The Impact of Stock Option Plans on the Decision to Repurchase
Shares: Evidence from US over the 2003-2007 period
Elena Bulubas1 Master Thesis Finance Supervisor: Prof. Dr. R.E. Wessels
Abstract
This paper examines the impact of stock option plans on the decision to repurchase shares using data on 1231 US firms over the 2003-2007 period. We find that both executive and employee stock options, are positively related to share buybacks. Thus, the firms which allocate stock options as part of their compensation plan are more likely to repurchase shares, after controlling for a series of firm characteristics. This result could be explained by the desire of a company to avoid the dilution effect associated with the exercise of employee stock options. The obtained results support the hypothesis that the growth in stock options could be a partial explanation for the drastic increase in the volume of stock buybacks in the recent decades.
JEL classification: G30, G35
Keywords: stock buybacks, executive stock options, employee stock options
2
1. Introduction
Before the ‘80s share repurchases were heavily regulated, but since then the process has been liberalized and the volume of stock buybacks on the world market has drastically increased. For instance, the volume of share repurchases on the US market rose from US$5 billion in 1980 to US$349 billion in 2005, and nearly US$600 billion of buybacks are forecasted for 2014 (Financial Times, 2013; Business Insider, 2014). Grullon and Michaely (2002) state that in 1999 and 2000 the volume of open market repurchases exceeded the volume of dividend payments for the first time.
There are several reasons why a firm may repurchase shares. The first reason is the distribution of excess capital. Stock buybacks are a form of dividend payments to shareholders (Bhargava, 2010). Share buybacks have several advantages over dividends. First, share repurchases are more flexible, as there are no expectations from shareholders or market that they will occur on a regular basis as the dividend payments do (Dittmar, 2000). Furthermore, the stock repurchases are advantageous for shareholders that seek better tax treatment than dividends, as in general the tax rate for capital gains is lower than the income tax (Quiry, Le Fur, Salvi, Dallochio and Vernimmen, 2011).
The second reason is the undervaluation hypothesis also called the signaling theory. It states that the executives, by repurchasing shares, signal to the market their belief in the firm’s future positive earnings (Kahle, 2002). Therefore, the market perceives the buybacks as a sign that the firm’s stock is undervalued (Dittmar, 2000). Several studies show that the announcements of share repurchases are followed by abnormal positive returns (see Vermaelen, 1981; Comment and Jarrell, 1991; Stephens and Weisbach, 1998; Grullon and Michaely, 2004).2
Another reason that could explain share buybacks is the optimal leverage hypothesis, which claims that the firm can use the stock buybacks to mechanically increase the leverage ratio until it reaches a certain target or optimal level (Dittmar, 2000; Benton and Doowoo, 2001). When the excess capital is distributed to shareholders the equity falls, while the debt stays constant, which thus increases the leverage ratio. According to Grullon and Michaely (2004), the firms that carry out share repurchase in order to attain an optimal leverage, manage to reduce the cost of capital.
2 The papers find that on average the repurchase announcements are followed by a significant abnormal price
3 The last motive would be the stock option plans. An employee stock option is a contract between the company and its employees, which gives the right to employees to buy firm’s shares at the exercise price at the maturity (Hite and Long, 1982). The capital gains realized are computed roughly as the difference between the market stock price at the time of exercise and the strike price stated in the contract and adjusted for the tax rate specified by law. The higher is the market stock price from the strike price, the higher is the spread and thus the capital gains for the option holder (Summa, 2015). According to Hall and Murphy (2003), the companies generally use the cashless settlement in which the option holders will pay nothing for the stock options, but will receive the option-spread in either cash or firm stock. The stock options prior to their exercise have a potential dilutive effect, because to provide the shares underlying the stock option plan, the company usually will issue new stock or draw on a reserve of shares. As a result, the firm increases the number of shares in circulation, which leads to a decrease or dilution of the value of shares already in circulation and earnings-per-share (EPS) (Olagues and Summa, 2010). In contrast buying back shares on the market in order to provide the required number of shares, increases the EPS. The companies could use the share repurchase as a mechanism to offset the earnings-per-share dilution brought by the exercise of stock options and avoid the potential stock price depression brought by missing the analyst’s EPS forecasts (Bens, Nagar, Skinner, and Wong, 2003; Lazonick, 2014). A number of academic papers identify stock options as one of the main reasons for the surge in the volume of buybacks in the last decades, and particularly on this hypothesis will focus our paper.
4 shares and 37% were paid out as dividends. That leaves a small share of 9% for investment and growth projects. Lawrence Fink, CEO and chairman of BlackRock, highlighted this idea in his open letter to Corporate America. He states “too many companies have cut capital expenditure and even increased debt to boost dividends and increase share buybacks”.
Another drawback of the stock buybacks is the probability of “misuse” of the buybacks by management, as they could be tempted to manage the EPS through the repurchases in order to meet the quarterly analysts’ forecast (Kahle, 2002; Bens et al., 2003; Hribar, Jenkisn, and Johnson, 2006).
Most of the research has used event studies to determine if the share repurchases announcements lead to abnormal returns (Grullon and Michaely, 2002; Akyol and Foo, 2011). A small number of studies though have analyzed the impact of the stock-based compensations on the decision to repurchase shares. Our paper adds to this literature and relies on the agency theory to find possible explanations for the increase in share buybacks.
Our findings are as follows. First, both executive stock options and employee stock options are positively related to the decision to repurchase shares, thus, the use of these incentive tools increases the odds of the company carrying out stock buybacks. We also provide evidence that the decision to repurchase stock is significantly related to other firm characteristics. Consistent with the signaling theory, the probability of share repurchases increases for firms with poor recent stock performance. We find evidence supporting the free cash flow hypotheses as well. Thus, firms with high levels of free cash flows are more likely to buy back stock. Our findings also show a negative relation between the buybacks and leverage and a positive one with the firm size. However, we find no evidence that share repurchases crowd-out the investments as there is no significant relation between the share buybacks and capital expenditures. An interesting result is the relation between the buybacks and dividends. Even though some academic papers theorize about a possible substitution between these two methods of payout (e.g. Grullon and Michaely, 2002), our findings suggest the two methods are possible complements as the contemporaneous dividend payments increase the chance of share buybacks.
5
2. Previous research and hypotheses development
Kahle (2002) mentions that starting with ’90s the volume of share buybacks drastically increased, but at the same time there was no decrease in the shares in circulation. One possible explanation could be the increase in stock option plans reported in the paper.
As employees exercise their stock options, the firms can provide the required shares by either issuing new shares or repurchasing stock (Bens et al., 2002). The first method is not favored by the companies because it leads to dilution of share value which is why empirical research suggests that firms choose to buyback in order to offset the dilutive effect of stock options exercise (Bens et al., 2002; Kahle, 2002).
2.1. Literature review
Several papers research the relationship between stock options and share repurchases. The obtained results are mixed. Kahle (2002) researches the impact of stock options on the decision to repurchase shares. The study finds that the companies announce buybacks when the managers have a high number of stock options outstanding and the rest of employees hold a significant number of exercisable stock options. Moreover, the decision to repurchase shares is positively and significantly related to the total number of exercisable stock options, but it is not related to the options awarded to executives. Another interesting result provided by Kahle (2002) is that the market seems to discount repurchases that seek to offset the dilutive effect of stock options and reacts less positively to their announcement.
Dittmar (2000) researches the share buybacks on US market over the 1977-1996 period and asks the general question: Why firms repurchase shares? The paper looks at the general theories behind the share repurchases, namely excess capital hypothesis, signaling theory, optimal leverage ratio hypothesis, and determines the probability of share buybacks. The paper finds that the firms repurchase shares when they are undervalued, have excess capital available for distribution, need to increase the net leverage and handle the dilutive effects of stock option plans. Dittmar claims that the use of stock options may increase the number of firms that choose share repurchases over the dividend distributions, as the later lead to the dilution of per-share value of the stock.
6 compensation plan is directly related to the increased stock repurchases. Weisbenner (2002) does not find a relationship between the executive stock options and share repurchases, but he does find that the higher the number of executive stock options, the higher is the probability that the firm will retain earnings instead of distributing them to shareholders.
The study by Babenko (2009) finds that managers are more likely to carry out opportunistic share repurchases if the company awards many employee stock options. According to Babenko (2009), the announcement abnormal returns are positively related to the size of the repurchase program. Also the paper finds the presence of blockholders and awarding of a significant number of employee stock options push the firms to choose the stock repurchases over the dividend payments.
Feng and Liang (2001) study how the payout policy is affected by the managerial stock incentives. They find that the companies with potential agency problems, defined as firms with low insider stock ownership and high free cash flow, show a positive association between the executive stock options and the size of payouts. Moreover, the paper provides evidence of a negative correlation between executive stock options and dividends; and of a positive relation with the stock repurchases.
Bartov, Krinsky and Lee (1998) study how firms choose between dividends and open-market share repurchases. They analyze 3 factors that may stay behind the decision, namely the equity undervaluation, the stock-option based compensation plans and the extent of holdings by institutional investors. They claim that the managers who hold a significant package of stock options may choose the share repurchases over the dividend distribution, as unlike buybacks that do not affect the stock option value, the dividend payments decrease their value. Bartov et al. (1998) find that the higher is the number of executive option grants, the higher is the likelihood that the firm will carry out stock repurchases instead of dividend payouts.
7 stock option compensation plans. The study concludes that the firm’s compensation policy is an important incentive for the announcement of stock repurchases.
El Houcine and Boubaker (2014) research the share repurchases on a sample of 77 French firms during the 2003-2008 period. The paper finds a positive relation between the expected future exercises of stock options and the stock buybacks. Moreover, the study provides evidence that a significant number of executive stock options could encourage the firm to carry out share repurchases. At the same time the study shows a negative relation between the use of employee stock options and the increase of dividends.
2.2. Hypotheses
Kahle (2002) and Jolls (1998) make a distinction between executive and employee stock options in their research of share repurchases. Kahle (2002) states that the decision to buy back shares is motivated by two different stock option related theories. The first theory is called the option funding hypothesis. Kahle claims that as the exercise of employee stock options increases, the firm resort to repurchases in order to have shares available for distribution. The underlying reason for this move is to offset the dilutive effect that stock option exercises have on the share value and EPS. In general, potential investors view the stock buybacks as a positive event, a signal that the company’s insiders think that the firm is undervalued and that it will outperform in the future period (Comment and Jarrell, 1991; Benton and Doowoo, 2001; Grullon and Michaely, 2004;). However, Kahle (2002) shows that the market reacts less positively to the stock buyback announcements if the reasons behind the buybacks activity are less clear and there is a large number of stock options outstanding.
8 shares. Moreover, the magnitude of the repurchase is associated with EPS dilution by stock options, namely two-thirds of the executives consider the dilutive effect of stock options a very important factor affecting the decision to repurchase shares.
The above considerations lead to the following hypothesis, which is stated in its alternative form:
H1: Firms with high number of vested employee stock options have a higher probability of doing stock repurchases, after controlling for firm characteristics.
The second theory proposed by Kahle (2002) is called the substitution hypothesis. Jolls (1998) looks at the share repurchases as a “byproduct” of the agency problem. She states that the current rise in the level of the buybacks is conditioned by the incentives granted to the agents who run the firm. According to Hall and Murphy (2003), the stock option plans seek to mitigate the agency problems between the agent and the shareholders by tying the compensation to the stock price. However, it seems that the stock option plans encourage the executives to pursue value maximization, but also encourage managers to choose stock buybacks over the dividend payments, as the repurchases have the advantage of not diluting the per-share value of the stock (Jolls, 1998; Kahle, 2002). In contrast, the dividend payments decrease the value of a share of stock and thus decrease the wealth of the stock option holders (Lambert, Lanen, and Larker, 1989; Murphy, 1998). Moreover, Murphy (1998) finds that only 2% of 618 companies that use stock option plans have protection against dividend payments. Therefore, the executive stock options offer additional incentives to repurchase shares which are positively linked to the decision to buy back stock (see Jolls, 1998; Fenn and Liang, 2001; Kahle, 2002). Moreover, we expect that the granting of stock options is negatively related to dividend payments. Lambert et al. (1989) provide evidence that the decrease in the dividend payouts is more pronounced for firms that grant a significant number of executive stock options as part of their compensation plan.
Based on the above arguments we test the following hypothesis (stated in the alternative form):
9
3. Data and methodology
3.1. Data sources and sample selection
To investigate the impact of stock options on the share repurchase decision, we use the data provided by Compustat, Center for Research in Security Prices (CRSP) and ExecuComp. Compustat offers data regarding the stock buybacks, employee stock options and firm financial characteristics. Execucomp provides data regarding the executive stock options and from CRSP we get the data about the stock returns. As a result of merging the data from the 3 databases mentioned above, we get a sample of 1231 US firms listed on NASDAQ and NYSE over the 2003-2007 period, which provides us with 5639 firm-year observations. Based on the SIC codes we exclude the banks, transportation and utilities companies, as these firms are heavily regulated, and the repurchasing activity could be different from that of non-regulated companies (Fenn and Liang, 2001; Bens et al., 2003; Hribar et al., 2006). The sample data takes the form of an unbalanced panel data, as we tried to avoid any sample selection bias towards healthy companies and included the companies which went through mergers, acquisitions or bankruptcy during the analyzed period.
Of the 1231 firms observed between January 2003 and December 2007 fiscal years, more than half (946 firms or 77%) carried out repurchases of stock at least once during the studied period. During the same period, 1208 (or 98.1%) firms used stock options in their compensation plans. Figure 1 details the growth trend for both stock repurchases and stock options for the sample companies between 2003 and 2007.
Fig.1. Dollar value of repurchases and stock options over the 2003-2007 period. 0 100000 200000 300000 400000 500000 600000 700000 0 50000 100000 150000 200000 250000 300000 350000 400000 2003 2004 2005 2006 2007
Stock options Repurchases
10 Table 1 represents the industry distribution of share repurchases. The table shows evidence that the magnitude of stock buybacks increased during the 2003-2007 period, thus, the sample companies bought back shares in value of $413 million in 2003 and increased this amount to $1,282 million in 2007. In the final year researched, the industries with the highest buybacks were retail trade, finance and insurance, business services, industry machinery and electronic equipment.
Table 1
Industry distribution of share repurchases
Industry distribution of share repurchases for the sample of 1231 US firms over the 2003-2007 period. The industries are classified based on the 2-digit SIC code. N stands for the number firms. Rep. is the value of share repurchases ($MM) carried out in a given industry.
Year 2003 2004 2005 2006 2007
Variable
2-digit
SIC N Rep. N Rep. N Rep. N Rep. N Rep
Mining 10,12,14 12 0.43 12 1.9 11 8.56 10 10.92 9 13.76
Oil and gas extraction 13 43 8.6 37 5.72 35 54.36 30 24.84 31 9.84
Construction 15,16,17 15 8.75 15 4.03 15 12.95 15 16.56 15 4.87
Food and kindred products 20 38 18.04 37 19.13 37 31.02 37 37.47 37 36.01
Tobacco products 21 3 1.79 3 1.7 3 1.25 3 1.25 3 4.16
Textiles and apparel products 22,23 24 2.25 24 1.9 22 8.67 20 8.07 20 21.88 Wood and paper products 24,25,26 36 6.56 37 16.6 36 28.84 36 25.38 35 33.65 Printing and publishing 27 21 19.53 20 20.72 20 25.64 17 21.34 12 25.22 Chemicals and allied products 28 91 24.62 91 23.61 88 78.5 84 47.13 78 77.7
Petroleum products 29 8 0.06 8 1.73 8 7.25 8 7.79 7 9.48
Rubber and plastics products 30 10 3.63 10 6.25 10 7.74 10 5.44 10 9.34
Leather products 31 8 5.5 8 8.29 8 6.24 8 5.14 7 6.3
Glass and concrete products 32 9 2.88 9 0.01 9 0.01 9 0.3 9 0
Metal products 33,34 46 11.11 43 9.07 43 22.09 39 22.96 35 26.98
Industrial machinery 35 90 19.01 91 35.42 82 62.5 80 90.24 76 126.9 Electronic equipment 36 105 13.38 105 19.64 104 38.32 101 43.52 97 110 Transportation equipment 37 36 9.94 36 15.47 36 27.43 35 37.7 34 27.1 Measuring /medical equipment 38 83 23.09 81 27.38 81 51.16 77 53.75 74 72.1 Misc. manufacturing industries 39 11 1.78 11 1.07 11 1.96 11 4.99 10 13.8
Wholesale trade 50,51 38 6.01 37 9.87 37 15.42 34 19.85 33 42.2
Retail trade 52-59 124 68.05 124 83.22 114 111.6 109 117.74 105 193.7 Finance and insurance 61-65,67 133 68.57 131 96.77 128 118.4 124 179.5 123 192.4 Business services 73 133 64.7 130 69.5 125 108.8 112 135.5 105 141.9 Personal services 70-87 77 22.87 77 36.42 74 43.3 72 79.4 65 80.86
Miscellaneous 99 4 2.18 4 2.38 4 2.74 4 1.02 4 1.49
Total - 1198 413 1181 518 1141 875 1085 998 1034 1282
Source: own representation of Kahle, 2002, table 2, p. 248.
11 3.2. The variables
The dependent variable is the share repurchases which is calculated as in Kahle (2002), the Purchase of Common and Preferred stock (Compustat item #115) minus the Purchase of Preferred stock (Compustat item #130). In the model we use the binary variable that takes value 1 if the firm carried out stock buyback during the fiscal year and 0 if it did not.
The explanatory variables we are most interested in are the stock options. We divide the stock options into 2 groups based on the holder, namely we study the effect of executive stock options and the stock options awarded to the rest of the employees on the decision to buy back shares. Generally, the stock options are granted under certain restrictions, namely the vesting period restriction (Summa, 2015). The vesting period is the time period the holder must wait before exercising the stock options and it usually follows a certain vesting schedule (Olagues and Summa, 2010). According to Kahle (2002) the actual repurchases are related to the currently exercisable or vested options, as namely this type of stock options leads to a dilution effect in the current period. That is why when we talk about stock options in our model we refer to the currently exercisable or vested options. To counter any issues resulting from the variation in firm size in our sample, we divide the stock options variable by the number of shares outstanding (Bartov et al., 1998). The executive stock options variable is computed as the currently exercisable stock options of 5 main executives divided by the shares in circulation. The second variable is computed as the currently exercisable employee stock options divided by shares outstanding.
To take into consideration the other motives behind the stock repurchases, we include a set of control variables in our regressions.
We include the contemporaneous stock return variable in our models to account for the signaling theory. The managers may decide to repurchase shares if the stock price had poor performance in the current period, and they want to express or signal to the market their confidence in the firm’s future positive earnings (Bens et al., 2003).
12 high debt are less likely to carry out payouts to shareholders (Bagwell and Shoven, 1988; Kahle, 2002).
The model also includes a size control variable measured by the natural logarithm of total assets. Vermaelen (1981) states that the information asymmetry is more prominent in small firms, as these companies are not as closely tracked by the market and analysts as the large ones are. Large firms have lower information asymmetry, strong cash flows and lower financing costs (Kahle, 2002). According to Dittmar (2002), namely because of information asymmetry the small firms are most likely to be undervalued and resort to buybacks as a signaling mechanism.
To account for the excess capital theory, we include the variable of net operating cash flow in our regressions. According to Jensen (1986), the firms repurchase the share in order to distribute the excess cash to the shareholders, and thus, reduce the agency conflicts between shareholders and managers. We expect a positive relation between the net operating cash flow and share repurchases.
According to Bens et al. (2002) the firms with large stock options exercises, choose to repurchase shares and use the resources that otherwise would be spent on real investments. We include the capital expenditures divided by total assets in our model to account for this effect, as the companies with high capital expenditures would have better investment opportunities and less free cash flow to distribute (Kahle, 2002).
Lastly, there have been some research done on the possible substitution between the dividends and the share repurchases (e.g. Grullon and Michaely, 2002). The theory behind this concept stems from the advantages the buybacks have over the dividend payments. Even though Grullon and Michaelly (2002) show a negative relation between the 2 methods of payout, Bhargava (2010) cautions that this type of research is subject to bias from endogeneity and should be considered critically. Moreover, the study claims that the firms paying dividends are reluctant to cut or suspend the payments in favor of repurchases as it would send a bad signal to the market. Bhargava (2010) finds that dividends are negatively associated with the share buybacks, but he does not find an opposite correlation, when adding the share repurchases as an explanatory variable to a model for dividends. In order to take into account a possible impact of dividends on share buybacks, we add the dividend dummy variable in our model. It takes value 1 if the company paid dividends in the fiscal year and 0 if it did not.
13 Table 2
Definition of variables
Description of the variables used in the regression model and their way of measurement.
Variable Definition
REPUR Dummy variable that takes value 1 if the company repurchased share and 0
otherwise. The share repurchases are calculated as in Kahle (2002), the Purchase of Common and Preferred stock (Compustat item #115) minus the Purchase of Preferred stock (Compustat item #130).
OPTDIR Currently exercisable stock options granted to 5 main executives divided by
the number of shares outstanding.
OPTEMP Currently exercisable stock options granted to employees, and divided by the
number of shares outstanding.
OPTTOTAL The total number of currently exercisable stock options in a company divided
by the number of shares outstanding.
LOG(TA) Logarithm of total assets.
OPERCF The ratio of operating cash flow to total assets.
DIV Dummy variable that takes value 1 if the company paid dividends during the year, otherwise it takes value 0.
LEVER The ratio of long term debt and in current liabilities to total assets.
CAPEX Capital expenditures divided to total assets.
RETURN The stock return.
3.3. Estimation model
We test the 2 hypotheses discussed in the previous section using the logistic regression (Logit). We use the Logit model because we want to find out if the use of stock option plans has an impact on the stock repurchase decision and because our dependent variable is a binary variable.
REPURi,t = β0 + β1 OPTDIRi,t + β2 OPTEMPi,t + β3 DIVi,t + β4 CAPEXi,t + β5 LEVERi,t +β6 LOG(TA)i,t + β7 OPERCFi,t +β8 RETURNi,t + year + industry + εi,t
(1)
14 The firms have an average of 18.1 million of stock options outstanding. Moreover, the sample firms grant an average of 1.7 million employee stock options per year and have an average of 7.7 million exercisable stock options.
Table 3
Descriptive statistics for the sample firms
The sample statistics computed over the 2003-2007 period for the sample of 1231 US firms. The banks, transportation and utilities companies are excluded. Data on firm characteristics and employee stock options are provided by Compustat database. Std. dev. is the standard deviation. P25 and P75 are the 25th and 75th percentiles.
Variable Mean Std. Dev. P25 Median P75
Firm characteristics:
Total assets ($MM) 10,657 74,984 431 1,133 3,590
Turnover ($MM) 4,657 15,633 376 1,022 3,138
Net income ($MM) 315 1,333 11 54 194
Net operating cash flow ($MM) 504 2,999 29 99 338
Repurchases ($MM) 186 911 0 0.2 55
Number of shares outstanding (‘000) 184,934 639,654 25,727 50,374 121,376
Employee stock option characteristics:
Number of exercisable options (‘000) 7,702 31,897 0 1,280 4,269 Number of options granted (‘000) 1,713 6,328 0 215 1,038 Number of outstanding options (‘000) 18,058 62,872 2,018 4,399 11,350 Source: own representation of Bens et al, 2003, table 1, p. 65.
Appendix A, Table A presents the correlation statistics for the variables used in the empirical models. The p-values, which are written in parentheses, are displayed bellow the correlation coefficients. As we can see from the table, no significant high degree of correlation is found between 2 exogenous variables of the same model. Statistical theory states that multicollinearity is a problem if the bivariate correlation exceeds 0.8 (Bens et al., 2003, citing Kennedy, 1992).
15 Table 4
Descriptive statistics of variables
The sample statistics computed over the 2003-2007 period for the sample of 1231 US firms. The banks, transportation and utilities companies are excluded. Repur is a dummy variable calculated based on the repurchase of common stock. Optdir, optemp and opttotal are the executive stock options, employee stock options and total stock options respectively, as reported in Compustat and ExecuComp. Return is the stock return collected from CRSP. Opercf is the operational cash flow divided by total assets. Log(TA) is the natural logarithm of total assets. Capex is the capital expenditures divided by total assets. Div is the dummy variable that takes value 1 if the company paid dividends during the fiscal year and 0 otherwise. Std. dev. is the standard deviation. P25 and P75 are the 25th and 75th percentiles.
Variables Mean Std. dev. P25 Median P75
REPUR 0.530 0.499 0 1 1 OPTDIR 0.016 0.019 0.003 0.011 0.023 OPTEMP 0.029 0.036 0 0.018 0.044 OPTTOTAL 0.046 0.043 0.014 0.034 0.065 RETURN 0.018 0.037 -0.001 0.016 0.033 OPERCF 0.095 0.123 0.051 0.094 0.144 LOG(TA) 3.123 0.733 2.636 3.054 3.555 LEVER 0.187 0.179 0.015 0.158 0.297 CAPEX 0.045 0.053 0.015 0.029 0.056 DIV 0.530 0.499 0 1 1
4. Results and discussion
4.1. Empirical results
The results from estimating the panel data logit model for share repurchases are shown in Table 5 for the sample of 1231 US firms. Moreover, we provide the results for the regression both with and without the industry and year fixed-effects in order to validate our estimation. After running additional tests, we made sure that the models provide consistent, unbiased and efficient estimates.
16 probabilities in logistic regression and through mathematical calculations can be transformed into odds ratios.
Table 5
Determinants of the decision to repurchase shares
The dependent variable is a binary variable. It takes value 1 if the company repurchased shares during the year and 0 if it did not. The share repurchases are calculated as in Kahle (2002), the Purchase of Common and Preferred stock (Compustat item #115) minus the Purchase of Preferred stock (Compustat item #130). Optdir, optemp and opttotal are the executive stock options, employee stock options and total stock options respectively, as reported in Compustat and ExecuComp. Return is the stock return collected from CRSP. Opercf is the operational cash flow divided by total assets. Log(TA) is the natural logarithm of total assets. Capex is the capital expenditures divided by total assets. Div is the dummy variable that takes value 1 if the company paid dividends during the fiscal year and 0 otherwise. The number in parentheses is the p-value.
***, **, * - means that the coefficients are statistically significant at 1%, 5% and 10% levels.
Variable Predicted sign (1) (2) (3)
OPTDIR
(+)
8.2135*** (0.010) 8.3937*** (0.007) 10.1383*** (0.001) OPTEMP(+)
3.8883** 3.4495** 4.2154*** (0.027) (0.020) (0.004) DIV(-)
1.2005*** (0.000) 1.1232*** (0.000) 1.0322*** (0.000) CAPEX(-)
-0.4002 (0.794) (0.840) 0.3002 -1.5513 (0.219) LEVERAGE(?)
-1.9032*** -1.8666*** -1.9274*** (0.000) (0.000) (0.000) LOG(TA)(-)
1.1920*** 1.3187*** 1.2482*** (0.000) (0.000) (0.000) OPERCF(+)
6.1961*** 6.3786*** 6.5726*** (0.000) (0.000) (0.000) RETURN(-)
-6.7089*** (0.000) -9.3453*** (0.000) -9.7538*** (0.000)Industry controls Yes Yes No
Year controls Yes No No
χ2 467.46*** 406.12*** 333.17***
(0.000) (0.000) (0.000)
Sample size 1231 1231 1231
17 exercise and finds that the buybacks are positively affected by the desire of executives to manage the diluted earnings-per-share.
The second hypothesis implies that the management stock options bring additional incentives to resort to share repurchases, beyond those of employee stock options. Namely, because the majority of companies do not protect the stock option value against the dividend payments, the executives can have the incentive to carry out repurchase of shares as a mean of earnings distribution (Kahle, 2002). Therefore, we expect that the executive stock options lead to higher odds that the company will buy back stock. As we can see from the results for the Equation 1, there is a positive relation between the exercisable executive stock options (OPTDIR) and the probability of share repurchase. Namely, at a 1% significance level, a unit increase in the exercisable executive stock options will raise the probability of share buybacks by 8.21 log odds, with 4.32 log odds more than in the case of employee stock options.
Table 6
Dividends and executive stock options
Dividends (DIV) are the dependent variable, which is a dummy variable that takes value 1 if the company paid dividends during the fiscal year and 0 otherwise. OPTDIR are the executive stock options, as reported in ExecuComp, divided to shares in circulation. OPTEMP are the employee stock options, as reported in Compustat, divided by shares outstanding. The number in parentheses is the p-value.
DIVi,t = β0 + β1 OPTDIRi,t + ∑ βi Firm characteristicsi,t + year + industry + εi,t (4) ***, **, * - means that the coefficients are statistically significant at 1%, 5% and 10% levels.
Variables (4)
OPTDIR -23.0606***
(0.006)
OPTEMP 3.4503
(0.341)
Firm characteristics Yes
Industry controls Yes
Year controls Yes
χ2 385.20***
(0.000)
18 The second hypothesis also implies that because the managers favor buybacks to dividends, as the last ones decrease the value of stock options, there is a negative relation between the dividends and stock options. Table 6 provides additional evidence of a negative relation between the executive stock options and dividend payments. Thus, a unit increase in the managerial stock options holdings will decrease the probability of dividend payments by 23.06 log odds.
There are several possible reasons for our results. It appears that the companies carry out stock buybacks in order to handle the dilution effect from the exercise of stock options, and here an important concept is the dilution of EPS. The firms seek to continuously meet the analysts’ forecasts, as not doing that could put a downward pressure on the firm’s price (Bens et al. 2003; Hribar et al. 2006). Myers, Myers, and Skinner (2007) analyze a number of firms with consecutive EPS increases and conclude that “managers appear to strategically time stock repurchases to boost reported EPS when the string would otherwise be broken”. The executive stock options are awarded to align the managers’ financial goals to those of shareholders, thus, we could assume that in a small part the repurchases are the product of agency theory, as buybacks are a form of earnings distribution (Jolls, 1998).
Another motivation to repurchase shares, which is stemmed from the use of stock option plans, is the dividend policy. The majority of companies do not have protections against dividend payments (Murphy, 1998), and the managers are more likely to choose the repurchases as a mean of earnings distribution as the buybacks do not decrease the share-value (Fenn and Liang, 2001).
Table 5 offers results regarding the association between the decision to repurchase stock and certain firm characteristics. We find a negative significant relation between the level of debt and share buybacks. This result is statistically significant at a 1% level and consistent with the findings of Kahle (2002) and Bagwell and Shoven (1988). It confirms the hypothesis that the firms with high debt have less resources to distribute to shareholders and are less likely to carry out share repurchases. Also, Kahle (2002) claims that this result could stem from the situation that firms with high leverage have higher financing costs and resort less to stock buybacks.
19 We did not find evidence supporting the theory that the buybacks crowd-out the investments, as the coefficient of capital expenditure variable is not significant.
According to Vermaelen (1981), the small companies are more likely to deal with information asymmetry, so we expected for the company size to be negatively related to the share repurchases. Our results do not support this assumption, as the odds of repurchases increase as the firm size grows. However, this finding is consistent with the evidence provided by Dittmar (2000).
Stephens and Weisbach (1998) find evidence that the firm repurchase shares following a period of poor stock performance. This finding is consistent with our results, as we find that low stock returns lead to higher odds that the firm will choose to buy back shares.
Based on the previous research (e.g. Grullon and Michaely, 2002) we expected a negative correlation between the stocks repurchases and dividend payments, however our findings suggest the two are possible complements, as the dividends are positively related to share buybacks. In his research, Lintner (1956) claims that the firms are reluctant to reduce the dividend payments and came up with a model that explains the dividends as a function of “permanent” earnings. The companies tend to pay the dividends regularly out of “permanent” earnings, while the repurchases are used to distribute the “temporary” earnings (Dittmar, 2008). In this case, the dividends and repurchases could be seen as complementary.
4.2. Robustness check
We perform a robustness check by running a panel data regression with random effects. We include only the firms that have carried out share repurchases. The sample comprises 946 firms, which is equivalent to 2983 firm-year observations. The results of the model estimation are presented in Appendix B, Table B.
20 4.3. Research limitations
This paper has several limitations. First, we used the secondary data provided by Compustat and ExecuComp for stock options variables. Kahle (2002) questions the reliability of these data. She collected the data on both outstanding and exercisable executive stock options, and then she compared the degree of correlation between her variables and the Compustat-based proxies for the same variables used by Bartov et al. (1998). She finds that the correlation is less than 0.30.
Second, the endogeneity problem is not completely tackled. We used the industry and year effects to take into account the omitted variable bias, but the question remains about the endogeneity occurred due to simultaneity. Kahle (2002) mentions that both share repurchases and stock options are endogenous variables, as they are affected by similar factors, e.g. increase in cash flow or stock returns. Moreover, these factors are part of our models as control variables. The problem of endogeneity plagues many papers and dealing completely with it, in our case, is not feasible.
Because the stock repurchases and leverage variables are endogenous as well and the leverage variable may be correlated to the regression error term, we re-estimate our regression without the leverage variable. Consistent with endogeneity checks in other papers, our initial regression results are not significantly different from the leverage-restricted models (Fenn and Liang, 2001; Kahle, 2002).
5. Conclusion
This paper has analyzed the relationship between the employee stock option plans and the firm’s decision to repurchase shares. We predicted that the recent drastic growth in the level of share repurchases can, in part, be explained by the use of the stock-option- based compensation plans. We find strong and significant evidence that the use of both executive and employee stock options increase the likelihood that firms will carry out share repurchases.
We also find evidence to support the free cash flow, optimal leverage, and signaling theories that provide additional motivation to firms to buy back stock. Our empirical models find that firm size, net operating cash flows, and dividend payments are positively related to the decision to repurchase shares. Leverage and stock returns have a negative relation to the stock buybacks.
21 stock options, are endogenous, as they are influenced by the same external factors, e.g. corporate governance, positive stock returns etc. The issue of endogeneity is present in much research in the field of corporate governance and corporate finance, and dealing with it is beyond the scope of this paper. Thus, our results should be interpreted with caution.
22
Appendix A
Table A
Correlation matrix
The correlation statistics calculated for the variables employed in the study, computed over the 2003-2007 period for the sample of 1231 US firms. The banks, transportation and utilities companies are excluded. Repur is a dummy variable calculated based on the repurchase of common stock. Optdir, optemp and opttotal are the executive stock options, employee stock options and total stock options respectively, as reported in Compustat and ExecuComp. Return is the stock return collected from CRSP. Opercf is the operational cash flow divided by total assets. Log(TA) is the natural logarithm of total assets. Capex is the capital expenditures divided by total assets. Div is the dummy variable that takes value 1 if the company paid dividends during the fiscal year and 0 otherwise. The p-value is indicated in parentheses. There is no high degree of correlation between 2 variables of the same model.
REPUR OPTDIR OPTEMP OPTTOTAL CAPEX DIV LEVER LOG(TA) OPERCF RETURN
---23
23
Appendix B
Table B
Panel data random effects estimation
The share repurchases, the dependent variable, is calculated as in Kahle (2002), the Purchase of Common and Preferred stock (Compustat item #115) minus the Purchase of Preferred stock (Compustat item #130). In the model we use the natural logarithm of share repurchases. Optdir, optemp and opttotal are the executive stock options, employee stock options and total stock options respectively, as reported in Compustat and ExecuComp. Return is the stock return collected from CRSP. Opercf is the operational cash flow divided by total assets. Log(TA) is the natural logarithm of total assets. Capex is the capital expenditures divided by total assets. Div is the dummy variable that takes value 1 if the company paid dividends during the fiscal year and 0 otherwise. The number in parentheses is the p-value.
***, **, * - means that the coefficients are statistically significant at 1%, 5% and 10% levels.
Variable Predicted sign (1) (2)
24
24
References
Akyol, A.C., Foo, C.C., 2011. Share repurchase reasons and the market reaction to actual share repurchases: Evidence from Australia. International Review of Finance 13:1, 1–37.
Babenko, I., 2009. Share repurchases and pay-performance sensitivity to employee compensation contracts. The Journal of Finance 69, 117-151.
Bagwel, L., Shoven, J., 1988. Share repurchases and acquisitions: an analysis of which firms participate. In: Auberbach, A. Corporate Takeovers: Causes and Consequences. University of Chicago Press, Chicago, IL, 191-213.
Balachandran, B., Chalmers, K., Haman, J., 2008. On-market share buybacks, exercisable share options and earnings management. Accounting and Finance 48(1), 25-49.
Bartov, E., Krinsky, I., Lee, J., 1998. Evidence on how companies choose between dividends and open-market stock repurchases. Journal of Applied Corporate Finance 11, 89–96.
Bens, D.A., Nagar, V., Wong, M.H.F., 2002. Real investment implications of employee stock option exercise. Journal of Accounting Research 40 (2), 359-393.
Bens, D.A., Nagar, V., Skinner, D.L., Wong, M.H.F., 2003. Employee stock options, EPS dilution, and stock repurchases. Journal of Accounting and Economics 36, 51-90.
Benton, E., Doowoo, N., 2001. Stock buybacks, corporate performance, and EVA. Journal of Applied Corporate Finance 14(1), 99-110.
Bhargava, A., 2010. An econometric analysis of dividends and share repurchases by US firms. Journal of the Royal Statistical Society A 173, 631-656.
Brav, A., Harvey, C.R., Graham, J.R., Michaely, R., 2005. Payout policy in the 21st century. Journal of
Financial Economics 77(3), 483-527.
Business Insider, 2014. The rate of share buybacks has doubled in the last decade. Retrieved from: http://www.businessinsider.com/sp-500-stock-buyback-history-2014-4
Comment, R., Jarrell, G.A., 1991. The relative signaling power of dutch-auction and fixed-price self-tender offers and open-market share repurchases. Journal of Finance 46, 1243-1271.
Dittmar, A.K., 2000. Why do firms repurchase stock?. Journal of Business 73 (3), 331-355.
Dittmar, A.K., 2008. Corporate cash policy and how to manage it with stock repurchases. Journal of Applied Corporate Finance 20(3), 22-34.
El Houcine, R., Boubaker, A., 2014. Share repurchases and the policies of stock-options. International Journal of Academic Research in Accounting, Finance and Management Sciences 4(1), 175-187. Fenn, G., Liang, N., 2001. Corporate payout policy and managerial stock incentives. Journal of Financial
Economics 60, 45-72.
Financial Times, 2013. US share buybacks at record amount. Retrieved from: http://www.ft.com/intl/cms/s/0/c630b3e6-8735-11e2-bde6-00144feabdc0.html#axzz3AXtMIoDj Grullon, G., Michaely, R., 2002. Dividends, share repurchases, and the substitution hypothesis. Journal of
Finance 57, 1649–1684.
Grullon, G., Michaely, R., 2004. The information content of share repurchase programs. Journal of Finance 59, 651–680.
Hall, B.J., Murphy, K.J., 2003. The trouble with stock options. The Journal of Economic Perspectives 17 (3), 49-70.
Hite, G.L., Long, M.S., 1982. Taxes and executive stock options. Journal of Accounting and Economics 4, 3-14.
Hribar, P., Jenkisn, N.T., Johnson, W.B., 2006. Stock repurchases as an earnings management device. Journal of Accounting and Economcis 41, 3-27.
25
25
repurchases. Journal of Financial Economics 39(2-3), 181-208.
Jensen, M., 1986. Agency costs of free cash flow. American Review 76, 323-329.
Jolls, C., 1998. Stock repurchases and incentives compensation. Unpublished working paper No. 6467. NBER.
Kahle, K.M., 2002. When a buyback isn’t a buyback: Open market repurchases and employee options. Journal of Financial Economics 63(1), 235-261.
Kennedy, P., 1992. A Guide to Econometrics. MIT Press, Cambridge, MA.
Lambert, R.A., Lanen, W.N., Larker, D.F., 1989.Executive stock option plans and corporate dividend policy. Journal of Financial and Quantitative Analysis 24 (4), 411- 425.
Lazonick, W., 2014. Profits without prosperity. Harward Business Reviews Spring, 46-55.
Lintner, J., 1956. Distribution of incomes of corporations among dividends, retained earnings, and taxes. American Economic Review 46(2), 97-113.
Murphy, K., 1998. Executive compensation. Unpublished working paper. University of Southern California.
Myers, J., Myers, L., Skinner, D., 2006. Earnings momentum and earnings management. Journal of Accounting, Auditing and Finance 22(2), 249-284.
Olagues, J., Summa, J., 2010. Getting Started In Employee Stock Options. Wiley, 1st ed.
Quiry, P., Le Fur, Y., Salvi, A., Dallochio, M., Vernimmen, P., 2011. Corporate Finance: Theory and Practice. John Wiley & Sons.
Reuters, 2013. When a buyback looks like executive pay. Retrieved from: http://www.reuters.com/article/2013/05/29/us-buybacks-execpay-saft-idUSBRE94S17W20130529 Stephens, C.P., Weisbach, M.J., 1998.Actual share reacquisitions in open-market repurchase programs.
Journnal of Finance 53, 313-334.
Summa, J. 2015. Employee stock options (ESO). Retrieved from:
http://www.investopedia.com/university/employee-stock-options-eso/
The Wall Street Journal, 2012. The Pros and Cons of stock buybacks. Retrieved from: http://online.wsj.com/news/articles/SB10001424052970203824904577213891035614390
Vermaelen, T., 1981. Common stock repurchases and market signaling: An empirical study. Journal of Financial Economics 9, 139-183.
Voss, J., 2012. Why do firms repurchase stock?. Major themes in Economics, 58-75.
Weisbenner, S., 2000. Corporate share repurchases in the 1990s: What role do stock options play?. Unpublished working paper. Federal Reserve Board.
Yi-Mien, L., Yu-Cheng, C., Shwu-Jen, Y., Ruey-Jane,C., 2009. Stock repurchases, stock options, and earnings management: An empirical analysis of US firms. Asia Pacific Management Review 14(3), 349-362.
