Equity Based Compensation: The effect of Restricted Stock on Acquisition
Stock Performance.
Using a sample of 2558 mergers and acquisitions from 2001 till 2007, I find a significant negative relation between restricted stock compensation and acquisition announcement stock performance. Because of the increased use of restricted stock in equity based compensation, the total equity based compensation also shows a negative relation with the announcement stock performance. The results are stronger for CEOs of large firms and CEOs in the middle two quartiles of ownership.
Universiteit van Amsterdam MSc Finance, Corporate Finance Master Thesis
Quinten Duin 10658300 July 2017
Statement of Originality
This document is written by Student Quinten Duin who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is 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 3 Literature Review 5 Methodology 7 Sample Characteristics 10 Results 14 Conclusion 22 Bibliography 24
1. Introduction.
Mergers & Acquisitions are large investments which can create shareholder value. There are various reasons for CEOs to participate in acquisitions. The most important incentives are hubris, personal risk and wealth-based incentives. However, these investments can lead to conflicts of interest between the CEOs and shareholders. Executive compensation packages are often used to align the incentives of CEOs and shareholders.
The effectiveness of executive compensation packages as alignment tool is well studied. Early research shows that incentive based compensation should increase value-maximizing behavior of managers (Shleifer & Vishny, 1988). More recent research examines the effect of equity based compensation on M&A stock performance. The results from Datta et al. (2001) show that higher equity based compensation results in higher stock returns at acquisition announcements for both the short- and long-run. Similar research from Minnick et al. (2011) shows a positive relation between pay-for-performance sensitivity (PPS) and acquirer stock returns. This is due to PPS reducing the incentives for CEOs to engage in value destroying acquisitions.
Most prior research uses the newly granted options as factor for equity based compensation. However, research shows that companies start to use less options grants as compensation and start to use more restricted stock grants. Lord & Saito (2015) show that in 2001 the use of restricted stock was only 5% of total compensation and option grants was 43% of total compensation. In 2007 25% of total compensation was restricted stock and only 21% was option grants. Similar results are shown by Frydman & Jenter (2010).
This shows that restricted stock has become a more important part of equity based compensation. However, restricted stock and option grants are not valued similarly. Irving et al. (2011) show that restricted stock is on average valued negatively by the stock market. This is in contrast with stock options which are valued positively. This is due to the fact that restricted stock lacks the same positive incentive effects as option grants. Additionally, Hodge et al. (2009) show that managers value stock options greater than restricted stock. Longstaff (2001) and Kahl et al. (2003) show that this negative valuation is because of the liquidity costs of the restriction. Lastly, Bryan et al. (2000) show that restricted stock makes managers less likely to engage in risky but value-enhancing investments.
This paper uses the rise in use of restricted stock and the valuation differences between restricted stock and stock options to re-examine the effect of equity based compensation on acquisition stock performance. Previous research used only newly granted
options as equity based compensation, whereas this paper uses both newly granted options and newly granted restricted stock.
This paper uses a sample of 2558 acquisitions in the period from January 1, 2000 till December 31, 2007. This time period shows a large increase in the use of restricted stock and large decrease in the use of stock options. The results show positive cumulative abnormal returns for the entire sample. However, it shows weakly significant higher returns for firms with high equity based compensation and for firms with high restricted stock. Using a multivariate regression analysis I find a significant negative coefficient for restricted stock and insignificant coefficients for options. The total equity based compensation is weakly significant and is because of the significance of the restricted stock part. Using restricted stock as fraction of equity based compensation, instead of total compensation, also shows a negative coefficient. These results are in line with previous research showing that restricted stock is valued negatively. Therefore, this shows that restricted stock becoming a larger part of equity based compensation results in lower acquisition stock returns. Controlling for ownership it shows that the results are the strongest in the middle two quartiles. Additionally, after controlling for firm size it shows stronger results for large firms. The results clearly show that the composition of executive compensation affects shareholder wealth. Therefore it adds to the existing literature on executive compensation by showing the impact of restricted stock on shareholder wealth in M&A’s.
The remainder of this thesis is organized in the following way. Part 2 describes the current existing literature. The sample formation and research methods are explained in part 3. Next, the sample characteristics and results are shown in part 4 & 5 respectively. Part 6 concludes, gives limitations of this thesis and recommendations for additional research.
2. Literature review.
There are various studies on incentives for managers to participate in mergers. Generally the incentives are categorized in hubris, personal risk and wealth-based incentives. First of all, Roll (1986) shows that Hubris is an important incentive for managers to participate in mergers. Another well studied incentive is the diversification of personal risk. First, Amihud & Lev (1981) find that managers use mergers to decrease their “employment risk”. They use the diversification effect of mergers to reduce this risk. Further there are studies that show wealth-based incentives for managers to participate in mergers. So do Harford & Li (2007) show that bidding CEOs are rewarded with substantial new stock and option grants. These large grants offset negative stock returns of poor-acquisition stock performance. So the CEOs are insensitive to poor acquisition performance while they are sensitive to good acquisition performance. Where negative stock returns are considered as poor performance and positive stock returns as good performance. Additionally, Bliss & Rosen (2001) perform a similar research for bank mergers and acquisitions. Their results show that CEOs compensation increases significantly after a merger even if there are negative stock returns. This is mainly due to the increase in firm size.
Overall these studies show that there are incentives for CEOs to participate in mergers. However, these incentives of CEOs might not be aligned with the incentives of stockholders. Generally executive compensation packages are a good way to align management and stockholder incentives. This alignment has a potential influence on corporate take-over decisions. Early research from Jensen & Ruback (1983) contributes to this discussion by examining the relation between acquiring management compensation to stock price effects of an acquisition. They find that corporate take-overs generate gains. The target shareholders earn benefits and the shareholders of the acquirer do not lose (Jensen & Ruback, 1983). Shleifer & Vishny (1988) state that incentive based compensation should increase value-maximizing behavior of managers. Additionally, Mehran (1995) examines the impact of the executive compensation structure on firm performance. He shows that firm performance is positively related to the percentage of equity based compensation.
More recent research from Datta et al. (2001) contributes to this discussion by examining the effect of the compensation structure prior to the acquisition on the acquirer stock performance. They use the newly granted options as fraction of total compensation as a variable for equity based compensation. Their results show a positive relation between stock
prices around the acquisition announcement and executives’ equity-based compensation in the period 1993-1998. Additionally they find that high EBC (equity-based compensation) managers pay significantly lower acquisition premiums. Their research is based on non-financial firms. Minnick et al. (2011) contributes to the paper from Datta et al. by doing a similar research for financial firms. Their results are very similar. Financial institutions CEOs with higher pay-for-performance sensitivity (PPS) have significantly higher stock returns around acquisition announcements. They find, similar as Shleifer and Vishny that this is because PPS reduces the incentives for value destroying investments (Minnick et al., 2011).
The previous research use only newly stock options granted for equity based compensation. However, since then the use of restricted stock has increased largely. So do Lord & Saito (2015) show that the compensation structure changed in the more recent years. Since the internet bubble the total compensation increased up through 2007. In this period the use of stock options decreased notably, mostly replaced by the use of restricted stock. They find that restricted stock only represents 5% of total compensation in 2001 while it increased to 25%. In contrast, in the same period option grants decreased from 43% to 21%. Additionally, Frydman & Jenter (2010) also show that restricted stock replaced stock options as most popular form of equity compensation.
The increased use of restricted stock shows that it is become a more important factor in equity based compensation. However, various studies show that restricted stock is not similarly valued as stock options. A study from Irving et al. (2011) shows that “restricted stock is on average valued negatively by the stock market. While, contrary to the restricted stock, stock options are valued positively”. This is because restricted stock doesn’t have the same positive incentive effects as stock options and therefore are seen as liabilities. Similar to these findings, Hodge et al. (2009) find that managers value stock options greater than their Black-Scholes value and also greater than their fair-value-equivalent of restricted stock. They also find that managers’ subjective valuation decreases when the vesting is extended and increases if the expiration date is extended. This is consistent with managers being impatient for future rewards. Additionally, Longstaff (2001) shows that investors with restricted stock give up a large portion of their wealth to avoid the risks of being constrained from selling. This results in substantial discounts for illiquid stocks. Further, research from Kahl et al. (2003) shows that the liquidity costs of the restriction can be substantial. Restricted stock holders even prefer to receive a smaller fraction of the share in cash instead of receiving a restricted stock. Next, the paper from Silber (1991) shows that companies must offer discounts of more than 30 percent to sell blocks of restricted stock. This shows that the
illiquidity of restricted stocks is valued negatively. Additionally, Bryan et al. (2000) show that restricted stock makes CEOs even less willing to take on risky, but value enhancing, investments. This is due to the linear payoff of restricted stock and the concave utility function of CEOs. Overall, the results from earlier research shows that restricted stocks are valued negatively and costly compared to stock options.
3. Methodology.
A. Sample Formation
The ThomsonOne database is used to download data on mergers and acquisitions during the period January 1, 2000, to December 31, 2007. Transactions are included if it is listed as completed with an effective date and an announcement date during the sample period. Further, the transaction is included if it is a merger or acquisition of majority interest. This database includes data on transaction value and method of payment. Next, the data on executive compensation is retrieved from the S&P ExecuComp database. The total compensation, salary, bonus, stock options and restricted stock can be retrieved from this database for firms in S&P indexes and firms that are no longer in the S&P indexes but were in the past. This database has two different variables for both restricted stock and options paid. They have a variable for before and after the FASB 123R change in 2005. The data for these different variables is combined into a new variable with the data over the entire sample period. Finally, stock price data is retrieved from the CRSP database. Merging these databases leaves a sample of 2558 mergers and acquisitions.
B. Research method
This research is based on event study methodology. Abnormal returns are used examine the stock price reaction around the acquisition announcement. To calculate the abnormal returns the predicted returns are estimated using the market model. Brown & Weinstein (1985) show that there is limited value to use more complicated values than the market model. Therefore the market model is used to estimate the predicted returns. The estimation period is from 200 to 60 trading days prior to the event. Next the abnormal returns are calculated as the difference between the actual return and the predicted return.
𝑖𝑡 = 𝑖+ 𝑖 𝑚𝑡 + 𝜀𝑖𝑡
𝑖𝑡 = 𝑖𝑡− ̂𝑖 − ̂𝑖 𝑚𝑡
Finally, the cumulative abnormal returns are calculated as sum of the abnormal returns in the event period. The event period is a three-day period from day 0 till day 2. The three-day (0,2) cumulative abnormal return is calculated for the entire sample, as well as segments of the sample. The sample is split based on equity based compensation as fraction of total compensation, newly granted options as fraction of total compensation and newly granted restricted stock as fraction of total compensation. Equity based compensation (EBC) is the total of all equity compensation (options and restricted stock). This split is based on whether or not the EBC, options and restricted stock are above the median values. Additionally, for control the sample is divided based on ownership and method of payment. The CAR and the difference of CAR between High and Low EBC, Restricted Stock and Options are tested using a t-test. Next, a cross-sectional regression analysis is used to examine the effect of these variables on the stock price response. The dependent variable is the Three-day (0,2) cumulative abnormal return. Various regressions on the following general model are done:
= ( 𝑖 𝑖 , 𝑎, 𝑖 ,, 𝑎 𝑖 𝑖 ,𝑖 𝑖 , 𝐸 ,𝑖 𝐸 , 𝐼ℎ𝑖 , , 𝑌 𝑎 )
The main important variables are EBC, Restricted Stock, Options and Restricted EBC. EBC is the natural logarithm of 1+ equity based compensation as fraction of total compensation. Options is the natural logarithm of 1 + newly granted options as fraction of total compensation. Restricted Stock is the natural logarithm of 1 + newly granted restricted stock as fraction of total compensation. Additionally, there is the variable Restricted EBC. This is the natural logarithm of 1 + restricted stock as fraction of equity based compensation. An increase in this variable means that restricted stock increases more than stock options. This variable is important since previous literature shows that Restricted stock increased largely while Options decreased and therefore shows an increase in this variable. It shows if this change affects the stock price reaction at acquisitions. The natural logarithm for these variables is used to enhance the skewness and to mitigate effects of outliers.
Size is a variable for the natural logarithm of the market capitalization of the acquirer. This control variable is added because previous research shows that the size affects announcement returns. Moeller et al. (2004) show that smaller acquirers have higher announcement returns. This result is not affected by the method of payment or whether the firm is private or public. Bajaj & Vijh (1995) also show higher announcement returns for smaller firms. This is because there is little news produced during non-announcement periods for these firms. Payment is a binary variable which takes the value 1 if the acquisition was financed with 100 percent cash. Travlos (1987) shows that the method of payment has a significant effect on the acquisition type. This is due to a signaling effect. These results are independent from the method of acquisition. Therefore the method of payment is included as a control variable. Previous Options is the variable of the natural logarithm of 1 + the previous granted options as fraction of the total shares outstanding. Guay (1999) shows that the sensitivity of managers’ wealth to firm performance increases with previous granted options. Next, Ownership is the natural logarithm of 1 + the sum of previously acquired stock and restricted stock owned as fraction of the shares outstanding. This control variable is used because previous research from Ofek & Yermack (2000) shows that the effectiveness of equity based compensation is affected by their ownership. They show equity compensation does well in incentivizing managers with low-ownership. However, managers with high-ownership reduce this effect by selling previously owned shares. Lastly, a coefficient for the relative size of the acquisition is added. This is the natural logarithm of 1 + transaction value divided by acquirer market value. Kusewitt (1985) shows that the relative size and acquirer performance are negatively related. Year is to control for time trends and Industry is to control for industry effects. To examine the robustness of the findings, the regressions are re-estimated by slicing them into quartiles of ownership and slicing them in subsamples of firm size.
The main contribution of this research is to estimate if restricted stock affects acquisition stock price performance. The mentioned literature in part 2 shows that restricted stock becomes a more important factor of equity based compensation and that restricted stock is valued negatively. Therefore the hypothesis states that the use of restricted stock in executive compensation decreases the acquisition stock price performance. Since restricted stock becomes a more important part of equity based compensation it is also expected to be observed in the coefficient for equity based compensation.
4. Sample Characteristics.
First, table I shows descriptive statistics for the 2558 acquisitions in the sample during the period 2001 to 2007. Panel A shows that the frequency of acquisitions is relatively similar for all the years. Further, panel A also shows that the average transaction value is non-monotonic. It shows a high value with $720.48 million in 2005 but it decreased since then. This makes for an average transaction value of $423.79 million over the entire sample period. Panel B shows that 44.29% of the sample is paid with 100% cash and that only 6.1% is paid with 100% stock. On average 55% of the deal is paid with cash and 11.3% is paid with stock. Therefore, in this sample the largest part of acquisitions is paid with cash. There could be several explanations for the large use of cash in this sample. According to Amihud et al. (1990), acquiring firms with higher managerial ownership are more likely to pay with cash. This is because managers are afraid to lose control. Additionally, Eckbo et al. (1990) show that announcement returns increase in the fraction of cash paid. However, the largest part 50.39% is mixed pay. This is a combination between cash pay, stock pay, other pay and unknown pay. A large part of the sample is paid with by ThomsonOne classified “unknown pay”.
Table I : Descriptive Statistics Acquisitions
The sample is based on 2558 mergers & acquisitions in the period 1-1-2001 till 31-12-2007. The acquisitions are listed in the ThomsonOne database for mergers and acquisitions. The executive compensation data is listed in the S&P ExecuComp database. The deal value is in $ million. Cash indicates acquisitions paid with 100% cash. Stock indicates acquisitions paid with 100% stock. Mixed is a mix between cash, stock, other and unknown pay as indicated by the ThomsonOne database.
Panel A: Frequency
Year Number of
acquisitions
% of sample Avg. transaction value
2001 361 14.11% 422.39 2002 355 13.88% 351.45 2003 355 13.88% 217.67 2004 373 14.58% 417.65 2005 379 14.82% 720.48 2006 388 15.17% 504.23 2007 347 13.57% 332.67 Total 2558 100% 423.79
Table I–Continued Panel B: Method of Payment
Method of payment Number of
acquisitions
% of sample Average
Cash 1133 44.29% 54.95%
Stock 156 6.10% 11.30%
Mixed 1289 50.39%
Table II shows descriptive statistics for the executive compensation. First, the average total compensation is $7.7 million with a median of $3.5 million. On average only 0.67 million is paid with salary and 0.94 million is paid with bonus. Equity based compensation (the combination of Options and Restricted Stock) makes up for 50.95% of the total compensation. Options are on average the largest part with 40.98% while restricted stock is 9.97% of the total compensation. The median of restricted stock is 0%, therefore more than half the CEO’s don’t receive restricted stock as compensation. Panel B shows the change in equity based compensation over time. This shows, similar to Lord and Saito (2015), that the use of options decreased largely while the use of restricted stock increased. In the year 2000 the average restricted stock granted was only $466 thousand. Since then it increased to almost $2 million in 2006. On the other hand, the average options granted decreased from $7.6 million to 2.5$ million. As a proportion of total compensation this means that restricted stock increased from 5% to 21.65%. Similar, options granted decreased from 48% of total compensation to only 32%. This equal to an increase of 310% in restricted stock granted and a decrease of 33% of options granted. This shows that Restricted Stock has become a larger part of the equity based compensation.
Table II: Descriptive Statistics CEO Compensation
The sample is based on 2558 mergers & acquisitions in the period 1-1-2001 till 31-12-2007. The acquisitions are listed in the ThomsonOne database for mergers and acquisitions. The executive compensation data is listed in the S&P ExecuComp database. The Total Compensation, Salary, Bonus, Restricted and Options are in $ thousand. %EBC indicates the equity based compensation as percentage of the total compensation. %Options and %Restricted indicate the newly granted options and the newly granted restricted stock as percentage of the total compensation. Compensation data is from fiscal year prior to acquisition announcement.
Panel A: Executive Compensation Statistics
Average Median Minimum Maximum
Total Compensation $ 7,692.89 $ 3,549.84 $ 0.001 $ 600,347.40 Salary $ 671.09 $ 616.89 $ 0.00 $ 5,806.65 Bonus $ 937.61 $ 487.50 $ 0.00 $ 43,511.53 Restricted $ 948.28 $ 0.00 $ 0.00 $ 47,880.00 Options $ 4,481.24 $ 1,186.57 $ 0.00 $ 600,347.40 % EBC 50.95% 53.72% 0.00% 323.24% % Options 40.98% 38.80% 0.00% 315.14% % Restricted Stock 9.97% 0.00% 0.00% 222.50%
Panel B: Compensation Change Over Time
Year Avg. Restricted
stock
Avg. Options Avg. %
Restricted stock Avg. % Options 2000 $466.22 $ 7,563.26 5.28% 47.89% 2001 $548.68 $ 7,321.55 4.34% 49.53% 2002 $605.24 $ 3,962.35 6.79% 45.07% 2003 $701.53 $ 3,452.91 7.93% 41.16% 2004 $1,135.99 $ 3,512.70 10.17% 38.03% 2005 $1,369.90 $ 2,905.65 16.00% 31.91% 2006 $1,994.27 $ 2,449.12 21.65% 32.07%
Figure I: CEO compensation structure over time
Components of average total compensation from 2000 till 2006. Compensation data is from fiscal year prior to acquisition announcement. Data is collected from the S&P ExecuComp database. All components are in $ thousand.
Figure I shows the average level of CEO compensation for the period 2000 till 2006. The total compensation is split in the five categories classified by the ExecuComp database. CEO compensation data is from the fiscal year prior to the acquisition announcement. The figure shows that Total Compensation had a large drop in 2002. This drop is mostly caused by a large drop in Options paid in 2002. As seen in the figure the newly granted options kept declining. However, Restricted Stock increased over this time period. The dollar value of the salary has been relatively constant. Additionally, the value of the Bonus paid increased a bit in 2004 and 2005 but decreased again in 2006. The component for Other compensation varies year-by-year and therefore does not show a real pattern. The most important part of Figure I and Table II is that the use of Options decreased and Restricted Stock increased. Additionally, Total Compensation decreased and therefore Restricted Stock became a larger component of Total Compensation.
0 2 ,0 0 0 4 ,0 0 0 6 ,0 0 0 8 ,0 0 0 1 0 ,0 0 0 2000 2001 2002 2003 2004 2005 2006 Salary Bonus
Options Restricted Stock Other
5. Results.
A. Executive compensation & Abnormal returns
First, the three-day (0,2) cumulative abnormal return (CAR) for the acquisition announcement are reported in table III. The mean returns are reported for the entire sample and for parts of the sample based on the mode of payment and ownership. Mode of payment is based on cash or noncash, where cash means a 100% cash payment. Further, the sample is split between high and low EBC, options and restricted stock firms. High is classified as above the median and low is below the median. Panel A shows statistical significant abnormal returns for the entire sample. The table further shows that the CAR of 0.31% is insignificant for firms with high EBC and with 1.61% is significant at the 5%-level for low EBC firms. This means that low EBC firms on average have cumulative abnormal returns of 1.61% around an acquisition announcement. The difference between the CAR’s of high and low EBC firms is with a t-statistic of -1.74 only weakly significant. Therefore, firms with lower EBC have higher abnormal returns than high EBC firms. Firms with high option based compensation only have a weakly significant CAR of 0.49%. Low option firms have an average CAR of 1.43% which is significant at the 5% level. However, the difference in CAR between high and low option firms is with a t-statistic of -1.26 not significant. Therefore there is no statistical difference in abnormal returns between high and low option firms. Additionally, high restricted stock firms have an insignificant CAR of 0.29%, while low restricted stock firms CAR of 1.28% is significant at the 5%-level. The t-statistic of the difference between high and low restricted stock is -1.68 and therefore only weakly significant. Although it is only weakly significant, it shows that high restricted stock firms have lower cumulative abnormal returns around the announcement than low restricted stock firms. This is consistent with the hypothesis that restricted stock compensation has a negative relation with acquisition announcement returns.
A.1. Method of Payment
Panel B shows the cumulative abnormal returns for the sample split by method of payment. The method of payment is based on whether the acquisitions are paid with 100% cash or not. For the full sample both cash (1.36%) and noncash (0.38%) CARs are weakly significant. First, for both cash and noncash the difference between high EBC CARs and low EBC
CAR’s are estimated. As seen in panel B, for both Cash and Noncash there is no significant difference in the CAR for high and low EBC. Next, for cash acquisitions with high option pay the CAR of 0.86% is significant at the 5% level. The CAR for low option pay is however not significant. With a t-statistic of -0.71 the difference in CAR between high and low option pay for cash acquisitions is not significant. Similarly, there is no statistical significant difference in the CAR between high and low option pay for noncash acquisitions. However, there is a statistical significant difference between high and low Restricted stock CARs for cash payments. This shows that for cash acquisitions the CAR is lower when the CEO has higher restricted stock pay. Additionally, panel B shows significant CARs for Non Cash acquisitions for high Restricted stock firms. The CAR difference between high and low Restricted stock firms for Non Cash acquisitions is however not significant.
A.2. Ownership
The level of CEO ownership is split in quartiles of ownership to examine whether or not it affects the incentive of equity based pay. Previous literature from Ofek & Yermack (2000) showed that the effectiveness of EBC is affected by the ownership. Their findings show that a higher level of ownership reduces the effectiveness of equity based pay. Panel C shows statistical significant difference between CARs of high and low EBC groups for the middle two quartiles (quartile 2&3). For both quartile 2 and 3, the cumulative abnormal return is statistically lower for high EBC groups. Therefore, high equity based compensation results in lower abnormal stock returns. Further, the difference in CAR between high and low restricted stock groups is significant for quartile 2 and weakly significant for quartile 3 & 4. Quartile 2 and 3 show that high restricted stock has lower CARs, but quartile 4 shows higher CARs for high restricted stock. Additionally, panel C shows no statistical difference in CARs for high and low option pay for any of the quartiles of ownership. Overall, the effect of difference between high and low EBC and Restricted Stock seems to be the strongest in the middle two quartiles. These results show that the abnormal returns for CEOs in the lowest quartile of ownership are not affected by difference in equity based compensation, newly granted restricted stock and newly granted stock options. The diminished effect of equity based compensation in the highest quartile of ownership is in line with the findings of Ofek & Yermack (2000).
Table III: Three-day (0,2) Cumulative Abnormal Returns
The three-day (0,2) cumulative abnormal returns are calculated for a sample of 2558 mergers and acquisitions using the market model. The estimation period is from 200- till 60 trading days before the acquisition announcement. All compensation data is from the fiscal year prior to the announcement. EBC is the equity based compensation consisting of new Options granted and Restricted stock granted. High refers to above the median value, and Low to below the median. T-stat difference is the t-statistic of the difference in the mean CARs between High and Low compensation. Ownership is the sum of the previous granted stock and restricted stock as fraction of the outstanding shares.
Panel A: CAR Categorized by Executive Compensation
Full sample High EBC Low EBC T-stat difference
Mean 0.009*** 0.003 0.016** -1.744*
[2558] [1279] [1279]
High Options Low Options T-stat difference
Mean 0.005* 0.014** -1.255
[1278] [1280]
High Restricted Low Restricted T-stat difference
Mean 0.003 0.013** -1.68*
[833] [1725]
Panel B: CAR Categorized by Method of Payment
Cash Non Cash
Full sample 0.014* 0.004* High EBC 0.004 0.002 Low EBC 0.024* 0.010 T-stat Difference -1.347 -1.14 High Options 0.009** 0.002 Low Options 0.019 0.011* T-stat Difference -0.712 -1.128 High Restricted -0.007 0.005** Low Restricted 0.019** 0.007 T-stat Difference -2.456** -0.247
Panel C: CAR Categorized by Ownership
Quartile 1 Quartile 2 Quartile 3 Quartile 4
Full sample 0.014 0.024*** 0.005 -0.002 High EBC -0.003 0.009 -0.001 0.004*** Low EBC 0.03 0.039*** 0.015*** -0.013 T-stat Difference -1.596 -2.214** -2.519** 0.561 High Options -0.002 0.014* 0.003 0.008 Low Options 0.033 0.035*** 0.008 -0.011 T-stat Difference -1.522 -1.505 -0.821 1.195 High Restricted -0.001 0.006*** -0.007 0.016*** Low Restricted 0.023 0.034*** 0.009*** -0.008 T-stat Difference -1.33 -2.581*** -1.74* 1.919*
B. Multivariate regression analysis
In this part the multivariate regression is used to estimate the systematic relation between equity based compensation and the stock performance around acquisition announcements. The main hypothesis is that restricted stock will have a negative effect on the stock returns. As previous literature has shown, restricted stock is valued negatively to stock options due to its illiquidity. Additionally, Bryan et al. (2000) show that CEOs with higher restricted stock compensation are less willing to pursue risky but value-enhancing deals. Therefore, lower abnormal returns are expected for CEOs with higher restricted stock compensation.
The results from column 1 in table IV show that the coefficient for equity based compensation (EBC) is negative and only weakly significant. A 1% increase in EBC results in a 0.035% decrease in cumulative abnormal returns. Further the coefficients for Size and Previous Options are negative and weakly significant. In column 2 the EBC is split in Restricted Stock and Options paid. The results of column 2 show that Restricted Stock is negative and significant, while Options paid is insignificant. The coefficient of -0.069 means that if Restricted Stock as fraction of total compensation increases with 1%, than the abnormal returns decrease with 0.069%. The negative coefficient for Restricted stock is in line with the hypothesis that higher restricted stock has a negative effect on the returns. It also is in line with the findings from Irving et al. (2011) which show that restricted stock is valued negatively by the stock market. Additionally, it is in line with Bryan et al. (2000) who show that because of restricted stock CEOs are less likely to take on risky but value enhancing deals. However, the insignificant coefficient for options paid is not in line with the prior literature from Datta et al. (2001) who find a significant positive coefficient. This might be caused by the decreased use of stock options in this sample compared to their sample. Further, in column 2 the variable for Size is weakly significant and the previous options granted are significant. Next, a variable for relative size is added to the regression. This is added since the relative size of the acquisitions should affect how large an impact the acquisition has on the firm. The addition of this coefficient gives very similar results in column 3 & 4 as in column 1 & 2. Restricted stock still has a significant negative effect on the CAR and the EBC is weakly significant. This shows that the weakly significant EBC coefficient is driven by the significant coefficient for Restricted Stock. The coefficient for the relative size of the acquisition is however not significant. Additionally, the Payment factor is insignificant in all the regressions and therefore does not affect the announcement returns. Lastly, the coefficient for Ownership negative in all those regressions. One possible
explanation for this is comes from Guay (1999), who shows that stock holdings, in contrast to option holdings, do not change the sensitivity of managers wealth to firm performance. The previous shown negative significant coefficients for Size are consistent with the literature from both Moeller et al. (2004) and Bajaj & Vijh (1995). These coefficients show that smaller firms have larger stock returns around acquisition announcements. All regressions are controlled for industry effects and time trends.
Model 5 is based on a different main variable. In this regression the main variable is Restricted EBC. Restricted EBC is a variable for the newly granted restricted stock as fraction of the equity based compensation. Equity based compensation is explained as the combination of newly granted options and newly granted restricted stock. Therefore this variable goes up if the newly restricted stock increases or the options decrease. Therefore an increase in this variable means that restricted stock at least increased more than options. The results in this column show that Restricted EBC is negative and significant. Column 5 shows a coefficient of -0.041. Therefore, a 1% increase in Restricted Stock as fraction of equity based compensation results in a 0.041% decrease in abnormal stock returns around the announcement. This shows that if restricted stock becomes a larger part of the equity based compensation the abnormal announcement returns decrease. Therefore the increase in use of restricted stock and the decrease of stock options as shown by Lord & Saito (2015) would result in lower acquisition announcement returns. Additionally, the coefficients for Size and Previous Options are significant and therefore good determinants of announcement stock returns. Similar to the previous regression the coefficients for Relative size, Payment and Ownership are insignificant.
Table IV: Multivariate regression around Three-day (0, 2) Cumulative Abnormal Return
The dependent variable is the three-day (0,2) cumulative abnormal return around the acquisition announcement, EBC is the natural logarithm of 1+ the total equity based compensation as fraction of the total compensation. Restricted Stock refers to the natural logarithm of 1+ newly granted restricted stock as fraction of the total compensation. Options is the natural logarithm of 1+ newly granted options as fraction of the total compensation. Restricted EBC is the natural logarithm of 1+ newly granted stock as fraction of equity based compensation. Size indicates the natural logarithm of the market value of the acquiring firm. Payment is a binary variable which takes the value 1 if the acquisition is paid with 100% cash. Previous Options refers to the natural logarithm of 1 + the previous granted options as fraction of the shares outstanding. Relative Size is the natural logarithm of 1 + transaction value as fraction of the market value of the acquiring firm. Industry refers to all the industry dummies based on the two-digit SIC code. Year is to capture time trend effects. The t-statistic is in parentheses.
Panel A: Multivariate Regression Analysis
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept 0.052 0.042 0.068 0.058 0.053 (0.59) (0.48) (0.77) (0.67) (0.62) EBC -0.035 -0.035 (-1.77)* (-1.80)* Restricted Stock -0.069 -0.070 (-2.18)** (-2.19)** Options -0.016 -0.017 (-0.78) (-0.82) Restricted EBC -0.041 (-2.00)** Size -0.006 -0.006 -0.006 -0.007 -0.007 (-1.75)* (-1.73)* (-1.83)* (-2.01)** (-2.30)** Payment 0.007 0.007 0.006 0.006 0.005 (0.84) (0.84) (0.68) (0.69) (0.64) Ownership -0.008 -0.008 -0.008 -0.008 -0.007 (-1.29) (-1.35) (-1.32) (-1.37) (-1.22) Previous Options -0.019 -0.022 -0.019 -0.022 -0.023 (-1.76)* (-2.04)** (-1.73)* (-1.97)** (-2.23)** Relative Size -0.043 -0.042 -0.038 (-1.52) (-1.51) (-1.41)
Industry Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes
R-squared 0.018 0.019 0.019 0.020 0.019
Observations 2293 2292 2293 2292 2381
Table V: Multivariate regression around Three-day (0,2) Cumulative Abnormal Returns for different sub-samples
The dependent variable is the three-day (0,2) cumulative abnormal return around the acquisition announcement, EBC is the natural logarithm of 1+ the total equity based compensation as fraction of the total compensation. Restricted Stock refers to the natural logarithm of 1+ newly granted restricted stock as fraction of the total compensation. Options is the natural logarithm of 1+ newly granted options as fraction of the total compensation. Size indicates the natural logarithm of the market value of the acquiring firm. Payment is a binary variable which takes the value 1 if the acquisition is paid with 100% cash. Previous Options refers to the natural logarithm of 1 + the previous granted options as fraction of the shares outstanding. Relative Size is the natural logarithm of 1 + transaction value as fraction of the market value of the acquiring firm. Industry refers to all the industry dummies based on the two-digit SIC code. Year is to capture time trend effects. The t-statistic is in parentheses. Sample is split based on quartiles of ownership and Firm Size.
Panel A: Multivariate Regression Split by Ownership
Quartile 1 Quartile 2 Quartile 3 Quartile 4
Intercept 0.107 0.179 0.147 0.084 -0.72 -1.03 (1.66)* -0.33 Restricted Stock -0.08 -0.085 -0.081 0.057 (-1.00) (-2.21)** (-2.46)** -0.6 Options -0.111 -0.09 -0.014 0.028 (-1.73)* (-0.25) (-0.80) -0.59 Size -0.014 -0.014 -0.002 0.002 (-1.54) (-2.24)** (-0.72) -0.28 Payment 0.032 -0.004 0.003 -0.014 -1.28 (-0.24) -0.45 (-0.62) Relative Size -0.035 -0.077 0.009 -0.053 (-0.40) (-1.68)* -0.35 (-0.80)
Industry Yes Yes Yes Yes
Year Yes Yes Yes Yes
R-squared 0.043 0.196 0.069 0.027
Observations 579 568 587 559
Panel B: Multivariate Regression Split by Firm Size
Small Medium Large
Intercept 0.055 0.129 -0.010 (0.37) (0.45) (-0.27) Restricted Stock -0.018 -0.150 -0.066 (-0.25) (-1.92)* (-3.54)*** Options 0.032 -0.041 -0.028 (0.81) (-0.75) (-2.29)** Payment -0.006 0.027 -0.007 (-0.36) (1.20) (-1.23) Ownership -0.001 -0.023 -0.003 (-0.04) (-1.51) (-0.59) Previous Options -0.019 -0.044 0.015 (-1.05) (1.64) (1.85)* Relative Size -0.018 -0.059 -0.036 (-0.41) (-0.91) (-1.55)
Industry Yes Yes Yes
Year Yes Yes Yes
R-squared 0.043 0.036 0.149
Observations 765 750 778
To test the robustness of the results the regressions are re-estimated based on different sub-samples. The sample will be split in quartiles of Ownership and on Firm Size as seen in table V. Panel A shows the multivariate regression segmented by the CEO ownership quartiles. With Quartile 1 being the lowest ownership and Quartile 4 being the highest ownership. The results show that in the highest quartile neither Restricted stock nor Options are significant. This is consistent with the findings of Ofek & Yermack (2000) that high ownership diminishes the effect of incentive based compensation on the CARs. Further, in the middle two quartiles Restricted stock has a significant negative effect on the cumulative abnormal returns. For CEOs in quartile 2 a 1% increase in Restricted Stock decreases the cumulative abnormal returns with 0.085%, for quartile 3 it results in an 0.081% decrease. Both coefficients are significant at the 5%-level. Additionally, the coefficient for Options paid is weakly significant for the lowest quartile of ownership and insignificant for all the other quartiles. The coefficient for Size is only significant in quartile 2. The Relative size is weakly significant in quartile 2, and the method of payment is insignificant for all quartiles of ownership.
Additionally, the regression is re-estimated based on sub-samples of Firm Size. The total sample is split in Small, Medium and Large firms based on market value. Panel B shows that Restricted Stock is a strongly significant negative coefficient for large firms. The result show that for large firms a 1% increase in restricted stock as fraction of total compensation decreases the abnormal returns with 0.066%. Additionally, Options is also a significant coefficient for large firms. A 1% increase in options as fraction of total compensation decreases the abnormal stock returns with 0.028%. For medium firms is Restricted Stock only weakly significant. However, for small firms neither Options nor Restricted Stock is significant. This can be explained by the findings from Baker & Hall (2004). Their results show that pay-for-performance sensitivity decreases in firm size. Therefore, the CEO is less dependent on firm performance for its compensation if the firm is large. Since their pay-for-performance sensitivity is low it is more affected by a change in their newly granted options and restricted stock. Therefore the returns are more affected by a change in newly granted options and restricted stock for CEOs of large firms.
6. Conclusion.
This thesis uses a sample of 2558 acquisitions made by U.S. firms in the period from 2001 to 2007. The results show weakly significant negative relation between equity based compensation and stock price reaction around acquisition announcements. When splitting equity based compensation in newly option grants and newly granted restricted stock, it becomes clear that this result is driven by the significant negative relation between newly granted restricted stock and the announcement stock return. Additionally, using restricted stock as fraction of equity based compensation, instead of total compensation, it also shows a significant negative relation. Therefore, when the increase in newly granted restricted stock is bigger than the increase in newly granted options it will result in lower announcement stock returns. Controlling for method of payment, firm size, ownership, previous granted options and relative size of the acquisition the results still hold. These results show that restricted stock does not incentivize CEOs to engage in value enhancing acquisitions. This is because of the illiquidity of restricted stock. This illiquidity decreases the sensitivity of their compensation to the stock performance of the firm. Therefore CEOs compensation is not affected much by the acquisition. Thus, Restricted Stock is not a good tool to align the incentives of CEOs and stockholders in the case of acquisitions.
These results add to the current discussion on the efficiency of equity based compensation. Equity based compensation is seen as a good way to align incentives between managers and stockholders. However, the results show that the use of restricted stock can harm shareholder wealth. Therefore, it contributes to the discussion on the effectiveness of restricted stock as incentive alignment tool. Additionally, it contributes to the effectiveness of mergers & acquisitions.
One implication of this research is that after splitting the sample in quartiles of ownership the results only hold for the middle two quartiles (quartiles 2 and 3). This shows that the results do not hold for CEOs in the quartile of highest ownership and the quartile of lowest ownership. The diminishing effect in the highest quartile is in line with the literature of Ofek & Yermack (2000), but the result in the lowest quartile is not. However, Guay (1999) showed that the stockholdings do not affect the sensitivity of managers’ wealth to firm performance. Therefore there might be an unknown problem in the dataset for the lowest quartile of ownership. Further, the split between small, medium and large firms is done based on percentiles of size in this sample. However, it might be more useful to use certain dollar
values as split-off between small & medium and medium & large. Certain firms which are considered small based on percentile might actually be considered medium size firms when looking at their value.
Another implication of this research is that the results are only for short term returns. CEOs with more restricted stock might be more concerned with the long term returns since their stock is restricted from selling in the short run. Therefore, the results could be different on the long run. Hence, additional research could examine the effect of restricted stock granted on the long run acquisition stock performance. Future research could also take into account how long the CEO is restricted from selling the stock. The restriction time affects the liquidity and consequently also the sensitivity of their compensation to the stock performance. If the stock is restricted for a longer period, than the CEO compensation is less sensitive to the short run stock performance. Therefore, CEOs whose stock is restricted for a longer period might not engage in short-run value enhancing deals since they are not affected by those deals.
Bibliography.
Amihud, Y., & Lev, B. (1981). Risk reduction as a managerial motive for conglomerate mergers. The bell journal of economics, 605-617.
Amihud, Y., Lev, B., & Travlos, N. G. (1990). Corporate control and the choice of
investment financing: The case of corporate acquisitions. The Journal of Finance, 45(2), 603-616.
Bajaj, M., & Vijh, A. M. (1995). Trading behavior and the unbiasedness of the market reaction to dividend announcements. The journal of Finance, 50(1), 255-279.
Bliss, R. T., & Rosen, R. J. (2001). CEO compensation and bank mergers. Journal of
Financial Economics, 61(1), 107-138.
Brown, S. J., & Weinstein, M. I. (1985). Derived factors in event studies. Journal of
Financial Economics, 14(3), 491-495.
Bryan, S., Hwang, L., & Lilien, S. (2000). CEO stock‐based compensation: An empirical analysis of incentive‐intensity, relative mix, and economic determinants. The Journal of
Business, 73(4), 661-693.
Datta, S., Iskandar‐Datta, M., & Raman, K. (2001). Executive compensation and corporate acquisition decisions. The Journal of Finance, 56(6), 2299-2336.
Eckbo, B. E., Giammarino, R. M., & Heinkel, R. L. (1990). Asymmetric information and the medium of exchange in takeovers: Theory and tests. The Review of Financial Studies, 3(4), 651-675.
Frydman, C., & Jenter, D. (2010). CEO compensation. Annu. Rev. Financ. Econ., 2(1), 75-102.
Guay, W. R. (1999). The sensitivity of CEO wealth to equity risk: an analysis of the magnitude and determinants. Journal of Financial Economics, 53(1), 43-71.
Harford, J., & Li, K. (2007). Decoupling CEO wealth and firm performance: The case of acquiring CEOs. The Journal of Finance, 62(2), 917-949.
Hodge, F. D., Rajgopal, S., & Shevlin, T. (2009). Do managers value stock options and restricted stock consistent with economic theory?. Contemporary Accounting Research, 26(3), 899-932.
Jensen, M. C., & Ruback, R. S. (1983). The market for corporate control: The scientific evidence. Journal of Financial economics, 11(1-4), 5-50.
Kahl, M., Liu, J., & Longstaff, F. A. (2003). Paper millionaires: How valuable is stock to a stockholder who is restricted from selling it?. Journal of Financial Economics, 67(3), 385-410.
Kusewitt, J. B. (1985). An exploratory study of strategic acquisition factors relating to performance. Strategic Management Journal, 6(2), 151-169.
Longstaff, F. A. (2001). Optimal portfolio choice and the valuation of illiquid securities. The
Review of Financial Studies, 14(2), 407-431.
Lord, R. A., & Saito, Y. (2015). Trends in CEO compensation and equity holdings for S&P 1500 firms: 1994-2007.
Mehran, H. (1995). Executive compensation structure, ownership, and firm performance.
Journal of financial economics, 38(2), 163-184.
Minnick, K., Unal, H., & Yang, L. (2011). Pay for Performance? CEO Compensation and Acquirer Returns in BHCs. Review of Financial Studies, 24(2).
Moeller, S. B., Schlingemann, F. P., & Stulz, R. M. (2004). Firm size and the gains from acquisitions. Journal of Financial Economics, 73(2), 201-228.
Ofek, E., & Yermack, D. (2000). Taking stock: Equity‐based compensation and the evolution of managerial ownership. The Journal of Finance, 55(3), 1367-1384.
Roll, R. (1986). The hubris hypothesis of corporate takeovers. Journal of business, 197-216. Shleifer, A., & Vishny, R. W. (1988). Value maximization and the acquisition process. The
Journal of Economic Perspectives, 2(1), 7-20.
Silber, W. L. (1991). Discounts on restricted stock: The impact of illiquidity on stock prices.
Travlos, N. G. (1987). Corporate takeover bids, methods of payment, and bidding firms' stock returns. The Journal of Finance, 42(4), 943-963.
