Executive compensation and the outcome of mergers

Executive compensation and the outcome of mergers

Ryan McKee

July 2015

Abstract

This paper investigates how the different components of CEO compensation incentivize and influence a CEO’s decision to undertake a merger. I use probit regressions to model the likelihood of a merger taking place with regards to the CEO’s compensation. I find that as CEO option holdings (ownership) increase(s) a firm has a higher (lower) probability of acquiring another firm. These results remain even after controlling for specific CEO-financial-firm- and governance control variables. Overall I contribute by showing that CEO compensation is influential in incentivizing CEOs to (not) engage in mergers and add to the debate on how to properly structure CEO compensation to align the interests of the management team with that of the shareholders.

University of Amsterdam

MSc Business Economics, Finance track Master Thesis

Statement of Originality

This document is written by Student Ryan McKee who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Table of Contents

Introduction ...4

Literature Review ...8

Methodology ...13

Data and descriptive statistics ...16

Results ...20

Robustness Specifications ...22

Conclusions ...25

References ...28

Introduction

This paper brushes on the subject of compensation and mergers and acquisition literature and aims to combine the two fields. Existent literature on compensation advocates that shareholders should incentivize managers to take on risky, but positive NPV projects (Guay, 1999). The CEO of a company is responsible for a firm’s investment decisions and is ultimately responsible for maximizing the value of a firm. This can be done in various ways. One of the options a CEO has to maximize firm value is performing a merger to obtain the resources and capabilities of another firm. Due to the heavy amount of responsibility that is laid upon him/her, the CEO is offered compensation by the board. This incentive pay (compensation) is offered in different shapes and sizes ranging from cash bonuses, stock grants and stock options. According to existent compensation literature these mechanisms should be interchangeable and have been treated as such in various studies (Walkling and Long, 1984; Yermack, 1995). However, recent studies have also indicated that options in particular can lead to excessive risk taking (Burns and Kedia, 2006). This paper wishes to bridge the gap between compensation and mergers. CEOs cannot always guarantee the outcome of a merger; however, given the proper incentive and motivation he/she can influence the outcome. This means that a CEO could possibly benefit from a merger because of the favourable impact on his/her earnings. Therefore this paper focuses on how CEO compensation influences the merger and acquisition activity of the CEO/firm. This relationship could be subjected to endogeneity and so I use a variety of firm-, year-, CEO- financial-, and governance-specific control variables in order to account for this possible endogeneity. The question this paper ultimately aims to answer is: does CEO compensation influence the merger activity of a company?

Previous studies investigating the relationship between CEO compensation and merger activity mainly focus on the target firms incentives (Walkling and Long, 1984). There has been little research done on CEO wealth and its impact on mergers. CEO wealth means the combined value of a CEO’s stock options and the value of his/her ownership in the company. If options and ownership are interchangeable both should align CEO incentives with that of the shareholders according to optimal contract theory. However, previous studies have shown that high levels of stock ownership cause managers to become more risk averse as they have “more skin in the game” (Ross, 2004). In contrast, stock options have shown to lead to more risk seeking behaviour (Burns and Kedia, 2006). I wish to combine these previous findings and examine if they also hold in a CEO’s decision to undertake mergers and acquisitions. By doing so, we are able to show that certain compensation ‘packages’ could lead to unfavourable behaviour from the CEO and be harmful for the company’s shareholders.

Mergers are on average not value creating (Moeller, Schlingemann and Stultz, 2004). CEO compensation, however, generally increases after mergers (Grinstein and Hribar, 2004) and is not subjected to post-merger performance (Harford and Li, 2007). The CEO could thus be stimulated to increase firm size by engaging in more mergers, which in effect increases his/her compensation (Grinstein and Hribar, 2004). This paper begins by analyzing the effect of CEO wealth on mergers outcomes. Afterwards I separate the two mechanisms stock ownership and stock option holdings, and analyze these variables individually.

In order to analyze the relationship between compensation and merger activity I use a probit model to find the increase in likelihood of undertaking a merger given certain compensation structure. This compensation consists of four main variables: CEO cash compensation, CEO

wealth, CEO ownership and CEO option holdings; the latter two being the focus of this study. CEO cash compensation consists of purely cash payouts such as salary and bonuses. CEO ownership is the ownership the CEO has presented as a percentage of the total outstanding stock. CEO option holdings are comprised of both the exercisable and un-exercisable options plus the unvested stock. CEO wealth is defined as the combination of the value of CEO ownership and CEO option holdings. I use a sample data of 32,948 observations of CEO compensation over the years and combine them with 6,095 completed mergers ranging from 1994 to 2014. It is possible that omitted variables are drivers of a firm’s or CEO’s merger strategy. I use a range of governance, financial and CEO specific controls to account for such omitted variables.

I find in a sample of 32,948 observations with 5,500 different CEOs at over 3000 different companies combined performing more than 6,000 mergers that the structure of CEO compensation in fact influences a CEO’s and ultimately the firm’s decision to engage in more merger and acquisition activity. I observe that CEO option holdings significantly increase a CEO’s decision to perform mergers. This is also prevalent during the recent financial crisis. CEO stock ownership validates previous findings in the sense that it acts as a deterrent for CEOs to partake in risky behaviour. Cash compensation, I find, also leads to less merger activity by CEOs.

On the whole, I contribute to the merger and compensation literature by showing that the structure of CEO compensation influences real firm effects in terms of mergers. My research is similar to that of Cai and Vijh (2007) with some distinct differences. The paper by Cai and Vijh (2007) also focuses on compensation and its effect on merger activity, however, they only focus on CEO wealth. I too look at CEO wealth but also distinguish between the two

elements of CEO wealth; namely stock ownership and stock option holdings. Furthermore I look at a more comprehensive data set comprising over 6,000 completed mergers while Cai and Vijh (2007) look at 250 merger deals between 1993-2001. In contrast to other papers I also investigate the merger behaviour during periods of financial recessions and again link this to compensation. The global financial crisis of 2008 acts an exogenous shock to the economy and provides a natural experiment to explore whether such a shock has implications in a CEO’s investment decisions regarding mergers.

The remainder of this paper is structured as follows. Section I presents the literature review and hypotheses development. Section II discusses the proposed methodology. Section III provides data description and descriptive statistics. Section IV provides the results of the analysis. Robustness checks, including the financial crisis are discussed in Section V. Section VI concludes.

Literature Review

Optimal contract theory advocates boards to offer CEO compensation contracts to maximize shareholder wealth (Grossman and Hart, 1983). However, numerous studies have shown that there may be possible agency conflicts between managers and shareholders (Jensen, 1986). If executive managers have too much influence over boards, they can indirectly influence/decide how much executive compensation they receive. According to Bebchuk and Fried (2003) CEOs with more managerial power receive higher compensation and as a result they perform more mergers in order to justify this higher compensation. Existing literature provides other examples of cases where managers have the opportunity to personally gain and are in conflict with the interest of shareholders. Hartzell et al. (2004) show that target CEOs are willing to accept lower merger premiums if they receive higher personal payouts. Acquirer CEOs with a larger control over the board also receive larger compensation in forms of bonuses even if their performance is poor (Grinstein and Hribar, 2004).

The previous mentioned studies focus primarily on CEO compensation and not overall CEO wealth. They define CEO compensation as salary and any potential cash bonuses he/she might receive. CEO wealth is defined as value of the stocks and options holdings the CEO receives. This is potentially the largest component of income for a CEO as this type of compensation is linked to firm value. There are numerous amounts of studies that demonstrate that CEO stock ownership is positively tied to firm value (Jensen and Meckling, 1976). On the contrary, there are also documented cases in which high levels of stock ownership lead to a decrease in firm performance due to managerial entrenchment (Stulz, 1988; Wang, 2011). There has also been a noticeable downwards trend in CEO ownership in public traded firms (Core and Guay, 2002) suggesting that there is demand for alternative

devices to align CEOs’ and shareholders’ interests. Previous studies advocate incorporating stock options as part of CEO compensation incentivizing managers to increase firm and options value (Yermack, 1995).

However, stock options may also incentivize managers to take on more risk. According to Guay (1999) options are subject to both change in performance and risk of the stock. Stock options are measured by its delta and vega. Delta is the change in option value which comes from a change in the underlying value of the stock; the vega is the change in option value due to the change of the underlying volatility of the stock. Coles et al., (2006) investigate the nature of an options delta and vega and show that increases in vega leads CEOs to riskier choices. They find the delta to be unclear; a high delta could incentivize a CEO to take on riskier projects but could also lead to more risk aversion, dependent on the CEOs preferences.

Ex ante, however, it is difficult to differentiate between good and bad projects. This uncertainty combined with large enough risk-based incentives could incentivize CEOs to partake in a larger number of risky projects (i.e. mergers) than is optimal. More importantly, CEOs do not bear the cost of the risk associated with poor performance. Studies show that CEOs rarely receive a decline in compensation after poor performance (Grinstein and Hribar, 2004; Harford and Li, 2007). Summarizing this means that CEO wealth increases with good outcomes of risky investments, while the consequences of the bad outcomes are borne by others.

Most studies on CEO compensation look at ways to alleviate conflicts of interest between shareholders and managers. By increasing CEO stock ownership agency problems due to separation of ownership and control can be minimized and thus increase firm value (Jensen

and Meckling, 1976). Again, however, Stultz (1988) provides evidence that increased stock ownership may also increase risk aversion in managers. Options are hypothesized to alleviate this risk aversion (Smith and Stulz, 1985). Options grants could provide incentives for managers to take on risky, positive NPV projects (Morgan and Poulsen, 2001).

The main hypothesis of this paper tests how the different types of compensation a CEO receives, influences his behaviour in regards to merger and acquisition activity. Following the alignment incentive hypothesis, CEOs with compensation existing of large option holdings and high stock ownership before the merger should engage in firm value maximization (Himmelberg et al., 1999; Core and Guay, 1999; and Rajgopal and Shevlin, 2002). There are many empirical studies supporting this claim. Hanlon et al. (2004) provide empirical evidence that stock options provide positive future performance. Brickley et al. (1985), Mehran (1995), and Hall and Liebman (1998) find that higher equity-based compensation leads to improved firm performance. Lewellen, Loderer, and Rosenfeld (1985) show that CEOs with high equity based wealth are less inclined to perform value destroying merging activity. Stock options have also been found to create incentives for managers to reduce firm risk Lambert et al. (1991) and Carpenter (2000). Therefore, under the alignment incentive hypothesis, the stock ownership and option compensation the CEO has/receives prior to the merger should be interchangeable as they both motivate the CEO with reduced risk. Empirical predictions following from this hypothesis would be to observe CEOs with high levels of stock ownership and large option value to engage in fewer mergers.

Another hypothesis is based on the risk-incentive effect. Firms with high stock ownership might experience an increase in CEO risk aversion and as a result will also not perform risky, but positive, NPV projects leading to underinvestment. This stems from the fact that CEOs

are primarily under diversified than other shareholders leading to sub-optimal investing performance. Nevertheless, if the CEO’s decisions are based on the risk part of the options then this could lead to increased merger activity at the expense of the shareholders. Armstrong and Vashishtha (2012) document that CEO option holdings increases the likelihood of CEO’s to perform more risky business decisions suggesting a positive relationship between CEO option holdings and the increased incentive to perform more mergers. Empirical predictions following from the risk-incentive effect would be to observe CEOs with high stock ownership to still be less likely to engage in mergers while CEOs with a large portion of option wealth to be more willing to engage in mergers. The central hypothesis of this paper is as follows:

H0: The height and composition of the CEO’s compensation does not influence the firm’s

merger and acquisition activity.

H1: The height and composition of the CEO’s compensation does influence the firm’s merger

and acquisition activity.

From the two leading theories we can further construct two more specific hypotheses regarding CEO stock ownership and CEO option holdings namely:

H0: The height and composition of CEO stock ownership does not influence the firm’s merger

and acquisition activity.

H1: The height and composition of CEO stock ownership does influence the firm’s merger

and acquisition activity.

H0: The height and composition of CEO option holdings does not influence the firm’s merger

H1: The height and composition of CEO option holdings does not influence the firm’s merger

and acquisition activity.

The table below provides the hypothesis development overview based on each incentive.

Alignment Incentive Risk-based Incentive

 Premise: High option and stock wealth should lead to firm maximization (Himmelberg et al., 1999)

 Premise: stock ownership lead to risk aversion due to undiversified CEOs (Stultz, 1985). However, studies show CEO option wealth increases risk taking behavior (Burns and Kedia, 2006)

 Supporting studies: Hanlon et al. (2004), Brickley et al. (1985), Mehran (1995) and Rosenfeld (1985)

 Supporting studies: Coles et al. (2006) Armstrong and Vashishtha, (2012)

 Conclusion: stock ownership and option grants interchangeable

 Conclusion: stock ownership and option holdings behave differently  Expectations:

 High option holdings and/or stock ownership should not lead to different merger activity

 Expectations:

 CEO stock ownership → fewer mergers

Methodology

As previously explained this paper investigates whether certain executive compensation instruments create incentives for CEOs to conduct more merger and acquisition activity. I will need data on all mergers conducted by firms listed on both the NYSE and NASDAQ stock exchanges. Furthermore, I need data on executive compensation for CEOs who are under contract and receive compensation from publicly traded firms listed on the NYSE and NASDAQ. I will run two types of regressions. I will model the relationship using a probit model and a linear model. The main focus of the paper is the regression outputs obtained from the probit model as it provides the best statistical and economic insight. Afterwards (in the robustness section) I will also discuss the linear probability model. The models are specified below.

Probit model:

Prob(Acquire|Compensation) = φ(α + β1 Cash compensation + β2 CEO wealth +W1Controls)

&

Prob(Acquire|Compensation) = φ(α + β1 Cash compensation + β2 CEO option holdings + β3

CEO ownership +W1 Controls)

Linear regression model:

Acquiring activity = α + β1 Cash compensation + β2 CEO wealth +W1 Controls

&

Acquiring activity = α + β1 Cash compensation + β2 CEO option holdings + β3 CEO ownership

+W1 Controls

The dependent variable is a dummy variable which equals one if a merger has taken place in any given year ranging from 1994 to 2014. This variable represents the acquiring activity

following from the compensation the senior official receives. I incorporate the probit model as it gives insight to the probability of an acquisition happening based on the structure of the compensation. I will be focussing on CEO cash compensation, CEO wealth, CEO option holdings and CEO ownership. CEO cash compensation is comprised of salary and bonuses. CEO option holdings are the value of the CEO option holdings (unexercised exercisable and unexercisable options) plus the unvested stock. CEO ownership is the amount of stock the CEO owns as a percentage of total outstanding shares. The variables are discussed in further detail in section III: data and descriptive statistics.

The regressions could possibly suffer from an endogeneity problem as the height and structure of CEO compensation can be determined by other omitted factors (e.g. the longevity of the CEO or the size of the firm). To reduce the amount of endogeneity I will need to include a number of control variables. I will be using similar control variables as proposed in the paper by Grinstein and Hribar (2004) who focus on acquiring CEO’s payment after the acquisition has taken place. These control variables are to control for CEO and firm specific characteristics. The control variables for CEOs will be: age, gender, and I also add the longevity of the CEO to this list i.e. how long the CEO has been at the company. To control for firm specific variables I will include the firm size, and the specific industry the firm operates in. I also include a control variable for the general state of the economy as a CEO also takes into account the state of the current economy when acquiring a new firm (Maksimovic, Vojislav et al., 2013). I have also added governance controls as CEO influence over the board has been shown to influence CEO compensation and merger activity (Grinstein and Hribar, 2004). These controls include CEO age, CEO duality and if the company has a classified board. I expect all these control variables to be significant in determining acquisition activity as they have been proven to be important factors in deciphering the

different factors involved in CEOs making merger investment decisions. The main independent variables are CEO total wealth, cash compensation, option and equity value, and percentage ownership.

After making the necessary adjustments I can run the regressions. The probit model provides the most economic insight. The probit model provides evidence on whether the probability of acquiring is altered depending on the composition of the compensation. Given the extant literature and following from the risk based incentive hypothesis I expect the percentage CEO option wealth to be a significant factor in increasing the likelihood of a CEO to engage in more merger activity. The coefficient of CEO option wealth is therefore theorized to be a positive value. The alignment incentive hypothesis does not answer whether CEO options holdings should have a negative or positive impact on merger activity but it does state that they should be significant in determining merger activity. More importantly, following the alignment incentive hypothesis, we expect to see that CEO option holdings and CEO stock ownership should lead to the same incentive as according to optimal contract theory they should be interchangeable and lead to firm value maximization. Following the risk incentive hypothesis I would expect the variable percentage CEO ownership to have a negative effect on merger activity as it argues that CEOs become more risk averse with increasing ownership. As a result the coefficient of this variable is expected to be negative.

Data and descriptive statistics

To be able to conduct the research I will need data on both acquisition activity and executive compensation. For data on executive compensation I will need to access the Execucomp dataset which is provided by Wharton Research Data Services. From Execucomp I collect CEO, firm and year information ranging from January 1994 to January 2014 (35,576 observations) and obtain data on the CEO annual salary, bonus, other annual compensation, long-term incentive payments, restricted stock grants, restricted stock holdings, option values (unexercised exercisable and unexercisable options) number of options granted, and percentage CEO ownership. After doing this data streaming the number of observations comes to 32,948.

Data on mergers and acquisitions can be accessed from the Thomson One dataset. All data will be quarterly data. The sample will incorporate mergers performed by publicly listed firms on the NYSE and NASDAQ between January 1994 and January 2014. I take a sample period of 20 years to ensure that I have a large enough data set to obtain significant results (if applicable). In addition, the last 7 years of the data set contain the largest economic crisis since the great depression; so this paper will differentiate these time periods within the data set. Therefore, the paper investigates the data as a whole but also distinguishes a pre-, (post-) and crisis period. All data regarding non-US mergers and acquisitions, leveraged buy-outs, spin-offs, carve buy-outs, and liquidations are excluded from this research. The Thomson One dataset provides names of firms, cusips, macro/micro industry and codes, percentage shares acquired, deal value, date announced and completed for all acquirer and target companies.

As stated before I also use governance control variables as Grinstein and Hribar (2004) have shown that increased CEO control and managerial power over the board can influence (his own) CEO compensation and merger activity. I obtain the governing characteristics from the Institutional Shareholder Services (formerly Risk Metrics). I include measures for classified board and CEO duality (whether CEO is also Chairman of the board) and CEO age. CEO age is an important CEO characteristic as financial security becomes more important as a CEO ages and so on average elder CEOs treat risky behaviour and strategies more conservatively (Wiersema and Bantel, 1992). The final sample combining all Execucomp and Thomson data consists of 32,948 observations with 6,095 mergers completed by 5,029 CEOs ranging between 1994 and 2014.

B. Variable construction

The main variables are CEO wealth, cash compensation, option holdings and percentage ownership. We first need to properly construct these variables. CEO wealth is defined as the sum of the value of CEO ownership, restricted stock holdings and the value of exercisable and unexercisable options. % Cash Compensation is the value of the cash compensation (CEO salary and bonuses) divided by the total annual compensation of the previous year (TDC2). % CEO option holdings is defined as the sum of the value of exercisable and unexercisable options – determined by Black and Scholes formula – divided by the market capitalization 30 days prior to the merger announcement. We take a 30 day announcement window as previous studies by Haw, Pastena and Lilien (1990) have shown that the share price, and thus the market cap, are influenced by rumours leading up to the official announcement date. The % CEO ownership is simply the percentage of stocks held by the

CEO as a total of outstanding market capitalization. This last variable indicates how much ownership the CEO has of the total outstanding equity.

C. Summary statistics

Table 1 provides the summary statistics for all of the 3,122 firms listed on the NYSE and NASDAQ that combined conducted 6,095 mergers over the time span of twenty years ranging from 1994 until 2014. The sample includes firms that conducted mergers (one or more) and firms that did not conduct any merger activity whatsoever. Panel A describes the merger characteristics. My sample selection exists out of 6,095 completed public mergers. Of those 6,095 mergers almost 80% (77.26%) are deals with firms that are in the same industry. Nearly 80% (77.46%) of the deals are acquisitions done with other U.S. based firms while the remainder of the mergers are conducted with firms on an international level. We find that of our sample 28% of the mergers were conducted during the years of the financial crisis. Table 1 also provides information on the percentage of shares acquired, owned and sought after a merger. The % of shares acquired is the percentage of shares the firm actually acquired after completing the acquisition. The % of shares owned after represents the total percentage amount of share holdings the company has of the other firm after the merger. % sought represents the amount the acquiring firm wished to acquire during the merger process. The mean of these variables all lie around the high seventies mark with 74%, 75% and 79% respectively. A high standard deviation combined with a median and 90th percentile of 100% for all three variables suggests that the majority of the mergers were complete takeovers of the acquiring firms. The average transaction value is around 800 million dollars. This variable is however highly skewed with a median value of 104 million and a standard deviation of 3.6 billion. This is probably down the fact that of the 6,095 mergers there are 71

instances where the value of the transaction was above 10 billion U.S. dollars. If we ignore these 71 instances the standard deviation drops dramatically to around 700 million (not presented) compared to the earlier 3.6 billion. Panel B describes the CEO compensation measures; our independent variables. % CEO cash compensation represents the cash compensation a CEO receives as a percentage of annual compensation. Over 52% of a CEO annual pay comes from salary and bonuses. CEOs hold on average 2.3% of total stock holdings with the median being 0.16%. CEO option holdings are on average 0.60% of total compensation with the median lying at (0.20%). Examining total wealth we find that on average stock ownership accounts for 4-5 times the amount of option holdings. This holds under the notion that CEOs will exercise their option holdings as soon as they become available in order to diversify their total wealth position. Panel C describes the governance measures used. The average (median) age is relatively old 65.8 (66). The average amount of years a CEO is at the company is around 7.5. In 37% of the cases the CEO is also chairman of the board and 25% of the firms have a classified board.

Results

This section presents the results from the analysis. Here I discuss whether and how compensation structure affects the merger activity of a firm. I begin by investigating the relationship of CEO wealth on merger activity and subsequently follow with analyzing the percentage of CEO ownership and the percentage of CEO option holdings. Table 2 presents the results of the probit regression approach that was implemented during the analysis. The four important variables are %CEO cash compensation, %CEO option holdings, %CEO ownership and %CEO wealth. Control variables include CEO age, Chairman/CEO duality, classified board, longevity and firm-, year-, and industry fixed effects.

Column 1 of table 2 presents the results in which the independent variable of interest is %CEO wealth. This variable is used to observe whether I obtain similar results as previous conducted studies, in particular that of Cai and Vijh (2007). Model 1 shows that as the percentage of total CEO wealth increases the probability of undertaking a merger in fact declines. The results deem to be statistically significant. This is in contrast with previous findings. Previous papers report a positive relationship. The differences in result could be caused by different sample selection and size used, as well as the time period. I conduct my research with a time period of 20 years (compared to 10 years) and furthermore I use a sample selection of 6,095 merger deals conducted from various industries compared to the 250 deals analyzed by Cai and Vijh (2007).

I next examine the two mechanisms guiding CEO wealth namely: stock ownership and option holdings. I analyze how these two independent variables affect the probability of a firm/CEO engaging in a merger. Column 2 shows that the probability of a firm engaging in an acquisition declines as CEO ownership increases. This corresponds with both the alignment

incentive hypothesis and risk-incentive hypothesis as stated in the literature review. This result is in line with previous literature in particular the work done by Stultz (1985). A decrease in probability of takeovers with increasing CEO stock ownership leads to believe that CEOs become more risk averse if they have a larger stake in company ownership. According to my regression results this concept extends through to a CEO’s decision of acquiring other firms. It is important for board members who package CEO compensation to realize that offering too much stock ownership could lead to too much risk aversion. The next independent variable of interest is the CEO option holdings. I find that the probability of acquiring increases significantly as CEO option holdings increase. The result is deemed statistically significant at the 1% level. This finding is in contrast with the alignment incentive hypothesis and the many papers arguing this hypothesis. The alignment incentive hypothesis reasons that CEO stock ownership and option holdings are interchangeable as they both incentivize the CEO to engage in firm value maximization (Himmelberg et al., 1999, Hanlon et al., 2004, Brickley et al., 1985). My results show that when it comes to the probability of undergoing a merger CEOs do not regard stock ownership and option holdings as interchangeable and in fact act differently according to the type of compensation they receive. The increasing probability of acquiring as CEO option holdings increase conforms to the risk-incentive hypothesis. According to this hypothesis CEO options are tied to risk taking behavior (Burns and Kedia, 2006) and we would expect to see an increase in the amount of mergers. This is what the regressions results confirm. Furthermore stock ownership has a negative relationship with the probability of executing a merger. This also ties in with the risk incentive hypothesis arguing that stock ownership causes more risk aversion in managers (Stultz, 1985).

Robustness Specifications

In the main analysis of this paper we use a probit model to determine the relationship between CEO compensation and firm merger and acquisition activity. The results from this analysis can be found in table 2 and have been discussed in section IV. Here we find that both CEO ownership and CEO cash compensation have a negative effect on merger activity while CEO option holdings have a positive effect. I also run the regression of the probit model without a constant. The reason for doing so is that the variables I specify, cash compensation, option holdings and CEO ownership, would never all simultaneously have a zero value as a CEO will always receive some sort of compensation (at least this is the case in the sample set). This means that including a constant could lead to a dummy variable trap as it has no specific economic (or statistic) meaning. If all dependent variables are zero the value of the constant should give an indication as to the probability of a merger happening regardless of compensation, however, economically speaking this does not make sense as previously stated a CEO always receives some sort of compensation. Therefore I also ran the regression without a constant in the model. The results of this regression are presented in table 3. Model 1 of table 3 models the relationship of CEO wealth with the merger activity. This variable, CEO ownership is still deemed significant and shows that as CEO wealth increases merger activity decreases. Model 2 splits the two determinants of CEO wealth into CEO option holdings and CEO (stock) ownership. The two variables are yet again significant (as in table 2) however the coefficients are slightly altered (0.73 to 0.81 and -0.83 to -0.69 respectively). The model used in table 3 confirms the models used in table 2 and provides

evidence that an increase in CEO option holdings increase merger activity while CEO stock ownership decreases merger activity.

B. Financial crisis

Exogenous shocks have shown to cause a change in a firm’s investment decisions and CEO compensation (Mitchell and Mulherin, 1996). The recent financial crisis (the global financial crisis of 2008) provides a perfect example of a recent exogenous shock to not only the US economy but the global economy. The financial crisis therefore provides a natural experiment to see how CEOs alter their risk perceptions and their investment decisions in terms of merger activity. Columns 3 and 4 of table 1 provide the regression results of using a financial crisis dummy. Overall the dependent variables CEO wealth, cash compensation, option holdings and ownership appear to have the same effect as during no crisis period. Columns 3 and 4 show that the dummy variable ‘financial crisis’ is not significant. This means that CEOs do not necessary alter their merger decisions during a (large) financial crisis. However, the interaction terms provide more insight into CEO merger decisions. Column 3 of table 1 models the effect of CEO wealth on merger activity during the financial crisis and it shows that CEO wealth does not have a (significant) influence. The height of a CEO’s wealth (option and stock ownership combined) does not affect his/her decision to undertake a merger. Column 4 of table 1 models CEO option holdings and CEO ownership against merger activity. As previously stated these variables remained predominately unchanged (slight change in coefficient value but negligible). However, the interaction term of CEO option holdings and financial crisis is in fact very significant as is the coefficient of the variable. This interaction terms shows that CEO option holdings were an important factor during the recent financial crisis. CEOs with large(r) options holdings were incentivised to engage in

more mergers. Column 4 also presents the interaction term of CEO ownership with the financial crisis dummy variable. This coefficient of this variable is negative but is considered not significant. CEO (stock) ownership did not affect firm merger decisions during the recent financial crisis.

C. Linear probability model

As stated in the methodology section, I also use the linear probability model to model the relationship between merger activity and CEO compensation. I have included the linear probability model (LPM) in the robustness section as the probit model provides the most economic and statistical insight and as such is the main focus of the paper. To clarify, the regression coefficient β is the change in probability that the dependent variable (i.e. mergers) equals one given a unit change in the regressor (holding other regressors constant) in the LPM. The results of the linear regression are posted in table 4. I find that in terms of significance the LPM provides the same outcomes as posted by the probit model. The difference lies in the interpretation of the coefficients of the regressors and their effect on the dependant variable mergers. Model 1 and model 2 of table 4 are the LPM equivalent of models 1 and 2 of table 1 respectively. Model 1 of table 4 shows that CEO wealth has a negative impact on a CEO’s merger activity. Cash compensation is again (in both models) an incentive for many CEOs to engage in less merger activity. Model 2 of table 4 shows the effect of option holdings and ownership. The LPM confirms the probit model in the sense that CEO option holdings increase merger activity while CEO ownership in fact decreases the activity.

Conclusions

I examine how the merger and acquisition activity of a firm is influenced by the structure and height of CEO compensation. I use a probit model to find the relationship between the merger activity of a firm and the height and structure of the CEO compensation. This compensation is broken down into four main variables. These are cash compensation, CEO option holdings, CEO stock ownership and CEO wealth. Note here that CEO wealth is comprised out of both CEO option holdings and CEO stock ownership (in accordance with previous literature). Subsequently the variable mergers (a dummy variable indicating whether a merger took place in a given year) is regressed using the probit model against the previous mentioned compensation variables.

I observe that the different types of CEO compensation affect the merger activity of a firm differently. I begin by examining the relationship between the merger activity and CEO wealth. I find that as CEO wealth increases, the probability of a firm acquiring another firm is reduced. This is in contrast with previous literature in particular work done by Cai and Vijh (2007). They find in their paper on CEO compensation and the relationship to mergers that as CEO wealth increases the probability of a firm becoming an acquirer increases. They account this due to CEO stock and options illiquidity and argue that mergers act as a diversification tool for undiversified CEOs. The difference in results could also be attributed to the sample size used. Afterwards I separate CEO wealth into the two driving mechanism behind it: CEO option holdings and CEO ownership. I find that CEO ownership reduces the probability of a CEO conducting a merger. This result already provides evidence that my first hypothesis should be rejected saying that the height and composition of CEO compensation does not influence merger activity. This result also provides evidence that my second

hypothesis should also be rejected. This hypothesis states that the height and composition of CEO stock ownership does not influence the firm’s merger and acquisition activity. As previously mentioned, I provide statistical evidence that CEO ownership does influence merger activity in a negative way. I also looked at the influence of CEO option holdings on the merger activity of a firm. According to the regression results following from the probit model I have statistical evidence that CEO option holdings decrease the probability of a firm performing a merger. I can therefore reject my third hypothesis stating that the height and composition of CEO option holdings does not influence the firm’s merger and acquisition activity. I provide evidence that it does influence the activity. My findings support the risk incentive theory and are in contrast with previous work arguing the alignment incentive theory. According to the alignment incentive theory CEO option holdings and ownership should provide the same direction of incentive for CEOs to perform mergers as stock ownership and option holdings should be interchangeable and lead to firm value maximization. This is, however, not the case. Option holdings increase the probability of a merger while ownership decreases it. My findings hold more according to risk based incentive. The risk based incentive argues that CEOs respond differently to certain compensation due to the difference in risk perception associated with the type of compensation. CEO stock ownership is accompanied by risk aversion while option holdings increases risk taking behaviour according to this hypothesis. My results confirm that there is indeed a difference in behaviour with respect to certain compensation. These results maintain when controlling for the financial crisis.

There is a slight limitation to the methodology and data I use to investigate this relationship. I look at whether compensation affects the merger activity of a firm but little has been discussed on the risk involved when conducting a merger. The assumption used throughout

this paper is that performing a merger is in itself a risky business decision. Previous papers document that CEO ownership could cause risk aversion while stock options increase seeking behaviour. When combined with my results saying that CEO ownership reduces merger activity while stock options increase merger activity one could be inclined to think that a merger is therefore a risky business decision. However, a merger per definition does not need to risky. My results show that CEO option holdings improve the probability of a CEO conducting a merger. This decision could also of course improve firm value through added resources and capabilities. A CEO with stock options could be more inclined to perform the value-creating merger as he/she too has the possibility to gain from it personally. Stock options in this case act as a positive mechanism to align CEO and shareholder interests and maximize firm value. The most important implication from this study is that the different components constituting CEO compensation are not interchangeable mechanisms when incentivizing CEOs with regards to mergers. By separating the components I can differentiate between the effects of each compensation component on a CEO’s investment decision (i.e. mergers) and can aid in the debate of how to align interest with that of the shareholders. In conclusion, this study has policy implications in particular for businesses and shareholders in determining the right compensation to match the business strategy of the company.

Overall I illustrate that the composition of a CEO’s compensation has an economically and statistically significant effect on the merger activity of a firm. As previously stated this paper lacks in determining risk. Future research could focus on isolating the effects of the specific risk elements associated with stock options, such as the delta and vega, and explore whether these variables help better understand the determinants of CEOs to engage in mergers with respect to their compensation. Future research could investigate whether the effects of CEO ownership and option holdings are also present in other business decisions.

References

Alexandridis, G., Petmezas, D., and Travlos, N. G., 2010 Gains from Mergers and Acquisitions Around the World: New Evidence, Financial Management, Vol.39 (4), pp.1671-1695

Armstrong, C., and Vashishtha R., 2012, Executive stock options, differential risk-taking incentives, and firm value, Journal of Financial Economics 104, 70-88

Bebchuk, L., and Fried J., 2003, Executive compensation as an agency problem, Journal of Economic Perspectives 17, 71-92

Brickley, J., Bhagat, S. and Lease, R., 1985, The impact of long-range managerial compensation plans on shareholder wealth, Journal of Accounting and Economics, Vol. 7 No. 1-3, pp. 115-29.

Cai, J., and A. Vijh, 2007, Incentive effects of stock and option holdings of target and acquiring CEOs, Journal of Finance, 2007, Vol.62(4), pp.1891-1933

Carpenter, J., 2000, Does option compensation increase managerial risk appetite?, Journal of Finance, Vol. 55 No. 5, pp. 2311-31.

Coles, J., N. Daniel, and Naveen L., 2006, Managerial incentives and risk-taking, Journal of Financial Economics 79, 43`-468

Core, J. and Guay, W., 1999, The use of equity grants to manage optimal equity incentive levels, Journal of Accounting and Economics, Vol. 28 No. 2, pp. 151-84.

Core, J., and Guay, W., 2002, Estimating the value of employee stock option portfolios and their sensitivities to price and volatility, Journal of Accounting Research 40, 613-630.

Grinstein, Y., and Hribar, P., 2004, CEO compensation and incentives: Evidence from M&A bonuses, Journal of Financial Economics 73, 119–143.

Grossman, S., and Hart, O., 1983, An analysis of the principal-agent problem, Econometrica 51, 7-45 Guay, W.R., 1999, The sensitivity of CEO wealth to equity risk: an analysis of the magnitude and determinants, Journal of Financial Economics 53, 43–71.

Hagendorff, J. and Vallascas F., 2011, CEO pay incentives and risk-taking: Evidence from bank acquisitions, Journal of Corporate Finance 17, 1078-1095

Hall, B.J., and Liebman, J.B., 1998, Are CEOs really paid like bureaucrats?, Quarterly Journal of Economics, Vol. 113 No. 3, pp. 653-91.

Hanlon, M., Rajgopal, S. and Shevlin, T., 2004, Are executive stock options associated with future earnings? Journal of Accounting and Economics, Vol. 36 No. 1, pp. 3-43.

Harford, J., and K. Li, 2007, Decoupling firm performance and CEO wealth: The case of acquiring CEOs, Journal of Finance 62, 917-949

Hartzell, J., Ofek, E., and Yermack, D., 2004, What’s in it for me? CEOs whose firms are acquired, Review of Financial Studies 17, 37-61`

Haw, I., Pastena, S., and Lilien, S., 1990, Market Manifestation of Nonpublic Information Prior to Mergers: The Effect of Ownership Structure, The Accounting Review, Vol.65 (2), pp.432-451

Himmelberg, C., Hubbard, G., and Palia, D., 1999, Understanding the determinants of managerial ownership and the link between ownership and performance, Journal of Financial Economics, Vol. 3 No. 3, pp. 353-84.

Jensen, M., Meckling W., 1976, Theory of the firm: Managerial behavior, agency costs, and ownership structure, Journal of Financial Economics 3, 305-360

Jensen, M., 1986, Agency costs of free cash flow, corporate finance, and takeovers, American Economic Review 76, 323-329

Lambert, R., and Larcker, D., 2004, Stock options, restricted stock and incentives’’, available at SSRN: http://ssrn.com/abstract =527822, date last visited: 06/04/2015

Lewellen, W., C. Loderer, and A. Rosenfeld, 1985, Merger decisions and executive ownership in acquiring firms, Journal of Accounting and Economics 7, 209-231.

Maksimovic, Vojislav; Phillips, Gordon; Yang, Liu Private and Public Merger Waves, Journal of Finance, 2013, Vol.68 (5), pp.2177-2217

Mehran, H., 1999, Executive compensation structure, ownership, and firm performance, Journal of Financial Economics, Vol. 38 No. 2, pp. 163-84

Moeller, S., Schlingemann, F., Stulz, R., Firm size and the gains from acquisitions, Journal of Financial Economics, 2004, Vol.73 (2), pp.201-228

Morgan, A., and Poulson, A., 2001, Linking pay to performance – compensation proposals in the S&P 500, Journal of Financial Economics 62, 489-523.

Rajgopal, S., and Shevlin, T., 2002, Empirical evidence on the relation between stock option compensation and risk taking, Journal of Accounting and Economics, Vol. 2 No. 33, pp. 145-71

Stulz, R., 1988, Managerial control of voting rights: Financing policies and the market for corporate control, Journal of Financial Economics 20, 25–54.

Tosi et al., 1989, The Decoupling of CEO Pay and Performance: An Agency Theory Perspective Administrative Science Quarterly, Vol.34 (2), pp.169-189

Yermack, D., 1995, Do corporations award CEO stock options effectively? Journal of Financial Economics 39, 237-269.

Wang, Hao, 2011, Managerial entrenchment, equity payout and capital structure, Journal of Banking and Finance, Vol.35 (1), pp.36-50

Walkling, R., Long, M., Agency Theory, Managerial Welfare, and Takeover Bid Resistance, The RAND Journal of Economics, 1 April 1984, Vol.15(1), pp.54-68

Wiersema, F., Bantel, A., Top Management Team Demography and Corporate Strategic Change, The Academy of Management Journal, 1 March 1992, Vol.35 (1), pp.91-121

Appendix

Summary Statistics

Table 1 provides the summary statistics for mergers between 1994 and 2014. Panel A describes the merger characteristics; the industry type, national/international merger and deal value. Panel B describes the independent variables, the CEO measures, for all available observations. Panel C provides governance and CEO characteristics. The data was obtained from the following sources: Thomson One (mergers and acquisitions), Execucomp (compensation) and Governance/CEO controls (Institutional Shareholder Services).

Obs. Mean Median

10th percentile 90th percentile Stan. Dev.

Panel A: Mergers characteristics

Completed 6,095

Same industry 6,095 77.26%

Masterdeal type 6,095 77.46%

Financial crisis 6,095 27.72%

% Shares acquired 6,095 73.94% 100.00% 37.56% 7.71% 100.00% % Shares owned after 6,095 79.10% 100.00% 35.92% 8.60% 100.00%

% Shares sought 6,095 74.98% 100.00% 36.89% 8.88% 100.00%

Value ($ millions) 6,095 779.81 104.00 3603.43 10.06 1477.70

Panel B: CEO measures

% CEO ownership 32,948 2.30% 0.16% 6.04% 0.00% 6.27%

% CEO cash compensation 32,948 52.33% 48.57% 33.84% 9.27% 98.72%

% CEO wealth 32,948 2.91% 0.66% 7.00% 0.02% 7.64%

% CEO option holdings 32,948 0.60% 0.20% 3.36% 0.00% 1.39%

Panel C: Governance/CEO controls

CEO Age 32,948 65.80 66 9.49 54 78

Longevity 32,948 7.51 5.00 7.54 1 17

GDP growth 32,948 89.26%

Gender 32,948 97.94%

CEO Duality 32,948 37.10%

Table 2

Probability acquiring determinants

Table 2 provides the probit regressions of the likelihood that a firm undertakes a merger in any given year ranging from 1994 to 2014. Governance controls as well as year fixed, financial and CEO controls are specified.

(1) (2) (3) (5)

Model 1 Model 2 Model 3 Model 4

% CEO cash compensation -0.27*** -0.27*** -0.27*** -0.27***

(0.03) (0.03) (0.03) (0.03)

% CEO option holdings 0.73*** 0.56***

(0.18) (0.18) % CEO ownership -0.83*** -0.85*** (0.17) (0.19) % CEO wealth -0.009*** -0.009*** (0.001) (0.001) Financial crisis -0.01 -0.03 (0.03) (0.03)

% CEO wealth * fin crisis 0.001

(0.003)

% CEO option holdings * fin crisis 4.97***

(1.01)

% CEO ownership * fin crisis -0.002

(0.003)

Constant 86.20*** 85.86*** 85.21*** 85.27***

(3.98) (3.99) (5.77) (5.78)

Governance controls Yes Yes Yes Yes

Financial controls Yes Yes Yes Yes

Year fixed effects Yes Yes Yes Yes

CEO controls Yes Yes Yes Yes

Observations 32,948 32,948 32,948 32,948

R-squared 0.056 0.056 0.056 0.056

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

Table 3

Probability acquiring determinants

Table 3 provides the probit regressions of the likelihood that a firm undertakes a merger in any given year ranging from 1994 to 2014 without using a constant. Governance controls as well as year fixed, financial and CEO controls are specified.

(1) (2) (3) (5)

Model 1 Model 2 Model 3 Model 4

% CEO cash compensation -0.05* -0.05* -0.19*** -0.19***

(0.0264) (0.026) (0.03) (0.03)

% CEO option holdings 0.81*** 0.58***

(0.17) (0.18) % CEO ownership -0.69*** -0.69*** (0.17) (0.19) % CEO wealth -0.007*** -0.007*** (0.001) (0.001) Financial crisis -0.33*** -0.35*** (0.02) (0.02)

% CEO wealth * fin crisis 0.0001

(0.003)

% CEO option holdings * fin crisis 5.03***

(1.02)

% CEO ownership * fin crisis -0.004

(0.004)

Governance controls Yes Yes Yes Yes

Financial controls Yes Yes Yes Yes

Year fixed effects Yes Yes Yes Yes

CEO controls Yes Yes Yes Yes

Observations 32,948 32,948 32,948 32,948

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

Table 4

Linear probability acquiring determinants

Table 4 provides the regression output of the linear probability that a firm undertakes a merger in a given year ranging from 1994 to 2014. Governance controls as well as year fixed, financial and CEO controls are specified.

Model 1 Model 2 % CEO cash compensation -0.07*** -0.07***

(0.007) (0.007)

% CEO option holdings 0.29***

(0.06)

% CEO ownership -0.192***

(0.04)

% CEO total wealth -0.002***

(0.0004)

Constant 21.02*** 20.85***

(1.02) (1.02)

Governance controls Yes Yes

Financial controls Yes Yes

Year fixed effects Yes Yes

CEO controls Yes Yes

Observations 32,948 32,948

R-squared 0.060 0.061

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1