Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Director Compensation Incentives and AcquisitionPerformance
Ismail LAHLOU∗, Patrick NAVATTEUniversité de Rennes 1 & CREM
January 4, 2017
Abstract
This paper investigates the relation between director compensation structure and share-holder interests in the context of acquisitions. Our evidence suggests that acquirer firmsthat compensate their directors with a higher proportion of incentive-based compen-sation have significantly higher stock returns around the announcement. Compared toacquirers in the low equity-based compensation group, acquirers in the high equity-based compensation group outperform by 9.54% in a five-day period surrounding theannouncement date. These results hold even after controlling for endogeneity issues. Wefurther find that acquirers with higher equity-based pay exhibit greater improvementsin stock price and operating performance in the three-years following acquisitions. Anincrease in director equity-based pay also results in a lower acquisition premium fortargets. These results indicate that equity-based compensation provides incentives fordirectors to make decisions that meet the interests of shareholders.
JEL classification: G30, G34.
Keywords : Board of directors; Compensation; Director incentives; Acquisitions; Bargainingpower; Agency theory
Postal address: The Graduate School of Management, IGR-IAE de Rennes, Université de Rennes 1,11 Rue Jean Macé, CS 70803, 35708 Rennes, Cedex 7, France
E-mail addresses: [email protected], [email protected]∗Corresponding author
1. Introduction
Nowadays, more than eight decades after the Berle and Means (1932)’ pioneering dis-
cussion of the modern corporation, executive compensation is still one of the most debated
topics in corporate finance. This question is probably even more important today given the
major changes occurring in the economy and the protests from several constituent groups
to focus more attention on this issue. Besides executive compensation, directors’ pay has
gradually received more and more attention over the past few years, especially after the
Enron and Worldcom scandals and the resulting Sarbanes Oxley Act of 2002.
Over the last years, the level of director compensation has increased substantially.
For example, among S&P 1500 firms, the total remuneration per director increased from $35
thousand in 1996 to $190 thousand in 2014. The increase in the level of director compensation
is largely due to an increase in director equity-based pay, as the value of stock and options
granted to directors soared from less than $9 thousand to near $109 thousand during the
same period.
Also, acquisitions may present a particular setting in which equity-based compensation
can help strengthen director effort. For example, an increase in directors’ equity ownership
(via new stocks and/or option grants) may provide directors with direct incentives to engage
in active oversight, which can alleviate agency problems. As a consequence of such improved
oversight, directors would approve only value-enhancing acquisitions and prevent managers
from engaging in bad ones, which may lead to higher acquisition performance.
Acquisitions can also provide directors with opportunities that might exacerbate the
potential conflicts of interests between directors and shareholders. Compared with other
corporate investment decisions, acquisitions represent a relatively quick way to increase firm
size. Given that director compensation is positively associated with firm size (Ryan and
Wiggins (2004) and Engel et al. (2010)), one could reasonably expect such compensation
to increase following acquisitions. Moreover, while a decline in stock price might reduce
both directors’ compensation and wealth, these losses may be offset by potential increases
in directors’ compensation as acquiring firm’s directors may receive new grants of stock and
options following acquisitions (see Harford and Li (2007) for evidence on CEO compensa-
1
tion1). Hence, directors desiring a rapid increase in compensation may have strong incentives
to approve acquisitions even if they do not serve shareholders’ interests.
In view of the conflicting predictions regarding the relation between director compen-
sation and acquisition outcomes, we empirically investigate its direction using a sample of
6840 acquisitions completed by S&P 1500 firms between 1998 and 2014.
First, in an attempt to understand how director equity-based pay affects acquisition
decisions, we examine the announcement returns around acquisitions. We show that acquir-
ers with higher equity-based compensation experience higher announcement returns. After
controlling for the CEO’s equity-based compensation2 and a wide range of firm and deal
characteristics, we find that a one standard deviation increase in the percent of director
equity-based pay is associated with an increase of about 2.65% in announcement returns,
which corresponds to an increase of $346 million in shareholder value considering an average
acquirer market capitalization of $13.06 billion. Additionally, when we split acquirers into
three groups based on director equity-based pay, we find that acquirers with high equity-
based compensation outperform their counterparts with low equity-based pay by 9.54% in
terms of the five-day window surrounding the announcement.
Like in many empirical studies, including our own, there is a possibility that our re-
sults are driven by endogeneity. Our analysis provides a number of techniques aimed at
thwarting these concerns. For example, in order to alleviate endogeneity problems which
may arise when our left and right-hand side variables are jointly determined by potentially
unobservable (or uncontrolled for) factors, we employ a two-stage instrumental variables
(IV) approach. The results of the 2SLS estimates show that director equity-based compen-
sation continues to have a positive effect on the acquirer’s abnormal returns. Moreover, to
control for potential endogeneity issues arising from reverse causality, we introduce changes
in equity-based compensation (%) as an alternative proxy in our model. The results are
once again in line with those presented previously. As a further precaution against potential
endogeneity concerns, we include in our models dummy variables to control for industry and1 Harford and Li (2007) provide evidence that bidding firm CEOs are rewarded with substantial new
grants of stock and options after acquisitions. This flow of new grants may offset the negative effect of poormerged-firm stock performance on the CEO’s pre-acquisition portfolio and reduce the sensitivity of his/hercompensation and wealth to negative stock performance.
2 The CEO’s equity-based compensation has been used in order to isolate the effect of director equity-based pay from the effect induced by the CEO’s equity-based pay.
2
year fixed effects, and use robust standard errors.
Second, in order to capture the real economic gains associated with acquisitions, we
investigate how director equity-based pay is related to changes in firm performance following
acquisitions. Our results reveal that acquirer firms with higher equity-based pay exhibit
significant improvements in stock returns, operating performance and efficiency in the 3-
years following the acquisition. We also note that these results are robust across the 2SLS
method.
Last, to explore the source of acquisition value gains, we examine the offer premiums
paid by acquiring firms.3 Our results suggest that directors with higher equity-based pay are
more likely to prevent managers from overpaying for targets. Interestingly, a one standard
deviation increase in director equity-based pay reduces the bid premium by approximately
4.77% which translates into considerable savings ($29.6 million) for the average target, whose
market capitalization amounts to $620 million. These results clearly show that providing
equity-based incentives to outside directors can have a non-trivial effect on shareholders’
wealth.
This study contributes to the literature in an important way. We extend the body of
research that highlights the importance of equity-based compensation by providing evidence
that the use of incentive-based compensation schemes to reward directors also matters. While
prior studies (see, Datta et al. (2001); Minnick et al. (2011)) find that high managerial equity-
based pay leads to improved stock price performance around and following the announcement
date of the acquisition, our article yields analogous results using director equity-based pay.
The remainder of this article is organized as follows. Section 2 reviews the literature
and develops our hypotheses. Section 3 describes the sample and provides summary statis-
tics. In Section 4, we examine how director equity-based pay affects acquisition decisions.
Conclusions are drawn in Section 5.3 The acquisition premium is defined as the ratio of the final offer price to the target share price four
weeks prior to the acquisition announcement, minus one.
3
2. Literature review and hypotheses development
Ensuring that managers act in the best interests of shareholders is definitely one of
the most important duties of corporate boards. Nevertheless, it has been argued that,
instead of acting as stewards of shareholder interests, outside board members may collude
with management, which, in turn, may lead to agency problems between directors and
shareholders (Fama and Jensen (1983)). As pointed out by Jensen and Meckling (1976), such
agency concerns can be mitigated through a number of mechanisms, including an increase
in equity ownership. Accordingly, equity-based compensation plans can be an effective way
to align directors’ interests with those of shareholders.
Acquisitions constitute an additional setting that could potentially compromise board
members’ independence, since acquisitions can provide them with a variety of benefits, es-
pecially in the form of higher compensation. Moreover, impaired directors’ independence
may be harmful for shareholders in the context of acquisitions, because acquisitions are po-
tentially associated with large wealth destruction (Moeller et al. (2004)). At the same time,
incentive-based compensation awarded to directors can increase director effort (Hermalin
and Weisbach (1998)), which can ultimately translate into value creation. Additionally, the
board of directors has a critical role to play in such important corporate decisions as ac-
quisitions (Byrd and Hickman (1992) and Subrahmanyam et al. (1997)), as directors have a
fiduciary responsibility to evaluate investment proposals presented by management to ascer-
tain whether they are in the best interests of shareholders (Weiss (1991)). Hence, one can
reasonably expect the board of directors to have a larger impact on acquisition performance
than on the firm’s overall performance, which is contingent on a broader set of environmental
and organizational factors (Hermalin and Weisbach (2003)).
There is a substantial body of corporate governance literature that focuses on the
relation between CEO compensation and corporate acquisitions (see, e.g., Datta et al. (2001);
Bebchuk and Fried (2003); Grinstein and Hribar (2004) and Minnick et al. (2011)).
Some authors find a positive relation between acquiring managers’ equity-based pay and
long-run performance following acquisitions (Datta et al. (2001)), which support the hypoth-
esis that incentive compensation aligns the interests of managers with those of shareholders.
Consistent with this, Minnick et al. (2011) show that incentive-based compensation reduces
4
the likelihood of engaging in value-destroying acquisitions and promotes value-improving
acquisitions.
With respect to incentive compensation provided to corporate directors, one may ar-
gue that director equity-based pay can also help to align directors’ interests with those of
shareholders. Given that their stock ownership will increase with equity-based pay, directors
will experience in a more direct way the effects of their actions.
Additionally, Bebchuk and Fried (2003) argue that while acquisition decisions that
are value-decreasing for shareholders would reduce the value of the current options owned
by managers, such acquisitions may raise managers’ future compensation by an even larger
amount, which can be justified by the increase in firm size. This higher pay might encourage
managers to engage in value-destroying acquisitions. In the same vein, Harford and Li (2007)
demonstrate that “even in mergers where bidding shareholders are worse off, bidding CEOs
are better off three quarters of time”. The same authors further argue that, following a
merger, acquiring CEOs are rewarded with substantial new option and stock grants, making
their total remuneration and overall wealth “markedly” insensitive to performance. These
findings complement those of Bliss and Rosen (2001), who find that CEO pay and wealth
typically increase following bank mergers even if the acquirer’s stock price falls. Overall,
these studies suggest that CEO compensation increases after acquisitions.
Hence, since it has been shown that directors’ compensation is positively related to
CEO compensation (Brick et al. (2006)) as well as firm size (Ryan and Wiggins (2004) and
Engel et al. (2010)), it would be logical to expect directors’ compensation to increase after
acquisitions. If directors anticipate such an increase in their compensation, they may be
inclined to ratify acquisitions even if they fail to serve the interests of shareholders.
3. Sample Selection and Data
This section is partitioned into two parts. The first one describes the sources of data
while the second one discusses descriptive statistics for the key variables included in the
econometric models.
5
3.1. Sources of data
Data were assembled from several sources. Director compensation data are drawn from
the Standard and Poor’s ExecuComp database. From this database, we are able to measure
the percent of director equity-based pay that we define as the value of stock and options
granted by a firm to an average director, divided by total director compensation. Total
director compensation is the average annual compensation that a firm pays each director
and is calculated by dividing the total amount that a company pays its board in a given year
(total board compensation) by the number of outside directors in that year.4 To measure
total board compensation, we use the sum of annual cash retainer, fees paid to directors
for attending board meeting, value of stock grants and value of option awards. We use the
Black and Scholes (1973)5 option valuation model to compute the value of option awards
assuming that the strike price is equal to the closing stock price at the end of the previous
fiscal year, the time to maturity is 7 years and the risk-free rate is the 7-year Treasury bond
yield (item Risk_Free_Rate in ExecuComp). We use the standard deviation of stock prices
over 60 months (item BS_Volatility in ExecuComp) and the company’s average dividend
yield over a 3-year period (item BS_Yield in ExecuComp) in our calculations.6 The value
of stock grants is computed as the number of shares granted multiplied by the closing stock
price at the end of the previous fiscal year.
We also gather from this database information on CEO characteristics and compensa-
tion. We extract data on board attributes from the RiskMetrics Director Database, financial
statement data from the Compustat database. We obtain anti-takeover provision data from
the RiskMetrics Governance Database.
After merging the above sources of data, we follow standard practice and exclude
financial and utility firms. This leaves us with a final sample of 13,802 firm-years for 1,356
unique firms over the course of our 1998-2014 sample period. We also note that the sample4 One might suppose that inside (or employee) directors are paid for their service on the board. Nev-
ertheless, this compensation is not separately disclosed and is presumably included in their executive payand therefore unobservable (Fedaseyeu et al. (2013)). For convenience purposes, we use the term “directorcompensation” to refer to outside (or non-employee) director compensation.
5 As modified by Merton (1973) to adjust dividend payouts.6 We acknowledge that the Black-Scholes model suffers from some limitations, the basic one being that it
is based on some unrealistic assumptions. Despite its flaws, the BS model is very useful for approximatingthe price of an option.
6
size depends on data availability and therefore varies according to variables used in each
regression.
We collect our acquisition sample from the SDC platinum database. The sample covers
acquisitions announced by S&P 1500 firms between January 1, 1998 and December 31, 2014.7
3.2. Summary statistics
Table 1 reports descriptive statistics for the primary variables used in our analysis.
The appendix A provides detailed definitions and sources of all variables.
[Insert Table 1 near here]
The Panel A of Table 1 summarizes the board and CEO characteristics. It indicates
that the mean (median) firm in our sample has 8.89 (9) board members. The mean (median)
proportion of independent directors is 0.69 (0.73). Additionally, we find that, while boards
are dominated by independent directors, there is a relatively large variation among the
sample firms, with a standard deviation amounting to 18%. Female directors make up 10%
of the board. The average director is 60 years old and has served on the board for 8 years.
We also observe that the CEO and Chairman positions are combined in 56% of firm years.
The mean and median CEO age are both equal to 55. The average CEO tenure is 7.05
years. This panel also presents descriptive statistics on director and CEO compensation for
our sample firms. It reveals that the mean equity-based compensation granted to directors
(the CEO) is 45.4% (56.7%) of directors’ (the CEO’s) total compensation.
Panel B of Table 1 reports summary statistics on firm and ownership characteristics.
It shows that our sample firms are fairly large, with a mean value of total assets of $9.32
billion and a mean market capitalization of $13.06 billion. The mean leverage, calculated
as long-term debt divided by total assets is 0.183. The mean MTB ratio is 3.93. Panel B
also indicates that the percentage of outstanding shares owned by all independent directors
is about 1.2%, while CEO ownership averages 3%.7 We require such deals to meet the following criteria: (a) the transaction is completed; (b) the deal value
is greater than $1 million; (c) the acquirer owns at least 50% of the target company after the transaction;(d) the acquirer has director compensation data available from the ExecuComp database one year prior tothe acquisition announcement. Applying these criteria, we end up with a sample of 6840 acquisitions
7
Panel C provides summary statistics of variables capturing deal characteristics. It
illustrates that the mean transaction value is 13% of the bidder’s pre-announcement market
capitalization. This is consistent with the findings of Masulis et al. (2007), who study
a sample of acquisitions completed by firms included in the IRRC antitakeover provision
database between 1990 and 2003. Moreover, about 13% of the acquisitions in our sample
involve publicly traded firms, and 59% involve firms in the same industry.
Table 2 reports the correlation matrix for the independent variables used in our multi-
variate analyses. None of these correlations appears large enough to cause multicollinearity
issues for our analyses. Indeed, the highest absolute correlation magnitude between variables
is 0.457, which is below the threshold of 0.8 beyond which multicollinearity problems arise.
We find, for example, that the Pearson correlation coefficient between director equity-based
compensation and CEO equity-based compensation is 0,34. To double check for any multi-
collinearity issue, we also perform the Variance-Inflation-Factor (VIF) analyses for all tests.
We find that the highest VIF is 2.06 which is considerably less than the 10 threshold above
which multicollinearity could be an issue (see Gujarati (2003)).
[Insert Table 2 near here]
In the next section, we investigate whether the use of equity-based incentives in the
compensation packages of directors aligns their interests with those of shareholders in the
specific context of acquisitions.
4. Director equity-based pay and corporate acquisitions
This section is divided into three parts. The first part examines the effect of director
equity-based compensation on stock returns accompanying acquisition announcements. The
second part investigates how director equity-based pay is related to changes in operating
performance after acquisitions while the last part performs some robustness checks.
8
4.1. Announcement stock returns
We compute the acquirer’s cumulative abnormal returns over the five-day event win-
dow; CAR [-2, 2], where day 0 denotes the announcement date as reported in the SDC
database.8 The cumulative abnormal returns are calculated using the standard market model
procedure with parameters estimated over a period of 255 days (-301, -46), preceding the
date of announcement.
To investigate the relation between director equity-based pay and acquisition announce-
ment returns, we estimate an ordinary least squares (OLS) regression, including year and
industry fixed effects, where the dependent variable is the five-day cumulative abnormal
returns. Our key explanatory variable is the percent of equity-based compensation for di-
rectors, which is defined as the value of stock and options awarded to an average director
divided by the total pay. We also control for a wide array of variables identified in prior
literature as having significant effects on acquisition returns.
These variables include the acquirer’s size, leverage, free cash-flow, the strength of
shareholder rights, the target’s size relative to the acquirer, the form of payment, whether
the bidder and target operate in the same industry, whether the target is a private or a public
company, whether the deal is hostile or friendly, board size, board independence, CEO age
and tenure, the average age of directors.
In addition, Datta et al. (2001) provide evidence that stock price performance around
acquisition announcements is positively related to acquiring managers’ equity-based pay.
Therefore, in order to isolate the effect of director equity-based pay from the effect induced
by the CEO’s equity-based pay, we use the percentage of equity-based compensation for
the CEO as well as the fraction of shares held by the latter as additional control variables.
We also include the average director tenure on the board as an additional control variable,
as it is reasonable to expect that long-tenured directors have accumulated more specialized
knowledge about the firm’s business environment and operations (Vafeas (2003)), and can
therefore help managers make better M&A deals. Moreover, motivated by the insights of8 Given that the announcement dates provided by the SDC database are not always accurate, we follow Cai
and Sevilir (2012) and Masulis et al. (2007), and use a five-day window centered on the announcement dateto estimate abnormal announcement returns. Indeed, using a random sample of 500 acquisitions announcedfrom 1990 to 2000, Fuller et al. (2002) find that the announcement date recorded by SDC was correct at92.6% of the sample, while it was off by no-more than 2 days in the remaining cases.
9
earlier studies on the impact of gender diversity in the boardroom and firm performance (see,
Adams and Ferreira (2009); Dowling and Aribi (2013); Levi et al. (2014), among others), we
control for the fraction of female directors on the board.
Table 3 displays the results of our regressions. In the first column of the table, we
use equity-based compensation (%) as our key explanatory variable. In the second column,
we divide acquirers, in each year, into two groups based on the equity-based compensation.
"High EBC", "Medium EBC" and "Low EBC" are dummy variables that take the value
of one if the acquirer is in the top, middle or bottom third of all firms (acquirers and non-
acquirers) in a given year.
[Insert Table 3 near here]
The first column of Table 3 shows that the coefficient on equity-based compensation
is positive and significant, suggesting that acquirers with higher equity-based compensation
tend to outperform acquirers with low equity-based pay. Moreover, when we divide acquirers
into three groups using the equity-based compensation, we observe that the acquirers in the
high equity-based compensation group outperform their counterparts with low equity-based
compensation group by 9.54% in terms of the five-day period surrounding the announcement.
Moreover, Table 3 indicates that CEO equity-based compensation is negative and sig-
nificant at the 10% level, indicating that, in order to increase their compensation, CEOs are
perhaps inclined to pursue acquisitions even if they fail to serve shareholders’ interests.
4.2. Performance change after acquisitions
The results obtained so far in this section indicate that firms with higher director
equity-based pay have significantly better abnormal returns around the announcement date.
Now, to assess the real economic consequences associated with acquisitions, we examine
how director equity-based compensation is related to changes in firm performance after the
acquisition. To this end, we use the following performance measures. The long-run abnormal
stock returns, return on assets as well as an efficiency ratio.
We measure long run abnormal returns using two methods. The first one is the long-
10
run cumulative abnormal returns for each acquiring firm, calculated over a (0, 36) month
horizon, as illustrated in the following equation:
CARi =T∑t=1
(Ri,t −Rbenchmark,t), (1)
where
Ri,t is the monthly return of a sample firm,
Rbenchmark,t is the monthly return of an appropriately matched control firm.9
Barber and Lyon (1997) as well as Kothari and Warner (1997) argue that event studies
that focus on long-run CARs are misspecified, and identify three reasons to such misspec-
ification, namely, a new listing bias, a rebalancing bias and a skewness bias. Barber and
Lyon (1997) document that the test statistics used to evaluate the buy-and-hold abnormal
returns calculated using an appropriately matched control firm yields well specified results.
We therefore use the buy-and-hold abnormal returns as an alternative measure to the use of
cumulative abnormal returns. To measure the buy-and-hold abnormal returns, we follow the
same procedure as in Barber and Lyon (1997) and Kothari and Warner (1997). We begin
by compounding monthly returns for each acquiring firm over a (0, 36)-month time horizon,
to obtain monthly buy-and-hold returns.10 We then compute buy-and-hold returns of an
appropriately matched control firm over the same time period. As a final step, we compute
buy-and-hold abnormal returns by subtracting buy-and-hold returns of the benchmark from
the buy-and-hold returns of the sample firms, as shown in the following formula:
BHARi =T∏t=1
(1 +Ri,t)−T∏t=1
(1 +Rbenchmark,t). (2)
9 To compute abnormal returns, we follow Barber and Lyon (1997) and use a control firm approach,where sample firms are matched to a control firm based upon size and book to market. The pool of potentialcontrol firms is first screened on the basis of size (market value of equity between 70% and 130% of thesample firm’s market value of equity). The firm with the closest book-to-market ratio to that if the samplefirm is then selected as a control firm.
10We also note that if a sample firm is delisted during the measurement time interval, the buy-and-holdreturn of that particular firm is computed over the time period for which return data are available in thecompustat database.
11
We compute return on assets as operating income before depreciation divided by total
assets. We then compute the change in this measure to proxy for the change in operating
performance. Changes in return on assets is calculated using quarterly data to more accu-
rately reflect the timing of acquisitions (see also Minnick et al. (2011)). Specifically, this
change is calculated using the difference between return on assets in the first quarter follow-
ing the completion of the acquisition and twelve quarters later. We also proxy for employee
productivity using an efficiency ratio, calculated as total assets divided by the number of
full-time employees (see Cornett and Tehranian (1992)). Given that quarterly data on the
number of full-time employees are not available in the compustat database, we use annual
data to calculate the change in the efficiency ratio.
The results are tabulated in Table 4. The dependent variables in the regressions pre-
sented in columns 1-4 of this table are the long-run cumulative abnormal returns, and buy-
and-hold abnormal returns, the change in return on assets and efficiency ratio, respectively.
While our primary explanatory variable of interest is director equity-based compensation
(%), we include several control variables that could potentially affect the change on acquir-
ers’ performance following acquisitions. We also introduce in these regressions year and
industry fixed effects and use robust standard errors.
[Insert Table 4 near here]
The results indicate that the coefficients on equity-based compensation are positive
and significant in all of the four models. We find, for example, that director equity-based
compensation is significantly related to the three-year long-run abnormal return (see the first
two columns of Table 4). The third column of this table also shows that firms with higher
director equity-based pay experience a significant increase in their operating performance
after acquisitions. These results suggest that high director equity-based pay (%) may benefit
not only equity holders but other stakeholders, including creditors, employees, etc. The
results reported in the last column of Table 4 also reveal a positive relationship between
director equity-based pay and the change in the efficiency ratio, suggesting that higher
directors’ incentives may also have a positive impact on employee productivity.
12
4.3. Robustness tests and additional analyses
In this subsection, we conduct a number of robustness tests and additional analyses to
deal with various economic and econometric issues. The subsection is therefore divided into
two parts. In the first part, we attempt to control for endogeneity using several approaches.
The second part investigates whether director equity-based compensation is related to the
acquisition premium paid for the target firm.
4.3.1. Estimations that account for endogeneity
A common concern that may plague our empirical work is the question of the possible
endogeneity. Thus, before reaching the conclusion that high equity based pay causes higher
stock returns around the announcement, we must first address the potential endogeneity
problems. In what follows, we present the results from some approaches aiming to deal with
these issues.
Some unobservable (or uncontrolled for) factors might be responsible for both the
fraction of director equity based pay and acquisition announcement returns. To tackle this
issue, we use a two-stage estimation procedure which enables us to disentangle the effect
of director equity-based pay on announcement returns. It has been argued that a firm’s
managerial pay may "crucially" depend on the compensation provided by competitors in the
same industry (Adams et al. (2011)). Following these authors and extending their line of
reasoning, we use the industry median equity-based compensation (%) as an instrumental
variable, as a firm’s board member compensation may also be affected by the industry.
Our first stage model involves regressing director equity-based pay (%) on the afore-
mentioned instrumental variable along with all other variables listed in Table 3, while the
second-stage model predicts the five-day cumulative abnormal returns using a fitted value
from the first stage as an independent variable. The results of the first and second stage
regressions are presented in the first and second columns of Table 5, respectively. The sec-
ond column of Table 5 shows that director equity-based pay continues to exhibit a positive
relation with the acquirer’s abnormal returns.
[Insert Table 5 near here]
13
Moreover, to explore the robustness of our findings concerning the effect of director
equity-based compensation on changes in performance after acquisitions, we perform a two-
stage least squares estimation using the instrumental variable described above, namely, the
industry median equity-based compensation (%). The results of the second stage regression
are reported in Table 6. These results are largely consistent with those reported in Table 4.
[Insert Table 6 near here]
In addition to omitted variable bias, another possible source of endogeneity may stem
from reverse causality. To rule out this possibility, we include changes in equity-based pay
(%) as an alternative proxy in our model. Hence, if an increase in this measure (i.e., an
increase in director incentive alignment) results in higher announcement returns, the effect
of equity-based compensation on the five-day cumulative abnormal returns can be interpreted
as causal. The results are displayed in Table 7. The results are once again highly consistent
with those presented in Table 3.
[Insert Table 7 near here]
As a further safeguard against endogeneity arising from reverse causality, we also es-
timate a model using lagged values of the explanatory variables. The results with lagged
variables are shown in Table 8. These results do not deviate from the outcomes presented
in Table 3.
[Insert Table 8 near here]
4.3.2. Acquisition premiums
Roll (1986) argues that hubris infected managers attempt to maximize value, but over-
estimate the value of takeovers and ultimately pay too much for target firms. In contrast,
Shleifer and Vishny (1988) suggest that managers overpay for targets not because they make
valuation mistakes, but rather to derive personal benefits from acquisitions that do not create
value for acquiring firm’s shareholders. One would therefore expect that if equity-based pay
increases directors’ monitoring efforts, then directors with higher equity-based compensation
14
may prevent managers from overpaying for targets. To verify this assumption, we investigate
the relation between equity-based pay (%) and the acquisition premium, which is calculated
as the ratio of the final offer price to the target share price four weeks prior to the acquisition
announcement, minus one.
Table 9 presents estimates from OLS regressions with both industry and year fixed
effects, where the dependent variable is the acquisition premium. The main explanatory
variable is director equity-based pay. We also include several control variables that are
likely to affect the acquisition premium. For example, Moeller et al. (2004) show that larger
acquirer pay, on average, higher premiums since large firm managers are more likely to be
driven by hubris. We therefore use the natural logarithm of the market capitalization as
a proxy for the acquirer size. Officer (2003) argues that premiums in intra-industry deals
are larger than those paid in inter-industry mergers. To control for this possibility, we
include a dummy variable that equals one if the target and acquirer share the same two-
digit SIC code and zero otherwise. In addition, it has been documented that cash financed
acquisitions usually have higher premiums than those paid by stock, since target shareholders
are expected to be compensated for the immediate tax liabilities associated with cash offers
(Savor and Lu (2009)). To test for this effect, we add a dummy variable: All Cash, that
takes the value of one if the deal is entirely paid in cash, and zero otherwise.11
[Insert Table 9 near here]
The first column of Table 9 reveals that the coefficient on director equity-based compen-
sation (%) is negative and significant, indicating that higher equity-based pay is associated
with lower acquisition premiums. This is consistent with the findings of Nguyen (2014) who
showed that paying a higher proportion of equity-based compensation to directors is posi-
tively related to monitoring activity. Furthermore, when we split acquirers into three groups
based on director equity-based pay, we find that the negative effect of director equity-based
compensation on the offer premium increases monotonically. These results add further cre-11 The remaining firm-specific and governance related control variables are motivated by the findings of
Alexandridis et al. (2013); Chen et al. (2007); Cotter et al. (1997); Datta et al. (2001); ?); Shivdasani (1993).These variables include the acquirer’s leverage, board size, board independence, an indicator variable forwhether the board is busy, the entrenchment index, CEO age, CEO tenure, CEO duality, CEO ownership,the fraction of equity-based compensation in the CEO pay package, and the fraction of female directors onthe board. We also include year and industry fixed effects and use robust standard errors.
15
dence to the view that equity-based pay is an effective means of aligning directors’ interests
with those of shareholders.
5. Conclusion
The principal objective of this paper is to investigate whether equity-based incentives
for directors align their interests with those of shareholders in the specific context of acqui-
sitions.
Specifically, we analyze how director equity-based compensation affects the bidders’
stock prices around the time of the announcement, the acquisition premium paid for target
firms and changes in performance following the acquisition.
Our results provide evidence that firms with higher director equity-based pay experi-
ence higher abnormal stock returns around the announcement. On average, acquirers in the
high equity-based compensation group outperform their counterparts in the low equity-based
compensation group in a five-day window centered on the announcement date.
Furthermore, firms in which directors’ interests are better aligned with those of share-
holders pay significantly lower acquisition premia. Our results also clearly show that acquir-
ers with higher equity-based compensation exhibit greater improvements in stock price and
operating performance following acquisitions, suggesting that various stakeholders benefit,
not only shareholders.
In sum, our results indicate that equity-based compensation provides effective motiva-
tion for directors to make decisions consistent with value maximization.
16
Acknowledgement
We want to thank John Bizjak, Pascal Dumontier, Edith Ginglinger for their construc-
tive comments at the EUROFIDAI and AFFI meetings.
Moreover, the corresponding author is also greatly indebted to Prof. Ulrich HEGE,
who accepted to invite him at HEC Paris under the Individual Research Program. He
is extremely thankful to Prof. HEGE for useful discussions and clever advice as well as
for providing him access to all campus infrastructure, including the library facilities in the
Finance Department.
Besides this, we gratefully acknowledge financial support from the CREM (UMR CNRS
6211) and the IGR Foundation. All remaining errors are ours.
17
ReferencesAdams, M., Lin, C., and Zou, H. (2011). Chief executive officer incentives, monitoring, and
corporate risk management: Evidence from insurance use. Journal of Risk and Insurance,78(3):551–582.
Adams, R. B. and Ferreira, D. (2009). Women in the boardroom and their impact ongovernance and performance. Journal of Financial Economics, 94(2):291–309.
Alexandridis, G., Fuller, K. P., Terhaar, L., and Travlos, N. G. (2013). Deal size, acquisitionpremia and shareholder gains. Journal of Corporate Finance, 20:1–13.
Barber, B. M. and Lyon, J. D. (1997). Detecting long-run abnormal stock returns: The empir-ical power and specification of test statistics. Journal of Financial Economics, 43(3):341–372.
Bebchuk, L. A. and Fried, J. M. (2003). Executive compensation as an agency problem.Journal of Economic Perspectives, 17:71–92.
Berle, A. A. and Means, G. G. C. (1932). The modern corporation and private property.Macmillan, New York.
Black, F. and Scholes, M. (1973). The pricing of options and corporate liabilities. TheJournal of Political Economy, pages 637–654.
Bliss, R. T. and Rosen, R. J. (2001). CEO compensation and bank mergers. Journal ofFinancial Economics, 61(1):107–138.
Brick, I. E., Palmon, O., and Wald, J. K. (2006). Ceo compensation, director compensation,and firm performance: evidence of cronyism? Journal of Corporate Finance, 12(3):403–423.
Byrd, J. W. and Hickman, K. A. (1992). Do outside directors monitor managers?: Evidencefrom tender offer bids. Journal of Financial Economics, 32(2):195–221.
Cai, Y. and Sevilir, M. (2012). Board connections and m&a transactions. Journal of Finan-cial Economics, 103(2):327–349.
Chen, X., Harford, J., and Li, K. (2007). Monitoring: Which institutions matter? Journalof Financial Economics, 86(2):279–305.
Cornett, M. M. and Tehranian, H. (1992). Changes in corporate performance associatedwith bank acquisitions. Journal of Financial Economics, 31(2):211–234.
Cotter, J. F., Shivdasani, A., and Zenner, M. (1997). Do independent directors enhance tar-get shareholder wealth during tender offers? Journal of Financial Economics, 43(2):195–218.
Datta, S., Iskandar-Datta, M., and Raman, K. (2001). Executive compensation and corpo-rate acquisition decisions. The Journal of Finance, 56(6):2299–2336.
Dowling, M. and Aribi, Z. A. (2013). Female directors and UK company acquisitiveness.International Review of Financial Analysis, 29:79–86.
18
Engel, E., Hayes, R. M., and Wang, X. (2010). Audit committee compensation and thedemand for monitoring of the financial reporting process. Journal of Accounting andEconomics, 49(1):136–154.
Fama, E. F. and Jensen, M. C. (1983). Separation of ownership and control. The Journalof Law & Economics, 26(2):301–325.
Fedaseyeu, V., Linck, J. S., and Wagner, H. F. (2013). The determinants of director com-pensation. Available at SSRN.
Fuller, K., Netter, J., and Stegemoller, M. (2002). What do returns to acquiring firms tell us?evidence from firms that make many acquisitions. The Journal of Finance, 57(4):1763–1793.
Grinstein, Y. and Hribar, P. (2004). Ceo compensation and incentives: Evidence from m&abonuses. Journal of Financial Economics, 73(1):119–143.
Gujarati, D. N. (2003). Basic econometrics. 4th. New York: McGraw-Hill.
Harford, J. and Li, K. (2007). Decoupling ceo wealth and firm performance: The case ofacquiring ceos. The Journal of Finance, 62(2):917–949.
Hermalin, B. E. and Weisbach, M. S. (1998). Endogenously chosen boards of directors andtheir monitoring of the ceo. American Economic Review, 88(1):96–118.
Hermalin, B. E. and Weisbach, M. S. (2003). Boards of directors as an endogenously deter-mined institution: A survey of the economic literature. Technical report.
Jensen, M. C. and Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agencycosts and ownership structure. Journal of Financial Economics, 3(4):305–360.
Kothari, S. and Warner, J. B. (1997). Measuring long-horizon security price performance.Journal of Financial Economics, 43(3):301–339.
Levi, M., Li, K., and Zhang, F. (2014). Director gender and mergers and acquisitions.Journal of Corporate Finance, 28:185–200.
Masulis, R. W., Wang, C., and Xie, F. (2007). Corporate governance and acquirer returns.The Journal of Finance, 62(4):1851–1889.
Merton, R. C. (1973). Theory of rational option pricing. The Bell Journal of Economicsand Management Science, pages 141–183.
Minnick, K., Unal, H., and Yang, L. (2011). Pay for performance? ceo compensation andacquirer returns in bhcs. Review of Financial Studies, 24(2):439–472.
Moeller, S. B., Schlingemann, F. P., and Stulz, R. M. (2004). Firm size and the gains fromacquisitions. Journal of Financial Economics, 73(2):201–228.
Nguyen, N. Q. (2014). On the compensation and activity of corporate boards. Journal ofCorporate Finance, 29:1–19.
Officer, M. S. (2003). Termination fees in mergers and acquisitions. Journal of FinancialEconomics, 69(3):431–467.
19
Roll, R. (1986). The hubris hypothesis of corporate takeovers. Journal of Business, pages197–216.
Ryan, Jr, H. E. and Wiggins, III, R. A. (2004). Who is in whose pocket? director com-pensation, board independence, and barriers to effective monitoring. Journal of FinancialEconomics, 73(3):497–524.
Savor, P. G. and Lu, Q. (2009). Do stock mergers create value for acquirers? The Journalof Finance, 64(3):1061–1097.
Shivdasani, A. (1993). Board composition, ownership structure, and hostile takeovers. Jour-nal of Accounting and Economics, 16(1):167–198.
Shleifer, A. and Vishny, R. W. (1988). Value maximization and the acquisition process. TheJournal of Economic Perspectives, 2(1):7–20.
Subrahmanyam, V., Rangan, N., and Rosenstein, S. (1997). The role of outside directors inbank acquisitions. Financial Management, pages 23–36.
Vafeas, N. (2003). Length of board tenure and outside director independence. Journal ofBusiness Finance & Accounting, 30(7-8):1043–1064.
Weiss (1991). The board of directors, management, and corporate takeovers: Opportunitiesand pitfalls. Arnold Sametz (Ed.), The battle for corporate control (Business One Irwin,Homewood, IL.).
20
Table 1: Descriptive statistics
The following table shows the mean, standard deviation, first quartile, median, third quartile and number of observation (N)for the primary variables used in our analysis. The sample consists of 13,802 firm year observations for 1,356 firms between1998 and 2014. We obtain director compensation data from ExecuComp, board data from RiskMetrics Directors Database.Financial statements are taken from Compustat Database. We gather acquisition data from the SDC platinum database.Panel A summarizes board and CEO characteristics. Panel B reports summary statistics of firm and ownership characteristicsand Panel C shows summary statistics of deal characteristics.
Variables Mean SD Q1 Median Q3 N
Panel A: Board and CEO characteristicsBoard size 8,89 2,47 7 9 10 10110Board independence 0,69 0,18 0,60 0,73 0,83 10110Female Director 0,10 0,10 0 0,10 0,17 9919Director Age 60 4,78 57 60 63 9919Director Tenure 8,06 3,97 5 8 10 7897Director EBC 0,454 0,268 0,218 0,500 0,642 7292CEO Duality 0,56 0,50 0 1 1 10110CEO Age 55 7,57 50 55 60 9530CEO Tenure 7,05 7,18 2 5 10 9588CEO EBC 0,567 0,312 0,372 0,665 0,812 9716
Panel B: Firm & ownership characteristicsTotal Assets ($ billions) 9,32 36,67 0,44 1,36 4,83 11287Market capitalization ($ billions) 13,06 42,40 0,53 1,60 5,87 12139Leverage 0,18 0,21 0,01 0,15 0,28 11242Return On Assets 4,23 16,09 2,33 5,54 9,16 11285MTB ratio 3,93 22,39 1,59 2,47 4,02 10672CEO Ownership 0,030 0,068 0,003 0,009 0,025 7909Directors Ownership 0,012 0,045 0,001 0,003 0,008 7909
Panel C: Deal characteristicsRelative Size 0,13 1,07 0,01 0,03 0,10 7775Public Target 0,13 0,33 0 0 0 20689Intra-industry 0,59 0,49 0 1 1 20689
21
Table 2: Pearson correlation matrix
The following table shows Pearson pair-wise correlation matrix. P-values are given in parentheses. The variables are defined in appendix A.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1 Director EBC
2 Firm size0,046(<0,05)
3 Leverage-0,041 -0,067(<0,05) (<0,05)
4 Relative size-0,022 -0,135 0,041(0,073) (<0,05) (<0,05)
5 Intra-industry0,028 -0,059 -0,114 0,006(<0,05) (<0,05) (<0,05) (0,529)
6 Public target-0,069 0,174 -0,055 0,052 0,109(<0,05) (<0,05) (<0,05) (<0,05) (<0,05)
7 Board size-0,143 0,449 -0,046 -0,043 -0,024 0,197(<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05)
8 Board indep.0,226 0,124 -0,047 -0,053 -0,080 -0,013 0,112(<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (0,241) (<0,05)
9 Busy Board-0,055 0,457 -0,008 -0,032 -0,088 0,068 0,371 0,200(<0,05) (<0,05) (0,475) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05)
10 CEO age-0,145 0,070 0,069 -0,003 -0,026 0,048 0,147 0,011 0,033(<0,05) (<0,05) (<0,05) (0,735) (<0,05) (<0,05) (<0,05) (0,387) (<0,05)
11 CEO duality-0,073 0,083 0,016 0,009 -0,035 0,030 0,082 0,056 0,075 0,232(<0,05) (<0,05) (0,15) (0,408) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05)
12 CEO EBC0,340 0,277 0,002 -0,044 -0,008 0,047 0,085 0,246 0,195 -0,052 -0,018(<0,05) (<0,05) (0,815) (<0,05) (0,409) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (0,103)
13 CEO Own.0,002 -0,143 -0,084 0,022 0,054 -0,015 -0,096 -0,167 -0,103 0,057 0,134 -0,239(0,878) (<0,05) (<0,05) (0,184) (<0,05) (0,337) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05)
14 Director age0,053 0,066 0,114 -0,034 -0,076 0,008 0,189 0,327 0,072 0,421 0,002 0,051 -0,044(<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (0,417) (<0,05) (<0,05) (<0,05) (<0,05) (0,791) (<0,05) (<0,05)
15 Dir. tenure-0,080 -0,039 -0,007 -0,031 0,000 0,002 0,051 -0,133 -0,110 0,196 -0,025 -0,134 0,151 0,433(<0,05) (<0,05) (0,551) (<0,05) (0,934) (0,804) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05)
16 FCF-0,003 0,176 -0,014 -0,092 -0,038 0,004 0,059 0,082 0,065 0,037 -0,018 0,028 0,023 0,080 0,059(0,782) (<0,05) (0,159) (<0,05) (<0,05) (0,652) (<0,05) (<0,05) (<0,05) (<0,05) (0,107) (<0,05) (0,175) (<0,05) (<0,05)
17 CEO tenure0,044 -0,070 -0,018 -0,017 0,028 -0,008 -0,053 -0,109 -0,151 0,375 0,231 -0,111 0,334 0,098 0,305 -0,009(<0,05) (<0,05) (0,084) (0,099) (<0,05) (0,443) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (<0,05) (0,388)
22
Table 3: Acquirer Stock Returns
The sample is a panel of S&P 1,500 companies between 1998 and 2014. We employ event study methodology to estimatethe market model for each acquisition over a period of 255 days (-301, -46) preceding the announcement date. We then useestimated parameters to calculate the acquirer’s cumulative abnormal returns (CAR) over a period of five days centered on theevent date. The following table shows OLS model estimation results where the dependent variable is the acquirer’s five-daycumulative abnormal returns, CAR [-2, 2], around the acquisition announcement. "High Director EBC" and "Medium DirectorEBC" are dummy variables that take the value of one if the acquirer has Equity-based compensation (%) in the top- or middle-third among all firms in a given year, respectively. The table reports the estimated coefficients and the Absolute t-Statisticsbased on Huber-White robust standard errors. The variables are defined in Appendix A. ***, ** and * denote significance atthe 1%, 5%, and 10% level, respectively.
Variables
CAR [-2, 2]( 1 ) ( 2 )
Coef (t-Stat) Coef (t-Stat)
Director EBC 0.138*** (2.631)High Director EBC 0.0954*** (2.750)Medium Director EBC 0.110*** (3.212)Firm Size -0.00908 (-1.154) -0.00602 (-0.771)leverage -0.0354 (-0.531) -0.0522 (-0.785)Relative size 0.0113 (0.215) 0.0136 (0.258)Intra-industry 0.0246 (1.299) 0.0260 (1.373)Public target -0.00945 (-0.360) -0.0123 (-0.468)Board size 0.0104* (1.652) 0.00936 (1.490)Board independence 0.263*** (2.911) 0.273*** (3.027)Busy board 0.00584 (0.262) 0.00109 (0.0489)CEO age 9.12e-05 (0.0540) -9.41e-05 (-0.0559)CEO duality 0.00315 (0.162) -0.00399 (-0.204)CEO EBC -0.0638* (-1.668) -0.0663* (-1.727)CEO ownership -0.0289 (-0.154) -0.0595 (-0.317)Female directors -0.105 (-0.913) -0.130 (-1.133)Director age -0.00245 (-0.741) -0.00350 (-1.061)Director tenure -0.0122*** (-3.524) -0.0123*** (-3.548)Acquisition dummy 0.00477 (0.249) 0.00549 (0.286)FCF 0.122 (1.118) 0.109 (1.001)CEO tenure 0.00376** (2.177) 0.00458*** (2.640)Constant 0.0332 (0.0976) 0.123 (0.363)Year Fixed-effects Yes YesIndustry Fixed-effects Yes YesObservations 1,000 1,000Pseudo R² 0.069 0.072
23
Table 4: Performance change following acquisitions
The sample is a panel of S&P 1,500 companies between 1998 and 2014. The following table reports OLS estimation results where the dependent variables are the long-run cumu-lative abnormal returns and buy-and-hold abnormal returns, the change in return on assets and efficiency ratio. We compute abnormal returns using a (0, 36)-month window,where month 0 represents the month of the acquisition announcement. We calculate changes in return on assets and the efficiency ratio using a (0, 3)-year window, where year0 represents the year in which the acquisition is completed. The table presents the estimated coefficients and their t-Statistics. For the sake of brevity, we do not report the es-timated coefficients of year and industry dummy variables. The variables are defined in Appendix A. ***, ** and * denote significance at the 1%, 5%, and 10% level, respectively.
Variables
Cumulative Buy-and-hold Return on Assets Efficiency Ratioabnormal returns abnormal returns
( 1 ) ( 2 ) ( 3 ) ( 4 )Coef (t-Stat) Coef (t-Stat) Coef (t-Stat) Coef (t-Stat)
Director EBC 1.418*** (2.676) 0.271* (1.925) 0.0480** (2.337) 0.705** (2.079)Firm Size -0.0525 (-1.250) 0.00395 (0.190) 0.00171 (0.868) 0.00297 (0.0955)Leverage 0.533 (0.917) 0.236 (1.451) 0.0330 (1.627) -0.445 (-1.165)FCF -0.760 (-0.748) -0.0934 (-0.245) 0.865*** (20.22) 0.604 (0.812)Relative Size 0.300 (0.607) -0.117 (-0.744) -0.00925 (-0.523) 0.116 (0.357)Intra-industry -0.0433 (-0.261) -0.0832 (-1.266) 0.00316 (0.460) 0.140 (1.217)Public Target 0.274 (1.108) 0.0440 (0.484) -0.00369 (-0.356) 0.229 (1.387)Board Size 0.0543 (1.125) -0.0232 (-1.515) -0.00417** (-2.101) 0.0847** (2.490)Board Independence 2.522*** (3.156) 0.282 (0.979) 0.0158 (0.468) -0.0340 (-0.0618)Busy Board 0.0496 (0.278) -0.0189 (-0.302) 0.00552 (0.726) 0.00682 (0.0564)CEO Age -0.0169 (-1.257) -0.00367 (-0.755) -0.00117** (-2.202) 0.0111 (1.295)CEO Tenure 0.0136 (1.043) -0.00408 (-0.864) 0.000796 (1.490) -0.00254 (-0.292)CEO Duality 0.269 (1.574) 0.0598 (0.975) 0.0129* (1.862) -0.00615 (-0.0551)CEO Ownership -0.733 (-0.491) -0.581 (-0.953) 0.0657 (1.185) -0.784 (-0.898)CEO EBC -0.680** (-2.019) -0.0331 (-0.276) -0.00460 (-0.329) -0.142 (-0.635)Constant -2.061 (-0.908) -0.807 (-1.002) 0.0407 (0.832) -1.327 (-1.365)Year FEs Yes Yes Yes YesIndustry FEs Yes Yes Yes YesObservations 685 627 333 326Pseudo R² 0.199 0.171 0.628 0.811
24
Table 5: Acquirer Stock Returns: Robustness checks
The sample is a panel of S&P 1,500 companies between 1998 and 2014. We employ event study methodology to estimatethe market model for each acquisition over a period of 255 days (-301, -46) preceding the announcement date. We then useestimated parameters to calculate the acquirer’s cumulative abnormal returns (CAR) over a period of five days centeredon the event date. This table presents the results of two-stage least squares regressions where the dependent variable, theacquirer’s five-day cumulative abnormal returns around the acquisition announcement, is regressed on the predicted values ofDirector EBC derived from the first stage. The table reports the estimated coefficients and the Absolute t-Statistics based onHuber-White robust standard errors. The variables are defined in Appendix A. ***, ** and * denote significance at the 1%,5%, and 10% level, respectively.
Variables
2SLSFirst stage Second stageDirector EBC CAR [-2, 2]
Coef (t-Stat) Coef (t-Stat)
Fitted Director EBC 1.360* (1.779)Industry Median EBC 0.251** (2.034)Firm Size 0.0320*** (7.255) -0.0476* (-1.875)leverage -0.0642* (-1.709) 0.0448 (0.540)Relative size 0.0131 (0.446) 0.000311 (0.00576)Intra industry -0.0156 (-1.440) 0.0459** (1.996)Public target -0.0237 (-1.561) 0.0204 (0.621)Board size -0.0146*** (-4.177) 0.0286** (2.220)Board independence -0.0399 (-0.835) 0.320*** (3.310)Busy board 0.0204 (1.610) -0.0137 (-0.495)CEO age -0.00196** (-2.037) 0.00261 (1.172)CEO duality -0.0131 (-1.192) 0.0208 (0.936)CEO EBC 0.0759*** (3.473) -0.161** (-2.256)CEO ownership -0.0177 (-0.190) -0.0201 (-0.106)Female directors -0.245*** (-3.809) 0.181 (0.832)Directors age -0.00635*** (-3.453) 0.00421 (0.707)Director tenure -0.00447** (-2.268) -0.00655 (-1.339)Acquisition dummy 0.0178 (1.618) -0.0168 (-0.718)FCF -0.204*** (-3.334) 0.356* (1.910)CEO tenure 0.00564*** (5.781) -0.00343 (-0.746)Constant 0.933*** (6.878) -1.217 (-1.443)Year Fixed-effects Yes YesIndustry Fixed-effects Yes YesObservations 1,023 974Pseudo R² 0.334 0.066
25
Table 6: Performance change following acquisitions: Robustness checks
The sample is a panel of S&P 1,500 companies between 1998 and 2014. The following table reports results from the second stage of the IV 2SLS estimation. The dependentvariables, i.e., the long-run cumulative abnormal returns and buy-and-hold abnormal returns, the change in return on assets and efficiency ratio, are regressed on the predictedvalues of Director EBC derived from the first stage. We compute abnormal returns using a (0, 36)-month window, where month 0 represents the month of the acquisitionannouncement. We calculate changes in return on assets and the efficiency ratio using a (0, 3)-year window, where year 0 represents the year in which the acquisition iscompleted. The table presents the estimated coefficients and their t-Statistics. For the sake of brevity, we do not report the estimated coefficients of year and industry dummyvariables. The variables are defined in Appendix A. ***, ** and * denote significance at the 1%, 5%, and 10% level, respectively.
Variables
Cumulative Buy-and-hold Return on Assets Efficiency Ratioabnormal returns abnormal returns
( 1 ) ( 2 ) ( 3 ) ( 4 )Coef (t-Stat) Coef (t-Stat) Coef (t-Stat) Coef (t-Stat)
Fitted Director EBC 6.299* (1.907) 1.275** (2.026) 0.445*** (4.464) 4.183** (2.534)Firm Size -0.288*** (-2.721) 0.00551 (0.197) -0.000111 (-0.0555) -0.00321 (-0.0978)Leverage 0.888* (1.800) 0.0493 (0.224) 0.0867*** (3.720) 0.0569 (0.132)FCF 1.099* (1.651) 0.271 (0.670) 0.777*** (17.86) 0.323 (0.416)Relative Size 0.656 (1.123) -0.189 (-0.787) 0.00173 (0.0912) 0.160 (0.445)Intra-industry -0.212 (-1.425) -0.138** (-2.146) 0.000180 (0.0258) 0.108 (0.900)Public Target 0.289 (1.197) -0.00974 (-0.0971) -0.00419 (-0.405) 0.219 (1.283)Board Size 0.135** (2.028) -0.00212 (-0.111) -0.00251 (-1.267) 0.106*** (3.076)Board Independence 1.789 (1.591) 0.328 (0.992) -0.0148 (-0.418) -0.639 (-1.090)Busy Board 0.358*** (2.957) -0.0511 (-0.760) 0.00130 (0.169) -0.0768 (-0.610)CEO Age 0.00913 (0.896) -0.00868* (-1.685) 0.000876 (1.346) 0.0251** (2.314)CEO Tenure 0.0113 (1.629) 0.000466 (0.0893) -0.000248 (-0.422) -0.0107 (-1.096)CEO Duality 0.195 (1.366) -0.0398 (-0.584) 0.0131* (1.871) 0.0494 (0.426)CEO Ownership -1.117 (-1.029) -0.410 (-0.656) 0.0354 (0.609) -1.113 (-1.185)CEO EBC -1.312** (-2.182) 0.138 (1.032) -0.0455** (-2.236) -0.594* (-1.770)Constant -0.944 (-0.902) -0.447 (-0.819) -0.239*** (-2.918) -3.515** (-2.433)Year FEs Yes Yes Yes YesIndustry FEs Yes Yes Yes YesObservations 716 425 372 363Pseudo R² 0.082 0.048 0.538 0.880
26
Table 7: Acquirer Stock Returns: Other robustness checks
The sample is a panel of S&P 1,500 companies between 1998 and 2014. We employ event study methodology to estimatethe market model for each acquisition over a period of 255 days (-301, -46) preceding the announcement date. We then useestimated parameters to calculate the acquirer’s cumulative abnormal returns (CAR) over a period of five days centered on theevent date. The following table shows OLS model estimation results where the dependent variable is the acquirer’s five-daycumulative abnormal returns, CAR [-2, 2], around the acquisition announcement. The table reports the estimated coefficientsand the Absolute t-Statistics based on Huber-White robust standard errors. The variables are defined in Appendix A. ***, **and * denote significance at the 1%, 5%, and 10% level, respectively.
VariablesCAR [-2, 2]
Coef (t-Stat)
Change of EBC 0.306*** (2.835)Firm Size -0.00706 (-0.297)leverage -0.116 (-0.559)Relative size 0.0212 (0.131)Intra industry 0.0582 (1.037)Public target -0.0333 (-0.386)Board size 0.0319 (1.559)Board independence 0.964*** (3.325)Busy board -0.0271 (-0.399)CEO age -0.00207 (-0.415)CEO duality 0.0326 (0.572)CEO EBC -0.107 (-0.960)CEO ownership -0.0545 (-0.116)Female directors -0.645* (-1.771)Directors age -0.00630 (-0.569)Director tenure -0.0342*** (-3.186)Acquisition dummy 0.000933 (0.0167)FCF 0.139 (0.361)CEO tenure 0.0114** (2.259)Constant 0.0293 (0.0383)Year Fixed-effects YesIndustry Fixed-effects YesObservations 335Pseudo R² 0.190
27
Table 8: Acquirer Stock Returns: Additional test
The sample is a panel of S&P 1,500 companies between 1998 and 2014. We employ event study methodology to estimatethe market model for each acquisition over a period of 255 days (-301, -46) preceding the announcement date. We then useestimated parameters to calculate the acquirer’s cumulative abnormal returns (CAR) over a period of five days centered on theevent date. The following table shows OLS model estimation results where the dependent variable is the acquirer’s five-daycumulative abnormal returns, CAR [-2, 2], around the acquisition announcement. The table reports the estimated coefficientsand the Absolute t-Statistics based on Huber-White robust standard errors. The variables are defined in Appendix A. ***, **and * denote significance at the 1%, 5%, and 10% level, respectively.
VariablesCAR [-2, 2]
Coef (t-Stat)
Director EBC (t - 1) 0.104** (2.494)Firm Size (t - 1) 0.000752 (0.119)leverage (t - 1) -0.113** (-2.143)Relative size (t - 1) 0.0441 (0.903)Intra industry (t - 1) -0.00956 (-0.604)Public target (t - 1) 0.00878 (0.410)Board size (t - 1) 0.00762 (1.542)Board independence (t - 1) -0.0342 (-0.500)Busy board (t - 1) -0.0232 (-1.213)CEO age (t - 1) -5.06e-08 (-3.53e-05)CEO duality (t - 1) 0.00845 (0.494)CEO EBC (t - 1) 0.00122 (0.0367)CEO ownership (t - 1) -0.00853 (-0.0566)Female directors (t - 1) 0.0363 (0.402)Directors age (t - 1) -0.000675 (-0.252)Director tenure (t - 1) 0.00146 (0.532)Acquisition dummy (t - 1) -0.0207 (-1.275)FCF (t - 1) -0.133 (-1.511)CEO tenure (t - 1) -0.000163 (-0.106)Constant 0.0543 (0.236)Year Fixed-effects YesIndustry Fixed-effects YesObservations 481Pseudo R² 0.118
28
Table 9: Acquisition premium
The sample is a panel of S&P 1,500 companies between 1998 and 2014. The following table shows OLS model estimationresults where the dependent variable is the acquisition premium, defined as the ratio of the final offer price to the target shareprice four weeks prior to the acquisition announcement, minus one. "High EBC" and "Medium EBC" are dummy variablesthat take the value of one if the acquirer has Equity-based compensation (%) in the top- or middle- third among all firms ina given year, respectively. The table reports the estimated coefficients and the Absolute t-Statistics based on Huber-Whiterobust standard errors. The variables are defined in Appendix A. ***, ** and * denote significance at the 1%, 5%, and 10%level, respectively.
Acquisition premiumVariables ( 1 ) ( 2 )
Coef (t-Stat) Coef (t-Stat)
EBC -0.112*** (-2.689)High EBC -0.0914*** (-2.827)Medium EBC -0.0592* (-1.771)Firm Size 0.00683 (0.623) 0.00532 (0.484)Leverage 0.118 (1.167) 0.121 (1.193)ROA 0.331* (1.668) 0.346* (1.730)Relative Size -0.00828 (-0.443) -0.00767 (-0.410)Intra-industry 0.0332 (1.218) 0.0339 (1.239)All Cash 0.0294 (0.959) 0.0319 (1.040)Hostile deal 0.0445 (0.477) 0.0339 (0.363)Number of bidders 0.00956 (0.246) 0.0110 (0.281)Board Size -0.000220 (-0.0276) 0.000546 (0.0684)Board Independence 0.0490 (0.449) 0.0375 (0.343)Busy Board 0.299*** (2.699) 0.296*** (2.662)E-Index -0.0118 (-1.040) -0.0122 (-1.072)Female Directors -0.0778 (-0.451) -0.0994 (-0.576)CEO Age -0.00253 (-1.032) -0.00218 (-0.889)CEO Tenure 0.000161 (0.0828) -8.62e-05 (-0.0442)CEO Duality 0.0280 (0.968) 0.0290 (1.001)CEO Ownership 0.0182 (0.103) 0.0492 (0.277)CEO EBC -0.0108 (-0.155) -0.00745 (-0.106)Constant 0.457** (1.986) 0.384* (1.679)Year Fixed-effects Yes YesIndustry Fixed-effects Yes YesObservations 455 455Pseudo R2 0.224 0.224
29
Appendix A. Variable Definition
Variables Definitions/Calculations Data Source
AcquisitionActivity
A dummy that equals one for years following thecompletion of an acquisition
SDC Platinum
AcquisitionPremium
Is the ratio of the final offer price to the targetshare price four weeks prior to the acquisition an-nouncement, less one
SDC Platinum
All Cash A dummy variable that takes the value of one if thedeal is entirely paid in cash, and zero otherwise
SDC Platinum
Board Inde-pendence
The percentage of independent directors on theboard
RiskMetrics DB
Board Size The total number of directors on the board RiskMetrics DB
Busy Board A dummy variable that equals one when a major-ity of independent directors serve on three or moreboards, zero otherwise
RiskMetrics DB
Cash Com-pensation
is the natural log of (one plus) the annual retainerplus fees paid to an average director for attendinggeneral board meetings
ExecuComp DB
CEO Age Is the CEO’s age in years ExecuComp DB
CEO Duality A dummy variable that equals one when the CEOand COB positions are combined and zero other-wise
RiskMetrics DB
CEO Owner-ship
The percentage of company’s shares owned by theCEO
RiskMetrics DB
CEO Tenure Is the number of years the CEO has held that po-sition
ExecuComp DB
CEO EBC Is the natural log of (one plus) the value of stockoptions and restricted stock awarded to the CEOduring the year, divided by total CEO compensa-tion
ExecuComp DB
30
Director age Is the average age of the board of directors RiskMetrics DB
Directortenure
Is the average number of years a director has beenon the board
RiskMetrics DB
Director EBC Is the natural log of (one plus) the annual stock op-tion compensation (measured as per share Black-Scholes option value times the number of stock op-tions granted) plus the annual stock compensation(measured as the number of shares awarded timesthe closing stock price at the end of the previousfiscal year) that a firm pays to an average director,divided by total director compensation
ExecuComp DB
E-Index Is defined as the sum of the 6 following provi-sions (dummy variables): staggered board, poisonpills, golden parachutes, limits to shareholders by-law amendments, supermajority requirements formergers and charter amendments.
RiskMetrics DB
Female direc-tors
Is the number of female directors on the board di-vided by board size
RiskMetrics DB
Firm size Is the natural log of the market capitalization Compustat DB
FCF Is free cash flow scaled by total assets Compustat DB
Hostile deal Is a dummy variable that equals one if the trans-action is classified as hostile in the SDC Platinumdatabase, and zero otherwise
SDC Platinum
Intra-industry
Is a dummy variable that equals one when the tar-get and acquirer share the same two-digit primarySIC code, zero otherwise
SDC Platinum
Investmentopportunities
Is the ratio of capital expenditures to total sales Compustat DB
31
Leverage Is the ratio of total debt to total assets Compustat DB
MTB Market value of equity divided by book value ofassets
Compustat DB
Numberof bidders
Is the number of firms competing to acquire thetarget
SDC Platinum
Percentage ofcash
Is the percentage of the deal value paid in cash bythe acquirer
SDC Platinum
Public Target Is a dummy variable that equals one when the tar-get is publicly held, zero otherwise
SDC Platinum
Relative Size Is the ratio of the deal value to the acquirer’s mar-ket capitalization at the end of the year prior tothe deal
SDC Platinum
ROA Is the return on assets Compustat DB
32