Changes in timing of stock option grants after the credit crisis
Bachelor thesis
Serge Schoberer 9761896
Faculty of Economics and Business Specialisation Finance and Organisation Thesis supervisor: dr. P.J.P.M. Versijp Final version July 14, 2015
1 1. Introduction
The last decades of the twentieth century CEO compensation has increased considerably (Murphy, 1999), until 2002 this was especially the case with stock-based compensation (Cohen, Dey and Lys, 2007). This is probably because stock-based compensation is considered to be a solution to the agency problem that firms are confronted with (Jensen and Meckling, 1976). Remuneration packages containing this type of compensation are meant to align the interests of the agent (CEO) and the principal (stockholders) by making the agent shareholders themselves. This view is called the optimal contracting view.
Another view on executive compensation comes from Bebchuck and Fried (2004), this managerial power view argues that CEO’s have influence on their own board of directors and are able to set pay in their own interest. This results in higher salaries that are less sensitive to firm performance. Over time CEO´s have become more able at timing awards to their advantage (Lie, 2005). By effective timing their stock options, they are able to earn personal benefits at the expense of the companies’ shareholders.
The purpose of this study is to investigate if there has been a change in this opportunistic behaviour by CEO’s around stock option grants, after the occurrence one of the most important economic events of the last decades, the financial crisis of 2008. This opportunistic behaviour can occur in three different ways, by timing of information disclosure around option grants, by timing of the option grants themselves and by backdating. In order to uncover these practises average cumulative abnormal returns around CEO stock option grants are examined in this paper. A decline in stock prices in the period before a stock option grand date and/or an increasing stock price after the grant date indicates opportunistic behaviour (Lie, 2005). This is because CEO´s prefer a stock price that is as low as possible on the date of the stock option grant, so that they can take full advantage of future rises in price of the stock.
As mentioned before, there are several ways in which timing can take place. One way is by using their influence to manipulate the timing of good and bad news (Chauvin and Shenoy, 2001). Bad news about a company will have a negative effect on the stock price whereas good news will increase the stock price. So an opportunistic acting CEO will disclose bad news prior to the option grant date and good news is revealed afterwards, resulting in abnormal low stock prices before and abnormal high prices after the grant date. This method can be used when option grants are scheduled.
2
When grants are unscheduled, CEO’s have to act in another way to achieve personal gain. Instead of manipulating the timing of good and bad news they can manipulate the date of the grant itself. By setting the grant date on a date when the stock price is low, for example before “good” earning announcements (spring loading) or when the CEO knows that at that moment the market is under valuating the company (Yermack, 1997). When this method is used abnormal high returns can be found in the period following the award.
A third method to influence their payoff is called backdating. Grant dates are set back in time to days on which the stock price is low (Heron and Lie, 2007). So negative abnormal returns are found before and positive abnormal returns are found after the grant date. In this way CEO’s are able to receive stock options at very favourable terms and earn maximal profits, because they have the benefit of hindsight. Backdating of stock options does not have to be illegal, it only becomes illegal when the shareholders of the company are misled by it.
As a reaction to a number of corporate and accounting scandals concerning some major U.S. companies like Enron, Tyco International and WorldCom the U.S. government introduced the Sarbanes- Oxley Act (SOX) in July 2002. The main reason to enact SOX was to reinforce the confidence of investors in capital markets. Disclosure requirements, supervision of financial reporting and penalties for misreporting were increased (Carter, Lynch and Zechman, 2007). To prevent executives from backdating SOX requires that companies have to report option grants within two business days following the grant. Ribstein (2005) argues that SOX may result only in significant costs for companies and may not be able to avoid future corporate scandals. As a result of increased liability on executives required by SOX Cohen et al. (2007) find changes in the structure of executive compensation. They find an increase in base salary as well as an increase in bonus compensation and a decrease of the value of option grants. Also the ratio of incentive compensation (the sum of bonus, options and restricted stocks) to fixed salary decreased. With overall compensation levels remaining equal this indicates that CEO’s were paid less incentive compensation in the period following SOX. This and the switch from option grants to more bonus awards indicates that firms shield executives from some of the risks imposed by SOX.
In 2006 the Securities and Exchange Commission (SEC) tightened up the reporting rules even more. To eliminate backdating companies were required to report the date of the grant, its fair value, the date of the compensation committee’s meeting and closing market price on the grant date. This new regulation made backdating almost impossible but did not end opportunistic behaviour by CEO´s around stock option grants. They shifted from backdating
3
unscheduled awards to timing of information disclosure around scheduled awards (Daines, McQueen and Schonlau, 2014).
In 2008 the United States were hit by the most severe financial crisis since the Great Depression of the 1930´s. This crisis led to a decrease of GDP, falling stock prices and higher unemployment rates. It also brought the issue of CEO compensation, a topic of great debate, more than ever in the centre of attention. Attention came not only from the press and the public but also from politicians. Politicians’ interest in regulating executive compensation is mainly driven by the contention that lack of oversight and poor incentives based on pay packages may have been an important factor in worsening this crisis (Keller and Stocker, 2008). As a result of this interest by politicians the Dodd-Frank Wall Street Reform and Consumer Protection Act (“Dodd-Frank Act”) was signed into law by President Obama on July 21 2010. The main goals of the Dodd-Frank Act are to decentralize the power of decision concerning executive pay and giving public shareholders information and power to influence corporate decision making (Masterson, 2010).
This study investigates, if there are abnormal stock returns around the grant dates of stock options awarded to CEO’s and if there are differences between the period before the beginning of the crisis (2005-2007), the year 2008 in which the crisis started (Ivashina and Scharfstein, 2010 and Subrahmanyam, 2009) and after (2009-2011). In addition, the period after the beginning of the crisis will be split in a period before and after the enactment of the Dodd-Frank Act on July 21, 2010, to find out if the act has been of influence on the abnormal stock returns.
The following hypotheses will be tested:
H1: There is a difference in cumulative average abnormal returns in the pre- and post-crisis period around CEO stock option grant dates. Where cumulative average abnormal returns in the period before the crisis are expected to be decreasing more in the days before the grant date and increasing more in the days after the grant than in the period after the crisis. When put in a graph, this would result in a steeper negative slope in the days before the grant and steeper positive slope in the days after the grant for the period before the crisis.
This hypothesis is expected to hold because of the fact that the proportion of stock options in the remuneration packages has been decreasing in the years following the beginning of the crisis (de Groot and Qin, 2011) it becomes less interesting for CEO’s to put effort in the timing of the grant dates. Also public outrage, media attention and political attention are
4
expected to have influence on opportunistic behaviour by CEO’s that leads to this V-shaped pattern.
H2: There is a difference between cumulative average abnormal returns pre- and post- SEC rules of December 2006 period. Where cumulative abnormal returns in the pre- SEC rules period are expected to be decreasing more in the days before the grant date and increasing more in the days after the grant than in the period after the new rules. When put in a graph, this would result in a steeper negative slope in the days before the grant and steeper positive slope in the days after the grant for the period before the introduction of the SEC rules.
H3:There is a difference between cumulative average abnormal returns in the pre- and post-Dodd-Frank period around CEO stock option grant dates. Where cumulative average abnormal returns in the pre- Dodd-Frank period are expected to be decreasing more in the days before the grant date and increasing more in the days after the grant than in the post-Dodd-Frank period. When put in a graph, this would result in a steeper negative slope in the days before the grant and steeper positive slope in the days after the grant for the period before the Dodd-Frank Act.
Both these hypotheses are expected to be true because these new rules and act make it more difficult to camouflage opportunistic behaviour. Also monitoring of CEO behaviour by shareholders is expected to be stricter after Dodd-Frank.
Recently there has been a number of studies that investigated the effect of the crisis on CEO remuneration. Many of them like Fahlenbach and Stulz (2011) and Vemala, Nguyen, Nguyen and Kommasani (2013) focus on changes in the way in which CEO’s are paid. This research contributes to prior literature on CEO compensation and CEO timing of stock options because it investigates if the financial crisis and the subsequent reaction of the public and politicians are of influence on CEO behaviour with respect to the timing of stock options and not only on changes in the way CEO´s are paid.
The remainder of this research will proceed as follows. In chapter 2 previous literature on this topic will be discussed. In chapter 3 sample and method will be presented. Chapter 4 will discuss the results. Finally chapter 5 concludes the paper and will contain a recommendation for future research.
5 2. Literature review
In this section previous literature on abnormal return patterns around stock option grants, backdating, regulation (SOX) and the financial crisis of 2008 will be discussed.
Timing of option grants and timing of information disclosure
There has been prior research on the topic of stock option timing by CEO´s. Yermack (1997), with his study of 619 stock option awards to CEO’s of largest U.S. firms, was one of the first to study this topic. He finds that companies that awarded stock options to their CEO’s outperform the market in a fifty day period after the reward date with more than 2%. With respect to the period preceding the grant date he finds no significant abnormal returns. He interprets this as proof that CEO’s have influence on the structure and timing of their compensation.
Chauvin and Shenoy (2001) find that for a sample of 783 stock option grants to CEO’s cumulative average abnormal returns during 10 days prior to the grants is -0.57%, meaning that there is a significant stock price decrease in the days before the grants. For the scheduled part of the sample they find average CAR’s of -0.79% indicating an even larger decrease than for the overall sample. Grants are assumed to be scheduled when they occur 365 days, plus or minus 5 days following the previous grant (Chauvin and Shenoy, 2001). They assume that these grants are made on annual compensation committee meetings and that CEO’s were not able to influence the timing of these grants. Because for this part of the sample it is not plausible to argue that CEO’s influence the dates of the grants, they state that these results are consistent with manipulation of the timing of information around the grant dates.
Aboody and Kasznik (2000) also examine the timing of information releases around stock option grants. In order to exclude the possibility of opportunistic timing of the award dates they only use scheduled awards for their research. Like Yermack (1997) they find significant positive abnormal returns in the period following the grant date but no significant negative abnormal returns in the preceding period. From these findings they conclude that CEO’s use their influence to opportunistically time the release of private information. Lie (2005) points out that, although both these studies state that there is enough evidence to conclude that CEO’s actively manipulate the timing of information disclosure, their results are somewhat conflicting. He explains this by the different sample periods used in the studies and by the way they compiled their samples.
6
Backdating
Lie (2005) finds that abnormal returns are negative before the grant dates and positive afterward. In his study these abnormal returns are significant around both scheduled and unscheduled stock option awards, but for unscheduled awards they are much more pronounced. Because he thinks that it is not plausible that executives are able to make exact forecasts about future market returns, he comes with a new hypothesis to explain the abnormal return patterns around award dates. His backdating hypothesis states that award dates are set retroactive. This means that CEO´s use hindsight to set the grand date to an earlier date on which the price was exceptionally low. The found return pattern, that shows a decline in the 30 day period before the grant date and an increase after the grant date, is consistent with this hypothesis. He also finds that the patterns around unscheduled grand dates get more pronounced over time, which could imply that CEO´s are gradually getting better in timing of the award dates. The reason the pattern is stronger for unscheduled awards than for scheduled awards also comes from the fact that backdating only makes sense for unscheduled awards. Around scheduled awards executives can only influence the timing of the release of either positive or negative information about the company to the public to profit from stock price movements (Lie, 2005).
Regulation (SOX)
The preceding studies are all about the period before the enactment of the Sarbanes-Oxley Act (SOX) in 2002. Following some large accounting scandals this act was installed to re-establish confidence in capital markets. Especially Section 403 of SOX that requires the acceleration of the disclosure of stock option grants to executives is of interest. Instead of reporting stock option grants to executives on a monthly basis within 10 calendar days of the following month or even on an annual basis within 45 days after the end of a company’s fiscal year when they met certain conditions (Collins, Gong and Li, 2005) which provided executives with an reporting lag that gave them opportunity for backdating, stock option grants now had to be reported within two business days after the grant date. This reduced the possibility of backdating significantly. In their research Collins et al. (2005) examine the effect of SOX on the timing manipulation of CEO stock option awards. They compare stock option grants to CEO´s in a pre-SOX period from January 1999 to July 29 2002 with those in a post-SOX period from August 29, 2002 to December 31, 2003. This research shows that SOX has had a deterring effect on timing of information and manipulating grant dates around good or bad news announcements, but above all it greatly reduces the backdating of stock
7
options. Heron and Lie (2007) also find that the enactment of SOX has had a strong decreasing effect on backdating. Just like Collins et al. (2005) they compare abnormal return patterns around grant dates in a pre-SOX period with a post-SOX period. They find that around 80% of the abnormal returns present in the pre-SOX period have disappeared in the post-SOX period. From this they conclude that backdating was the major reason of abnormal returns around grant dates before the enactment of SOX, but although backdating has been diminished (curbed) by its accelerated reporting requirements it has not been eliminated completely. To achieve this they suggest that reporting deadlines have to get even more rigid. Besides the accelerating reporting requirement of Section 403 of SOX the act also requires that CEO´s certify the correctness of the company´s financial statements and penalizes them when these statements do not comport with all the requirements set forth by SOX. This and other mandates (e.g. clawbacks of executive compensation for misconduct) in SOX impose additional liabilities on CEO´s. These extra risks for CEO´s could persuade firms to alter their compensation plans to protect them from it. Cohen, Dey and Lys (2007) investigated if there are changes in the way CEO’s are paid after the passage of SOX. They found that after SOX salary and bonus compensation increased whereas option-based compensation decreased. Another finding was that the ratio of incentive-based compensation to fixed salary decreased. From this they conclude that firms do shield their CEO´s from this extra risk imposed by SOX.
After the new reporting rules that the SEC adopted in 2006 researchers like Sen (2009), Narayanan and Seyhun (2008) concluded that these reporting changes together with increased public scrutiny and the Sarbanes-Oxley Act ended the problem of CEO manipulation of stock option grants. Opposing these conclusions Daines, McQueen and Schonlau (2014) found that instead of ending CEO option manipulation the new rules the problem only shifted from unscheduled grants to scheduled grants. For scheduled option grants they find a declining stock price before grant dates and an increasing price after the grant. Besides this they also find that the majority of companies moved from unscheduled to scheduled grants after the new regulations.
Financial crisis
After 2008, when the U.S. and in their slipstream many other countries were hit by the biggest financial crisis since the great depression of the 1930s, CEO compensation attracted more and more attention from many different sides like investors, the media, the public, regulators and
8
financial economists. Vemala, Nguyen, Nguyen and Kommasani (2013) tried to find out if firms have reacted to the persistent criticism and attention by the public. They concentrated their study on Fortune 500 firms to examine whether there has been a change in the level of CEO pay as well as in the way they are paid. They found that there was a significant decline in their cash compensation but also that this decline was more than offset by an increase in their long-term compensation, so that total compensation in the post-crisis period exceeded the pre-crisis period. De Groot and Qin (2011) in a comparable research on changes in CEO compensation during the crisis which they focused on the financial sector did find decreasing total compensation during the crisis. They saw a rise in level of base salary and a fall of the variable component of CEO compensation (i.e., bonuses, stocks and stock options). Especially the decrease of stock and stock option pay are the cause of a lower total compensation during the crisis.
3. Sample & Method
To investigate the abnormal returns around CEO´s stock option grant dates, a sample of stock option grants was collected. The sample used for this research consists stock option grants to CEO´s of companies that made part of the Fortune 500 list in the years from 2005 to 2011. This period is chosen because it runs from three years before the start of the financial crisis of 2008 till three years after. The Fortune 500 list consists of the 500 largest U.S. based companies measured by total revenues.
Fortune 500 firms were chosen because large companies have more visibility and scrutiny (Heron and Lie, 2007) and therefore their CEO´s are expected to be more sensitive for public outrage and legislation. Another reason for choosing Fortune 500 firms instead of other large international firms is that it is easier to obtain the necessary data for US companies. Only companies that made part of the list in all consecutive research years were selected. Companies that were not present in the list in all seven years from 2005 onward to 2011 and companies that do not have publicly traded stock were deleted. Eventually 324 firms did meet these criteria and were used in this study.
The necessary data was extracted from the Thomson and Reuters database for the executive stock option grant dates and the Center for Research in Security Prices (CRSP) database for the daily stock return data. Both these databases are accessible through Wharton Research Data Services (WRDS). In the period from 2005 to 2011 there were in total 293,296 awards, because this research is about stock option awards to CEO’s, only awards to CEO’s and with
9
code A (Grant or award transaction pursuant to Rule 16b-3(c)) were used. Further, only transactions with derivative types that constitute options were included and all duplicates were excluded. This resulted in a total of 1584 option grants to CEO’s in the sample period that are used in this research.
To determine the abnormal returns around the grant dates an event-study is performed. The event date used is the day on which the stock options were granted. The stock returns in a period 30 days prior to 30 days after the grant date were examined, so the event window is a period of 61 days. For the estimation period this research follows Lie (2005) so it consists of a 250 day period starting 300 trading days before and ending 50 days before the event date. The definition of an abnormal return is the actual return minus a normal or benchmark return.
= - (1)
To determine the normal return, in this research the Ordinary Least Squares (OLS) method is used to estimate the market model. The expected stock return for the estimation period of 250 days is estimated by the subsequent equation:
= + + (2)
Here is the expected return on i at time t, is the CRSP value-weighted stock market return at time t and and are the OLS estimated parameters.
Using equation 1, abnormal returns are:
A = - = – ( + ) (3)
And cumulative abnormal returns (CAR’s) for all stocks in the event windows starting at and ending at are:
= A
(4)
Because movements in stock prices can be caused by information or events that are not related with the studied event, it is more informative to investigate the abnormal returns and cumulative abnormal returns for all option grants in the sample as a whole than for each separate option grant. These average abnormal returns and cumulative average abnormal returns are obtained by the following equations:
10
= (6)
Besides the main event window of -30 to 30 other event window intervals will be investigated. These other intervals are (-30,0), (-30,-11), (-10,-6), (-5,0), (1,5), (6,10), (11,30) and (1,30).
To test for significance of the different CAAR intervals standard t-tests have been conducted. The formula for this test is:
=
4. Results
This chapter will discuss the results of the research that has been described in the previous chapter. First, the results of the periods before, during and after 2008 (the year that according to Subrahmanyam (2009) the crisis started) will be discussed. Second, the period before and after the instalment of the stricter SEC reporting rules of 2006 will be discussed. Third, attention will be paid to the results of the split period before and after the Dodd-Frank Act of 2010 came into force. Finally, the results of the different periods will be compared with each other to come to a conclusion.
Figure 1 -0,025 -0,02 -0,015 -0,01 -0,005 0 0,005 0,01 0,015 0,02 -30 -28 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 CA AR
Days relative to grant date
CAAR's around option grant dates
11 Table 1
Before 2008
CAAR period N Mean St Dev t-value [-30, +30] 738 0,0010914 0,117437 0,2525 [-30-, -1] 738 -0,0031314 0,083099 -1,0237 [-30, -10] 738 -0,0049528 0,060135 -2,2374* [-10, -5] 738 0,0018878 0,032219 1,5918 [-5, -1] 738 -0,0000664 0,036124 -0,0499 [0, +6] 738 0,0021061 0,035529 1,6104 [+6, +11] 738 0,0013305 0,026555 1,3611 [+11, +30] 738 0,0007862 0,058732 0,3637 [0 , +30] 738 0,0042228 0,074915 1,5313 * Denotes two-tail statistical significance at 5%
Table 2
After 2008
CAAR periode N Mean St Dev t-value [-30, +30] 571 -0,0157206 0,18548 -2,0253* [-30-, -1] 571 -0,0123199 0,13396 -2,1976* [-30, -10] 571 -0,0118154 0,111223 -2,5385* [-10, -5] 571 -0,0009137 0,051218 -0,4263 [-5, -1] 571 0,0004092 0,051108 0,1913 [0, +6] 571 -0,0045184 0,059417 -1,8171 [+6, +11] 571 0,0023908 0,044754 1,2765 [+11, +30] 571 -0,001273 0,107502 -0,283 [0 , +30] 571 -0,0034007 0,115525 -0,7034 * Denotes two-tail statistical significance at 5%
Figure 1 shows the average cumulative abnormal returns (CAAR´s) over a period from 30 trading days before to 30 days after the option grant dates for the periods before 2008, after 2008 and 2008. Abnormal returns are defined as the difference between stock returns of the selected firms and a benchmark return or normal return (de Jong, 2007). To calculate the normal return the market model is used. The graph shows that for the periods before and after 2008 CAAR’s are negative in the intervals preceding the grant date. The period from -30 to -10 days before the grant date shows the most significant negative results. For both periods before and after 2008 this period is significant negative at the 5% level (see table 1 and 2). For
12
days following the grant date CAAR’s are all positive before 2008, after 2008 only days 6 to 11 show some positive return. The fact that CAAR’s are negative before the grant date and positive after could indicate opportunistic behaviour by CEO’s in the pre-crisis period. The post-crisis period gives negative CAAR’s for the days before as well as for the days after the grant date. For 2008 the CAAR’s are slightly positive for the interval before the grant date and positive for the interval after the grant date, where days 0 to 6 show the most significant positive CAAR’s (see table 3).
Table 3
2008
CAAR periode N Mean St Dev t-value [-30, +30] 204 0,0075975 0,184569 0,5879 [-30-, -1] 204 0,0015244 0,135887 0,1602 [-30, -10] 204 0,0004877 0,104678 0,0665 [-10, -5] 204 0,0018536 0,064637 0,4096 [-5, -1] 204 -0,0008168 0,065507 -0,1781 [0, +6] 204 0,0089767 0,056603 2,2651* [+6, +11] 204 0,0042018 0,050074 1,1985 [+11, +30] 204 -0,0071054 0,105524 -0,9617 [0 , +30] 204 0,0060731 0,117044 0,7411 * Denotes two-tail statistical significance at 5%
To find out if the difference between the results in the period before and after 2008 has something to do with the stricter SEC rules of December 2006 the period before 2008 is split in a period before and after the new rules. The graphs of both these periods are displayed in figure 2.
13 Table 4
Before Dec.15, 2006
CAAR periode N Mean St Dev t-value [-30, +30] 488 -0,0031046 0,119725 -0,5728 [-30-, -1] 488 -0,0070701 0,085498 -1,8267 [-30, -10] 488 -0,0080316 0,063053 -2,8139* [-10, -5] 488 0,0016004 0,030082 1,1752 [-5, -1] 488 -0,0006389 0,037419 -0,3772 [0, +6] 488 0,0003641 0,037749 0,2131 [+6, +11] 488 0,0017854 0,02615 1,5083 [+11, +30] 488 0,001816 0,059408 0,6753 [0 , +30] 488 0,0039655 0,075909 1,154 * Denotes two-tail statistical significance at 5%
Table 5
After Dec. 15, 2006 and before 2008 CAAR periode N Mean St Dev t-value [-30, +30] 249 0,0094893 0,112568 1,3302 [-30-, -1] 249 0,0046794 0,077768 0,9495 [-30, -10] 249 0,0011725 0,053561 0,3454 [-10, -5] 249 0,0024187 0,036101 1,0572 [-5, -1] 249 0,0010882 0,033501 0,5126 [0, +6] 249 0,0055464 0,030517 2,868* [+6, +11] 249 0,0004761 0,027368 0,2745 [+11, +30] 249 -0,0012125 0,057456 -0,333 [0 , +30] 249 0,0048099 0,07307 1,0387 * Denotes two-tail statistical significance at 5%
-0,015 -0,01 -0,005 0 0,005 0,01 0,015 -30 -28 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 CA AR
Days relative to the grant date
CAAR's around option grant dates
14
Before December 15, 2006 the CAAR is significantly negative (see table 4) for the days -30 to -10 period before the grant, where after this date this period only gives a non-significant positive CAAR. For the days after the grant both periods have a positive CAAR for the 30 day period. The only significant positive CAAR is found in the period after December 2006 from days 0 to 6 (see table 5). Because the shape of the graph in the period after December 2006 is more in line with the complete pre-crisis period than with the period after 2008, the absence of the stricter SEC rules before December 15, 2006 seem not to be of influence on the difference between the period before 2008 and the period after.
On July 21, 2010 the Dodd-Frank Act was signed into law to give public shareholders more power to influence executive pay and corporate decision making (Masterson, 2010). To check if this act has been of influence on the difference between CAAR results of the period before and after 2008, the period after 2008 is split into a period before and a period after the act. Figure 3 -0,05 -0,04 -0,03 -0,02 -0,01 0 0,01 0,02 0,03 0,04 -30 -28 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 CA AR
Days relative to the grant date
CAAR's around option grant dates
15 Table 6
After 2008 and before July 21, 2010 CAAR periode N Mean St Dev t-value [-30, +30] 355 -0,0415173 0,212943 -3,6735* [-30-, -1] 355 -0,0289733 0,153794 -3,5495* [-30, -10] 355 -0,0230631 0,127876 -3,3982* [-10, -5] 355 -0,0014211 0,058379 -0,4587 [-5, -1] 355 -0,0044892 0,057237 -1,4778 [0, +6] 355 -0,0078379 0,071643 -2,0613* [+6, +11] 355 0,0020873 0,051178 0,7685 [+11, +30] 355 -0,0067934 0,129651 -0,9872 [0 , +30] 355 -0,012544 0,13436 -1,7591 * Denotes two-tail statistical significance at 5%
table 7
After July 21, 2010
CAAR periode N Mean St Dev t-value [-30, +30] 215 0,0263006 0,116457 3,3115* [-30-, -1] 215 0,0138032 0,084026 2,4087* [-30, -10] 215 0,0056602 0,071299 1,164 [-10, -5] 215 0,0000287 0,036577 0,0115 [-5, -1] 215 0,0081143 0,037405 3,1808* [0, +6] 215 0,0011466 0,029079 0,5782 [+6, +11] 215 0,0035656 0,02995 1,7457 [+11, +30] 215 0,0077853 0,052973 2,1549* [0 , +30] 215 0,0124974 0,07148 2,5636* * Denotes two-tail statistical significance at 5%
In figure 3 the period after 2008 is split into two different graphs, one for the period before the Dodd-Frank Act and one for the period after. In the period before the act the graph is almost constant downward sloping, where the graph is upward sloping for the period after the act. Most intervals in both periods are significant. The only interval that gives an opposing result is the days 6 to 11 interval for the period before the Act. For this interval the CAAR is very small and not significant (see table 6 and 7). Therefore the Dodd-Frank Act seems to have had a positive influence on stock returns of the selected companies. However, although the
16
graphs of both periods run in opposite directions there is no reason to assume that there has been a change in opportunistic behaviour by CEO’s around stock option grant dates as a result of the Dodd-Frank Act. This is because in both periods there is no change in the direction of the graphs around the grant dates that could indicate this type of behaviour.
5. Conclusion and recommendation
The results show that there are less positive abnormal returns in the days following the grant date in the years after the crisis year 2008 than in the years before. The CAAR is increasing from day 0 to day 30 in the period before and decreasing in the period after 2008. For the days preceding the grant date the graphs and tables show decreasing CAAR’s for both periods. The decrease in the period after 2008 is more pronounced than in the period before. This gives the period before 2008 more typical V-shaped pattern belonging to opportunistic behaviour of CEO’s (Lie, 2005). The differences between these two patterns could be caused by the absence of the stricter SEC rules in a large part of the pre-crisis period. However no evidence is found for this assumption. In fact the part after the instalment of the rules show more differences with the post crisis period than the earlier part. Compared with the previously mentioned periods 2008 displays the most noticeable CAAR increase after the grant date. This and the fact that the CAAR for days -30 to 0 is very small could also indicate opportunistic behaviour.
The Dodd-Frank Act of 2010 could be another reason for the observed differences between the periods. To check whether this is case, the period after 2008 is split into two separate parts that can be compared with each other. The two graphs move in opposite directions but both have no characteristic shape that indicates opportunistic behaviour around stock option grant dates, so nothing can be said about the effect of the Dodd-Frank Act on this type of behaviour. Hypotheses’ 2 and 3 both do not hold, and thus the new SEC rules nor the Dodd-Frank ACT seem to have significant influence on the observed results in both the pre- and post-crisis period. Hypothesis 1 however seems at least partly correct, because the period before the crisis shows the most distinctive shape belonging to opportunistic behaviour by CEO’s. So this research gives some small evidence that the financial crisis of 2008 and its consequences like public outrage and media attention, has been of influence on CEO behaviour. To get more conclusive results further research should be done. This research should make a distinction between scheduled and unscheduled grants, because the SEC rules, the Dodd-Frank Act and other changes in legislation have different effects on both types of option grants.
17 Reference list
Aboody, D., & Kaznik, R. (2000). CEO stock option awards and the timing of corporate voluntary disclosures. Journal of Accounting and Economics, 29, 73-100
Bebchuck, L., & Fried, J. (2004). Pay without performance: the unfulfilled promise of executive compensation. Harvard University Press.
Carter, M.E., Lynch, L.J., & Zechman, L.C. (2007). Changes in bonus contracts in the post-Sarbanes-Oxley era. Working Paper, University of Pennsylvania and University of Virginia. Chauvin, K.W. , & Shenoy, C. (2001). Stock price decreases prior to executive stock option grants. Journal of Corporate Finance, 7, 53-76.
Cohen, D.A., Dey, A., & Lys, T.Z. (2007). The Sarbanes Oxley Act of 2002: implications for compensation contracts and managerial risk-taking. Working Paper, New York University. Collins, D.W., Gong, G., & Li, H. (2005). The effect of the Sarbanes-Oxley Act on the timing manipulation of CEO stock option awards. Working paper, University of Iowa.
Daines, R.M., McQueen, G.R., & Schonlau, R.J. (2014). Right on schedule: CEO option grants and opportunism. Available at SSRN: http://dx.doi.org/10.2139/ssrn.2363148.
De Groot, A. and B. Qin. 2011. The changes of CEO compensation in the US financial sector during the credit crisis. Management Control and Accounting (MCA), 2011(2), 34-39.
De Jong, F. (2007). Event Studies Methodology. Lecture notes written for the course Empirical Finance and Investment Cases. Unpublished, Tilburg University.
Fahlenbrach, R. & Stulz, R.M. (2011). Bank CEO incentives and the credit crisis. Journal of Financial Economics, 99(1), 11-26.
Heron, R.A., & Lie, E. (2007). Does backdating explain the stock price pattern around executive stock option grants? Journal of Financial Economics, 83, 271-295.
Ivashina, V. & Scharfstein, D. (2010). Bank lending during the financial crisis of 2008. Journal of Financial Economics, 97(3), 319-338.
Jensen, M.C., & Meckling, W.H. (1976). Theory of the firm: managerial behaviour, agency cost and ownership structure. Journal of Finance 3, 305-360.
18
Keller, C, & Stocker, M. (2008). Executive compensation’s role in the financial crisis. The National Law Journal, Nov. 18, accessed March 20, 2010. Available at http://www.law.com/jsp/ihc/PubArticleIHC.jsp?id=1202426091714.
Lie, E. (2005). On the timing of CEO stock option awards. Management Science, 51(5), 802-812.
Masterson, J. (2010). Dodd-Frank Act ‘could be a home run’. BIZTIMES (Sep. 3, 2010). Available at http://www.biztimes.com/news/2010/9/3/dodd-frank-act-could-be-a-home-run. Murphy, K.J. (1999). Chapter 38 – Executive compensation. Handbook of Labor Economics, volume 3 (pp. 2485-2563). Elsevier.
Narayanan, M., & Seyhun, H.N. (2008). The dating game: do managers designate option grant dates to increase their compensation? The Review of Financial Studies, 21, 1907-1945. Ribstein, L. (2005). Sizing up SOX. Working paper, University of Illinois.
Sen, R. (2008). The returns to spring-loading. EFA 2008 Athens meetings paper, Hong Kong University of Science and Technology.
Subrahmanyam, G. (2009). Global financial crisis of 2008-2009: triggers, trails, travails and treatments. The IUP journal of applied economics, 8(5-6), 32-47.
Vemala, P., Nguyen, J., Nguyen, D., & Kommasani, A. (2014). CEO compensation: does financial crisis matter? International Business Research, 7(4), 125-131.
Yermack, D. (1997). Good timing: CEO stock option awards and company news announcements. The Journal of Finance, 52(2), 449-476.
