of debt and is comparable with some other
reported effects (e.g., Jung, Kim, and Stulz, 1996; Berger,
Ofek, and Yermack, 1997).
6. Conclusion
The scientic basis for the hypothesis examined in this
paper is an extensive set of well-cited studies on behavioral
consistency theory, i.e., the notion that individuals tend to
exhibit consistent behaviors across different situations.
While behavioral consistency has the potential of explaining, at least in part, a broad set of corporate nance
decisions, in this paper we take a rst step by examining
the relation between CEO personal and corporate leverage.
We nd that this is a promising empirical approach because
rms behave consistently with how their CEOs behave
personally, at least in the context of leverage choices.
We nd that CEOs personal debt preferences carry
over to the corporate domain so that CEOs who do not
seem to like debt personally manage rms with signicantly less corporate leverage, all else equal. In terms of
explaining the variation in observed capital structure,
personal leverage is comparable to several standard
determinants of capital structure such as rm size and
protability, and explains more of the cross-sectional
variation in capital structures than does any one of a
dozen personal CEO characteristics. One economic
mechanism behind our results appears to be endogenous
matching of CEOs to rms, whereby rms seeking, e.g.,
conservative capital structures match with top executives
with similar debt preferences. We also nd support for an
alternative mechanism. Particularly when corporate governance is weak, we nd that CEOs appear to more
signicantly imprint their personal debt preferences on
corporate capital structures.
The broader contribution of our paper is to show
empirically that personal behaviors of CEOs can be a
valuable basis to predict corporate nancial behavior of
the rms they manage. In other elds of economic research,
Heckman and Rubinstein (2001) and Heckman, Stixrud, and
Urzua (2006) show the predictive power of personal characteristics for non-nancial economic outcomes. It is therefore a fruitful avenue for future research to examine
additional questions related to CEOs personalities, personal
characteristics, and corporate decision-making.
Appendix A. Sample construction
The starting point is all CEOs of the largest U.S. public
rms, the Standard and Poors (S&P) 1,500 set of rms, in
2004. We identify the CEOs using the ExecuComp database.
There are 1,699 CEOs as the index was revised during the
year. We drop all nancial and utility rms (SIC codes 4813,
4911, 4931, 6020, 6311, and 6331) as they are subject to
capital structure regulations (339 rms), and nine rms
headquartered outside the U.S. because of lack of real estate
data. Data availability varies across states. The following

states do not provide public records of mortgages or other
data required for the computation of our personal leverage
measure: Alaska, Delaware, Iowa, Idaho, Indiana, Kansas,
Louisiana, Maine, Minnesota, Missouri, Mississippi, Montana, Nebraska, New Hampshire, Nevada, Oklahoma, Oregon, Texas, Utah, and Wisconsin. As a result, the sample is
reduced to 1,003 CEOs.
We hand-collect data on these CEOs primary homes
and mortgages using several data sources and following
Liu and Yermack (2007). We mainly use the Nexis online
database of public records, www.nexis.com/research. In
this database, we are able to search tax assessment, deed
transfer, and mortgage records for each CEO. We supplement these data by searching various county assessor,
auditor, and recorder Web sites. For each CEO, we start by
performing a name search using the rst name, middle
initial, and last name. We restrict this search to individuals with age 71 year of the CEOs age because some of
them have common rst and last names. Most of the CEOs
and their residences were identied in this manner. In a
few cases where there are several individuals with exactly
the same name and age, we use Security and Exchange
Commission (SEC) lings and voter registration records to
try to identify the CEOs home. For example, there are
eight CEOs with the last name Smith in our database,
and we are able to identify the primary home for six of
them. For estate planning, tax, or other reasons, a trust is
sometimes recorded as owner of the CEOs home. When
the trust has a different name than the CEO, he or his
spouse are recorded as sellers of the property or as
trustees and thus are still in the database. In addition,
listed on some records may be the name of the CEOs
spouse, commonly with the label Husband and Wife.
Spousal names may in some cases be found in the rms
SEC lings. Additional records are in some cases located
through a search based on these trust or spousal names.
In cases of intra-family real estate transactions, we search
until we nd an arms-length transaction.
We focus on a CEOs primary home, as it constitutes
the vast majority of most CEOs real estate holdings. In
many cases, the primary residence is listed as Owner
Occupied. Listed on all records is the mailing address for
tax purposes, which is often the CEOs primary home
address. If a CEO owns multiple homes in the area of the
corporate headquarters, then we classify the largest
property as the primary home provided we do not nd
information from other data sources suggesting otherwise. By their specic location, some homes are determined to be recreation homes or the like, such as a golf
community condominium. By reviewing all the records
for a CEO, we are able to determine the primary residence
of 757 CEOs (75.4% of the sample). Once all primary home
and mortgage records are located, we collect data on the
purchase price of each CEOs most recent primary home,
as well as details regarding mortgages and renancings.
(We recorded executive loans from the company, but
found them to be very rare in our sample, probably
because Sarbanes-Oxley (SOX) bans such loans to CEOs:
from the 514 pre-2002 CEO home purchases, we found
only ve such loans.) Only 10.6% of the CEO homes are
39
Table A1
Summary statistics: CEOs primary homes.
The table reports summary statistics for characteristics of primary homes of CEOs for a sample of S&P 1,500 rms. Data on CEO home characteristics
were collected primarily from the Lexis-Nexis public documents database, which includes national coverage of mortgage records, deed transfers, and tax
assessor records. The U.S. median data are tabulated from 2005 data provided by the Federal Housing Finance BoardPeriodic Summary Tables and the
2005 American Community Survey Subject Tables. Purchase prices are reported in 2005 home price dollars, and adjusted using the Ofce of Federal
Housing Enterprise Oversights National Home Price Index.
Home size (sq ft)

Land size (acres)
Year built
Total rooms
Bedrooms
Bathrooms
Purchase price ($1000s)
MED
MEAN
STD
MIN
MAX
5,154
1.1
1989
10.0
4.0
5.0
1,585
5,658
3.4
1975
10.9
4.5
5.1
2,155
2,852
9.7
34
3.5
1.4
2.0
1,929
785
0.1
1740
5.0
0.0
1.0
114
22,371
140.0
2008
36.0
16.0
17.0
14,643
647
604
676
396
520
622
641
U.S. MED
1975
5.0
3.0
new construction homes. These observations have more

complications when it comes to determining the purchase
price for the home in addition to the land. For new
construction homes, we use as the purchase price, the
cost of the land plus the construction cost, when
available, and otherwise the improvement value as
stated in assessment records.
Table A1 reports descriptive statistics for the database
of CEOs primary homes. The number of observations (N)
varies across variables because property records are
sometimes incomplete. We nd that the median CEO
home is large at 5,154 square feet and was built on 1.1
acres of land in 1989. We coded condominiums as having
zero land size and they are therefore not included in the
land size statistics. Land sizes close to zero are townhouses. The median CEO home has ten rooms, whereof
four are bedrooms, and in addition, there are ve bathrooms. There is signicant variation in home size because
the standard deviation is 2,852 sq. ft. All of the distributions of house size or estimated market values are somewhat skewed to the right.
The table also reports data on purchase prices. The
median CEO purchase price is $1.585 million in 2005
home price dollars. The purchase prices have been
adjusted using the Ofce of Federal Housing Enterprise
Oversights National Home Price Index. Current market
values are very difcult to estimate without actual real
estate transactions for the properties, in particular for
high-priced CEO homes for which the market is illiquid
and there are not many reasonable benchmark transactions. In the last column of the table, we compare the CEO
homes to those of the median U.S. household based on
data from the Bureau of Census 2005 American Community Survey. These data show that the median home in the
U.S. has ve rooms, whereof three are bedrooms. At the
median, these homes are 14 years older (built in 1975)
than a CEOs primary home.
found but if any one of the mortgage amount, purchase

price, or the improvement cost (if the home is new
construction) is not available, then HomeLev is set to
missing.
Mortgage (Mort): An indicator variable that is one if the
CEO uses a mortgage at the time of the purchase of his
primary residence, and zero otherwise.
Mortgage or renancing (MortRe): An indicator variable that is one if the CEO uses a mortgage at the time of
purchase or any other time, and zero otherwise.
Zero leverage (ZeroPersLev): An indicator variable that
is one if the CEO never used any mortgage, revolving
credit home equity lines/loans, or other forms of shortterm debt home nancing, and zero otherwise.
Total home leverage (TotHomeLev): Is the valueweighted average of the loan-to-value ratios of all homes
owned by the CEO simultaneously to his primary residence (including the primary residence). Observations for
which the loan-to-value ratio cannot be computed for all
homes are set to missing. The same rules for computing
HomeLev apply to these ratios.
Appendix B. Variable denitions
B.3. Control variables
B.1. Measures of personal leverage
Market-to-book ratio (Mktbk): Mktbk is the ratio of

market value of assets (MVA) to Compustat item 6, assets.
MVA is obtained as the sum of the market value of equity
(Compustat item 199, price-close Compustat item 54,
shares outstanding)Compustat item 34, debt in current
Home leverage (HomeLev): The sum of the primary and

secondary mortgage liens, at the time of the home
purchase, divided by the purchase price. If a mortgage is
B.2. Measures of corporate leverage

Total debt/market value of assets (TDM): TDM is the ratio
of total debt (Compustat item 34, debt in current liabilitiesCompustat item 9, long-term debt) to MVA, market
value of assets. MVA is obtained as the sum of the market
value of equity (Compustat item 199, price-close
Compustat item 54, shares outstanding)Compustat item
34, debt in current liabilitiesCompustat item 9, longterm debt Compustat item 10, preferred-liquidation
value, Compustat item 35, deferred taxes and investment tax credit.
Total debt/assets (TDA): TDA is the ratio of total debt
(Compustat item 34, debt in current liabilitiesCompustat item 9, long-term debt) to Compustat item 6, assets.
40
liabilitiesCompustat item 9, long-term debtCompustat

item 10, preferred-liquidation value, Compustat item 35,
deferred taxes and investment tax credit.
Log of assets (Assets): Assets is the log of Compustat
item 6, assets.
Protability (Prot): Prot is the ratio of Compustat
item 13, operating income before depreciation, to Compustat item 6, assets.
Tangibility (Tang): Tang is the ratio of Compustat item
8, net property, plant, and equipment, to Compustat item
6, assets.
Median industry leverage (IndusLev): IndusLev is the
median of total debt to market (book) value of assets
(TDM) ((TDA)) by four-digit SIC code. The use of either
TDM or TDA corresponds with the dependent variable in
the regression analysis.
References
Allport, G.W., 1937. Personality: A Psychological Interpretation. H. Holt
& Company, New York, NY.
Allport, G.W., 1966. Traits revisited. American Psychologist 21, 110.
Altman, E., 1968. Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. Journal of Finance 23, 589609.
Barsky, R.B., Juster, F.T., Kimball, M.S., Shapiro, M.D., 1997. Preference
parameters and behavioral heterogeneity: an experimental
approach in the health and retirement study. Quarterly Journal of
Economics 112, 537579.
Berger, P., Ofek, E., Yermack, D., 1997. Managerial entrenchment and
capital structure decisions. Journal of Finance 52, 14111438.
Bertrand, M., Schoar, A., 2003. Managing with style: the effect of
managers on rm policies. Quarterly Journal of Economics 118,
11691208.
Borghans, L., Duckworth, A.L., Heckman, J.J., Weel, B., 2008. The economics and psychology of personality traits. Unpublished NBER working
Paper No. 13810.
Cadenillas, A., Cvitanic, J., Zapatero, F., 2004. Leverage decision and
manager compensation with choice of effort and volatility. Journal of
Financial Economics, 7192.
Chava, S., Purnanandam, A.K., 2010. CEOs vs. CFOs: incentives and
corporate policies. Journal of Financial Economics 97, 263278.
Chyz, J.A., 2010. Personally tax aggressive managers and rm level tax
avoidance. Unpublished working paper, University of Arizona.
Coles, J., Daniel, N., Naveen, L., 2008. Boards: Does one size t all. Journal
of Financial Economics 87, 329356.
Endler, N.S., Magnusson, D., 1976. Toward an interactional psychology of
personality. Psychological Bulletin 83, 956974.
Epstein, S., 1979. The stability of behavior: on predicting most of the
people much of the time. Journal of Personality and Social Psychology 37, 10971126.
Epstein, S., 1980. The stability of behavior. American Psychologist 35,
790806.
Fahlenbrach, R., 2009. Founder-CEOs, investment decisions, and stock
market performance. Journal of Financial and Quantitative Analysis
44, 439466.
Fama, E.F., French, K.R., 1997. Industry costs of equity. Journal of
Financial Economics 43, 153193.
Frank, M.Z., Goyal, V.K., 2009a. Capital structure decisions: Which factors
are reliably important? Financial Management 38, 137.
Frank, M.Z., Goyal, V.K., 2009. Corporate leverage: How much do
managers really matter? Unpublished working paper, University of
Minnesota and Hong Kong University of Science and Technology.
Funder, D.C., Colvin, C.R., 1991. Explorations in behavioral consistency:
properties of persons, situations, and behaviors. Journal of Personality and Social Psychology 60, 773794.
Gilson, S.C., 1989. Management turnover and nancial distress. Journal
of Financial Economics 25, 241262.
Gilson, S.C., 1990. Bankruptcy boards, banks and block-holders: evidence

on changes in corporate ownership and control when rms default.
Journal of Financial Economics 27, 355387.
Gilson, S.C., Vetsuypens, M.R., 1993. CEO compensation in nancially
distressed rms: an empirical analysis. Journal of Finance 48,
425458.
Gompers, P.A., Ishii, J., Metrick, A., 2003. Corporate governance and
equity prices. Quarterly Journal of Economics 118, 107155.
Graham, J.R., Harvey, C.R., Puri, M., 2009. Managerial attitudes and
corporate actions. Unpublished working Paper, Duke University.
Hackbarth, D., 2008. Managerial traits and capital structure decisions.
Journal of Financial and Quantitative Analysis 43, 843882.
Heckman, J.J., Rubinstein, Y., 2001. The importance of noncognitive
skills: lessons from the GED testing program. American Economic
Review 91, 145149.
Heckman, J.J., Stixrud, J., Urzua, S., 2006. The effects of cognitive and
noncognitive abilities on labor market outcomes and social behavior.
Journal of Labor Economics 24, 411482.
Hong, H., Kostovetsky, L. Red and blue investing: values and nance.
Journal of Financial Economics, in press. doi:10.1016/j.jneco.2011.
01.006.
Hutton, I., Jiang, D., Kumar, A., 2010. Do Republican managers adopt
conservative corporate policies? Unpublished working paper, Florida
State University and University of Miami.
Jensen, M., Meckling, W., 1976. Theory of the rm: managerial behavior,
agency costs, and capital structure. Journal of Financial Economics 3,
305360.
Jung, K., Kim, Y.-C., Stulz, R.M., 1996. Timing, investment opportunities,
managerial discretion, and the security issue decision. Journal of
Korteweg, A.G., 2010. The net benets to leverage. Journal of Finance 65,
21372170.
Leary, M.T., Roberts, M.R., 2005. Do rms rebalance their capital
structures? Journal of Finance 60, 25752619.
Lemmon, M.L., Roberts, M.R., Zender, J.F., 2008. Back to the beginning:
persistence and the cross-section of corporate capital structure.
Journal of Finance 63, 15751608.
Liu, C.H., Yermack, D., 2007. Where are the shareholders mansions?
CEOs home purchases, stock sales, and subsequent company performance. Unpublished working paper, Cornell University and New
York University.
Loughran, T., 2008. The impact of rm location on equity issuance.
Financial Management 37, 121.
Malmendier, U., Tate, G., 2005. CEO overcondence and corporate
investment. Journal of Finance 60, 26612700.
Malmendier, U., Tate, G., Yan, J., 2010. Managerial beliefs and corporate
nancial policies. Unpublished NBER Working Paper No. 15659.
Mischel, W., 1968. Personality and Assessment. Wiley, New York, NY.
Modigliani, F., Miller, M., 1958. The cost of capital, corporation nance
and the theory of investment. American Economic Review 48,
261297.
Opler, T.C., Titman, S.D., 1994. Financial distress and corporate performance. Journal of Finance 49, 10151040.
Parsons, C.A., Titman, S., 2008. Capital structure and corporate strategy.
In: Eckbo, B.E. (Ed.), Handbook of Corporate Finance: Empirical
Corporate Finance, Handbooks in Finance Series, vol. 2. Elsevier,
North-Holland, pp. 203234 (Chapter 13).
Slovic, P., 1972a. Information processing, situation specicity, and the
generality of risk-taking behavior. Journal of Personality and Social
Psychology 22, 128134.
Slovic, P., 1972b. Psychological study of human judgment: implications
for investment decision making. Journal of Finance 27, 779799.
Warner, J.B., Watts, R.L., Wruck, K.H., 1988. Stock prices and top
management changes. Journal of Financial Economics 20, 461492.
Weisbach, M.S., 1988. Outside directors and CEO turnover. Journal of
White, H., 1980. A heteroskedasticity-consistent covariance matrix
estimator and a direct test for heteroskedasticity. Econometrica 48,
817838.
Yermack, D.L., 1996. Higher market valuation of companies with a small
board of directors. Journal of Financial Economics 40, 185211.

Journal of Financial Economics

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Journal of Financial Economics

Hochgeladen von

Copyright:

Verfügbare Formate

Journal of Financial Economics 103 (2012) 2040

Contents lists available at SciVerse ScienceDirect

Journal of Financial Economics

Behavioral consistency in corporate nance:

I just dont like to owe money.

Fatma Sonmez, Meir Statman, Per Stromberg,

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

nancial economists have devoted signicant effort to

evidence is ambiguous. For example, Bertrand and

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

the overall indebtedness of the CEO. Based on behavioral

numerous robustness checks, including using personal

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

2.1. Behavioral consistency

not employed by another public rm for at least three

3. Personal leverage of CEOs in the U.S.

Number of CEO home purchases

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

We nd that in terms of both corporate leverage and rm

3.3. Determinants of CEOs personal leverage

1970 1975 1980 1985 1990 1995 2000 2004 2008

3.2. Summary statistics

Why do some CEOs have a higher personal home

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

Mortgage usage at purchase

compensation in the form of non-tax deferrable income

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

4.1. Regression results

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

4.1.2. Industry xed effects

CEOs during three years prior to 2004, in total 84

4.1.6. Robustness checks

4.1.4. Excess personal leverage

This result is robust to including year xed effects (untabulated).

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

4.2. Economic magnitude of estimated effect

4.3. Persistence in corporate leverage

The economic magnitude of the estimated effect is

Fig. 2. Absolute deviations from predicted corporate leverage. The gure

we control for the previous years leverage.21 The overall

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

signicantly different than those who are levered at some

capital structures compared to CEOs who do not use any

We do not predict a perfect positive correlation ( 1) between

Because we measure corporate leverage in year 2004,

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

the timing of CEOs personal home leverage choices to

4.6. CEO characteristics

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

papers.23 We also collect data on measures of CEO wealth

characteristics, and nd several to be signicant: CEOs

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

changes in the control variables, the corporate leverage in

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

In column 3, we decompose the change in home

dene IncentPay as the CEOs total compensation minus

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

CEOs ability to imprint their specic preferences on their

Finally, Table 10 examines valuation effects. In column 1,

H. Cronqvist et al. / Journal of Financial Economics 103 (2012) 2040

governance can reduce the value-destroying effect of CEOs