A window of opportunity : the effect of CEO pay slice on opportunistic insider trading

(1)

A Window of Opportunity

The effect of CEO pay slice on opportunistic insider trading

Master’s Thesis 15 ECTS

Author:

Ewout Baaij

MSc Business Economics,

11145811

Finance

(2)

Statement of Originality

This document is written by Student Ewout Baaij 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.

(3)

Abstract

In this thesis I try to find the effect of CEO pay slice on opportunistic insider trading. The main research question is whether there is a clear relationship between the size of the CEO pay slice and opportunistic insider trading. Based on the literature reviewed on this subject, the hypothesis that a firm with a higher value for the CEO pay slice would have a lower chance of an insider trade being an opportunistic insider trade, is supported by the results of the logistic regression model used. I use a large dataset of over 30 thousand insider trade observations and classify them as either a routine or opportunistic insider trade. I regress the effect of CEO Pay Slice on those trades, while controlling for executive gender, age and whether or not the CEO owns more than 20 percent of the company’s stock. The literature that corresponds with the research outcomes is first of all, the theory that more powerful CEOs are assumed to be more likely to conceal information, because they are well secured in their positions and thus are entrenched. The entrenched CEO may want to limit the amount of opportunistic insider trading in his firm, because he does not want the trades to signal information about the firms’ performance to the market. Secondly, the theory that supports this hypothesis is the one that the board of directors has introduced tournament incentives for the executives of their firm. If they would allow the non CEOs to increase their total compensation with insider trading, this would undo part of the incentive of the tournament theory for these executives.

(4)

1. Introduction

In the past decade there have been numerous scandals around inside information and insider trading in the United States. For example, the case of Sean Stewart in 2016, who was a former vice president at JPMorgan Chase and who was convicted on allegations he provided his father with confidential information on upcoming health-care deals. Moreover, the most recent case of insider trading in the U.S. centered on William T. Walters, a Las Vegas sports bettor. In 2017, he was convicted of making $40 million with inside information he obtained on a dairy

company named Dean Foods. The trial even involved professional golfer Phil Mickelson, who agreed to hand over the $1 million profit he had made.

The Research I will be conducting focuses on this insider trading. Not on all types of insider trading, as has been done numerous times before, but on opportunistic insider trading. This because although early academic research (Bhattacharya and Daouk, 2002) suggested that all insider trading is informative and by observing them an abnormal profit could be made, more recent research (Jeng, Metrick and Zeckhauser, 2003) showed no significant results on abnormal returns and market movements (Lakonishok and Lee, 2001). Restricting insider trading is very important to maintain a fair stock market for investors and new research in the field of insider trading focusses on the separation of liquidity insider traders that do not provide any information to investors, because they trade on a regular basis, and opportunistic insider traders, who trade large amounts of shares in a onetime transaction and thereby increase information asymmetry and improve price efficiency. Tirapat and Visaltanachoti (2010) and Cohen et al. (2012) find that the profitability from an opportunistic insider portfolio is

significantly higher than from a traditional insider portfolio of both liquidity and opportunistic traders.

To link opportunistic insider trading to corporate governance, Dai et al. (2014) found that insiders of poorer governed firms earn significantly larger abnormal profits from their sales transactions. They constructed some channels through which internal governance mechanisms affect the profitability of opportunistic insider trading. However, the authors do not take into account the importance of CEO power. Bebchuk et al. (2010) provides us with the important finding that a company with a larger CEO pay slice, a measure of CEO power and thus a poorer governed firm, leads to lower stock market returns and firm value. The CEO pay slice is

computed as the fraction of aggregate compensation of the top five members of the executive team that is captured by the CEO. The negative effect of CEO power has also been found by Bebchuck on a number of other important corporate outcomes, such as mergers and acquisitions and an increase in the opportunistic timing of option grants to CEOs.

(6)

Furthermore, other authors have also found significant effect of CEO power on corporate characteristics, such as capital structure (Chintrakarn & Jiraporn, 2012). Finally, Jiraporn & Withisuphakorn (2015) have found in their empirical note that more powerful CEOs are more likely to conceal information, as they may engage in agency-driven actions that maximize their private benefits at the expense of the shareholders. This leads to more information asymmetry and therefore lower stock price informativeness.

The gap in the existing literature that I would like to fill is to combine this recent research on opportunistic insider trading with that on CEO power, in the form of CEO pay slice, to see if there is a direct link between the two. The existing literature suggests that firms with a large CEO pay slice, a sign of bad corporate governance, will lead to less available information for investors, because the CEOs will try to conceal this. However, bad governed firms have also shown to have a higher amount and profitability of opportunistic insider trades and these opportunistic insider trades on the other hand, are expected to provide more information to investors. This interesting puzzle in the existing literature will be the starting point for my research. Therefore, my main research question is: How does the size of the CEO pay slice affect opportunistic insider trading in firms? My findings will have important implications for investors, regulators and corporate managers.

I will combine multiple data-sets, accessible through Wharton WRDS for the period of 2007 - 2014. Firstly, I will collect data on opportunistic insider trades. I will use the Center for Research in Security Prices (CRSP) database to obtain the required data for this variable. To properly code the transactions, the method of Cohen, Malloy and Pomorski (2012) is used. Moreover, I will collect data on CEO compensation from the Execucomp database, necessary to compute a measure for CEO pay slice. To construct this measure, the method of Bebchuck et al (2010) is used. The statistical analysis will be done using a Logit model.

The rest of this thesis is structured as follows: the upcoming chapter reviews the existing literature on opportunistic insider trading, corporate governance and CEO pay slice and how I used this to arrive at my hypothesis. Followed by a discussion of the methodology used for answering the research question. Thereafter, I discuss the data and the descriptive statistics. Subsequently, I provide the results from the empirical analysis. Finally, this thesis ends with a conclusion and discussion of its results.

(7)

2. Literature Review

This chapter will discuss the different theories and concepts used to answer the research question of this thesis. At the end of this chapter, the hypotheses used to answer the research question will be constructed.

2.1 Insider Trading

Preventing insider trading is an important matter in most of the stock exchanges around the world and many countries have insider trading laws in place (Bhattacharya and Daouk, 2002). They find that the cost of equity in a country does not change after the introduction of insider trading laws, but decreases significantly after the first prosecution. Regulators are convinced that insiders use private information to benefit from trading and that this creates an unfair market environment for other investors, however, the existing academic literature is far from being conclusive about the profitability of insider trading.

Early academic research showed that insider trading is informative, because outsiders could observe insider trade information and thereby earn an abnormal profit on their

investments (Rozeff and Zaman, 1988). However, recent research shows no significant

abnormal returns earned by insider traders (Jeng, Metrick and Zeckhauser, 2003). Furthermore, Lakonishok and Lee (2001) found no large market movements after insiders traded or reported their trade to the SEC. Beneish and Vargus (2002) find that insider trading is informative about earnings quality and the valuation implications of accruals. They suggest that managers’ contemporaneous trading activities can be used by market participants and researchers to ex ante asses the likelihood of earnings management and the quality of a firm’s accruals.

Continuing, Dai et al. (2015) find that media reduces insiders’ future trading profits by disseminating news on prior insiders’ trades available from regulatory filings. Their findings provide new insights into the real effect of news dissemination.

Moreover, Brochet (2010) has conducted research on the effect of the implementation of the Sarbanes-Oxley Act of 2002 (SOX) on the informativeness of insider trading. Especially on the Form 4 filings required by Section 403 of SOX, which required a more timely disclosure. He discovered that abnormal returns and trading volumes were significantly higher after than before the passage of SOX, when looking at insider purchases. For insider sales, he found that trading volumes were also higher after the implementation of SOX, but that it did not influence stock returns. Brochet identifies two factors that influence the difference in sale returns. First, a decrease in insiders’ propensity to sell closely to the release of bad news and second, the better spread of filings over time compared to before SOX.

(8)

Brochet (2010) delivers also a great contribution on the distinction between all insider trading and opportunistic insider trading and the role the latter plays in providing the market with information about the firm where the opportunistic trading occurs. He quotes Li and Zhang (2007) who find that after the implementation of SOX, opportunistic insider traders sold less stock shortly before accounting restatement announcements. This evidence suggests that after the implementation of SOX, managers have less opportunities and incentives to engage in opportunistic behavior.

2.2 Opportunistic Insider Trading

Tirapat and Visaltanachoti (2010) decided to build a framework to separate the insider traders into liquidity and opportunistic insiders. Opportunistic insider trade increases information asymmetry and improves price efficiency. They identify opportunistic trades as trades for which the probability of informed trading and the speed of convergence to market efficiency increase during the month after an inside trade. Furthermore, they find that the profitability from an opportunistic insider portfolio is significantly higher than from a traditional insider portfolio, which consists of both liquidity and opportunistic traders. They conclude by suggesting that one could apply their framework to examine the impact of corporate governance on insider trading in future research on this topic.

Cohen, Malloy and Pomorski (2012) also manage to separate the different kinds of insider traders. They define a routine trader as an insider who placed a trade in the same calendar month for at least a certain number of years in the past. All other traders for which there is no pattern in the past timing of trades are then identified as opportunistic. This

distinction is made each calendar year. They find that most informed opportunistic traders are local, non-executive insiders from geographically concentrated, poorly governed firms.

2.3 Corporate Governance and Insider Trading

Regarding the field of corporate governance, Dai et al. (2014) find that compared with insiders of poorer-governed firms, insiders of better-governed firms earn significantly smaller abnormal profits from their sales transactions, but not from their purchase transactions. They suggest that future insider trading research should consider the channels through which internal governance mechanisms affect informed insider trading and compute a valuable internal

corporate governance index, consisting of 13 items in the categories of independence, compensation, institution and one with the remaining items.

(9)

Moreover, Hodgson, Lhaopadchan & Treepongkaruna (2016) show the importance of corporate governance to limit the insider profits. They illustrate that abnormal returns after insider trading are generally positive over time, especially on Monday (insider sales) and Friday (insider purchase) for both executive and non-executive directors.

Morse, Wang & Wu (2015) study the paradox of executive lawyers serving multiple tasks inside the firm. They find that general counsels have an impact in outcomes associated with effort both in gatekeeping and strategic initiatives. General Counsels’ fixed effect explains 4% of variation in governance and 2.8% in investment across firms. The General Counsel has the objective to approve insider trades in a specific firm.

Furthermore, Ali and Hirshleifer (2015) find that firms with opportunistic insiders have higher levels of earnings management, restatements, SEC enforcement actions, shareholder litigation, options backdating, and excess executive compensation. Finally, Skaife et al. (2013) studied the relationship between ineffective internal control over financial reporting and the profitability of insider trading. They found that the profitability of insider trading is

significantly greater in firms disclosing material weaknesses in internal control compared to firms that have effective control in place.

2.4 CEO Pay Slice

Continuing, Bebchuk et al. (2010) find that the size of the CEO Pay Slice (CPS), a measure of CEO power which is the fraction of the aggregate compensation of the top-five executive team captured by the CEO, is negatively associated with firm value as measured by industry-adjusted Tobin’s Q. So, an increase in CPS would result in lower stock market returns. They also find that a large CPS leads to lower stock returns accompanying acquisitions announced by the firm and higher likelihood of a negative stock return accompanying such announcements, a higher chance of the CEO receiving a lucky option grant at the lowest price of the month and lower performance sensitivity of CEO turnover. Bebchuk and Fried (2003) use the managerial power approach to predict that CEO pay will be less sensitive to performance and also higher in firms in which managers have relatively more power.

Furthermore, other authors have also found significant effect of CEO power on

corporate characteristics. Chintrakarn & Jiraporn, (2012) find that capital structure, such as debt levels and also the cost of debt are affected by the CPS, because relatively less powerful CEOs exhibit risk aversion, resulting in less risky strategies and a lower cost of debt. Finally, Jiraporn & Withisuphakorn (2015) have found in their empirical note that more powerful CEOs are more likely to conceal information, because they are well secured in their positions and thus are

(10)

entrenched. These entrenched CEOs may engage in agency-driven actions that maximize their private benefits at the expense of the shareholders. This leads to more information asymmetry and therefore lower stock price informativeness. The entrenched CEO may want to limit the amount of opportunistic insider trading in his firm, because he does not want the trades to signal information about the firms’ performance to the market. However, even if a CEO is not

entrenched or if he is under pressure, for example because of poor performance, he might want to disallow opportunistic insider trading for fear of losing his own position.

Continuing, the link from CEO pay to opportunistic insider trading is found in another paper of Cohen et al. (2008). They state that before the implementation of SOX, there was a relationship between the proportion of option holdings in top executives’ total compensation and earnings management. However, the amount of accrual-based earnings management has significantly decreased after the passage of this new regulation on insider trading and firms seem to have shifted to real earnings management. This implies that opportunistic insiders were left with less possibilities to “pump and dump” their stock after the passage of SOX. Bartov and Mohanram (2004) found that top-level executives used earnings management and private

information to increase their total compensation. Cheng and Warfield (2005) arrived at the same conclusion and add that managers with consistently high equity incentives were less likely to report unexpected increases in earnings. Moreover, Huddart and Louis (2006) found the same results when analyzing the 1990’s market bubble. Managers tended to inflate the earnings of their firms prior to them selling their stock options. This happened off course more in

companies where the managers earned a larger percentage of their total compensation in stock options. Nowadays, in the post-SOX era, executives seem to have far less options to use earnings management in order to increase their total compensation. However, the option to use inside information to trade their stock at the right moment remains available to them.

Moreover, Henderson (2011) found evidence that firm executives discuss the profits they expect to make from trading in firm stock, with their board of directors. They negotiate the potential gains from trading in the firm stock and it leads to a reduction in other forms of pay. This is because the board of directors wants to pay their executives on a level that accounts for the extra profits they are expected to earn from informed trades.

A potential reason the board does this, is because they have introduced tournament incentives for the executives of their firm. If they would allow the non CEOs to increase their total compensation with insider trading, this would undo part of the incentive of the tournament theory for these executives. Eriksson (1999) finds in his research on tournament theory that there is a stable convex relation between pay and job levels. Moreover, he finds that the larger

(11)

the amount of managers with important responsibilities in a firm, the larger the wage spread between them. His findings support the tournament theory that there is a positive relationship between the number of participants and the prize of the tournament. The results of his research also support another key implication of the tournament theory, namely that a larger managerial pay spread leads to incentives to improve corporate performance. This might be the reason that the board of directors wants to maintain the salary gap between the top executives of the firm at the same level and do not want to decrease it.

On the contrary, one could argue that a CEO who receives a much higher compensation than his fellow executives, would allow the non-CEOs to increase their total compensation with insider trading and thereby keeping their loyalty. Denis and Xu (2013) found that the use of equity incentives is significantly greater in countries with stronger regulation on insider trading. These larger equity incentives lead to higher total compensation. For top executive pay, they observe also a significant increase in their compensation after the initial enforcement of new insider trading laws. This implies that executives use their equity with inside information to increase their total compensation. Once this is no longer possible, due to restrictions on insider trading, the executives demand and receive a higher compensation. Denis and Xu (2013) conclude by stating that insider trading laws are one of the possible explanations on large differences between countries on executive pay.

Finally, when examining CEO pay, on should be aware of the fact that there exists a large pay gap between male and female executives (Bell, 2005). According to this paper, women top executives earn 8 to 25 percent less than their male counterparts, after controlling for differences in industry, company size and occupational title. One could reason that female top executives would engage in opportunistic insider trading more often than male executives to compensate for this large pay gap.

2.5 Implications and Hypothesis

This section will derive the hypothesis, used for my research, from the literature review. The main question that I would like to answer is whether there is a relationship between the size of the CEO pay slice and opportunistic insider trading.

Firstly, there is the possibility that a firm with a higher value for the CEO pay slice would have a higher chance of an insider trade, made by an employee of that firm in its own stock, being an opportunistic insider trade. Which is forbidden by law. The literature we discussed that

supports this possibility is, first of all, the fact that opportunistic insider trading has proven to happen more often in poorly governed firms. Since a high value for the CEO pay slice is

(12)

considered an indication of more power for the CEO and thus a poorer governed firm, one could argue that we would expect more opportunistic insider trading in firms with a larger CEO pay slice. Subsequently, we found evidence that supports the theory that a CEO who receives a much higher compensation than his fellow executives, would allow the non-CEOs to increase their total compensation with insider trading and thereby keeping their loyalty. This could also lead to the expectation that there would be more opportunistic insider trading in firms with a larger CEO pay slice.

However, there is also the opposing possibility that a firm with a higher value for the CEO pay slice would have a lower chance of an insider trade, made by an employee of that firm in its own stock, being an opportunistic insider trade. The literature we discussed that supports this possibility is, first of all, the fact that more powerful CEOs are more likely to conceal

information, because they are well secured in their positions and thus are entrenched. These entrenched CEOs may engage in agency-driven actions that maximize their private benefits at the expense of the shareholders. This leads to more information asymmetry and therefore lower stock price informativeness. The entrenched CEO may want to limit the amount of

opportunistic insider trading in his firm, because he does not want the trades to signal information about the firms’ performance to the market. However, even if a CEO is not

entrenched or if he is under pressure, for example because of poor performance, he might want to disallow opportunistic insider trading for fear of losing his own position. This could lead to the expectation that there would be less opportunistic insider trading in firms with a larger CEO pay slice.

Moreover, the second theory that supports this possibility is the one that the board of directors has introduced tournament incentives for the executives of their firm. If they would allow the non CEOs to increase their total compensation with insider trading, this would undo part of the incentive of the tournament theory for these executives. This could also lead to the expectation that there would be less opportunistic insider trading in firms with a larger CEO pay slice.

Considering both possibilities and the amount of literature that support the different views, I expect the second possibility to be the outcome of my research. Therefore, the hypothesis becomes:

A firm with a higher value for the CEO pay slice has a lower chance of an insider trade being opportunistic.

(13)

3. Methodology

This chapter describes the methodology used to answer the main research question. First of all, the regression model used to test the effect of CEO pay slice on opportunistic insider trading is explained as well as the methods used to compute the variables for opportunistic insider trading and CEO pay slice. In the next chapter I will discuss the data and descriptive statistics used for my analysis.

3.1 Logit Model

The regression model I will use to answer the main research question is a binary logistic model. This because my dependent variable, opportunistic insider trading, has a binary value. It either takes the value 1 for an opportunistic insider trade or the value 0 for a non-opportunistic, or liquidity, insider trade.

𝑌 𝑖𝑠 𝑡ℎ𝑒 𝑏𝑖𝑛𝑎𝑟𝑦 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒 𝑓𝑜𝑟 𝑜𝑝𝑝𝑜𝑟𝑡𝑢𝑛𝑖𝑠𝑡𝑖𝑐 𝑖𝑛𝑠𝑖𝑑𝑒𝑟 𝑡𝑟𝑎𝑑𝑒 {_{𝑦 = 0 𝑖𝑓 𝑛𝑜 𝑜𝑝𝑝𝑜𝑟𝑡𝑢𝑛𝑖𝑠𝑡𝑖𝑐 𝑖𝑛𝑠𝑖𝑑𝑒𝑟 𝑡𝑟𝑎𝑑𝑒}𝑦 = 1 𝑖𝑓 𝑜𝑝𝑝𝑜𝑟𝑡𝑢𝑛𝑖𝑠𝑡𝑖𝑐 𝑖𝑛𝑠𝑖𝑑𝑒𝑟 𝑡𝑟𝑎𝑑𝑒 }

The logistic formulas are stated in terms of the probability that Y = 1, which is referred to as 𝑝. 1 − 𝑝 On the other hand stands for the probability that Y = 0. From here we derive the logistic regression equation as follows:

ln ( 𝑝

1 − 𝑝) = 𝛽0+ 𝛽1𝑋 + 𝜀

The ln symbol refers to a natural logarithm and 𝛽₀+ 𝛽₁𝑋 + 𝜀 is the standard regression equation, with 𝜀 the error term.

The independent variable that tests our hypothesis is CEO pay slice. Furthermore, we will control for the gender and age of the executive engaging in the insider trade. As well as for a dummy variable for CEO ownership, that takes value 1 if the CEO owns more than 20% of the outstanding shares of the company. Combining the variables stated above into a logistic regression equation gives:

(14)

The time-frame used to test this model is the period of 2007-2014, this is a period of 8 years. First of all, the dependent variable Y is the binary variable opportunistic insider trade, which takes the value of 1 if an observed insider trade can be classified as an opportunistic one. Secondly, CPS is a variable for CEO pay slice. It is defined as the percentage of the total compensation to the top five executives that goes to the CEO (Bebchuck, Cremers and Peyer, 2010). According to the observed literature and our hypothesis, we would expect a negative coefficient for this variable in our regression model. This since the entrenched CEOs may want to limit the amount of opportunistic insider trading in their firms, because they do not want the trades to signal information about the firms’ performance to the market. Moreover, the board of directors could have introduced tournament incentives for the executives of their firm. If they would allow the non CEOs to increase their total compensation with insider trading, this would undo part of the incentive of the tournament theory for these executives.

Subsequently, Gender is a dummy variable which takes the value of 1 if the executive is male. We would expect this variable to have a negative coefficient. This because there exists a large pay gap between male and female executives (Bell, 2005). According to this paper, women top executives earn significantly less than their male counterparts. One could reason that female top executives would engage in opportunistic insider trading more often than male executives to compensate for this large pay gap.

Continuing, Executive age is a dummy variable which takes the value of 1 if the

executive is older than 60. It is a common control variable when observing executive behavior, but there is no clear expectation of the sign that this control variable will have in our regression model. Furthermore, Ownership is a dummy variable which takes the value of 1 if the CEO owns more than 20% of the outstanding shares of the company. For this control variable there is also no clear indication of what the sign of the coefficient will be. However, Bebchuk et al. (2010) use it as a control variable for their CEO pay slice measure. They explain that by basing CPS on compensation information from executives that are all at the same firm, we control for any firm-specific characteristics that affect the average level of compensation in the firm’s top executive team. My regression model will however control for industry and year fixed effects.

3.2 Measuring opportunistic insider trading and CEO pay slice

The first objective of my research is to identify opportunistic insider traders.

I will identify opportunistic insider transactions using the methods of Cohen, Malloy and Pomorski (2012). They define a routine trader as an insider who placed a trade in the same calendar month for at least a certain number of years in the past. All other traders for which

(15)

there is no pattern in the past timing of trades are then identified as opportunistic. This distinction is made each calendar year.

By identifying opportunistic insiders, I will be able to identify the firms where opportunistic insider trading occurs as well, this is my second operational objective. Once I have identified these optimistic firms, I will be able to compute their value for the CEO pay slice (CPS) and observe how this value affects opportunistic insider trading, my third

operational objective. The CPS is defined as the percentage of the total compensation to the top five executives that goes to the CEO (Bebchuck, Cremers and Peyer, 2010). The measure is based on the total compensation to each executive, including salary, bonus, other annual pay, the total value of restricted stock granted that year, the Black-Scholes value of stock options granted that year, long-term incentive payouts, and all other total compensation.

(16)

4. Data and Descriptive Statistics

The data used for this research is obtained from multiple different data-bases, accessible through Wharton WRDS, for the period of 2007 - 2014.

4.1 Opportunistic Insider Trading

First of all, the data used to code the insider trading data into opportunistic and liquidity trades. I will use the Center for Research in Security Prices (CRSP) database to obtain the required data for this variable. To properly code the transactions, the method of Cohen, Malloy and Pomorski (2012) is used. They define a routine trader as an insider who placed a trade in the same calendar month for at least a certain number of years in the past. All other traders for which there is no pattern in the past timing of trades are then identified as opportunistic. This distinction is made each calendar year.

Continuing, all insider traders are thus labeled as either opportunistic or routine traders at the start of each calendar year, based on their past history of trades. From that point onwards, we observe their future trades to keep them classified as the correct kind of insider trader. All subsequent trades that are made after we label each insider as either opportunistic or routine are then placed into one of two buckets: a) “opportunistic trades” (i.e., all trades made by opportunistic traders) and b) “routine trades” (i.e., all trades made by routine traders). We keep only the sales transactions, because they are considered to provide more information to the market. Furthermore, for each calendar year, the first trade determines how the person is coded. This is thus either as an opportunistic or routine insider trader. Cohen, Malloy and Pomorski (2012) require an insider to make at least one trade in each of the three preceding years, in order to be defined as either a routine or opportunistic insider. This means that we have kept only insider trades when that insider had at least one trade in the 3 preceding consecutive years. Finally, we use a link table to link the trades that have only a Cusip code, to different companies, that are listed with a Gvkey code. After this, we have properly linked both the opportunistic and routine trades to the firms in which they occurred.

Note that this straightforward method for classifying opportunistic and routine trades is obviously a noisy proxy for actual insider trading. The strategy used will not be perfect and accurately classify every insider trade in the sample data set. However, for the majority of the trades, the trades made for diversification and liquidity reasons are more likely to be regular in their timing and the trades that provide information to the market are less likely to be regular in their timing.

(17)

4.2 CEO Pay Slice

Moreover, I collected data on CEO compensation from the Execucomp database, necessary to compute a measure for CEO pay slice. To construct this measure, the method of Bebchuck et al (2010) is used. Once I have identified these optimistic firms, I will be able to compute their value for the CEO pay slice (CPS). The CPS is defined as the percentage of the total compensation to the top five executives that goes to the CEO. The measure is based on the total compensation to each executive, including salary, bonus, other annual pay, the total value of restricted stock granted that year, the Black-Scholes value of stock options granted that year, long-term incentive payouts, and all other total compensation (as reported in ExecuComp item # TDC1). Observations were dropped that did not have an executive rank reported or one with a value above 6, since these were considered useless when computing the value of the CPS. The control variables on executive age, executive gender and CEO ownership were also computed from the ExecuComp database. By basing CPS on compensation information from executives that are all at the same firm, we control for any firm-specific characteristics that affect the average level of compensation in the firm’s top executive team.

When observing the values for different CEO pay slices, I decided to adjust this variable to Max CPS percentage in order to improve the results of my research. Max CPS percentage stands for the highest percentage of total compensation of the 5 top executives per company. I did this, because with the regular CEO pay slice variable, there were some CEO’s that received almost no compensation (i.e. sometimes only 2% of total compensation). Because this greatly influenced my results (significance), I chose to replace the CEO compensation for CEO’s with lower compensation than some of the executives in their company, with that of the highest compensated executive for that company. Off course, in the majority of the companies the CEO received the highest compensation. This means that I had no longer values for CPS that were lower than 20%. Observing the new data points for max CPS in a histogram showed that almost all values were between 20 and 60%, with most CEOs receiving around 40% of the top 5 executive compensation.

Finally, the two different data sets on opportunistic insider trading and CEO pay slice had to be merged. They were merged by gvkey and year. Observations of trades who were not merged to a value of CPS were dropped. The final result is a merged database of over 30 thousand observations on insider trades, spread over 8 years (2007-2014), that correspond to a company that also has a value of its CEO pay slice computed. Please find on the next page the tables with the variable descriptions and the summary statistics.

(18)

4.3 Descriptive Statistics

Variable Description

Opportunistic Dummy variable = 1 if insider trade is opportunistic

Routine Dummy variable = 1 if insider trade is routine

MaxCPSperc Highest percentage of total compensation of the 5 top executives per company

CPSper Percentage of total compensation of the 5 top executives per company earned by the CEO

Egender Dummy variable = 1 if executive is male

Eage Dummy variable = 1 if executive is older than 60

CEOownership Dummy variable = 1 if CEO owns more than 20% of the outstanding shares of the company

Table 1. Variable descriptions

Variable Obs Mean Std.Dev 25% 75%

Opportunistic 31.853 0,62 0,48 0 1 Routine 31.853 0,38 0,48 0 1 MaxCPSperc 31.853 0,43 0,10 0,36 0,47 CPSper 31.853 0,41 0,11 0,35 0,47 Egender 31.853 0,96 0,19 1 1 Eage 31.853 0,28 0,45 0 1 CEOownership 31.853 0,13 0,34 0 0

Table 2. Summary Statistics

Table 2 contains the summary statistics of both the dependent and independent variables used for this research, which are described in Table 1. All variables have 31853 observations. We observe that 62 percent of the insider trade data used in this research is classified as an opportunistic insider trade. Since insider trades are always classified as either opportunistic or routine, the remaining 38 percent of the trades is considered to be a routine trade.

Furthermore, the summary statistics on the independent variable Max CEO pay slice show that the average percentage of the total compensation to the top 5 executives, which is rewarded to the number one earning executive, is 43 percent. When we observe the 25% and 75% percentile for Max CEO pay slice, we see that they are also close to 40 percent. Compared to the regular CEO pay slice, containing values below 20 percent for the CEO, the regular variable has a mean only 2 percent lower than the adjusted one.

Moreover, when we examine the summary statistics of the different control variables, we see that 96 percent of all observations came from companies led by a male CEO.

(19)

Continuing, 28 percent of the CEOs is older than 60 years. Finally, 13 percent of the CEOs in this sample own more than 20 percent of the company’s stock.

(20)

5. Results

Logit Logit Logit

(1) (2) (3) MaxCPSperc -0,309 -0,401 -0,440 (0,120)** (0,121)*** (0,124)*** Egender -0,256 -0,288 (0,064)*** (0,066)*** Eage 0,176 0,183 (0,038)*** (0,039)*** CEOownership -0,032 0,008 (0,042) (0,043)

Industry Fixed Effects no no yes

Year Fixed Effects no no yes

Observations 31.853 31.828 31.828

Number of Years 8 8 8

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

Table 3: Regression output, dependent variable is Opportunistic Insider Trade

Table 3 contains the regression results of the logistic model used. In column 1, Max CPS percentage is the only independent variable and is regressed directly on opportunistic insider trade. In all three regressions, the dependent variable is binary and takes the value of 1 for an opportunistic insider trade and the value of zero for a routine insider trade, executed by an insider in the fiscal year. Column 1 contains the estimation on the Max CPS percentage variable and the effect it will have on the dependent variable opportunistic trade. The result is a negative coefficient and strongly significant. The coefficient is -0.31, this means that a one percentage increase in Max CPS results in a 0.31% decrease in the chance of an insider trade being an opportunistic one. Please note that in the regression in column 1, there are no control variables included in the regression. This means that this effect is probably not correctly estimated.

Continuing, for the regression in column 2, we add three control variables to the model. Executive age and CEO ownership are used as control variables by Bebchuck et al (2010). Gender is also often used as a control variable in the literature to observe the difference in results for men and women. First of all, executive gender is added and has a negative coefficient on opportunistic insider trade that is strongly significant. The variable of executive gender takes value 1 for male and this result implies that males are 26 percent less likely to engage in

opportunistic insider trading. This could be considered an interesting finding, even though according to the literature, there exists a large pay gap between male and female executives

(21)

(Bell, 2005). According to this paper, women top executives earn significantly less than their male counterparts. We could expect that female top executives would engage in opportunistic insider trading more often than male executives to compensate for this large pay gap.

Furthermore, executive age seems to have a positive and strongly significant effect on opportunistic insider trade. The variable of executive age takes value 1 for executives more than 60 years old. This result implies that older executives are 17.6% more likely to engage in opportunistic insider trade. I did not find a clear explanation for this result in the literature. However, perhaps CEOs that are approaching their retirement are tempted to increase their cash levels in their final years of service. This could be an interesting starting point for future research. Finally, the fact that a CEO owns more than 20 percent of a company’s stock does not seem to significantly influence the dependent variable of opportunistic insider trade. The variable of CEO ownership takes the value of 1 if the CEO owns more than 20 percent of the company’s stock. Moreover, due to the addition of the control variables, the coefficient of Max CPS takes a larger negative value and becomes even stronger significant.

Finally, the regression in column 3 adds industry fixed effects and year fixed effects to the regression model. They do not change the sign or significance of the coefficients. However, the three significant variables obtain larger negative or positive coefficients. A one percent increase in Max CPS now results in a 0.44 percent decrease in the chance of an insider trade being an opportunistic one. Furthermore, male executives are now 29 percent less likely to engage in opportunistic insider trading than their female counterparts. Concluding, the

probability that a CEO who is more than 60 years old engages in opportunistic insider trading is now 18.3% larger than a younger CEO.

(22)

6. Conclusion and Discussion

This thesis has studied the effect of the CEO pay slice on opportunistic insider trading. The main research question is whether there is a clear relationship between the size of the CEO pay slice and opportunistic insider trading.

Based on the literature reviewed on this subject, the hypothesis stated that a firm with a higher value for the CEO pay slice would have a lower chance of an insider trade, made by an employee of that firm in its own stock, being an opportunistic insider trade. The literature we discussed that supports this theory was, first of all, the fact that more powerful CEOs are assumed to be more likely to conceal information, because they are well secured in their positions and thus are entrenched. The entrenched CEO may want to limit the amount of opportunistic insider trading in his firm, because he does not want the trades to signal information about the firms’ performance to the market. However, even if a CEO is not

entrenched or if he is under pressure, for example because of poor performance, he might want to disallow opportunistic insider trading for fear of losing his own position. This could lead to less opportunistic insider trading in firms with a larger CEO pay slice.

Moreover, the second theory that supported this hypothesis is the one that the board of directors has introduced tournament incentives for the executives of their firm. If they would allow the non CEOs to increase their total compensation with insider trading, this would undo part of the incentive of the tournament theory for these executives. This leads to the expectation that there would be less opportunistic insider trading in firms with a larger CEO pay slice.

This hypothesis is confirmed by the empirical results of the research conducted in this paper. I have used a Logit regression model to find a statistically significant negative

relationship between an increase in the CEO pay slice and the chance of an insider trade being an opportunistic insider trade. I find similar and even stronger outcomes when including control variables for executive gender, executive age and CEO ownership. The same holds for adding year and industry fixed effects to the regression model. For the control variables, we find statistically significant results for two of them when regressed on opportunistic insider trade. Male executives appear to engage significantly less often in opportunistic insider trading. On the contrary, CEOs with an age above 60 engage significantly more into opportunistic insider trading.

These results provide interesting implications for future research. One could conduct research on the question if the negative effect of a larger CEO pay slice on the chance of an opportunistic insider trade arises from entrenchment of the CEO or perhaps not because of entrenchment, but still a result of CEO behavior. However, the other theory found in the

(23)

literature for this observation is the importance that boards of directors attach to the presence of tournament incentives in the boards of their firms.

Further implications of the results on the control variables suggest that companies or policy makers should decrease the salary gap between male and female executives. This since one could argue that female executives would compensate their lower salary by engaging in opportunistic insider trade. Moreover, perhaps firms and the government should come up with a solution to prevent older CEOs from engaging in opportunistic insider trades when they are approaching their retirement.

Finally, like all research, this thesis has a few limitations that need to be mentioned. First of all, please note that the method used for separating opportunistic and routine trades is a relatively straightforward one and that this is obviously a noisy proxy for actual insider trading. The strategy used will not be perfect and accurately classify every insider trade in the sample data set, however, the majority of the insider trades will be classified correctly. Still, the results are to be interpreted with caution. Furthermore, including some additional control variables from the literature and reporting some robustness checks on the logistic regression model would further increase the quality and accuracy of this research.

(24)

7. Reference List

Ali, U. & Hirshleifer, D. 2015. “Opportunism as a Firm and Managerial Trait: Predicting Insider Trading Profits and Misconduct.” Working paper. Available at:

http://papers.ssrn.com/sol3/Papers.cfm?abstract_id=2635257

Bartov, E. and Mohanram, P. 2004. "Private information, earnings manipulations, and executive stock-option exercises." The Accounting Review 79.4 : 889-920.

Bebchuk, L. and Fried, J. 2003. "Executive compensation as an agency problem." The Journal

of Economic Perspectives 17.3 : 71-92.

Bebchuk, L., Cremers, M. and Peyer, U. 2010. “The CEO pay slice”. Journal of Financial

Economics 102, no. 1, (October 2011): 199–221

Bell, L. 2005. "Women-led firms and the gender gap in top executive jobs." Working paper. Beneish, M. and Vargus, M. 2002. "Insider trading, earnings quality, and accrual

mispricing." The Accounting Review 77.4 : 755-791.

Bhattacharya, U. & Daouk, H. 2002. “The World Price of Insider Trading.”

Journal of Finance 57, 75-108.

Brochet, F. 2010. "Information content of insider trades before and after the Sarbanes-Oxley Act." The Accounting Review 85.2 : 419-446.

Chintrakarn, P., Jiraporn, P. & Tong, S. 2015. "How do powerful CEOs view corporate risk-taking? Evidence from the CEO pay slice (CPS)." Applied Economics Letters 22.2: 104-109.

Cohen, D., Dey, A. and Lys, T. 2008. "Real and accrual-based earnings management in the pre-and post-Sarbanes-Oxley periods." The accounting review 83.3 : 757-787.

Cohen, L., Malloy, C. & Pomorski, L. 2012. "Decoding Inside Information." Journal of

Finance 67, no. 3: 1009–1043.

Dai, L., Parwada, J. & Zhang, B. 2015. "The governance effect of the media's news

dissemination role: Evidence from insider trading." Journal of Accounting Research 53.2 : 331-366.

Dai et al. 2014. “Internal Corporate Governance and the Profitability of Insider Trading.”

Available at:

www.business.kaist.ac.kr%2Ffaculty%2Finmoo%2FArticles%2FDaiFuKangLee2015Jan.pdf&u sg=AFQjCNEok3inCkQ4zc93jVF_3P2uHV1gvg

Denis, D. and Xu, J. 2013. "Insider trading restrictions and top executive compensation." Journal of Accounting and Economics 56.1 : 91-112.

Eriksson, T. 1999. "Executive compensation and tournament theory: Empirical tests on Danish data." Journal of labor Economics 17.2 : 262-280.

(25)

Henderson, M. 2011. "Insider trading and CEO pay." Vand. L. Rev. 64 : 503.

Hodgson, A., Lhaopadchan, S., & Treepongkaruna, S. 2016. “Corporate Governance, Earning Quality and Information Asymmetry: Evidence from Insider Trading Around the Globe.”

Earning Quality and Information Asymmetry: Evidence from Insider Trading Around the Globe, February 1, 2016.

Huddart, S. and Louis, H. 2006. "Managerial stock sales and earnings management during the 1990s stock market bubble." Unpublished working paper, The Smeal College of Business, The

Pennsylvania State University.

Jeng, L., Metrick, A. & Zeckhauser, R. 2003. “Estimating the returns

to insider trading: A performance evaluation perspective.” Review of Economics and

Statistics 85(2), 453-471.

Jiraporn, P. & Withisuphakorn, P. 2015. “The Effect of CEO Power on the Informativeness of Stock Prices: An Empirical Note.” Available at SSRN 2555864.

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2555864

Lakonishok, J. & Li, I. 2001. “Are Insider Trades Informative?” Review of

Financial Studies 14(1), 79-111.

Morse, A., Wang, W. & Wu, S. 2015. “Executive gatekeepers: the paradox of lawyers in the firm.” Working Paper. Available at:

http://faculty.haas.berkeley.edu/morse/research/papers/MorseWangWu.pdf

Qiang, C. and Warfield, T. 2005. "Equity incentives and earnings management." The

accounting review 80.2 : 441-476.

Rozeff, M., and Zaman, M. 1988. “Market Efficiency and Insider Trading: New evidence.” Journal of Business 61, 25-44.

Skaife, A., Veenman, D. and Wangerin, D. 2013. "Internal control over financial reporting and managerial rent extraction: Evidence from the profitability of insider trading." Journal of

Accounting and Economics 55.1 : 91-110.

Tirapat, S. and Visaltanachoti, N. 2010. “Opportunistic Insider Trading.” Pacific-Basin Finance

A window of opportunity : the effect of CEO pay slice on opportunistic insider trading