Content uploaded by Maurice D. Levi
Author content
All content in this area was uploaded by Maurice D. Levi on Apr 16, 2014
Content may be subject to copyright.
Antitakeover Provisions, Managerial Incentives,
and Firm Value ∗
Feng Zhang†
Sauder School of Business
University of British Columbia
January 28, 2011
Abstract
This paper investigates the interaction between antitakeover provisions and
managerial ownership, two corporate governance mechanisms. Antitakeover provi-
sions weaken the incentive effect of managerial ownership and magnify its entrench-
ment effect, and thus will decrease the effect of managerial ownership on firm value.
I show that the effect of managerial ownership on firm value crucially depends on
the strength of antitakeover provisions. For firms with weak antitakeover provi-
sions, managerial ownership enhances firm value, unless the managers have very
high ownership. For firms with strong antitakeover provisions, however, increas-
ing managerial ownership always destroys firm value. Antitakeover provisions also
decrease the announcement returns of manager share purchases in the open mar-
ket. Because of the negative impact of antitakeover provisions on the value effect
of managerial ownership, firms with stronger antitakeover provisions have signif-
icantly lower managerial ownership. The findings are robust to various measures
of managerial ownership, different estimation methods, and endogeneity concerns.
The evidence reported in this paper is consistent with the hypotheses that anti-
takeover provisions decrease the value effect of managerial ownership and affect the
manager’s compensation contract.
∗I thank Rob Heinkel, Thomas Hellmann, and Kai Li for many insightful comments and guidance. In
addition, I am grateful to Jan Bena, Jason Chen, Glen Donaldson, Ron Giammarino, Alan Kraus, Ali
Lazrak, Mike Lemmon, Maurice Levi, Hernan Ortiz-Molina, Ning Tang, seminar participants at UBC,
University of Utah, University of Toronto, and conference participants at the 2010 Northern Finance
Association meetings for their comments and suggestions. Financial support from the Canadian Founda-
tion for Governance Research’s Robert Bertram Doctoral Research Awards is gratefully acknowledged.
All remaining errors are mine.
†Sauder School of Business, University of British Columbia, 2053 Main Mall, Vancouver,
BC, Canada V6T 1Z2. Email: feng.zhang@sauder.ubc.ca; Phone: +1-778-847-2158; Webpage:
http://finance.sauder.ubc.ca/∼zhang.
Antitakeover Provisions, Managerial Incentives,
and Firm Value
Abstract
This paper investigates the interaction between antitakeover provisions and
managerial ownership, two corporate governance mechanisms. Antitakeover provi-
sions weaken the incentive effect of managerial ownership and magnify its entrench-
ment effect, and thus will decrease the effect of managerial ownership on firm value.
I show that the effect of managerial ownership on firm value crucially depends on
the strength of antitakeover provisions. For firms with weak antitakeover provi-
sions, managerial ownership enhances firm value, unless the managers have very
high ownership. For firms with strong antitakeover provisions, however, increas-
ing managerial ownership always destroys firm value. Antitakeover provisions also
decrease the announcement returns of manager share purchases in the open mar-
ket. Because of the negative impact of antitakeover provisions on the value effect
of managerial ownership, firms with stronger antitakeover provisions have signif-
icantly lower managerial ownership. The findings are robust to various measures
of managerial ownership, different estimation methods, and endogeneity concerns.
The evidence reported in this paper is consistent with the hypotheses that anti-
takeover provisions decrease the value effect of managerial ownership and affect the
manager’s compensation contract.
1 Introduction
As essential parts of a corporate governance system, antitakeover provisions and manage-
rial ownership work together to affect governance in a firm. This study investigates their
interactions in affecting corporate governance and firm value. The specific questions of
interest are as follows. Do antitakeover provisions affect the incentives generated by man-
ager shareholdings? If so, how should shareholders adjust the manager’s compensation
contract accordingly? How do antitakeover provisions and managerial ownership interact
in affecting firm value? No paper to my knowledge has examined the combined effect
that managerial ownership and antitakeover provisions have on corporate governance and
firm value.1This study attempts to fill the void.
There are two channels of interaction between antitakeover provisions and managerial
ownership. First, antitakeover provisions and managerial ownership magnify each other’s
entrenchment effect. The voting power of managers’ shareholdings helps entrench the
managers (Stulz, 1988). Anittakeover provisions may also entrench the managers by
protecting them from takeovers. These two entrenchment mechanisms could complement
each other. For instance, before the passage of the second-generation antitakeover laws
in the 1980s, an acquirer can purchase a significant proportion of a firm’s shares and
then dismiss the firm’s managers with the voting power of the acquired shares, even if
the managers also own non-trivial ownership. The antitakeover laws usually require that
the acquirer must receive the approval of a supermajority (usually 75% or 80%) of the
target firm’s shareholders before the acquirer purchases significant amount of the target’s
shares. The supermajority requirement makes the target managers’ voting power more
pivotal, and thus makes it easier for the managers to block the acquisition with their
1The literature has examined related but different questions. Stein’s (1988) model shows that anti-
takeover provisions encourage managers to commit to long-term investments. Bertrand and Mullainathan
(1999, 2000) and Cheng and Indjejikian (2009) study how the second-generation state takeover laws affect
the level and structure of executive compensation.
1
voting power.
The second channel of interaction is that antitakeover provisions weaken the incentive
effect of managerial ownership. The intuition is as follows. When the manager exerts less
effort, the stand-alone value of the firm is lower and so acquiring this firm will generate a
greater potential synergy gain. That is, less managerial effort leads to greater synergies.
When the target firm is able to seize more of the synergy with the help of antitakeover
provisions, its manager has incentives to increase the size of the synergy by reducing effort,
ceteris paribus. This implies that antitakeover provisions reduce managerial effort; or put
another way, since managerial ownership leads to increased manager effort, antitakeover
provisions weaken the incentive effect of managerial ownership.
Taken together, antitakeover provisions magnify the entrenchment effect of manage-
rial ownership and weaken its incentive effect. These two channels of interaction predict
that antitakeover provisions will decrease the effect of managerial ownership on firm
value.
Employing a sample of almost 15,000 firm-year observations over the period 1992-
2007, I show that antitakeover provisions significantly, both economically and statistically,
reduce the effect of managerial ownership on firm value. The strength of antitakeover pro-
visions is proxied by the entrenchment index (E index hereafter) constructed by Bebchuk,
Cohen, and Ferrell (2009) using six antitakeover provisions. A greater E index indicates
stronger antitakeover provisions. Each additional antitakeover provision in the E index
reduces the marginal effect of managerial ownership on firm value by approximately 30
percent. The impact of antitakeover provisions is so large that the effect of managerial
ownership on firm value becomes negative when the E index is above the median. On
the other hand, managerial ownership enhances firm value when the E index is below
the median, unless the managers have very high levels of ownership. These results are
robust to a battery of robustness checks, including alternative measures of managerial
2
ownership and different econometric estimation methods. In addition, the antitakeover
provisions in the E index decrease the value effect of managerial ownership not only in
aggregate but also individually.
Aware of the negative impact of antitakeover provisions on the value effect of man-
agerial ownership, shareholders should grant fewer shares to the manager when there
are stronger antitakeover provisions. I show that managerial ownership decreases signifi-
cantly with the strength of antitakeover provisions. On average, the combined ownership
of the top five executives is 7.6% in the firms with the weakest antitakeover provisions,
and 2.1% in the firms with the strongest antitakeover provisions.
The empirical results are prone to endogeneity concerns because, arguably, both the
level of managerial ownership and the strength of antitakeover provisions are endoge-
nously determined. The last part of the paper deals with potential endogeneity issues
by controlling for firm fixed effects, conducting an event study, running instrumental
variable regressions, and excluding alternative explanations. The results of these analy-
ses uniformly support that antitakeover provisions weaken the value effect of managerial
ownership. For example, the event study shows that antitakeover provisions decrease the
announcement returns of manager share purchases in the open market.
This paper relates to the literature on managerial ownership, which documents an
inverse U-shaped relation between managerial ownership and firm value.2I show that the
inverse U-shaped relation holds only when antitakeover provisions are weak. With strong
antitakeover provisions, increasing managerial ownership always destroys firm value.
This paper also contributes to a growing literature on the interactions between corpo-
rate governance mechanisms. Cremers and Nair (2005); Cremers, Nair, and Wei (2007);
Huson, Parrino, and Starks (2001); and Kini, Kracaw, and Mian (2004) study how the
2See, among others, Morck, Shleifer, and Vishny (1988), McConnell and Servaes (1990), Hermalin
and Weisbach (1991), Kole (1995), and Holderness, Kroszner, Sheehan (1999), McConnell, Servaes, and
Lins (2008), and Fahlenbrach and Stulz (2009).
3
market for corporate control interacts with the board of directors and/or institutional in-
vestor monitoring. Cohn and Rajan (2010) model the interaction between the board and
activist investors. Giroud and Mueller (2010a, 2010b) study the interaction between anti-
takeover provisions and product market competition. Kim and Lu (2010) study whether
and how CEO ownership interacts with product market competition and institutional in-
vestor monitoring. Bertrand and Mullainathan (1999, 2000), and Cheng and Indjejikian
(2009) study how the second-generation state takeover laws affect the level and structure
of executive compensation. This paper studies the impact of antitakeover provisions on
managerial incentives and the manager’s compensation contract.
The rest of the paper is organized as follows. Section 2 presents a simple model to an-
alyze the impacts of antitakeover provisions on managerial incentives and the manager’s
compensation contract. Section 3 describes the data and econometric methodology. Sec-
tion 4 presents the empirical results and implements various robustness checks. Section
5 addresses potential endogeneity issues. Finally, section 6 concludes the paper.
2 Antitakeover Provisions and Managerial Incentives
2.1 Antitakeover Provisions and the Compensation Contract
The shareholders hire a professional manager to run their firm. They compensate the
manager with a salary, s, and a certain amount of ownership, α∈[0,1].3The shareholders
are assumed to be risk neutral.
The manager decides how much effort to exert given the compensation contract.
The effort exerted by the manager is neither observed by the shareholders, nor is it
contractible. The manager’s effort is denoted as e, and is costly to the manager. For
3This compensation contract may not be the optimal one. A linear contract is the second-best
solution under certain conditions, as shown by Holmstr¨om and Milgrom (1987). This linear contract
is used for its simplicity. A complex compensation contract would make it difficult to highlight the
economic intuition.
4
simplicity, the cost-of-effort function is assumed to be quadratic: C(e) = 1
2ke2, where
k > 0. A greater kindicates a lower cost of effort. The firm’s stand-alone value is equal
to the manager’s effort plus a noise: φ=e+ǫ, where ǫis normally distributed with zero
mean and variance σ2. The manager has a mean-variance risk preference represented by
the following utility function:
u(w, e) = E[w−C(e)] −η
2var(w),(1)
where wis the amount of compensation the manager receives, E(.) and var(.) are the
mean and variance operators, respectively, and η > 0 is the coefficient of risk aversion.4
The manager’s reservation utility is ¯u, which is known to the shareholders.
The above model setup is standard in the literature on linear compensation con-
tracts (Bolton and Dewatripont, 2005). New to the literature, this paper introduces
antitakeover provisions to the model and explores how they affect the contract and man-
agerial incentives. Antitakeover provisions are assumed to be exogenously given. This
simplifies the algebra and thereby allows me to clearly demonstrate the impact of anti-
takeover provisions on managerial incentives. Antitakeover provisions are endogenized in
Appendix A.
A raider tries to acquire the target firm mentioned before at the end of the model.
The value of the target firm to the raider is: τ= ¯v+ǫ, where ǫis the same noise that
affects the stand-alone value of the target firm, and ¯vis the intrinsic value of the target
firm to the raider.5Thus, the synergy gain associated with the takeover is: τ−φ= ¯v−e.
How the synergy is divided between the raider and the target depends on the target’s
bargaining position, which in turn is determined by the strength of the target firm’s
4When the amount of compensation follows a normal distribution, this mean-variance utility function
is equivalent to the constant absolute risk-averse (CARA) utility function with coefficient of absolute
risk aversion η:u(w, e) = −exp {−η[w−C(e)]}.
5The results are not affected if ¯vis assumed to depend on eas long as the synergy decreases with e.
The results are also not affected if τdoes not depend on ǫ.
5
antitakeover provisions. Stronger antitakeover provisions enable the target shareholders
to form a cartelized response to the takeover bid and therefore enhance the target’s
bargaining position. That is, stronger antitakeover provisions enable the target to seize a
larger share of the synergy. Section 2.2 discusses how antitakeover provisions enhance the
target’s bargaining position in more detail. Suppose the target’s antitakeover provisions
allow it to seize pof the synergy. In other words, the raider will pay a bid premium
p(¯v−e) to the target’s shareholders.
The shareholders choose a compensation contract (sand α) to maximize their ex-
pected wealth, subject to the manager’s optimal choice of effort and the manager’s par-
ticipation constraint:
max
s,α ψ= (1 −α)[e+p(¯v−e)] −s(2)
s.t. e ∈arg max s+α[e+p(¯v−e)] −1
2ke2−η
2α2σ2(3)
s+α[e+p(¯v−e)] −1
2ke2−η
2α2σ2≥¯u. (4)
The optimal managerial ownership and the optimal managerial effort are:
α∗=k(1 −p)2
k(1 −p)2+ησ2,(5)
e∗=kα(1 −p) = k2(1 −p)4
k(1 −p)2+ησ2.(6)
Proposition 1. Antitakeover provisions have the following impacts on managerial in-
centives:
(1) Antitakeover provisions reduce managerial effort ( de∗
dp ≤0).
(2) Antitakeover provisions decrease the marginal effect of managerial ownership on
firm value ( ∂2π
∂α∂p ≤0,where πdenotes firm value).
6
(3) Managerial ownership decreases with the strength of antitakeover provisions (dα∗
dp ≤
0).
Proof. See Appendix B. Q.E.D.
The intuition behind Proposition 1 can be explained as follows. Less managerial
effort leads to a greater synergy gain, which is divided between the target and the raider.
Stronger antitakeover provisions raise the target’s share of the synergy, and thus encour-
age the manager to exert less effort because more of the synergy accrues to the target
with stronger antitakeover provisions. That is, antitakeover provisions reduce manage-
rial effort, ceteris paribus. A high managerial ownership incentivizes the manager to
exert more effort. Antitakeover provisions weaken this incentive effect by reducing man-
agerial effort. Consequently, the marginal effect of managerial ownership on firm value
decreases with the strength of antitakeover provisions. Aware of this negative impact
of antitakeover provisions on the incentive effect of managerial ownership, shareholders
grant fewer shares to the manager when there are stronger antitakeover provisions.
2.2 Antitakeover Provisions and Target Firms’ Bargaining Po-
sition
This section discusses how antitakeover provisions enhance a target firm’s bargaining
position against an acquirer, which is a key assumption in the model.
Suppose a raider launches a takeover bid to acquire up to 100% of the target firm’s
shares. The bid succeeds if more than 50% of shares are tendered. Grossman and Hart
(1980) suggest that a necessary condition for the bid to succeed is that the offer price
must be greater than the stock price of the target after the bid succeeds. Otherwise, the
free rider problem arises and no shareholder will tender.
The target’s shareholders as a group have an incentive to not tender their shares (hold
7
up) in order to push up the bid price. If the shareholders are able to form a cartel then
no shareholder will tender at a low price, and consequently they will be able to bargain
for a higher offer. However on an individual level, each shareholder has an incentive to
tender at a low price because the bid premium will be lost if he or she holds up while
other shareholders tender more than 50% of the total shares of the firm. In other words,
individual shareholders have an incentive to rush, while as a group they have an incentive
to hold up.
DeAngelo and Rice (1983) suggest that antitakeover provisions help the target’s
shareholders form a cartelized response to a tender offer and thus enhance their bargaining
position. For example, supermajority provisions raise the level of ownership that must be
purchased in order to ensure subsequent shareholder approval of a merger, and therefore
encourage shareholders to hold out for a higher price.
Another example of antitakeover provisions is poison pills, which make it almost
impossible for a bid to succeed without the approval of the target’s board. Poison pills
can be removed by the board without the approval of shareholders. Therefore, one
way to circumvent poison pills is to initiate a proxy fight in the hope of replacing the
directors who resist the acquisition. The proxy fight may not succeed given that many
firms have staggered boards, which would prevent the raider from replacing the directors
all at once and would thus increase the cost of the proxy fight. Hence, poison pills
and staggered boards force the raider to negotiate directly with the target’s directors
on the bid premium. Since there are fewer directors than shareholders, and since the
directors meet frequently, they can collude at a lower cost and therefore bargain for
better acquisition terms. This enhanced bargaining position will increase the proportion
of the synergy that accrues to the target’s shareholders, provided that the board is acting
in the shareholders’ interest.
The above discussions suggest that antitakeover provisions enhance the bargaining
8
position of the target firm, and therefore enable the target to seize a larger share of the
synergy. Consistent with these discussions, Comment and Schwert (1995) and Heron and
Lie (2006) find that antitakeover provisions significantly increase the bid premiums in
large samples of takeovers.
2.3 Testable Implications
In summary, the implications of antitakeover provisions for managerial incentives and
firm value are:
(1) Antitakeover provisions decrease the effect of managerial ownership on firm value.
(2) Managerial ownership decreases with the strength of antitakeover provisions.
In the sections that follow, I first empirically evaluate the above predictions in a
multivariate context; I then account for potential endogeneity concerns about managerial
ownership and antitakeover provisions.
3 Methodology and Data
3.1 Methodology
Following the literature (e.g., Morck, Shleifer, and Vishny, 1988), firm value is proxied
by Tobin’s Q. Managerial ownership is calculated as the percentage ownership of the top
five executives covered in the Standard and Poor’s ExecuComp database.6Restricted
stocks are included in the calculation, while stock options are excluded. Including stock
options does not qualitatively affect the results, as will be shown in section 4.
6The literature uses various measures of managerial ownership. For example, Morck, Shleifer, and
Vishny (1988) and Hermalin and Weisbach (1991) use the ownership of the board of directors; McConnell
and Servaes (1990) and Fahlenbrach and Stultz (2009) use the ownership of insiders (both top managers
and directors). Directors do not necessarily share the same interest as professional managers. This
paper explores how antitakeover provisions affect the incentives of professional managers. Therefore,
managerial ownership is measured with the ownership of the top five executives.
9
The strength of a firm’s antitakeover provisions is proxied by the E index constructed
by Bebchuk, Cohen, and Ferrell (2009) using six antitakeover provisions: staggered
boards, poison pills, supermajority requirement for mergers, limits to amend bylaws,
limits to amend charter, and golden parachutes. The value of the E index increases by
one for each antitakeover provision in place. Therefore, it takes a value from zero (for
the weakest antitakeover provisions) to six (for the strongest antitakeover provisions).
The following econometric model is employed to test the implication of antitakeover
provisions for vale effect of managerial ownership:
Qit =β0+β1×Managerial ownershipit +β2×Managerial ownership2
it
+β3×E indexit +β4×Managerial ownershipit ×E indexit
+β5×Xit +uit.(7)
where Qit is the Tobin’s Q of stock iin year t;Xit is a vector of control variables; and
uit is the residual. The interaction variable of managerial ownership and the E index
captures the impact of antitakeover provisions on the effect of managerial ownership
on firm value. Following Morck, Shleifer, and Vishny (1988) and others, the square of
managerial ownership is also added as an explanatory variable.7
Following Himmelberg, Hubbard, and Palia (1999), I control for possible determi-
nants of managerial ownership in the regression. Firms whose assets are difficult to
monitor should provide greater incentives to their managers in order to mitigate the
moral hazard problem. Therefore, the following variables are included as proxies for the
scope of managerial discretionary spending: capital-to-sales ratio, research and develop-
ment (R&D) spending, advertisement spending, investment, and profit margin. The first
7See Footnote 3 for the literature on the inverse U-shaped relation between managerial ownership and
firm value. In unreported results, an interaction variable of the E index and the square of managerial
ownership is also added on the right hand side of the regression. The coefficient on this interaction
variable is statistically indifferent from zero. Also, adding this interaction variable does not qualitatively
change the results of this paper.
10
three variables are related to asset tangibility; profit margin measures the gross cash flows
available for operation; and investment measures the scope for discretionary projects. All
else being equal, managers in firms with greater idiosyncratic risks prefer lower ownership
for the reason of diversification. Therefore, idiosyncratic risk is also included as a control
variable. In addition, I control for firm size as measured by sales. Descriptions of these
variables can be found in Appendix C.
3.2 Data
The data on managerial ownership are retrieved from Standard and Poor’s ExecuComp
database, which provides compensation data for the top five executives of the S&P 1500
companies. The data are collected directly from each company’s annual proxy since 1992.
For each fiscal year from 1992-2007, managerial ownership of each firm is calculated
as the total number of shares owned by the top five executives divided by the total
number of shares outstanding. Stock return data are retrieved from the CRSP database,
accounting numbers from the Compustat database, and corporate governance data from
the RiskMetrics database. Utilities (SIC codes 4900-4949) and financial firms (SIC codes
6000-6999) are excluded from the sample. Including them in the sample yields similar
results throughout. The final sample has 14,962 firm-year observations.
Table 1 Panel A presents the summary statistics of managerial ownership over the
sample period. There are 524 firms in the sample in 1992. The number of firms increases
to more than 700 during the period 1993-1997, and reaches 1,000 in 1998. The number
stays stable at around 1,000 firms during the period 1998-2001, and further increases to
more than 1,100 firms after 2001. On average, the top five executives own 3.9 percent
of their companies. The distribution of managerial ownership is positive-skewed with a
median of 0.8 percent. The mean managerial ownership is more than 4 percent in the
1990s and gradually decreases to 2.7 percent in 2007.
11
RiskMetrics collects data on antitakeover provisions for the S&P 1500 firms in the
years of 1990, 1993, 1995, 1998, 2000, 2002, 2004, and 2006. For the years when the
antitakeover provision data are not updated, the most recent data are used following
Bebchuk, Cohen, and Ferrell (2009). Table 1 Panel B presents the summary statistics of
the E index for the years when the antitakeover provision data are updated.8The mean
E index is around 2.3 in the 1990s and increases slightly to around 2.5 in the 2000s. The
median E index is 2 or 3 over the sample period. The summary statistics are similar to
those in Bebchuk et al. (2009).
Table 1 Panel C presents the frequencies of the six antitakeover provisions in the E
index. Golden parachutes, staggered boards, and poison pills are the most commonly
adopted provisions. On average, 62 percent of firms use golden parachutes, 59 percent
use staggered boards, and 59 percent use poison pills. A supermajority requirement to
approve mergers and limits to amend bylaws are adopted by 18 percent of the firms.
Limits to amend charter is the least commonly adopted antitakeover provision among
the six: on average, only 2 percent of firms implement it. The frequencies of staggered
boards, poison pills, and limits to amend charter were stable during the sample period
with only small fluctuations over the years. The frequency of having a supermajority
requirement to approve mergers steadily decreased from approximately 40 percent in
1992 to less than 33 percent in 2006. On the contrary, limits to amend bylaws gained
popularity during this period—its frequency increased from 13 percent in 1992 to 19
percent in 2006. The most salient changes however happened to golden parachutes: the
proportion of firms that had golden parachutes increased from about 50 percent in 1992
to about 75 percent in 2006.
Table 2 Panel A presents the summary statistics of Tobin’s Q, managerial ownership,
the E index, and the control variables. The mean Tobin’s Q and industry-adjusted
8I also include the year of 1992, the first year of the sample period.
12
Tobin’s Q are 2.0 and 0.4, respectively. As shown in Table 1, the mean managerial
ownership is 3.9 percent; the mean E index is 2.4. The average firm has annual sales of
approximately $5 billion. On average, property, plant, and equipment (PPE) accounts
for 38 percent of annual sales. The mean idiosyncratic risk is about 2.4 percent, while the
mean profit margin is about 15 percent. On average, the sample firms spend 28 percent
of PPE on R&D, 9.5 percent on advertisement, and 23 percent on investment. Lastly,
the data on R&D spending and advertisement spending are available for 64 percent and
36 percent of the firms, respectively.
Before presenting the multivariate regression results, it is necessary to examine the
correlations between the variables, which are presented in Table 2 Panel B. Managerial
ownership is positively and significantly correlated with both Tobin’s Q and industry-
adjusted Tobin’s Q; while the E index is negatively and significantly correlated with
Tobin’s Q, industry-adjusted Tobin’s Q, and managerial ownership. The extent of the
correlation among most pairs of variables raises little concern for multicollinearity in the
regression analyses.
4 Empirical Results
4.1 Multivariate Regression Results
Table 3 presents the multivariate regression results of the model (7). Column (1) reports
the pooled OLS regression results. Column (2) controls for both industry fixed effects
and year fixed effects. The residuals in the regression may be correlated across firm or
over time, leading to over- or under-stated standard errors in the pooled OLS regressions.
Following the suggestions of Petersen (2009), I control for year fixed effects and cluster
the residuals by firm in column (3) in order to have robust standard errors. As an
alternative way to show the impact of antitakeover provisions on the effect of managerial
13
ownership on firm value, the sample is divided into two sub-samples based on the E
index: the firms in the first sub-sample have an E index between 0 and 2, while those in
the second sub-sample have an E index between 3 and 6.9Columns (4) and (5) of Table
3 present the regression results for these two sub-samples, respectively. In column (6),
the dependent variable is replaced with the industry-adjusted Tobin’s Q as an alternative
way to control for industry effects.
The results in column (1) indicate that managerial ownership is positively and signif-
icantly associated with Tobin’s Q at the one percent level; the coefficient on managerial
ownership squared is negative and statistically significant at the ten percent level; and
the E index is negatively and significantly associated with Tobin’s Q at the one percent
level. These findings are consistent with the previous literature. More interestingly, the
coefficient on the interaction variable of managerial ownership and the E index is negative
and statistically significant at the one percent level, suggesting that the marginal effect
of managerial ownership on firm value decreases with the E index. This interaction is
also economically significant. When the E index increases by one, the marginal effect of
managerial ownership on Tobin’s Q decreases by 0.552, which is approximately one third
of the coefficient on managerial ownership (1.635).
The results are robust when controlling for industry and year fixed effects and clus-
tering standard errors by firm: the results in columns (2), (3), and (6) are qualitatively
similar to those in column (1).
The results in columns (4) and (5) confirm that the effect of managerial ownership on
firm value depends significantly on the E index. The coefficient on managerial ownership
is positive and statistically significant at the five percent level in column (4), where the
firms have an E index from 0 to 2. It becomes negative and statistically significant at
the ten percent level in column (5), where the firms have an E index from 3 to 6.
9Figure 1 suggests that the effect of managerial ownership on firm value becomes negative for firms
with an E index above 2.
14
Figure 1 plots the fitted value of Tobin’s Q against managerial ownership at dif-
ferent E indexes. The fitted value of Tobin’s Q is generated using the regression re-
sults in column (3) of Table 3: Q= 3.574 + 1.858 ×Managerial ownership −1.360 ×
Managerial ownership2−0.044×E index−0.536×Managerial ownership×E index. Since
the top five executives in almost all firms have ownership below 20 percent, the figure
focuses on the relation between managerial ownership and Tobin’s Q over this empirical
range of managerial ownership.10 It shows that for the firms with an E index below 3,
firm value increases with managerial ownership over the empirical range of managerial
ownership. The effect of managerial ownership on firm value is indistinguishable from
zero for the firms with an E index of 3, and becomes negative as the E index reaches
above 3.11
Table 3 also shows that firms with smaller sales, less intangible assets, lower id-
iosyncratic risks, greater profit margins, more R&D spending, and more investments are
positively associated with Tobin’s Q. Moreover, sales squared and tangibility squared are
positively associated with Tobin’s Q.
4.2 Different Measures of Managerial Ownership
Stock options have been widely used to compensate executives since the early 1990s
(Hall and Liebman, 1998, and Aggarwal and Samwick, 2003). They are expected to
provide similar incentives for executives as stock grants. As a robustness check, in this
section I include stock options as part of managerial ownership. Usually the CEO makes
the most important corporate decisions and thus has the biggest impact on firm value.
Therefore as another robustness check, this section also singles out the CEO’s ownership
and investigates its effect on firm value.
10The 95th percentile of managerial ownership is just above 20 percent in the sample.
11Over the full range of managerial ownership (from 0 to 100 percent) there is an inverse U-shaped
relation between managerial ownership and Tobin’s Q for the firms with an E index below 3. For the
firms with an E index between 3 and 6, increasing managerial ownership always destroys firm value.
15
Pay-performance sensitivity (PPS) is commonly used to measure managers’ incen-
tives when both stocks and stock options are considered. For stocks, PPS is simply the
percentage stock ownership; for stock options, PPS equals the number of shares under-
lying the options times the delta of each option divided by the total number of shares
outstanding. The delta is defined as the partial derivative of the option value with re-
spect to stock price. Managerial PPS is computed following Guay (1999) and Core and
Guay (2002).
Table 4 Panel A reports the means of six measures of managerial ownership for
each year from 1992-2007. The six measures are managerial stock ownership, managerial
PPS of stock options, managerial PPS of both stocks and stock options, CEO stock
ownership, CEO PPS of stock options, and CEO PPS of both stocks and stock options.
The mean stock ownership of the top five executives was more than 4 percent in the
1990s and gradually decreased to 2.7 percent in 2007. In contrast with the declining
trend in managerial stock ownership, the mean managerial PPS of stock options gradually
increased from 0.8 percent in 1992 to 2.5 percent in the new millennium, and then slightly
decreased to 2.1 percent in 2007. The mean managerial PPS—calculated as the sum of
managerial stock ownership and managerial PPS of stock options—was 5 percent in 1992,
increased to 7 percent in the new millennium, and then decreased to 4.7 percent in 2007.
The CEO stock ownership was on average 2.9 percent in 1992, slightly increased to its
peak value of 3.2 percent in 1998, and then gradually decreased to 1.7 percent in 2007.
On average, the CEO PPS of stock options was merely 0.2 percent in 1992, gradually
increased to 1.2 percent in 2002, and then slightly decreased to 1.1 percent in 2007.
Finally, the mean CEO PPS was 3.1 percent in 1992, increased to more than 4.1 percent
during the “Internet Bubble” (1998-2000), and then decreased to 2.7 percent in 2007.
Table 4 Panel B reports the medians of the six measures of managerial ownership
for each year from 1992-2007. The medians are much lower than their respective means
16
in Panel A, but follow similar patterns as the means over the sample period. Taken
together, the period 1992-2007 has exhibited diminishing managerial stock ownership
and an increasing popularity in stock options as a component of executive compensation,
while the total managerial PPS remained stable over the sample period. The temporal
patterns observed in my sample are consistent with previous studies such as Hall and
Liebman (1998) and Aggarwal and Samwick (2003).
To assess the robustness of the results in Table 3, I estimate model (7) using three
alternative measures of managerial ownership: managerial PPS, CEO stock ownership,
and CEO PPS. The results are reported in Table 4 Panel C. All these three measures
of managerial ownership are positively associated with the industry-adjusted Tobin’s Q,
and all the associations are statistically significant. The coefficient on CEO stock own-
ership squared is negative and statistically significant at the ten percent level, whereas
the coefficients on the squares of the other two measures are negative but statistically
insignificant from zero. The E index is negatively and significantly associated with the
industry-adjusted Tobin’s Q in all three columns. In addition, the coefficients on the
interaction variables between the E index and the three measures of managerial owner-
ship are all negative and statistically significant at the one percent or five percent level.
Finally, the coefficients on the control variables have similar economic and statistical
significances across the three regressions. Overall, the results are robust to different
measures of managerial ownership.
4.3 Individual Antitakeover Provisions
The results in the last two sections suggest that the six antitakeover provisions in the
E index as a whole diminish the marginal effect of managerial ownership on firm value.
Does each individual antitakeover provision have the same effect? This section answers
this question.
17
Poison pills and staggered boards make it almost impossible for a takeover to succeed
without negotiating directly with the board of directors. The number of directors is much
smaller than the number of shareholders and the directors meet frequently. This makes
it easier for the board to form a cartelized response to the acquirer and thereby enhances
the target firm’s bargaining position. This enhanced bargaining position will increase
the proportion of the synergy that accrues to the target’s shareholders, and consequently
weakens the incentive effect of managerial ownership. Poison pills and staggered boards
may also magnify the entrenchment effect of antitakeover provisions. For example, a
staggered board allows the target firm’s managers to use their voting power to prevent
the acquirer from replacing the target directors who oppose the acquisition in more than
one year. Hence, poison pills and staggered boards are expected to decrease the effect of
managerial ownership on firm value.
Limits to amend bylaws and limits to amend charter usually require a supermajority
vote in order to pass an amendment. Both these types of limits as well as a superma-
jority requirement to approve mergers increase the size of the ownership stake that must
be purchased in order to ensure subsequent shareholder approval of the changes. From
the perspective of each individual shareholder, these supermajority provisions encour-
age shareholders to hold out for a higher price. Therefore, these three supermajority
provisions weaken the incentive effect of managerial ownership. These supermajority
provisions also magnify the entrenchment effect of managerial ownership by making the
managers’ voting power more pivotal. Hence, these three supermajority provisions are
expected to decrease the effect of managerial ownership on firm value.
Most managers lose their jobs after their firms are acquired, and fail to find a com-
parable job in other institutions (Hartzell, Ofek, and Yermack, 2004). This is probably
because they have firm-specific skills which are not desirable for other firms. Such per-
sonal losses could lead the target’s managers to resist profitable merger proposals. Golden
18
parachutes, by reducing their personal losses, may persuade them to agree with the ac-
quisition and thereby benefit the target’s shareholders. Harris (1990) shows that golden
parachutes increase the proportion of the synergy that accrues to the target firm using
the Nash bargaining solution. Hence, golden parachutes are expected to weaken the
incentive effect of managerial ownership.
To investigate the interactions between the six provisions and managerial ownership,
the E index in model (7) is replaced with dummy variables of the existence of each
antitakeover provision, one at a time:
Qit =β0+β1×Managerial ownershipit +β2×Managerial ownership2
it
+β3×Individual provisionit +β4×Managerial ownershipit ×Individual provisionit
+β5×Xit +uit,(8)
where Qit is the industry-adjusted Tobin’s Q of firm iin year t, and Individual provisionit
is a dummy variable that takes the value of one if firm ihas the provision in place in
year t, and zero otherwise.
Table 5 Panel A reports the regression results of model (8) for each of the six pro-
visions in the E index. To conserve space, only the coefficients and associated p-values
of the provisions and their interactions with managerial ownership are reported. The
coefficients on other variables are similar to those in Table 3. All six antitakeover provi-
sions reduce the marginal effect of managerial ownership on firm value, but the impacts
of poison pills and a supermajority requirement to approve mergers are statistically in-
significant. The coefficients on all six provisions are negative, but only those on golden
parachutes and poison pills are statistically significant. It is worth noting that staggered
board, limits to amend bylaws, and limits to amend charter significantly affect firm value
only through their interactions with managerial ownership.
19
The RiskMetrics database also collects information on eighteen other antitakeover
provisions in addition to the six provisions included in the E index. The twenty four
provisions constitute the G index developed by Gompers, Ishii, and Metrick (2003).
Bebchuk, Cohen, and Ferrell (2009) find that only the six provisions in the E index are
significantly associated with firm value, while the other eighteen provisions are mainly
“noises.” Their findings imply that managerial ownership will not interact significantly
with these eighteen provisions. To investigate this implication, I replace the E index in
model (7) with the G index and the difference between the G index and the E index,
respectively. The difference between the two indexes measures the strength of the eighteen
provisions other than the six provisions in the E index.
Table 5 Panel B presents the regression results. The results indicate that both the G
index and the difference between the G index and the E index are negatively associated
with the industry-adjusted Tobin’s Q, with the effects statistically significant at the
one percent and ten percent level, respectively. However, neither the G index nor the
difference between the G index and the E index significantly interacts with managerial
ownership.
Taken as a whole, Table 5 suggests that the antitakeover provisions in the E index
weaken the effect of managerial ownership on firm value not only in aggregate but also
individually.
4.4 Does Managerial Ownership Decrease with the Strength of
Antitakeover Provisions?
This section tests the second hypothesis, which states that managerial ownership de-
creases with the strength of antitakeover provisions.
Table 6 Panel A presents the summary statistics of managerial ownership for five
groups of firms based on the E index. The first four groups of firms have an E index of
20
0, 1, 2, and 3, respectively. The last group of firms has an E index between 4 and 6. The
firms with an E index of 5 and 6 account for only 2.5 percent of the sample, and thus are
grouped together with the firms with an E index of 4. Managerial ownership decreases
significantly as the E index increases. The mean managerial ownership is 7.6 percent for
the firms with an E index of 0, and 2.1 percent for the firms with an E index above 3.
Table 6 Panel B presents the regression results of the stock ownership and the PPS
of the top five executives and the CEO on the E index and the control variables. The E
index is negatively associated with all four measures of managerial ownership, and this
effect is statistically significant at the one percent level. In terms of economic significance,
each additional antitakeover provision in the E index decreases the ownership (options
excluded) of the top five executives and the CEO by 1.1 percentage points and 0.8
percentage points, respectively. Recall that the mean managerial ownership of the top
five executives is 3.9 percent, while mean ownership of the CEO is 2.6 percent.
5 Dealing with Endogeneity
The findings reported so far have established that managerial ownership, the E index and
their interaction are significantly associated with firm value. The statistical relations do
not necessarily imply that managerial ownership or the E index causes variations in
firm value. Neither does the statistical relation between managerial ownership and the
E index establish causality. Managerial ownership, antitakeover provisions, and firm
value may be simultaneously determined by some fundamental firm characteristics, as
shown in the model. These fundamental firm characteristics are usually unobserved by
researchers and thus result in the unobserved heterogeneity problem. If this problem is
indeed relevant, the statistical relations will disappear once all the relevant fundamental
firm characteristics are appropriately controlled for.
This section attempts to address this unobserved heterogeneity concern in four ways.
21
First, I use firm fixed effects to control for time-invariant firm characteristics. Second, I
investigate the impact of antitakeover provisions on the announcement returns of man-
ager share purchases in the open market. Third, I construct instrumental variables for
managerial ownership and the E index. Lastly, I propose and test an alternative expla-
nation for the above results.
5.1 Firm Fixed Effects
If the unobserved firm characteristics are constant over time, adding firm fixed effects
in the regression can effectively resolve the unobserved heterogeneity problem. This
observation motivates Himmelberg, Hubbard, and Palia (1999) to employ firm fixed
effects to deal with the endogeneity issue concerning managerial ownership.
I run OLS regressions with firm fixed effects for Tobin’s Q and the industry-adjusted
Tobin’s Q and present the results in the two columns of Table 7. Managerial ownership
is positively associated with Tobin’s Q and the industry-adjusted Tobin’s Q, and both
effects are significant at the one percent level. The coefficients on managerial ownership
squared in the two columns are negative and statistically significant at the one percent
level. The coefficients on the E index become statistically indifferent from zero. They are
statistically significant at the one percent level in Table 3 where firm fixed effects are not
controlled for. The coefficients on the interaction variable of managerial ownership and
the E index are negative and statistically significant at the five percent and one percent
level in the two columns, respectively. These results indicate that antitakeover provisions
affect firm value only through their interactions with managerial ownership.
Firm fixed effects also influence the relation between managerial ownership and the
E index. Table 6 Panel C shows that the E index is still negatively associated with
managerial ownership after controlling for firm fixed effects. However, comparing Panels
B and C of Table 6 reveals that the negative association between managerial ownership
22
and the E index is significantly weakened by firm fixed effects. Without firm fixed effects,
each additional antitakeover provision reduces managerial ownership by about 1 percent-
age point; after controlling for firm fixed effects each additional antitakeover provision
reduces managerial ownership by about 0.1 percentage points. That is, approximately 90
percent of the effect of antitakeover provisions on managerial ownership is attributable
to firm fixed effects.
Zhou (2001) points out that managerial ownership changes slowly from year to year
within a company, and argues that small, one-year changes in ownership are unlikely to
provide sufficient incentives that would lead to substantive within-year changes in firm
value. By relying on within variations, regressions with firm fixed effects may not detect
an effect of managerial ownership on firm value even if one exists. This argument also
applies to the E index, which shows small variations over the sample period (Table 1).
Zhou’s argument implies that firm fixed effects militate against finding significant
effects of managerial ownership and the E index on firm value. Even so, I find that
antitakeover provisions still significantly weaken the effect of managerial ownership on
firm value after controlling for firm fixed effects. Meanwhile, the E index is no longer
directly associated with firm value after controlling for firm fixed effects. These results
indicate that antitakeover provisions affect firm value only through their interactions with
managerial ownership.
5.2 An Event Study
Another way to resolve the unobserved heterogeneity problem is to investigate how
changes in managerial ownership affect firm value. This method has two merits relative
to the firm fixed effects approach. First, by relating changes in managerial ownership to
changes in firm value, it controls for any unobserved firm-specific fixed effects. Second,
changes in firm value within a short event window contain less “noise” than the annual
23
changes in firm value using the firm fixed effects approach. Following McConnell, Ser-
vaes, and Lins (2008), I study the market reactions to announcements of manager share
purchases in the open market. Share sales are excluded because they usually occur after
option exercises and thereby are anticipated by the market (Ofek and Yermack, 2000).
I merge the ExecuComp database with the Thomson Reuters insider trading database
to retrieve open market stock purchases by the top executives whose shareholdings are
reported in the ExecuComp database. Multiple manager share purchase announcements
on the same day are combined into one. During the period 1993-2008, the top executives
announced 4,162 stock purchases of the firms they manage. The mean (median) size
of share purchases is 0.065 (0.008) percent of the total number of shares outstanding.
Similar to McConnell et al., I compute the 7-day ([-1, +5]) cumulative abnormal returns
(CARs) around the share purchase announcement. This 7-day window is chosen because
the information usually does not enter the public domain for several days after being
filed with the SEC (Lakonishok and Lee, 2001). The announcements are greeted with a
mean (median) 7-day CARs of 2.08 (0.86) percent. The 5-day ([-1, +3]) CARs are also
computed as a robustness check of the following regression analysis.
The following model is employed to investigate how antitakeover provisions influence
the effect of manager share purchases on firm value:
CARs =β0+β1×Manager share purchasesi
+β2×Pre-purchase managerial ownershipi+β3×E indexi
+β4×Manager share purchasesi×Pre-purchase managerial ownershipi
+β5×Manager share purchasesi×E indexi+ui.(9)
where Manager share purchases is the number of shares the manager purchases divided
by the total number of shares outstanding, and Pre-purchase managerial ownership is
24
the managerial ownership at the beginning of the fiscal year.12 The model is similar to
the one used by McConnell et al. except that I add the interaction variable of manager
share purchases and the E index. The first interaction variable in the model captures the
diminishing marginal effect of managerial ownership on firm value; the second interaction
variable captures the impact of antitakeover provisions on the value effect of manager
share purchases. If antitakeover provisions weaken the effect of managerial ownership on
firm value, β5will be negative.
Table 8 presents the regression results of the model (9). The dependent variable is
the 5-day CARs in the first column, and the 7-day CARs in the second column. The
results indicate that the amount of manager share purchases is positively associated with
the announcement returns. This effect is statistically significant at the five percent level
for 5-day CARs and at the one percent level for 7-day CARs. In terms of economic
significance, the purchase of one percent of a firm’s shares by its managers increases the
stock price by approximately 7 percent and 4 percent over the 7- and 5-day announce-
ment period, respectively. The E index is negatively associated with the announcement
returns, but this effect is statistically insignificant. The coefficient on the interaction
variable of manager share purchases and pre-purchase managerial ownership is negative
and statistically significant, suggesting that the marginal effect of managerial ownership
on firm value diminishes as managers hold more shares. More interestingly, the coefficient
on the interaction variable of manager share purchases and the E index is negative and
statistically significant at the ten percent and five percent level, respectively; indicating
that antitakeover provisions weaken the effect of managerial ownership on firm value.
12The regression results are qualitatively unchanged if the managerial ownership right before the
share purchase is used instead. This alternative measure requires tracking the changes in managerial
ownership from the beginning of the fiscal year to the open market stock purchase, which are very small
and thus do not change the results. The changes in managerial ownership during that period are hard to
compute because the Thomson Reuters insider trading database only records open-market transactions
and exercises of stock options, but does not include other changes in managerial ownership such as grants
of restricted stocks. Therefore, the managerial ownership at the beginning of the fiscal year is preferable.
25
5.3 Instrumental Variable Regressions
In this section I construct instrumental variables for managerial ownership and the E
index to deal with the endogeneity issue. The first instrument for the E index is the
firm’s E index in 1990. Table 1 shows that the E index has small time-series variations,
indicating significant correlations between the E index in 1990 and the E indexes in later
years. To be a valid instrument, the E index in 1990 should not be correlated with the
error term in the model (7). I argue that the E index in 1990 is unlikely to be correlated
with the error terms in the years far away from 1990. To make it a more valid instrument,
I also exclude the data before 1995 from the regressions to leave a longer gap between the
instrument year and the data year.13 Similarly, I use the managerial ownership in 1992
(the first year when the data on managerial ownership are available) as an instrument
for managerial ownership.
The second instruments for the E index and managerial ownership are the average
E index and the average managerial ownership of the firm’s industry peers, respectively.
Firms in the same industry may have a similar E index and managerial ownership level
because of common industry characteristics. Meanwhile, the E index and managerial
ownership of industry peers are unlikely to be correlated with the firm-specific error term
in the model. This makes them valid instruments.
Since each state has different antitakeover laws, I also use the average E index of the
firms incorporated in the same state as an instrument for the E index.
Columns (1) and (2) of Table 9 present the OLS regression results of managerial
ownership and the E index, respectively, on the instruments and control variables. The
results indicate that both managerial ownership and the E index are significantly and
13Excluding the data before 1995 leaves at least a five year gap between the instrument year and the
data year. On one hand, I want to make the gap as large as possible because the E index in 1990 is less
likely to be correlated with the error term in the more remote future. On the other hand, I need to keep
as many data points as possible for the results to be consistent. The empirical results are qualitatively
similar if I use the cut-off year of 1993-1997.
26
positively correlated with their respective instruments, suggesting that the proposed
instruments are strong ones.
Following the suggestions of Wooldridge (2000), I use the square of the fitted man-
agerial ownership from the regression in column (1) as an instrument for managerial
ownership squared, and the product of the fitted managerial ownership times the fitted
E index from the regression in column (2) as an instrument for the interaction variable
of managerial ownership and the E index.
Column (3) of Table 9 presents the GMM regression results where the dependent
variable is the industry-adjusted Tobin’s Q. The results indicate that managerial own-
ership is positively associated with the industry-adjusted Tobin’s Q and significant at
the one percent level. The coefficient on managerial ownership squared is negative and
statistically significant at the five percent level. The coefficient on the E index is positive
but statistically indifferent from zero. Finally, the coefficient on the interaction variable
of managerial ownership and the E index is negative and statistically significant at the
ten percent level. These findings again suggest that antitakeover provisions affect firm
value only indirectly through their impacts on the value effect of managerial ownership.
It is necessary to check the validity of the proposed instruments. The p-value of
the Sargen-Hansen J-test is 0.42, indicating that the instruments are not significantly
correlated with the error term. The p-values associated with the F-tests of the strength
of the instruments are all below the one percent level, indicating that they are strong
instruments.
5.4 Accounting for Management Quality
Management quality may simultaneously determine managerial ownership, the E index,
and firm value. High-quality managers should own more shares of their firms for two
reasons. First, they may use high ownership to signal their quality (Leland and Pyle,
27
1977). Second, Milbourn (2003) shows that high-quality managers should have more
ownership because they are less likely to be replaced. The lower probability of being
replaced implies that more of their efforts—and less of the efforts of other managers
who may replace them—are incorporated into stock prices. The more informative a
manager’s effort is, the more effective stock-based incentives are. Therefore, high-quality
managers are granted with more ownership, and thus have incentives to exert more
effort. Consequently, high-quality managers rely less on the bid premium to enhance firm
value, and are associated with weaker antitakeover provisions. Firms with high-quality
managers are more valuable because of high quality of management and high managerial
ownership. Overall, high-quality managers are associated with more ownership, weaker
antitakeover provisions, and greater firm value.
The management quality hypothesis can be empirically examined. Similar to Mil-
bourn (2003), I use the cumulative abnormal returns (CARs) during the past three years
as a proxy for management quality. The stock returns are adjusted with respect to the
Fama-French three factors. The management quality hypothesis predicts that past stock
returns are negatively associated with the E index, positively associated with managerial
ownership, and positively associated with the firm value.
Table 10 presents the regression results. The CARs in the past three years are nega-
tively, positively, and positively associated with the E index, managerial ownership, and
the industry-adjusted Tobin’s Q, respectively. These effects are all statistically significant
at the one percent level, consistent with the management quality hypothesis. However,
managerial ownership, the E index, and their interaction are still significantly associated
with the industry-adjusted Tobin’s Q after controlling for past stock returns. The re-
sults suggest that management quality is not the driving force behind the findings of this
paper.
28
5.5 Reverse Causality?
The above empirical analysis focuses on the direction of causation from managerial own-
ership and antitakeover provisions to firm value. The direction of causation, however,
may be the opposite way, i.e., from firm value to managerial ownership and antitakeover
provisions. For instance, managers may adjust their shareholdings in anticipation of
changes in firm value: they increase (decrease) their ownership if they expect high (low)
firm performance. Since the amount of synergy gain decreases with firm performance,
firms may adopt weaker (stronger) antitakeover provisions if they expect high (low) per-
formance. This reverse causality hypothesis is able to explain the negative association
between the E index and firm value, and the negative association between the E index
and managerial ownership. In addition, it is consistent with the positive relation between
managerial ownership and firm value for firms with low E indexes. However, this hypoth-
esis predicts that the positive relation between managerial ownership and firm value does
not depend on the strength of antitakeover provisions, and therefore fails to explain the
negative association between managerial ownership and firm value for firms with high E
indexes. Therefore, reverse causality cannot be the driving force behind the findings in
this paper.
6 Conclusion
This paper investigates the impacts of antitakeover provisions on the value effect of
managerial ownership and the manager’s compensation contract. Antitakeover provisions
enhance the bargaining position of the target firm against the acquirer and thereby
raise the bid premium. An enhanced bargaining position weakens the incentive effect
of managerial ownership because it encourages the target manager to rely more on the
bid premium, and thus less on managerial effort, to create value for shareholders, ceteris
29
paribus. Antitakeover provisions may also magnify the entrenchment effect of managerial
ownership. Aware of these negative impacts of antitakeover provisions on the effect of
managerial ownership on firm value, shareholders grant less ownership to the manager
when there are stronger antitakeover provisions.
The impacts of antitakeover provisions on the value effect of managerial ownership
are then empirically tested. I show that the effect of managerial ownership on firm value
decreases significantly with the strength of antitakeover provisions. For firms with weak
antitakeover provisions, managerial ownership and firm value are positively associated
with each other. For firms with strong antitakeover provisions, they become negatively
associated with each other. I also show that antitakeover provisions decrease the value
effect of managerial ownership not only in aggregate but also individually.
The analyses with firm fixed effects and instrumental variables show that antitakeover
provisions do not directly influence firm value: instead, they only influence firm value
indirectly by decreasing the effect of managerial ownership on firm value. This indirect
channel through which antitakeover provisions affect firm value sheds light on the cur-
rent debate over whether the relation between antitakeover provisions and firm value is
causal.14 This indirect channel implies that antitakeover provisions have causal effects
on firm value. However, unlike the studies in the preceding footnote suggest, such effects
are indirect. Since managerial ownership, antitakeover provisions, and firm value might
all be endogenously determined, it is difficult to establish causality between them. This
paper has endeavored to deal with endogeneity using firm fixed effects, an event study,
and instrumental variables. Future research is needed to shed more light on causality
between them.
Managerial ownership decreases significantly with the strength of antitakeover pro-
visions. Each additional antitakeover provision in the E index reduces the ownership of
14See Gompers, Ishii, and Metrick (2003), Bebchuk and Cohen (2005), Chi (2005), Core, Guay, and
Rusticus (2006), Lehn, Patro, and Zhao (2007), Bebchuk, Cohen, and Ferrel (2009) for this debate.
30
the top five executives by about 1 percentage point. However, about 90 percent of this
effect is attributable to firm fixed effects.
Overall, I conclude that the findings of this study are consistent with the hypotheses
that antitakeover provisions weaken the value effect of managerial ownership and affect
the manager’s compensation contract.
31
Appendix A: The Strength of Antitakeover Provisions
This section studies how shareholders determine the strength of antitakeover provisions.
The discussions in section 2.2 suggest that antitakeover provisions enhance the tar-
get firm’s bargaining position and therefore increase the target’s share of the synergy.
Accordingly, assume the proportion of the synergy that accrues to the target, p, is an in-
creasing function of the strength of antitakeover provisions of the target firm, n. That is,
p′(n)>0. Since there is a one-to-one mapping between nand p, choosing the strength of
antitakeover provisions is equivalent to choosing the optimal bargaining position. There-
fore, pis regarded as the decision variable.
Suppose the raider has to spend certain cost, δ, to acquire the target firm. The sources
of such cost include information production cost, legal cost, cost to hire advisors, and time
cost of the raider’s managers. Antitakeover provisions reduce the share of the synergy
that accrues to the raider, and thus discourage the raider from acquiring the target given
the acquisition cost. On the other hand, antitakeover provisions reduce managerial effort
and thus raise the amount of the potential synergy gain. This makes the target firm more
attractive. The overall effect of antitakeover provisions on the attractiveness of the target
depends on the magnitudes of these two opposite effects. While choosing the strength
of antitakeover provisions, the target’s shareholders must make sure that enough profit
is left to the raider to cover the acquisition cost. Otherwise, the raider will not initiate
the takeover. In other words, the raider’s participation constraint must be satisfied. The
raider is assumed to be risk neutral for simplicity.
The shareholders choose the strength of antitakeover provisions to maximize their
expected wealth subject to the optimal compensation contract and the raider’s partici-
32
pation constraint. The shareholders’ problem is:
max
pψ= (1 −α)[e+p(¯v−e)] −s(A-1)
s.t. e ∈arg max s+α[e+p(¯v−e)] −1
2ke2−η
2α2σ2(A-2)
α=α∗(A-3)
s+α[e+p(¯v−e)] −1
2ke2−η
2α2σ2= ¯u(A-4)
θ≡(1 −p)(¯v−e)−δ≥0,(A-5)
where the last two lines are the manager’s and the raider’s participation constraint, re-
spectively. There is no closed-form solution to the strength of antitakeover provisions.
The properties of the optimal strength of antitakeover provisions are stated in the fol-
lowing proposition.
Proposition 2. The shareholders are more likely to adopt antitakeover provisions when:
(1) The potential synergy gain is greater (¯vis greater).
(2) Firm value is more volatile (σ2is greater).
(3) The manager is more risk-averse (ηis greater).
(4) Managerial effort is more costly (kis smaller).
When shareholders adopt antitakeover provisions, they choose the strength at which the
raider’s participation constraint is binding.
Proof. See Appendix B. Q.E.D.
The intuition behind Proposition 2 can be explained as follows. The marginal benefit
of antitakeover provisions increases with the amount of synergy gain. Therefore, the
shareholders are more likely to adopt antitakeover provisions when synergy gain is greater.
33
Ownership is less desirable to a manager who is more risk-averse or for whom effort is
more costly. It is also less desirable if firm value is more volatile. Thus, managerial
ownership decreases with firm value volatility, the cost of managerial effort, and the
degree of managerial risk-aversion. Since antitakeover provisions weaken the incentive
effect of managerial ownership, it is less costly to adopt antitakeover provisions at low
levels of managerial ownership. Therefore, the shareholders are more likely to adopt
antitakeover provisions when firm value is more volatile, when the manager is more risk
averse, and when managerial effort is more costly.
Antitakeover provisions reduce managerial effort, but the speed of such reduction
slows down as antitakeover provisions become stronger. In other words, the marginal cost
of antitakeover provisions decreases with their strength. On the other hand, the marginal
benefit of antitakeover provisions stay constant at ¯v. Consequently, the shareholders set
the optimal strength of antitakeover provisions at the level where the raider’s participa-
tion constraint is binding. This is the strongest antitakeover provisions the shareholders
are able to adopt without keeping the raider out of the game. If the marginal cost of anti-
takeover provisions outweighs the marginal befit at this level of strength, the shareholders
choose not to adopt any antitakeover provisions.
34
Appendix B: Proofs of the Propositions
Proof of Proposition 1:
Proof. Solving the manager’s problem yields the optimal managerial effort:
e∗=kα(1 −p).(A-6)
In equilibrium, the shareholders offer the lowest salary to the manager so that the par-
ticipation constraint is binding. Then the participation constraint becomes:
s+α[e+ (1 −p)(¯v−e)] = ¯u+1
2ke2+η
2α2σ2.(A-7)
Plugging (A-7) and (A-6) into the shareholders’ objective function yields the following
problem:
max
αkα(1 −p)2+p¯v−¯u+k
2(1 −p)2α2+η
2α2σ2.(A-8)
The first-order condition of (A-8) yields the optimal managerial ownership in equilibrium:
α∗=k(1 −p)2
k(1 −p)2+ησ2.(A-9)
Then
e∗=kα(1 −p) = k2(1 −p)4
k(1 −p)2+ησ2,(A-10)
dα∗
dp =−2ησ2k(1 −p)
k(1 −p)2+ησ2≤0 (A-11)
de∗
dp =−2k2(1 −p)3[k(1 −p)2+ 2ησ2]
[k(1 −p)2+ησ2]2≤0.(A-12)
Note that the terms in the first bracket of (A-8) are the expected firm value which
equals to the target’s stand-alone value plus bid premium. Define the expected firm value
35
as π. Then equation (A-8) becomes:
max
απ−¯u+k
2(1 −p)2α2+η
2α2σ2.(A-13)
The first-order condition is:
∂π
∂α −k(1 −p)2α−ησ2α= 0.(A-14)
Differentiating both sides of (A-14) yields:
d∂π
∂α −(1 −p)2dk −k(1 −p)2dα −2(1 −p)dp −σ2αdη −ηαdσ2−ησ2dα = 0.(A-15)
Then:
d
dp ∂π
∂α =k(1 −q)2+ηασ2dα
dp −2(1 −p)
=−2ησ2k(1 −p)
k(1 −p)2+ησ2−2(1 −p)<0.(A-16)
Q.E.D.
Proof of Proposition 2:
Proof. Plugging (A-9) into (A-8) yields the expected shareholder wealth for a given com-
36
pensation contract:
ψ=k2(1 −p)4
k(1 −p)2+ησ2+p¯v−¯u+k3(1 −p)6
2 [k(1 −p)2+ησ2]2+ησ2k2(1 −p)4
2 [k(1 −p)2+ησ2]2
=k2(1 −p)4
[k(1 −p)2+ησ2]2k(1 −p)2+ησ2−k
2(1 −p)2−η
2σ2+p¯v−¯u
=k2(1 −p)4
[k(1 −p)2+ησ2]2k
2(1 −p)2+η
2σ2+p¯v−¯u
=k2(1 −p)4
2 [k(1 −p)2+ησ2]+p¯v−¯u. (A-17)
Therefore, the shareholders’ problem becomes:
max
pψ=k2(1 −p)4
2 [k(1 −p)2+ησ2]+p¯v−¯u(A-18)
s.t. θ ≡¯v−p¯v−k2(1 −p)4
k(1 −p)2+ησ2−δ≥0.(A-19)
Let us first focus on the shareholders’ objective function and solve the unconstrained
problem. The first-order condition is:
∂ψ
∂p =−4k2(1 −p)3[k(1 −p)2+ησ2] + 2k3(1 −p)5
2 [k(1 −p)2+ησ2]2+ ¯v
=2k2(1 −p)3{−2 [k(1 −p)2+ησ2] + k(1 −p)2}
2 [k(1 −p)2+ησ2]2+ ¯v
=−k2(1 −p)3[k(1 −p)2+ 2ησ2]
[k(1 −p)2+ησ2]2+ ¯v
=−k3(1 −p)5+ 2ησ2k2(1 −p)3
[k(1 −p)2+ησ2]2+ ¯v
≡ −MC +M B = 0,(A-20)
where MC and M B stand for the marginal cost and marginal benefit of antitakeover
provisions, respectively. The marginal benefit is constant (¯v), while the marginal cost
37
decreases with p:
dMC
dp =[−5k3(1 −p)4−6ησ2k2(1 −p)2] [k(1 −p)2+ησ2]
[k(1 −p)2+ησ2]4
+4k(1 −p) [k3(1 −p)5+ 2ησ2k2(1 −p)3] [k(1 −p)2+ησ2]
[k(1 −p)2+ησ2]4
=−k2(1 −p)2
[k(1 −p)2+ησ2]3nk(1 −p)2+ησ22+ησ2k(1 −p)2+ 5η2σ4o<0.
Therefore, the shareholders will adopt the strongest antitakeover provisions as long as
the raider’s participation constraint is satisfied. Denote ˆpas the pat which the raider’s
participation constraint is binding:
θ(ˆp) = ¯v−ˆp¯v−k2(1 −ˆp)4
k(1 −ˆp)2+ησ2−δ= 0.(A-21)
The shareholders will set the optimal antitakeover provisions at p∗= ˆpif the marginal cost
of antitakeover provisions is less than the marginal benefit at ˆp:MC(ˆp)<¯v. Otherwise,
they choose p∗= 0.
The marginal benefit of antitakeover provisions increases with ¯v, while the marginal
cost decreases with ηand σ2and increases with k:
dMC
dσ2=−2k2(1 −p)3η2σ2
[k(1 −p)2+ησ2]3<0 (A-22)
dMC
dη =−2k2(1 −p)3ησ4
[k(1 −p)2+ησ2]3<0 (A-23)
dMC
dk =k(1 −p)2[k2(1 −p)4+ 3k(1 −p)2ησ2+ 4η2σ4]
[k(1 −p)2+ησ2]3>0.(A-24)
Therefore, the shareholders are more likely to adopt antitakeover provisions when ¯v,η,
and σ2are greater and when kis smaller. Q.E.D.
38
Appendix C: Variable Descriptions
Variable Definition
Measures of Managerial Ownership and Strength of Antitakeover Provisions
Managerial
ownership
The total number of shares owned by the top five executives (including restricted
stocks but not options) divided by the total number of shares outstanding.
Pay-performance
sensitivity (PPS)
The sensitivity of executives compensation to stock price fluctuation. For stocks,
PPS is simply the percentage stock ownership; for stock options, PPS equals the
number of shares underlying the options times the delta of each option divided
by the total number of shares outstanding. The delta is defined as the partial
derivative of option value with respect to stock price. I follows Guay (1999) and
Core and Guay (2002) to calculate the delta using the ExecuComp database.
E index An index based on six antitakeover provisions in the RiskMetrics database: stag-
gered board, poison pills, supermajority requirement for mergers, limits to share-
holder bylaw amendments, limits to charter amendments, and golden parachutes.
The index value increases by one for each antitakeover provision in place.
G index An index based on twenty-four antitakeover provisions in the RiskMetrics
database. The value of the index increases by one for each antitakeover pro-
vision in place.
Accounting Variables
Tobin’s Q The market value of total assets divided by the book value of total assets. The
market value of assets is computed as the book value of total assets (data 6 in
the Compustat database) plus common shares outstanding (data 25) times stock
price (data 199).
Industry-adjusted
Tobins Q
Tobins Q minus the median industry Tobins Q. I use the Fama-French (1997)
industry classifications.
Sales The natural log of total sales (data 12).
Sales squared The square of Sales.
Tangibility The ratio of property, plant, and equipment (data 8) to sales (data12).
Tangibility squared The square of PPE/Sales.
Idiosyncratic risk The standard deviation of the residuals of the CAPM model estimated using
daily stock returns in the fiscal year.
Profit margin The ratio of operating income before depreciation (data 13) to total sales (data
12).
R&D The ratio of research and development expenses (data 46) to property, plant, and
equipment (data 8).
R&D dummy A dummy variable which takes the value of one if the R&D data are available,
and zero otherwise.
Advertisement The ratio of advertisement expenditures (data 45) to property, plant, and equip-
ment (data 8).
Advertisement
dummy
A dummy variable which takes the value of one if the advertisement expenses
data are available, and zero otherwise.
Investment The ratio of capital expenditure (data 128) to property, plant, and equipment
(data 8).
39
Variable Descriptions (Continued)
Variable Definition
Stock returns
CARs in past three
years
The cumulative abnormal stock returns over the 36-month period ending at the
beginning of the fiscal year. The returns are adjusted with respect to the Fama-
French three factors.
Announcement
returns of manager
share purchase in the
open market
The cumulative abnormal return over the trading days [-1, +5] or [-1, +3] around
the share purchase announcement. The returns are adjusted with the CAPM
model. The estimation window for the market model is [-252,-42] days prior to
the announcement.
40
References
Aggarwal, R. K., and A. A. Samwick, 2003, Performance incentives within firms: The effect of managerial
responsibility, Journal of Finance 58, 1613-1649.
Andreoni, J., W. Harbaugh, and L. Vesterlund, 2003, The carrot or the stick: Rewards, punishments,
and cooperation, American Economic Review 93, 893-902.
Bebchuk, L., and A. Cohen, 2005, The costs of entrenched boards, Journal of Financial Economics 78,
409-433.
Bebchuk, L., A. Cohen, and A. Ferrell, 2009, What matters in corporate governance?, Review of Financial
Studies 22, 783-827.
Bebchuk, L., and J. M. Fried, 2005, Pay without performance: Overview of the issues, Journal of Applied
Corporate Finance 17, 8-23.
Berle, A., and G. Means, 1932, The modern corporation and private property, Macmillan, New York.
Bertrand, M., and S. Mullainathan, 1999, Is there discretion in wage setting? A test using takeover
legislation, RAND Journal of Economics 30, 535-554.
Bertrand, M., and S. Mullainathan, 2000, Agents with and without principals, American Economic
Review 90, 203-208.
Bertrand, M., and S. Mullainathan, 2003, Enjoying the quiet life? Managerial behavior following anti-
takeover legislation, Journal of Political Economy 111, 1043-1075.
Betton, S., B. E. Eckbo, and K. Thorburn, 2008, Corporate takeovers, in: Eckbo, B.E. (Ed.), Handbook
of Corporate Finance: Empirical Corporate Finance Vol. II, Elsevier/North-Holland, 291-429.
Bhagat, S., and B. Bolton, 2008, Corporate governance and firm performance, Journal of Corporate
Finance 14, 257-273.
Bhagat, S., and R. Romano, 2001, Event studies and the law: Part II—Empirical studies of corporate
law, Yale University working paper.
Bolton, P., and M. Dewatripont, 2005, Contract Theory, The MIT Press.
Cheng, S., and R. J. Indjejikian, 2009, The market for corporate control and CEO compensation:
Complements or substitutes?, Contemporary Accounting Research 26, 701-728.
Cheng, S., V. Nagar, and M. Rajan, 2005, Identifying control motives in managerial ownership: Evidence
of anti-takeover legislation, Review of Financial Studies 18, 637-672.
Chi, J. D., 2005, Understanding the endogeneity between firm value and shareholder rights, Financial
Management 34, 65-76.
Cohn, J. B., and U. Rajan, 2010, Corporate governance and managerial reputational concerns, Univer-
sity of Texas Austin working paper.
Coles, J. L., M. L. Lemmon, and J. F. Meschke, 2007, Structural models and endogeneity in corporate
finance: the link between managerial ownership and corporate performance, Arizona State University
41
working paper.
Comment, R., and G. W. Schwert, 1995, Poison or placebo? Evidence on the deterrence and wealth
effects of modern antitakeover measures, Journal of Financial Economics 39, 3-43.
Core, J. E., and W. R. Guay, 2002, Estimating the value of employee stock option portfolios and their
sensitivities to price and volatility, Journal of Accounting Research 40, 613-630.
Core, J. E., W. R. Guay, and T. O. Rusticus, 2006, Does weak governance cause weak stock returns? An
examination of firm operating performance and investors’ expectations, Journal of Finance 61, 655-687.
Core, J., and D. Larcker, 2002, Performance consequences of mandatory increases in executive stock
ownership, Journal of Financial Economics 64, 317–340.
Cremers, K. J., and V. B. Nair, 2005, Governance mechanisms and equity prices, Journal of Finance
60, 2859-2894.
Cremers, K. J., V. B. Nair, and J. Wei, 2007, Governance mechanisms and bond prices, Review of
Financial Studies 20, 1359-1388.
DeAngelo, H., and E. M. Rice, 1983, Antitakeover charter amendments and stockholder wealth, Journal
of Financial Economics 11, 329-360.
Demsetz, H., 1983, The structure of ownership and the theory of the firm, Journal of Law and Economics
26, 375-390.
Demsetz, H., and K. Lehn, 1985, The structure of corporate ownership: Causes and consequences,
Journal of Political Economy 93, 1155-1177.
Demsetz, H., and B. Villalonga, 2001, Ownership structure and corporate performance, Journal of
Corporate Finance 7, 209-233.
Fahlenbrach, R., 2009, Shareholder rights, boards, and executive compensation, Review of Finance 13,
81-113.
Fahlenbrach, R., and R. M. Stulz, 2009, Managerial ownership dynamics and firm value, Journal of
Financial Economics 92, 342-361.
Faleye, O., 2007, Classified boards, firm value, and managerial entrenchment, Journal of Financial
Economics 83, 501-529.
Fama, E., and K. French, 1997, Industry costs of equity, Journal of Financial Economics43, 153-193.
Fama, E., and K. French, 2000, Forecasting profitability and earnings, Journal of Business 73, 161-175.
Fama, E., and M. Jensen, 1983, Separation of ownership and control, Journal of Law and Economics
26, 301-325.
Fama, E. F., and J. D. MacBeth, 1973, Risk, return, and equilibrium: empirical tests, Journal of Political
Economy 81, 607-636.
Fich, E. M., and A. Shivdasani, 2006, Are busy boards effective monitors?, Journal of Finance 61,
689-724.
42
Giroud, X., and H. M. Mueller, 2010a, Does corporate governance matter in competitive industries?,
Journal of Financial Economics 95, 312-331.
Giroud, X., and H. M. Mueller, 2010b, Corporate governance, product market competition, and equity
prices, Journal of Finance forthcoming.
Gompers, P., J. Ishii, and A. Metrick, 2003, Corporate governance and equity prices, Quarterly Journal
of Economics 118, 107-155.
Goyal V. K., and C. W. Park, 2002, Board leadership structure and CEO turnover, Journal of Corporate
Finance 8, 49-66.
Grossman, S. J., O. D. Hart, 1980, Takeover bids, the free-rider problem, and the theory of the corpo-
ration, Bell Journal of Economics 11, 42-64.
Guay, W. K., 1999, The sensitivity of CEO wealth to equity risk: An analysis of the magnitude and
determinants, Journal of Financial Economics 53, 43-71.
Hall, B., and J. Liebman, 1998, Are CEOs really paid like bureaucrats?, Quarterly Journal of Economics
113, 653-691.
Harris, E. G., 1990, Antitakeover measures, golden parachutes, and target firm shareholder welfare,
RAND Journal of Economics 21, 614-625.
Hartzell, J. C., E. Ofek, and D. Yermack, 2004, What’s in it for me? CEOs whose firms are acquired,
Review of Financial Studies 17, 37-61.
Hermalin, B., and M. Weisbach, 1991, The effects of board compensation and direct incentives on firm
performance, Financial Management 20, 101-112.
Heron, R. A., and E. Lie, 2006, On the use of poison pills and defensive payouts by takeover targets,
Journal of Business 79, 1783-1807.
Himmelberg, C. P., R. G. Hubbard, and D. Palia, 1999, Understanding the determinants of managerial
ownership and the link between ownership and performance, Journal of Financial Economics 53, 353-
384.
Holderness, C., R. Kroszner, and D. Sheehan, 1999, Were the good old days that good?: Evolution of
managerial stock ownership and corporate governance since the great depression, Journal of Finance 54,
435-469.
Holmstr¨om, B., and S. N. Kaplan, 2001, Corporate governance and merger activity in the U.S.: Making
sense of the 1980s and 1990s, Journal of Economic Perspectives 15, 121-144.
Holmstr¨om, B., and P. Milgrom, 1987, Aggregation and linearity in the provision of intertemporal
incentives, Econometrica 55, 303-328.
Huson, M. R., R. Parrino, and L. Starks, 2001, Internal monitoring mechanisms and CEO turnover: A
long term perspective, Journal of Finance 56, 2265-2297.
Jensen, M., 1993, The modern industrial revolution, exit and the failure of internal control systems,
Journal of Finance 48, 831-880.
43
Jensen, M., and W. Meckling, 1976, Theory of the firm: Managerial behavior, agency costs and ownership
structure, Journal of Financial Economics 3, 305-360.
Jenter, D., 2005, Market timing and managerial portfolio decisions, Journal of Finance 60, 1903-1949.
John, K., and S. Kedia, 2003, Design of corporate governance: Role of ownership structure, takeovers,
bank debt and large shareholder monitoring, New York University working paper.
Johnson, S. A., T. C. Moorman, and S. Sorescu, 2009, A reexamination of corporate governance and
equity prices, Review of Financial Studies 22, 4753-4786.
Kim, E. H. and Y. Lu, 2010, CEO ownership and external governance, University of Michigan working
paper.
Kini, O., W. Kracaw, and S. Mian, 2004, The nature of discipline by corporate takeovers, Journal of
Finance 59, 1511-1552.
Kole, S. R., 1995, Measuring managerial equity ownership: a comparison of sources of ownership data,
Journal of Corporate Finance 1, 413-435.
Kole, S. R., 1997, The complexity of compensation contracts, Journal of Financial Economics 43, 79-
104.
Lakonishok, J., and I. Lee, 2001, Are insider trades informative?, Review of Financial Studies 14, 79-111.
Lee, I., 1997, Do firms knowingly sell overvalued equity?, Journal of Finance 52, 1439-1466.
Lefanowicz, C. E., J. R. Robinson, and R. Smith, 2000, Golden parachutes and managerial incentives in
corporate acquisitions: Evidence from the 1980s and 1990s, Journal of Corporate Finance 6, 215-239.
Lehn, K., S. Patro, and M. Zhao, 2007, Governance indexes and valuation: Which causes which?, Journal
of Corporate Finance 13, 907-928.
Leland, H. E., and D. Pyle, 1977, Information asymmetries, financial structure, and financial interme-
diation, Journal of Finance 32, 371-387.
McConnell, J., and H. Servaes, 1990, Additional evidence on equity ownership and corporate value,
Journal of Financial Economics 27, 595-612.
McConnell, J., H. Servaes, and K. Lins, 2008, Changes in insider ownership and changes in the market
value of the firm, Journal of Corporate Finance 14, 92-106.
Morck, R., A. Shleifer, and R. Vishny, 1988, Management ownership and market valuation, Journal of
Financial Economics 20, 293-315.
Parrino, R., R. W. Sias, and L. T. Starks, 2003, Voting with their feet: Institutional investors and CEO
turnover, Journal of Financial Economics 68, 3-64.
Petersen, M. A., 2009, Estimating standard errors in finance panel data sets: Comparing approaches,
Review of Financial Studies 22, 435-480.
Stein, J. C., 1988, Takeover threats and managerial myopia, Journal of Political Economy 96, 61-80.
44
Stulz, R. M., 1988, Managerial control of voting rights: Financing policies and the market for corporate
control, Journal of Financial Economics 20, 25-54.
Weisbach, M. S., 1988, Outside directors and CEO turnover, Journal of Financial Economics 20, 431-
460.
Williamson, O. E., 1983, Organizational form, residual claimants, and corporate control, Journal of Law
and Economics 26, 351-366.
Wooldridge, J., 2000, Econometric Analysis of Cross Section and Panel Data, MIT Press.
Yermack, D., 1996, Higher valuation of companies with a small board of directors, Journal of Financial
Economics 40, 185-212.
Zhou, X., 2001, Understanding the determinants of managerial ownership and the link between ownership
and performance: comment, Journal of Financial Economics 62, 559-571.
45
Figure 1: Relation between Managerial Ownership and Tobin’s Q by the E Index
This figure plots the fitted value of Tobin’s Q against managerial ownership at different values of the
E index. The X-axis is managerial ownership (in percentage). The Y-axis is Tobin’s Q. The values
of Tobin’s Q are estimated from the regression results in column (3) of Table 3: Q= 3.574 + 1.858 ×
Managerial ownership−1.360×Managerial ownership2−0.044×E index−0.536×Managerial ownership×
E index. Managerial ownership ranges from 0 to 20 percent in the figure since almost all firms in the
sample have managerial ownership below 20 percent.
0 10 20
3.6
3.8
4E index = 0
0 10 20
3.6
3.8
4E index = 1
0 10 20
3.6
3.8
4E index = 2
0 10 20
3
3.5 E index = 3
0 10 20
3
3.5 E index = 4
0 10 20
3
3.5 E index = 5
0 10 20
3
3.5 E index = 6
46
Table 1: Managerial Ownership and the E Index
This table presents the summary statistics of managerial ownership (Panel A) and the E index (Panel
B). Panel C reports the frequencies of the six provisions in the E index. In Panels B and C, only the
years when the E index is updated are included. The sample consists of 14,962 firm-year observations
over the period 1992-2007. See the Appendix for the descriptions of the variables.
Panel A: Managerial Ownership, 1992-2007
Year N mean sd p5 p50 p95
1992 524 4.17% 10.12% 0.03% 0.54% 22.72%
1993 714 4.44% 9.44% 0.05% 0.87% 23.96%
1994 711 4.38% 8.88% 0.05% 0.88% 24.47%
1995 775 4.40% 8.77% 0.04% 0.80% 25.01%
1996 772 4.08% 8.53% 0.05% 0.69% 23.68%
1997 783 3.91% 7.91% 0.05% 0.73% 21.85%
1998 1073 4.77% 9.03% 0.06% 0.93% 24.92%
1999 1004 4.52% 8.71% 0.06% 0.85% 23.70%
2000 972 4.60% 8.83% 0.06% 0.91% 23.68%
2001 946 4.13% 8.45% 0.05% 0.79% 21.96%
2002 1145 3.98% 8.21% 0.05% 0.80% 20.09%
2003 1136 3.50% 7.32% 0.04% 0.72% 18.55%
2004 1169 3.36% 7.26% 0.05% 0.76% 16.90%
2005 1095 3.04% 6.53% 0.06% 0.75% 15.52%
2006 1104 2.94% 6.26% 0.05% 0.73% 15.45%
2007 1039 2.69% 6.00% 0.06% 0.70% 13.36%
Total 14962 3.87% 8.08% 0.05% 0.77% 21.39%
Panel B: E Index, 1992-2007
Year N mean sd p5 p50 p95
1992 524 2.27 1.42 0.00 2.00 5.00
1993 714 2.28 1.40 0.00 2.00 4.00
1995 775 2.27 1.36 0.00 2.00 4.00
1998 1073 2.17 1.31 0.00 2.00 4.00
2000 972 2.31 1.31 0.00 2.00 4.00
2002 1145 2.47 1.28 0.00 3.00 4.00
2004 1169 2.49 1.24 0.00 3.00 4.00
2006 1104 2.41 1.24 0.00 2.00 4.00
Total 7476 2.35 1.31 0.00 2.00 4.00
Panel C: Frequencies of Antitakeover Provisions
Year Staggered Poison Super- Limits to Limits to Golden
board pill majority amend amend parachutes
bylaws charter
1992 57.82% 62.40% 39.89% 12.98% 2.48% 51.72%
1993 59.38% 60.08% 38.94% 14.99% 2.66% 52.38%
1995 60.13% 57.94% 38.71% 14.58% 2.97% 52.52%
1998 57.04% 55.17% 34.30% 15.66% 2.33% 52.47%
2000 58.23% 58.85% 34.36% 17.28% 2.16% 60.08%
2002 60.52% 62.10% 33.80% 20.79% 1.40% 68.03%
2004 59.97% 61.68% 32.42% 20.87% 1.71% 72.11%
2006 55.53% 56.25% 32.52% 19.02% 1.81% 76.27%
Total 58.56% 59.15% 34.97% 17.60% 2.10% 62.37%
47
Table 2: Summary Statistics
This table presents the summary statistics (Panel A) and the correlation matrix (Panel B) of the vari-
ables. The p-values for statistical significance of the correlations are presented in the brackets in Panel
B. The sample consists of 14,962 firm-year observations over the period 1992-2007. See the Appendix
for the descriptions of the variables.
Panel A: Summary Statistics
Variable N mean sd p5 p50 p95
Tobin’s Q 14962 2.02 1.28 0.94 1.62 4.53
Industry-adjusted Q 14962 0.41 1.19 -0.73 0.08 2.72
Managerial ownership (%) 14962 3.87 8.08 0.05 0.77 21.39
E index 14962 2.35 1.31 0.00 2.00 4.00
Sales ($Bn) 14962 4.95 11.52 0.16 1.37 20.50
Tangibility 14962 0.38 0.53 0.04 0.22 1.37
Idiosyncratic risk (%) 14962 2.37 1.17 1.08 2.07 4.69
Profit margin 14962 0.15 0.15 0.01 0.13 0.39
R&D 14962 0.28 0.67 0.00 0.01 1.51
R&D dummy 14962 0.64 0.48 0.00 1.00 1.00
Advertising expense 14962 0.09 0.31 0.00 0.00 0.46
Advertising dummy 14962 0.36 0.48 0.00 0.00 1.00
Investment 14962 0.23 0.14 0.07 0.20 0.51
48
Panel B: Correlation Matrix
1 2 3 4 5 6 7 8 9 10 11 12 13
1 Tobin’s Q 1.00
2 Industry-adjusted Q 0.94 1.00
[0.00]
3 Managerial ownership 0.02 0.05 1.00
[0.00] [0.00]
4 E index -0.13 -0.13 -0.19 1.00
[0.00] [0.00] [0.00]
5 Sales -0.04 0.00 -0.15 0.02 1.00
[0.00] [0.80] [0.00] [0.07]
6 Tangibility -0.11 -0.10 -0.05 0.01 -0.08 1.00
[0.00] [0.00] [0.00] [0.20] [0.00]
7 Idiosyncratic risk 0.01 0.01 0.05 -0.09 -0.41 0.01 1.00
[0.53] [0.07] [0.00] [0.00] [0.00] [0.46]
8 Profit margin 0.22 0.21 -0.03 -0.01 0.15 0.28 -0.30 1.00
[0.00] [0.00] [0.00] [0.31] [0.00] [0.00] [0.00]
9 R&D 0.22 0.13 -0.06 -0.08 -0.29 -0.14 0.26 -0.21 1.00
[0.00] [0.00] [0.00] [0.00] [0.00] [0.00] [0.00] [0.00]
10 R&D dummy 0.12 0.08 -0.10 0.02 0.00 -0.23 0.07 -0.12 0.31 1.00
[0.00] [0.00] [0.00] [0.04] [0.59] [0.00] [0.00] [0.00] [0.00]
11 Advertising expense 0.08 0.08 0.05 -0.07 -0.02 -0.15 0.01 -0.06 0.16 -0.03 1.00
[0.00] [0.00] [0.00] [0.00] [0.00] [0.00] [0.21] [0.00] [0.00] [0.00]
12 Advertising dummy 0.09 0.09 0.02 -0.05 0.12 -0.15 -0.02 -0.04 0.08 0.06 0.41 1.00
[0.00] [0.00] [0.00] [0.00] [0.00] [0.00] [0.01] [0.00] [0.00] [0.00] [0.00]
13 Investment 0.30 0.25 0.03 -0.11 -0.15 -0.19 0.18 0.01 0.34 0.06 0.18 0.08 1.00
[0.00] [0.00] [0.00] [0.00] [0.00] [0.00] [0.00] [0.09] [0.00] [0.00] [0.00] [0.00]
49
Table 3: Determinants of Tobin’s Q
The dependent variable in the first five columns and the last column are Tobin’s Q and the industry-adjusted Tobin’s Q, respectively. The firms
that are used in the regression in column (4) have an E index between 0 and 2; the firms in column (5) have an E index between 3 and 6. The
sample consists of 14,962 firm-year observations over the period 1992-2007. See the Appendix for the descriptions of the variables. The p-values
are reported in the brackets. Significance levels are indicated by *, **, and *** for 10%, 5%, and 1%, respectively.
50
Dependent Variable Industry-adjusted
Tobin’s Q Tobin’s Q
Sample Whole sample E index ≤2 E index ≥3 Whole sample
(1) (2) (3) (4) (5) (6)
Managerial ownership 1.635*** 1.858*** 1.858** 1.891** -1.268* 2.218**
[0.000] [0.000] [0.046] [0.030] [0.076] [0.019]
Managerial ownership squared -1.091* -1.360** -1.360 -2.455 2.819* -2.159
[0.089] [0.028] [0.299] [0.130] [0.057] [0.102]
E index -0.050*** -0.044*** -0.044*** -0.044***
[0.000] [0.000] [0.007] [0.008]
Managerial ownership ×E index -0.552*** -0.536*** -0.536** -0.560**
[0.000] [0.000] [0.016] [0.013]
Sales -0.729*** -0.576*** -0.576*** -0.670*** -0.464*** -0.476***
[0.000] [0.000] [0.000] [0.000] [0.006] [0.000]
Sales squared 4.622*** 3.711*** 3.711*** 4.344*** 2.971*** 3.071***
[0.000] [0.000] [0.000] [0.000] [0.006] [0.000]
Tangibility -0.578*** -0.727*** -0.727*** -0.781*** -0.672*** -0.598***
[0.000] [0.000] [0.000] [0.000] [0.000] [0.000]
Tangibility squared 0.089*** 0.147*** 0.147*** 0.190*** 0.088** 0.110***
[0.000] [0.000] [0.000] [0.001] [0.013] [0.001]
Idiosyncratic risk -0.009 -0.050*** -0.050*** -0.048* -0.057*** -0.066***
[0.404] [0.000] [0.008] [0.084] [0.010] [0.000]
Profit margin 2.810*** 2.627*** 2.627*** 2.592*** 2.764*** 2.238***
[0.000] [0.000] [0.000] [0.000] [0.000] [0.000]
R&D 0.260*** 0.192*** 0.192*** 0.190*** 0.212** 0.138***
[0.000] [0.000] [0.000] [0.002] [0.012] [0.004]
R&D dummy 0.188*** 0.159*** 0.159*** 0.235*** 0.065 0.121**
[0.000] [0.000] [0.001] [0.001] [0.216] [0.011]
Advertising expense -0.021 -0.036 -0.036 0.062 -0.177 -0.004
[0.658] [0.452] [0.705] [0.510] [0.287] [0.960]
Advertising dummy 0.116*** 0.046** 0.046 0.027 0.079 0.156***
[0.000] [0.046] [0.325] [0.691] [0.184] [0.001]
Investment 1.742*** 1.837*** 1.837*** 1.940*** 1.644*** 1.484***
[0.000] [0.000] [0.000] [0.000] [0.000] [0.000]
Constant 4.035*** 3.574*** 3.574*** 3.831*** 3.057*** 1.785***
[0.000] [0.000] [0.000] [0.000] [0.000] [0.001]
Industry fixed effects No Yes Yes Yes Yes No
Year fixed effects No Yes Yes Yes Yes Yes
Cluster by firm No No Yes Yes Yes Yes
Observations 14962 14962 14962 7916 7046 14962
R20.209 0.299 0.299 0.305 0.315 0.176
51
Table 4: Different Measures of Managerial Ownership
Panel A presents the means of six measures of managerial ownership from 1992-2007; Panel B presents
the corresponding medians. For stocks, pay-performance sensitivity (PPS) is the percentage stock own-
ership; for stock options, PPS equals the number of shares underlying the options times the delta of
each option divided by the total number of shares outstanding. Panel C presents the regression results
of the industry-adjusted Tobin’s Q with three different measures of managerial ownership as explana-
tory variables. The sample consists of 14,962 firm-year observations over the period 1992-2007. See the
Appendix for the descriptions of the variables. The p-values are reported in the brackets. Significance
levels are indicated by *, **, and *** for 10%, 5%, and 1%, respectively.
Panel A: Mean Managerial Ownership
Year Managerial Managerial Managerial CEO CEO CEO
ownership PPS PPS ownership PPS PPS
exc. options options only inc. options exc. options options only inc. options
1992 4.17% 0.79% 4.96% 2.86% 0.23% 3.08%
1993 4.44% 1.13% 5.58% 2.76% 0.48% 3.19%
1994 4.38% 1.30% 5.68% 2.94% 0.57% 3.47%
1995 4.40% 1.45% 5.85% 2.91% 0.65% 3.51%
1996 4.08% 1.58% 5.66% 2.64% 0.73% 3.34%
1997 3.91% 1.77% 5.68% 2.62% 0.82% 3.42%
1998 4.77% 2.18% 6.95% 3.21% 0.97% 4.15%
1999 4.52% 2.29% 6.80% 3.09% 1.04% 4.10%
2000 4.60% 2.48% 7.08% 3.05% 1.13% 4.15%
2001 4.13% 2.49% 6.63% 2.83% 1.17% 3.95%
2002 3.98% 2.54% 6.52% 2.74% 1.21% 3.90%
2003 3.50% 2.45% 5.95% 2.44% 1.17% 3.59%
2004 3.36% 2.29% 5.65% 2.20% 1.10% 3.27%
2005 3.04% 2.04% 5.08% 1.93% 1.02% 2.93%
2006 2.94% 2.17% 5.08% 1.71% 1.10% 2.74%
2007 2.69% 2.06% 4.72% 1.72% 1.06% 2.69%
Total 3.87% 2.03% 5.90% 2.55% 0.97% 3.49%
Panel B: Median Managerial Ownership
Year Managerial Managerial Managerial CEO CEO CEO
ownership PPS PPS ownership PPS PPS
exc. options options only inc. options exc. options options only inc. options
1992 0.54% 0.45% 1.48% 0.16% 0.10% 0.38%
1993 0.87% 0.64% 2.12% 0.30% 0.20% 0.78%
1994 0.88% 0.78% 2.32% 0.33% 0.28% 0.94%
1995 0.80% 0.95% 2.42% 0.33% 0.35% 0.98%
1996 0.69% 1.02% 2.43% 0.28% 0.39% 1.00%
1997 0.73% 1.14% 2.56% 0.31% 0.47% 1.13%
1998 0.93% 1.57% 3.49% 0.38% 0.58% 1.45%
1999 0.85% 1.69% 3.35% 0.33% 0.67% 1.40%
2000 0.91% 1.87% 3.65% 0.34% 0.77% 1.55%
2001 0.79% 1.89% 3.58% 0.31% 0.81% 1.56%
2002 0.80% 1.99% 3.79% 0.33% 0.87% 1.67%
2003 0.72% 1.83% 3.50% 0.30% 0.81% 1.55%
2004 0.76% 1.77% 3.21% 0.31% 0.77% 1.44%
2005 0.75% 1.47% 2.84% 0.31% 0.63% 1.25%
2006 0.73% 1.66% 3.00% 0.29% 0.77% 1.29%
2007 0.70% 1.61% 2.75% 0.28% 0.73% 1.19%
Total 0.77% 1.43% 3.01% 0.31% 0.60% 1.30%
52
Panel C: Regression Results with Different Measures of Managerial Ownership
Dependent Variable Industry-adjusted Tobin’s Q
Measure of managerial ownership Managerial CEO CEO
PPS ownership PPS
inc. options exc. options inc. options
(1) (2) (3)
Managerial ownership 1.526* 2.793** 2.111*
[0.093] [0.022] [0.074]
Managerial ownership squared -0.656 -3.593* -1.770
[0.579] [0.095] [0.372]
E index -0.034* -0.053*** -0.048***
[0.067] [0.002] [0.006]
Managerial ownership ×E index -0.625*** -0.632** -0.687**
[0.003] [0.031] [0.016]
Sales -0.482*** -0.446*** -0.449***
[0.000] [0.001] [0.001]
Sales squared 3.083*** 2.871*** 2.874***
[0.000] [0.001] [0.001]
Tangibility -0.607*** -0.588*** -0.595***
[0.000] [0.000] [0.000]
Tangibility squared 0.111*** 0.107*** 0.108***
[0.001] [0.001] [0.001]
Idiosyncratic risk -0.066*** -0.067*** -0.069***
[0.000] [0.000] [0.000]
Profit margin 2.237*** 2.228*** 2.227***
[0.000] [0.000] [0.000]
R&D 0.131*** 0.139*** 0.132***
[0.007] [0.005] [0.007]
R&D dummy 0.116** 0.124** 0.119**
[0.015] [0.012] [0.015]
Advertising expense -0.003 -0.005 -0.004
[0.972] [0.954] [0.964]
Advertising dummy 0.157*** 0.161*** 0.163***
[0.001] [0.001] [0.001]
Investment 1.491*** 1.529*** 1.525***
[0.000] [0.000] [0.000]
Constant 1.821*** 1.807*** 1.830***
[0.001] [0.001] [0.001]
Year fixed effects Yes Yes Yes
Cluster by firm Yes Yes Yes
Observations 14962 14196 14346
R20.176 0.177 0.176
53
Table 5: Individual Antitakeover Provisions and the G Index
For each of the six antitakeover provisions in the E index, panel A reports the regression results of the
following model:
Qit =β0+β1×Managerial ownershipit +β2×Managerial ownership2
it
+β3×Individual provisionit +β4×Managerial ownershipit ×Individual provisionit
+β5×Xit +uit.
To conserve space, only the coefficients and associated p-values of the provision and its interaction with
managerial ownership are reported. Panel B reports the regression results for the G index. I control for
year fixed effects and cluster the residuals by firm in all the regressions. The sample consists of 14,962
firm-year observations over the period 1992-2007. See the Appendix for the descriptions of the variables.
Significance levels are indicated by *, **, and *** for 10%, 5%, and 1%, respectively.
Panel A: Individual Provisions in the E Index
Provision Ownership ×Provision
Coefficient p-value Coefficient p-value
Staggered board -0.006 [0.896] -1.220** [0.034]
Poison pill -0.094** [0.040] -0.238 [0.742]
Supermajority to approve merger -0.031 [0.495] -0.275 [0.633]
Limits to amend bylaws -0.065 [0.202] -1.150** [0.029]
Limits to amend charter -0.033 [0.723] -2.999*** [0.000]
Golden parachutes -0.137*** [0.003] -1.314*** [0.006]
54
Panel B: The G Index
Dependent Variable Industry-adjusted Tobin’s Q
(1) (2) (3)
Managerial ownership 2.218** 1.325 -0.329
[0.019] [0.244] [0.789]
Managerial ownership squared -2.159 -1.744 -1.610
[0.102] [0.185] [0.214]
E index -0.044***
[0.008]
Managerial ownership ×E index -0.560**
[0.013]
G index -0.022***
[0.007]
Managerial ownership ×G index -0.010
[0.921]
G index - E index -0.022*
[0.065]
Managerial ownership ×(G index - E index) 0.266
[0.139]
Sales -0.476*** -0.491*** -0.521***
[0.000] [0.000] [0.000]
Sales squared 3.071*** 3.223*** 3.424***
[0.000] [0.000] [0.000]
Tangibility -0.598*** -0.608*** -0.610***
[0.000] [0.000] [0.000]
Tangibility squared 0.110*** 0.113*** 0.113***
[0.001] [0.001] [0.001]
Idiosyncratic risk -0.066*** -0.067*** -0.063***
[0.000] [0.000] [0.001]
Profit margin 2.238*** 2.263*** 2.282***
[0.000] [0.000] [0.000]
R&D 0.138*** 0.146*** 0.148***
[0.004] [0.003] [0.002]
R&D dummy 0.121** 0.117** 0.113**
[0.011] [0.014] [0.018]
Advertising expense -0.004 -0.004 0.001
[0.960] [0.964] [0.991]
Advertising dummy 0.156*** 0.158*** 0.160***
[0.001] [0.001] [0.001]
Investment 1.484*** 1.466*** 1.482***
[0.000] [0.000] [0.000]
Constant 1.785*** 1.925*** 1.963***
[0.001] [0.000] [0.000]
Year fixed effects Yes Yes Yes
Cluster by firm Yes Yes Yes
Observations 14962 14962 14962
R20.176 0.171 0.171
55
Table 6: Relation between Managerial Ownership and the E Index
Panel A presents the summary statistics of managerial ownership for five subsamples based on the E
index. Panel B presents the regression results of four measures of managerial ownership on the E index
and other variables. Panel C presents the regression results of the four measures of managerial ownership
on the E index and other variables with firm fixed effects controlled for. The sample consists of 14,962
firm-year observations over the period 1992-2007. See the Appendix for the descriptions of the variables.
The p-values are reported in the brackets. Significance levels are indicated by *, **, and *** for 10%,
5%, and 1%, respectively.
Panel A: Managerial Ownership by E Index
N mean sd p5 p50 p95
E index = 0 1412 7.55% 11.70% 0.04% 1.33% 32.56%
E index = 1 2679 5.40% 10.49% 0.04% 0.92% 27.73%
E index = 2 3825 3.88% 7.78% 0.04% 0.78% 21.67%
E index = 3 3987 2.88% 5.92% 0.05% 0.73% 14.12%
E index ≥4 3059 2.14% 5.11% 0.09% 0.66% 9.81%
Total 14962 3.87% 8.08% 0.05% 0.77% 21.39%
56
Panel B: Regression Results of Four Measures of Managerial Ownership
Dependent Variable Managerial Managerial CEO CEO
ownership PPS ownership PPS
exc. options inc. options exc. options inc. options
(1) (2) (3) (4)
E index -0.011*** -0.010*** -0.008*** -0.008***
[0.000] [0.000] [0.000] [0.000]
Sales -0.001 -0.012 -0.001 -0.007
[0.873] [0.156] [0.832] [0.330]
Sales squared -0.069 -0.030 -0.040 -0.019
[0.189] [0.563] [0.357] [0.664]
Tangibility -0.020** -0.024*** -0.018** -0.018***
[0.025] [0.009] [0.012] [0.010]
Tangibility squared 0.003 0.004* 0.003 0.003
[0.165] [0.088] [0.138] [0.118]
Idiosyncratic risk -0.002 -0.001 -0.001 0.000
[0.195] [0.737] [0.701] [0.970]
Profit margin -0.001 -0.007 0.005 0.003
[0.909] [0.540] [0.521] [0.737]
R&D -0.011*** -0.010*** -0.007*** -0.007***
[0.000] [0.000] [0.001] [0.001]
R&D dummy -0.008 -0.006 -0.007* -0.006
[0.101] [0.194] [0.069] [0.129]
Advertising expense 0.007 0.008 0.005 0.005
[0.333] [0.294] [0.353] [0.309]
Advertising dummy 0.000 0.000 0.002 0.002
[0.935] [0.924] [0.487] [0.427]
Investment -0.010 -0.005 -0.011 -0.009
[0.336] [0.610] [0.202] [0.338]
Constant 0.136*** 0.194*** 0.100*** 0.128***
[0.000] [0.000] [0.001] [0.000]
Industry fixed effects Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes
Cluster by firm Yes Yes Yes Yes
Observations 14962 14962 14196 14346
R20.140 0.171 0.132 0.141
57
Panel C: Regression Results of Four Measures of Managerial Ownership: Firm Fixed Ef-
fects
Dependent Variable Managerial Managerial CEO CEO
ownership PPS ownership PPS
exc. options inc. options exc. options inc. options
(1) (2) (3) (4)
E index -0.001* -0.001 -0.002*** -0.001*
[0.065] [0.471] [0.002] [0.098]
Sales -0.008* -0.014*** -0.003 -0.006
[0.080] [0.006] [0.486] [0.172]
Sales squared -0.009 0.000 -0.032 -0.025
[0.793] [0.998] [0.239] [0.354]
Tangibility -0.002 -0.002 0.002 0.002
[0.731] [0.762] [0.614] [0.579]
Tangibility squared 0.000 0.001 -0.001 -0.001
[0.859] [0.502] [0.135] [0.419]
Idiosyncratic risk 0.001 0.002** 0.001 0.001
[0.237] [0.024] [0.319] [0.300]
Profit margin 0.013** 0.013** 0.019*** 0.020***
[0.012] [0.024] [0.000] [0.000]
R&D -0.003*** -0.003* -0.003** -0.003**
[0.008] [0.056] [0.014] [0.026]
R&D dummy 0.004* 0.003 0.004* 0.003
[0.053] [0.200] [0.079] [0.239]
Advertising expense 0.001 0.001 -0.007** -0.007**
[0.731] [0.741] [0.029] [0.026]
Advertising dummy 0.002 0.001 0.003* 0.002*
[0.313] [0.469] [0.050] [0.087]
Investment 0.006 0.005 0.006 0.005
[0.168] [0.319] [0.108] [0.193]
Constant 0.109*** 0.153*** 0.071*** 0.091***
[0.000] [0.000] [0.000] [0.000]
Firm fixed effects Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes
Observations 14962 14962 14196 14346
R20.785 0.779 0.770 0.759
58
Table 7: Determinants of Tobin’s Q: Firm Fixed Effects
The dependent variables in the two columns are Tobin’s Q and the industry-adjusted Tobin’s Q, re-
spectively. The sample consists of 14,962 firm-year observations over the period 1992-2007. See the
Appendix for the descriptions of the variables. The p-values are reported in the brackets. Significance
levels are indicated by *, **, and *** for 10%, 5%, and 1%, respectively.
Dependent Variable Tobin’s Q Industry-adjusted Tobin’s Q
(1) (2)
Managerial ownership 1.965*** 2.067***
[0.000] [0.000]
Managerial ownership squared -2.367*** -2.125***
[0.000] [0.001]
E index -0.019 -0.013
[0.217] [0.375]
Managerial ownership ×E index -0.281** -0.379***
[0.015] [0.001]
Sales -0.986*** -0.973***
[0.000] [0.000]
Sales squared 4.803*** 4.619***
[0.000] [0.000]
Tangibility -1.129*** -1.103***
[0.000] [0.000]
Tangibility squared 0.202*** 0.198***
[0.000] [0.000]
Idiosyncratic risk -0.056*** -0.060***
[0.000] [0.000]
Profit margin 2.438*** 2.259***
[0.000] [0.000]
R&D -0.074 -0.053
[0.167] [0.299]
R&D dummy -0.025 -0.011
[0.546] [0.780]
Advertising expense 0.009 0.009
[0.908] [0.902]
Advertising dummy -0.138*** -0.120***
[0.000] [0.000]
Investment 1.146*** 1.044***
[0.000] [0.000]
Constant 6.429*** 4.917***
[0.000] [0.000]
Year fixed effects Yes Yes
Firm fixed effects Yes Yes
Observations 14962 14962
R20.714 0.689
59
Table 8: An Event Study: Manager Share Purchases in the Open Market
This table presents the OLS regression results of the 5- and 7-day cumulative abnormal returns com-
mencing from the day before the manager share purchase announcement. The sample consists of 4,128
stock purchases announced by the top five executives of the firms in the ExecuComp database over the
period 1993-2008. See the Appendix for the descriptions of the variables. The p-values are reported in
the brackets. Significance levels are indicated by *, **, and *** for 10%, 5%, and 1%, respectively.
Dependent Variable CAR5 CAR7
(1) (2)
Manager share purchases 4.007** 7.071***
[0.020] [0.009]
Pre-purchase managerial ownership -0.014 -0.021
[0.408] [0.342]
E index -0.001 -0.002
[0.571] [0.201]
Manager share purchases ×Pre-purchase ownership -4.493* -8.634**
[0.066] [0.015]
Manager share purchases ×E index -0.584* -0.959**
[0.063] [0.025]
Constant 0.003 0.009*
[0.449] [0.096]
Industry fixed effects Yes Yes
Year fixed effects Yes Yes
Cluster by firm Yes Yes
Observations 4128 4128
R20.022 0.029
60
Table 9: Instrumental Variable Regressions
Columns (1) and (2) report the OLS regression results where the dependent variables are managerial
ownership and the E index, respectively. Column (3) presents the GMM regression results where the
dependent variable is the industry-adjusted Tobin’s Q. The instrumental variables for managerial own-
ership are the managerial ownership in 1992 and the average managerial ownership of industry peers;
the instrumental variables for the E index are the E index in 1990, the average E index of industry
peers, and the average E index of the firms incorporated in the same state; the instrumental variable
for managerial ownership squared is the square of the fitted managerial ownership from the regression
in column (1); the instrumental variable for the interaction variable of managerial ownership and the
E index is the product of the fitted managerial ownership from the regression in column (1) times the
fitted E index from the regression in column (2). See the Appendix for the descriptions of the variables.
The p-values are reported in the brackets. Robust standard errors are used to compute the p-values.
Significance levels are indicated by *, **, and *** for 10%, 5%, and 1%, respectively.
61
Dependent Variable Managerial E index Industry-adjusted
ownership Tobin’s Q
Stage of Estimation First stage First stage Second stage
Estimation Method OLS OLS GMM
(1) (2) (3)
Managerial ownership in 1992 0.431*** -0.756***
[0.000] [0.003]
Average industry managerial ownership 0.105* 0.975
[0.089] [0.328]
E index in 1990 -0.006*** 0.709***
[0.001] [0.000]
Average industry E index 0.002 0.192***
[0.515] [0.004]
Average state E index 0.007** 0.272***
[0.046] [0.000]
Managerial ownership 8.085***
[0.008]
Managerial ownership squared -11.860**
[0.035]
E index 0.004
[0.922]
Managerial ownership ×E index -1.587*
[0.075]
Sales 0.006 0.792*** 0.193
[0.679] [0.000] [0.486]
Sales squared -0.046 -5.421*** -0.810
[0.586] [0.000] [0.635]
Tangibility -0.004 -0.176 -0.705***
[0.662] [0.265] [0.000]
Tangibility squared 0.001 0.032 0.100
[0.722] [0.455] [0.110]
Idiosyncratic risk -0.001 0.053* -0.019
[0.819] [0.096] [0.546]
Profit margin -0.017 0.237 3.794***
[0.271] [0.420] [0.000]
R&D -0.003 0.102 0.304**
[0.393] [0.218] [0.044]
R&D dummy -0.014*** -0.079 0.197***
[0.008] [0.253] [0.005]
Advertising expense 0.025** -0.050 0.182
[0.047] [0.762] [0.400]
Advertising dummy -0.002 0.000 0.226***
[0.563] [0.998] [0.002]
Investment 0.001 0.102 0.798***
[0.957] [0.651] [0.008]
Constant -0.012 -3.047*** -1.610
[0.872] [0.000] [0.151]
P-value of Sargen-Hansen J-test 0.421
Year fixed effects Yes Yes Yes
Cluster by firm Yes Yes Yes
Observations 4850 4850 4850
R20.466 0.696 0.259
62
Table 10: Controlling for Management Quality
The cumulative abnormal return in the past three years is a proxy for management quality. The sample
consists of 14,936 firm-year observations over the period 1992-2007. See the Appendix for the descriptions
of the variables. The p-values are reported in the brackets. Significance levels are indicated by *, **,
and *** for 10%, 5%, and 1%, respectively.
Dependent Variable E index Managerial Industry-adjusted
ownership Tobin’s Q
(1) (2) (3)
CARs in past three years -0.115*** 0.005*** 0.393***
[0.000] [0.000] [0.000]
Managerial ownership 1.863**
[0.038]
Managerial ownership squared -1.663
[0.184]
E index -0.032**
[0.043]
Managerial ownership ×E index -0.530**
[0.013]
Sales 1.306*** -0.016* -0.503***
[0.000] [0.050] [0.000]
Sales squared -8.663*** 0.030 3.171***
[0.000] [0.564] [0.000]
Tangibility 0.174 -0.022** -0.581***
[0.299] [0.017] [0.000]
Tangibility squared -0.051 0.004 0.105***
[0.292] [0.115] [0.001]
Idiosyncratic risk -0.080*** -0.001 -0.017
[0.001] [0.579] [0.343]
Profit margin -0.736*** 0.004 1.962***
[0.000] [0.707] [0.000]
R&D 0.019 -0.011*** 0.126***
[0.652] [0.000] [0.006]
R&D dummy 0.063 -0.009* 0.085*
[0.402] [0.070] [0.064]
Advertising expense -0.229** 0.010 0.036
[0.011] [0.161] [0.672]
Advertising dummy -0.048 0.001 0.148***
[0.429] [0.825] [0.001]
Investment -0.191 -0.013 0.930***
[0.261] [0.233] [0.000]
Constant -2.257*** 0.163*** 1.925***
[0.000] [0.000] [0.000]
Industry fixed effects Yes Yes No
Year fixed effects Yes Yes Yes
Cluster by firm Yes Yes Yes
Observations 14936 14936 14936
R20.132 0.112 0.228
63
