Spin-offs and CEO compensation:
Does it pay to spin-off?
Master thesis Accountancy
2021
Niels J. Kamps s2735504
University of Groningen MSc A&C Accountancy
2
Spin-offs and CEO compensation:
Does it pay to spin-off?
Abstract
A spin-off is a type of corporate divestiture where a company separates a part of the firm to become a standalone entity. Given the increasing volume of spin-offs in the last decade, they have a large economic importance. Generically, current empirical evidence shows that spin-offs are value enhancing and are driven by increasing corporate focus and reducing information asymmetry. Due to the high discretion available to the management of the parent company regarding the details of a spin-off, CEO decisions are likely to heavily impact the outcome of the divestiture. In this paper, I examine if CEO compensation is affected subsequent to a spin-off. I use uses panel regressions to determine if there is support for this claim. Empirical results do not show a significant effect of doing a spin-off on CEO compensation. Nonetheless, cross-sectional tests suggest that a focusing and a small spin-off, decreases CEO compensation subsequent to a spin-off, whereas this effect is not found for a non-focusing and large spin-offs. A small spin-off likely is of small economic significance whereas a big spin-off is not, resulting in a decrease in compensation after a small spin-off. A decrease in CEO compensation after a focusing spin-off suggests that CEOs are punished for simplifying their firm, because the CEO is now left with a company that is easier to manage. This has important implication for compensation setters, since focusing spin-offs are value enhancing. Naturally, a CEO has no personal reasons to enhance focus, when it means he will be compensated less in return.
3
Table of contents
1. INTRODUCTION ... 5
2. THEORETICAL BACKGROUND ... 8
3. RELATED LITERATURE AND HYPOTHESIS DEVELOPMENT ... 10
3.1 Related literature ... 10
3.1.1 Firm size and CEO compensation... 10
3.1.2 Firm performance and CEO compensation ... 11
3.2 Hypotheses development ... 11
4. RESEARCH METHODOLOGY ... 14
4.1 Data ... 14
4.2 Methodology ... 15
5. RESULTS ... 18
5.1 Descriptive statistics and correlations ... 18
5.2 Main results ... 19 5.3 Robustness tests ... 21 5.4 Additional tests ... 22 6. LIMITATIONS ... 24 7. CONCLUSION ... 24 REFERENCES ... 26 APPENDIX ... 28 Variable definitions ... 28 Table 1 ... 30 Descriptive statistics ... 30 Table 2 ... 31 Correlations ... 31 Table 3 ... 32
4
Table 4 ... 33
The effect of focusing spin-offs ... 33
Table 5 ... 35
The effect of spin-off size ... 35
Table 6 ... 37
Robustness test: ROCE instead of ROE ... 37
Table 7 ... 39
Robustness test: Lagged control variables ... 39
Table 8 ... 41
Robustness test: Controlling for lagged CEO Compensation ... 41
Table 9 ... 43
Additional test: Salary as dependant variable ... 43
Table 10 ... 45
5
1. Introduction
A spin-off is a type of corporate divestiture where a company separates a part of the firm to become a standalone entity. The divesting firm, also called the parent, continues trading under the same listing and the spun-off company will trade under a new listing. The divestment results in an allocation of assets and liabilities to the spun-off firm, leaving the parent with less assets and liabilities than prior to the divestment (Hite & Owers, 1983). Spin-offs have a large economic impact; for example, Abott Laboratories conducted a spin-off worth 55 billion dollars in 2012. Overall, considering that the total value of spin-offs in the year 2015 was 257 billion dollars (Wachtell, Lipton, Rosen, & Katz, 2016), spin-offs are clearly very relevant in the current economic environment.
CEOs of listed firms have discretion about undertaking a spin-off. Some aspects of a spin-off are largely determined by the board and management of a parent firm. In the case of a spin-off through dividend, doing a spin-off does not require a shareholder vote. The decision to what businesses should be spun-off and the selection of managers for the spun-off company are, in practice, determined by the board and management (Wachtell et. al., 2016). A spin-off is value enhancing, so a good quality spin-off pays for a firm to undertake.
Since a spin-off has several potential benefits for the firm, one would expect that CEOs of firms are awarded for their efforts, like enhancing corporate focus. Compensating CEOs for spin-off decisions is especially important, since it leads to a simpler firm structure. The literature advances that executive compensation is positively related the size of the firm. A spin-off leads to a smaller firm, so if the effect of firm size after the divestiture on the compensation of a CEO results in lower payment, then CEOs might not be interested to divest, because it is not in their own interest. Not utilizing spin-offs in creating shareholder wealth would constitute an agency problem. Therefore, in this paper I assess the effect of spin-offs on the compensation of CEOs. My formal research question is: What is the effect of
6
offs on CEO compensation? The answer shows if it is worth for CEOs to undertake
spin-offs from the perspective of their compensation.
Generically, there are several reasons for firms to initiate spin-offs. For example, creating enhanced corporate focus (Habib, Johnsen, & Naik, 1997) or defining a distinct investment identity which leads to less information asymmetry (Best, Best, & Agapos, 1998; Krishnaswami & Subramaniam 1999). The former allows businesses to focus more on its own strategy, while the latter results in a more transparent business for potential investors (Wachtell et. al., 2016). There is academic evidence that these reasons for undertaking a spin-off have positive implications for the divesting firms. Desai & Jain (1999) found that a focus spin-off results in stock market gains. Spin-offs are also associated with increases in long-term performance and the empirical evidence shows that spin-offs are value enhancing (Chemmanur & Yan, 2004; John, 1993).
I argue that since CEOs have discretion over whether to undertake a off and spin-offs are likely to create value for the firm, they might want to undertake a spin-off to increase corporate focus and corporate performance. In turn, this increase in performance should lead to a positive impact on CEO compensation (Sun, Wei, & Huang, 2013). But the divestment results in a smaller firm and the effect of firm size on executive compensation should result in a decrease in compensation. The empirical evidence for this effect is very clear and remarkably stable across published US and UK studies (Haynes, Steve, Thompson, & Wright, 2007). It is interesting to research if undergoing a spin-off is associated with CEO pay, because of these two effects divestment has on the firm. On the one hand, this study could show that CEOs are in fact rewarded for their focus increasing and information asymmetry decreasing activities. Alternatively, the results could show that compensation does not increase, or even decrease, subsequent of a spin-off, due to a decrease in firm size. This would mean that incentives for CEOs are unsatisfactory concerning spin-offs, unless CEOs are not
7
undertaking spin-offs to increase their own compensation, and are solely doing them because they are beneficial to the firm. An theoretical explanation for this could be found in stewardship theory.
A decrease in CEO compensation is possible because of the decrease in firm size which leads to a decrease in complexity of the firm. This means that a CEO can run the firm more easily. This is important considering focusing spin-offs. When increasing corporate focus through a spin-off, the divesting firm not only becomes less complex due to size, but also due to less ambiguity in economic activities. The conflicting possible effects a spin-off has on CEO compensation therefore becomes even stronger when considering focusing spin-offs.
This study adds to research streams on both CEO compensation and spin-offs. There is not much known about the causal effect of spin-offs on CEO compensation. This study will add to the knowledge about the determinants for executive remuneration. Haynes et al., (2007), already studied the effect of doing a corporate divestment on executive remuneration. They studied a sample of 107 UK quoted firms in the period of 1988 to 1993. The results show that executives are expected to lose compensation when divesting because of the loss of sales and size.
Although Haynes et al. (2007) researched the effects of corporate divestitures on executive compensation, my study will be different in numerous ways. Firstly, the sample is different. Specifically, in this study I focus on a sample of non-financial US listed firms in the 1992 to 2019 period. Secondly, I focus solely on spin-offs instead of all corporate divestitures. Thirdly, the effects of doing a focus spin-off compared to a non-focus spin-off are not researched by Haynes, et. al. (2007). In this study, I analyse if doing a focusing spin-off results in incrementally different results compared to a non-focusing spin-off. Increasing focus is beneficial to shareholders, so this gives even more reason to believe that CEOs will
8
be rewarded for such decisions. These differences result in a deeper understanding of the determinants of CEO compensation.
My research also has implications for boards and remuneration committees. Because if it can be determined that spin-offs do not increase pay, there may be little incentives for CEOs to do spin-offs. On the other hand, if it does not matter whether a focusing spin-off is initiated for compensation to increase subsequent of a spin-off, then CEOs might only do spin-offs for opportunistic reasons. This could have a negative impact on interests of the firm. With the knowledge this research will provide, boards can adjust compensation plans surrounding spin-offs accordingly.
The rest of the thesis is structured as follows. In section 2, I discuss the theoretical framework this research builds upon. In section 3, the related literature and the hypothesis development of this study is described. In section 4, the research methodology that is used to answer the research question will be explained. Section 5 describes the results from the empirical tests and in section 6 and 7, I discuss the limitations and conclusions of this study, respectively.
2. Theoretical background
In the thesis, I investigate two potential effects, under the theoretical umbrellas offered by agency theory (Jensen & Meckling, 1976; Eisenhardt, 1989) and stewardship theory (Donaldson & Davis, 1991; Davis, Schoorman, & Donaldson, 1997). Firstly, spin-offs might have a positive effect on CEO compensation. Secondly, spin-offs might not positively influence CEO compensation.
An explanation can be found in agency theory. Agency theory states that managers will not act in the interests of the shareholders unless strong corporate governance structures are in place (Jensen & Meckling, 1976). Instead, agency theory assumes that people (and therefore managers) will act out of self-interest (Eisenhardt, 1989). Because of the positive
9
effects associated with a spin-off, a reward in compensation is expected post spin-off, so that CEOs have incentive to perform one, if looking through the lens of agency theory. It is especially important to incentivise management since a proportion of assets and sales are divested. Firm size and sales are positively associated with executive compensation, so this could actually give disincentives to do a divestments, even if it’s in the interest of the firm. This shows that there is a underlying agency problem concerning spin-offs and divestments in general.
So a consequence of a spin-off could be that the CEO now manages a smaller company than before. And if offs are not associated with better compensation post spin-off, then it is also possible that CEOs are not doing spin-offs out of self-interest. A theory that could explain this reasoning is stewardship theory. This theory states that to maximise shareholder return, a CEO must have enough power to take effective action (Donaldson & Davis, 1991). Stewardship theory defines situations where managers are not motivated by individual goals, but are stewards whose motives are aligned with their principles, the shareholders (Davis, et. al., 1997). Therefore, a CEO might not undertake a spin-off due to opportunistic reasons, but to be a good steward to their shareholders.
As mentioned in the introduction, spin-offs are value enhancing for the firm and firm performance is positively associated with CEO compensation. So undertaking a spin-off can be a way for a CEO to increase its own compensation, because incentives are in place. On the other hand, CEO compensation might not be positively associated with undertaking a spin-off and because of this, firm management is not seizing value enhancing opportunities. On the other hand, stewardship theory could explain that CEOs are undertaking spin-offs just for reasons of good stewardship.
10
3. Related literature and hypothesis development
3.1 Related literature
In this section the related literature will be discussed. First the relationship between firm size and CEO compensation is described. Second, the literature concerning the relationship between firm performance and CEO compensation is mentioned, also in relation to spin-offs.
3.1.1 Firm size and CEO compensation
Spin-offs, although having numerous potential benefits, result in a downsized firm. The evidence for the positive effect of firm size on executive remuneration is extensive, so the bigger the firm, the more compensation the CEO will receive. This positive correlation has been established by various papers. In 1990, Kostiuk found that the elasticity of executive compensation to firm size is the same as in the 1930’s. So the positive effect of firm size on executive compensation has remained stable over a long period of time.
In a more recent study, Gabaix, Landier, & Sauvagnat (2013) examined the effects of firm size on CEO compensation after the financial crisis of 2008. They still find evidence that firm size positively influences CEO compensation. In 2007 firm value decreased, and so did CEO pay, while in the year 2009 firm value increased and so did CEO pay. This shows that the effect of firm size on CEO compensation remains an important phenomenon when studying determinants of CEO compensation.
In 2020, Keller & Olney studied the effects of globalization on executive compensation. In their analysis they found that next to globalization of a company, firm size still has a positive effect on executive compensation. So it is clear that through time several studies find support for the hypothesis that firm size is indeed an important factor in the increase of CEO compensation and executive compensation in general.
11
3.1.2 Firm performance and CEO compensation
As mentioned earlier, doing a spin-off has positive effects on firm performance. It has a positive effect on stock market gains (Desai & Jain, 1999), but it also has positive effects in the form of long-term performance increase and enhanced firm value (Chemmanur & Yan, 2004; John, 1993)
The literature surrounding the effects of firm performance on CEO pay is extensive. Murphy (1985) found that executive compensation is positive related to firm performance when examining individual executives over time. Anderson, Banker, & Ravindran (2000) found that firm performance has a positive impact on shares and options awards. On the other hand, Jensen & Murphy (1990) do not find this association using accounting ratios as firm performance variables (Sun, et. al., 2013).
Although the association between spin-offs and firm performance has been clear, prior literature suggests mixed results for the relationship between firm performance and CEO compensation. For this study this could mean two things. On the one hand, subsequent of a spin-off, CEOs are rewarded for their value enhancing activities. On the other hand, it is possible that CEOs are not rewarded for their spin-off activities, because there is not a strong association between firm performance and the compensation of the CEO.
3.2 Hypotheses development
Under the assumption that spin-offs create wealth for shareholders and that firm management should be awarded for their successful activities creating this wealth, naturally the consequence should be that compensation increases. But there are several reasons why this might not be the case in practice, if you take into the account the effect of the decrease in firm size and the possibility that value enhancing activities by management can go unrewarded.
The effect a spin-off has on CEO compensation is therefore contradictory if you look at what the literature says. So the effect a spin-off has on CEO compensation can be either
12
way. It can increase due to the positive benefits it has on firm performance and value. It could decrease because of the fact that the CEO is running a smaller and simpler firm, which could be a reason to adjust compensation for this. If a positive effect for the following hypothesis is found, then this would indicate that CEOs are rewarded for their spin-off activities. A negative effect would indicate that CEOs are actually punished for their spin-off activities due to the decrease in firm size and complexity. The first hypothesis is:
(H1): CEO compensation is affected by a spin-off.
One of the reason spin-offs are done is to increase corporate focus. Increasing corporate focus through spin-offs can has several positive effects. Spin-offs result in an increase in asset allocation efficiency (Feldman, 2016). Focus enhancing spin-offs are also associated with larger long-run abnormal returns compared to non-focusing spin-offs (Desai & Jain, 1999). Therefore it should be researched what the moderating effect of a focusing spin-off has on the hypothesis that spin-offs affect CEO compensation. This way it can be determined in what situation it is best to perform a spin-off for a CEO. Results might show that a focus spin-off is a better way to increase compensation than a non-focus spin-off, or the other way around.
The possibility that firm size is a stronger determinant for CEO compensation after a spin-off is even more likely with focusing spin-offs then with non-focusing spin-offs. Literature indicates a stronger positive effect of focusing spin-offs on firms compared to non-focusing spin-offs, naturally this should hold a stronger effect on CEO compensation. But if you consider that a focusing spin-off not only results in a smaller firm, but also decreases the complexity of that firm, the negative effect of divestment on CEO compensation also
13
becomes even more likely. When looking at focusing spin-offs, the possible outcomes of a spin-off on CEO compensation becomes even more contradictory.
If the first hypothesis is not supported, but there is a positive effect on focusing spin-off, this might indicate that CEOs are only rewarded for their spin-off activity when it increases corporate focus. And if the first hypothesis is not support and a negative effect is found for the second hypothesis then this would indicate that CEOs are losing compensation due to the fact they are managing smaller and simpler firms after the spin-offs. If neither of the first two hypotheses is supported than this gives even more evidence that CEOs are not always rewarded for their efforts to create shareholder returns. The effect of focusing spin-offs on CEO compensation is studied using the following hypothesis:
(H2): CEO compensation will be incrementally affected by a focusing spin-off.
Finally the moderating effect of the size of a spin-off is researched. This will also extend the knowledge about what situation is best for a CEO to undertake a spin-off. Hope, Porumb, Rusanescu, & Vyas (2020) find that spin-off size has an effect on loan-pricing. If a spin-off is of high quality, then the bigger the spin-off, the bigger the positive effects should be. Therefore, the size of a spin-off can very well have an effect on the post spin-off compensation, because of the positive effects for shareholders associated with the spin-off. The size of the spin-off can be measured by the size of the spun-off firm in dollars relative to the size of the parent. On the other hand, literature is clear that firm size has a strong effect on CEO compensation, so if this effect is stronger than the effect of the spin-off, a larger spin-off would result in lower compensation. This situation would mean that CEOs would be discouraged to spin-off larger portions of the company. Therefore, the third hypothesis is:
14
(H3): CEO compensation will be incrementally affected by a large spin-off.
4. Research methodology
In this section we will briefly discuss the research methods that will be used to form an answer to the research question. The kind of data and the dependant and independent variables will be discussed.
4.1 Data
The sample used in this thesis consist of US based firms. Because, as mentioned earlier, the rationale for developing the research question starts with the notion that spin-offs create wealth. The empirical evidence shows that spin-offs of United States (US) based firms indeed create wealth. But in contrast to US spin-offs, Veld & Veld-Merkoulova (2004) did not find evidence that European spin-offs are associated with long-term superior performance. Therefore US spin-offs are more suited for this research in particular. Also, the study by Haynes et. al. (2007), researched the effect of spin-offs on executive compensation using UK firms, so it is also interesting to see if there are different results.
Data concerning CEO compensation is from ExecuComp. This database has data about different forms of CEO compensation. ExecuComp data is from the period 1992 to 2020. The extracted data is first altered so it only has CEO data, all other executive compensation data is dropped. Financial data is extracted from CompuStat. The financial data is merged with the compensation data period (1992-2020), after deleting duplicates on company identification number (gvkey) and fiscal year. Observations for only financial data are dropped in the sample.
Finally, the spin-off data consists of 773 spin-offs in the period of 1981 to 2018. In the dataset there are some firms that have done more than one spin-off. To analyse the effect of a spin-off on executive compensation, only one spin-off can be used. Therefore, for every
15
parent firm with multiple spin-offs, only one is kept. To be able to match the spin-off data with compensation and financial data, the choice was made to only keep the most recent spin-off. This leads to a spin-off data set of 634 spin-offs, that is merged with the compensation and financial data.
The above mergers results in a data set of 49,977 year observations, of which 43,520 are of non-spin-off firm and 6,457 are of spin-off firms. A firm is considered a spin-off firm if the firm is present in the spin-off dataset. The dependant variable of interest is total annual executive compensation (“TDC1” in ExecuComp). This variable consists of salary, bonus, restricted stock grants, long term incentive plan pay-outs, value of option grants, other annual and all other compensation.
All variables shown in this thesis are winsorised on 1% and 99%, to avoid the effect of outliers on the analysis. For the regression analysis, the variables are also standardised. The values shown in the descriptive statistics and correlation matrix are not standardised, but are winsorised.
4.2 Methodology
To test the hypothesis CEO compensation increases subsequent of a spin-off will be tested using regression analysis. The sample consists of firm year observations and the effect of an event, the spin-off, is studied. Because of this, the sample is marked as panel data where firm identification number is used to specify the panel and fiscal year as the time variable. By using panel data for the linear regression analysis, the results are reliable to conclude what the effect of the spin-off is on CEO compensation. The research methodology builds on the methodology of Haynes et al. (2007). They researched the effects of spin-offs on executive remuneration. In their analysis, important determinants for executive remuneration are firm performance and firm size. As for the independent variable of interest (the spin-off), a similar approach as Hope et al. (2020) is used. They researched the effect of spin-offs on loan pricing.
16
The independent variable of interest used in that study is a variable that is equal to 1 when to all loan facilities granted in the year prior to the spin-off and 0 otherwise. To study the effects of a spin-off on CEO compensation after the spin-off, my independent variable takes value of 0 in the period before the spin-off and a value of 1 in the period after the spin-off.
This independent variable of interest, I Spin-Off Year, takes the value of 0 in the year T-2, T-1, T and the value of 1 in the year T+1, T+2, T+3 where T is the spin-off year. Only the years surrounding the spin-off year are studied to be more certain that the effect on the dependent variable is from the spin-off and not because of the natural increase in compensation over time. To study hypothesis 2, a variable focus is made which take the value of 1 for every year observation of a firm when a focusing spin-off has occurred and a value of 0 when a non-focusing spin-off has occurred. A spin-off is considered a focusing spin-off when the first two digits of the SIC-industry code is different for the spun-off firm compared to the parent firm, as mentioned in the study by Desai & Jain (1999). For hypothesis 3 a variable spin-off size is made which takes the value of 1 when the spin-off size is above the median and 0 when spin-off size is below median, following the methodology of Hope et al., (2020). Spin-off size is computed by dividing total spun-off assets with total assets prior to the spin-off year.
Focus and spin-off size is used to split the data in groups. The regression analysis is
carried out for both groups. This method provides separate results for focusing and non-focusing spin-offs, so they can be compared. It provides the same separate results for small and large spin-off size.
The regression analysis will be controlled for firm size, since it is established that this is an important determinant for CEO compensation. Firm Size is computed by taking the natural logarithm of total assets. Another important control variable is firm performance. Following Haynes, et. al. (2007), return on equity (ROE) is used as an indicator for
17
performance. As a robustness test, the regressions are also performed with return on capital employed (ROCE).
Several other variables are also controlled for. I include gross Sales which is another indicator of firm size, which is used in the study by Haynes et al. (2007).
Market-To-Book-Ratio is included, since CEO compensation is related to market returns (Wilson & Higgins,
2001; Sun et al., 2013). Cash Flow (from operating activities) is included in the analysis to have a control variable for a performance indicator based on cash.
Tangibility, Leverage and negative net income (Loss) are included to control for firms
in financial distress, since financial distress risk affects CEO compensation (Chang, Hayes, & Hillegeist, 2016). A dummy variable which takes the value of 1 when a firm is audited by a Big 4 audit firm (Big 4 Auditor) and 0 otherwise. This variable is used as a proxy for audit quality, because audit quality affects dividend policy which in turn impacts stock market returns (Koo, Ramalingegowda, & Yu, 2017). Finally, the analysis is controlled for the variable Interlock which takes the value of 1 (and 0 otherwise) in the following situations: (1) “The officer serves on the board committee that makes his compensation decisions, or (2) The officer serves on the board (and possibly compensation committee) of another company that has an executive officer serving on the compensation committee of the indicated officer's company, or (3) The officer serves on the compensation committee of another company that has an executive officer serving on the board (and possibly compensation committee) of the indicated officer's company.”1 Controlling for these variables result in more reliable results concerning the effects of a spin-off on CEO compensation. Fixed effects for fiscal year and industry are also added in all regressions in this thesis.
ROCE is used as a robustness test for performance, as previously mentioned. Besides
this, more robustness tests are performed. The regressions are also performed with lagged
18
control variables. In this regression ROE, Sales, Market-To-Book-Ratio and Cash Flow are lagged, to control for the possibility that CEO compensation will increase after good results. Compensation setters may use last year results for setting compensation for the next year. Lastly, the regressions are performed with lagged total compensation (L.CEO Compensation) as an independent variable. This will show if compensation is consistent over time, since this variable is also included in the study by Haynes et al. (2007).
To determine which type of CEO compensation is affected by a spin-off (fixed or variable), the main regression is also done with fixed compensation (Salary) and variable compensation (Variable Compensation). Salary is taken directly from ExecuComp. Variable
Compensation is calculated by total compensation (TDC1) minus Salary.
5. Results
In this sections the results of the empirical analysis is discussed. First I go over the descriptive statistics and correlations. Second, the main results of this thesis are given. After that the outcomes of the robustness tests are discussed. Lastly, additional tests, where I analyse what part of CEO compensation is affected by the spin-off, is described.
5.1 Descriptive statistics and correlations
<Insert Table 1>
Table 1 provide the descriptive statistics for spin-off firms and non-spin-off firms for all variables used, except dummy variables. The variables in Table 1 are unstandardized. The results show differences in values between firms that had a spin-off and firms that did not. For all variables, the mean value is higher for spin-off firms. The median value is also higher for spin-off firms for all variables. The standard deviation is higher in all variables for spin-off firms except tangibility and leverage. This indicates that spin-off firms seem to have higher
19
variation in values than non-spin-off firms in the sample used. It is worth noting separately that the dependant variable in this thesis, CEO Compensation, is higher for spin-off firms.
<Insert Table 2>
In Table 2 the Pearson Pairwise correlations matrix is shown for all year observations in the whole sample. The variables are unstandardized for the correlation matrix. The correlation coefficients with the dependant variable show little unexpected outcomes. Firm
Size has the highest correlation with CEO Compensation. These variables are positively
correlated (0.531). The other financial variables (ROE, ROCE, Market-To-Book-Ratio and
Cash Flow) are also positively correlated, as expected. As for the financial structure ratios,
tangibility is negatively correlated and leverage is positively correlated with CEO
Compensation. Loss has a negative correlation, as expected. Interlock is negatively correlated
with CEO Compensation, which is odd, since you would expect that CEOs with more power concerning setting their own compensation, would be positively correlated with compensation.
Loss has a correlation with ROCE of -0.533, this is expected since a loss naturally
results in lower performance. Cash Flow is correlated with Sales for 0.800. This is a high correlation, but both variables will still be included in the regression analysis, otherwise there are no cash flow variables in the model which might impact results. Firm Size is correlated with Sales for 0.578 and with Cash Flow for 0.598, which is logical since a bigger firm has more sales.
5.2 Main results
<Insert Table 3>
In Table 3 the main results are presented. The results show a negative coefficient for I Spin-off
Year, but it is insignificant. This indicates that undertaking a spin-off is not associated with CEO compensation subsequent of a spin-off. Both Firm Size and Market-To-Book-Ratio have
20
a positive coefficient and are significant associated with CEO compensation. Next, the impact of doing a focusing spin-off is examined in a cross-sectional analysis.
<Insert Table 4>
Table 4 shows the cross-sectional regressions for non-focussing spin-offs in Column (1) and focusing spin-offs in Column (2). Column (1) shows a positive association between I
Spin-off Year and CEO Compensation, but the coefficient is not significant. This shows a
spin-off that is not focus increasing, does not have a significant effect on compensation. Column (2) gives a negative coefficient for I Spin-off Year and the results are significant (p<0.05). Therefore H2 is supported with the results of the regression analysis. So instead of an increase in CEO compensation, undertaking a focusing spin-off results in a negative effect. Although no effect is found for all spin-offs in the sample, when doing a cross-sectional analysis on focusing spin-offs, I do find supporting evidence for the hypothesis that CEO
compensation is affected by a spin-off.
This effect indicates that CEOs are not rewarded for their spin-off activities, even when it is a focus increasing spin-off, which is beneficial for shareholders. They are not rewarded, but actually punished for doing a focusing spin-off. An explanation for this could be the fact that a focusing spin-off results in a simpler firm, because the firm now has more cohesion in activities. The compensation of the CEO possibly decrease, because the firm is now easier to manage. This would also explain why no effect is found for non-focusing spin-offs, since a non-focusing spin-off does not make a firm less complex in terms of activities, but only in terms of size. Next the effect of size of the spin-off is examined.
<Insert Table 5>
Table 5 contains the results for the cross-sectional regressions when separating small spin-offs from large spin-offs. Column (1) shows the results for small spin-offs and Column (2) for large spin-offs. In Column (1) the coefficient for I Spin-off Year is negative and
21
significant (p<0.05), which indicates that CEO compensation decreases subsequent of a small spin-off. Column (2) shows a positive coefficient for I Spin-Off Year, but the effect is not significant. Therefore, I do not find evidence that supports the hypothesis that larger spin-offs have a greater effect on CEO Compensation than smaller spin-offs.
It is interesting though, that the results show a significant negative effect when examining small spin-offs. This seems contradictory to the notion that firm size affects CEO compensation, since a larger offs have a greater effect on firm size than smaller spin-offs. But considering the fact that significant negative effect is found when examining focusing offs but no significant effect was found when examining non-focusing spin-offs. Because this indicates that it is not the firm size that influences CEO compensation, but firm complexity. So a small off might be insignificant in creating value, but a large spin-off might be significant in creating shareholder wealth. To spin-spin-off a large part of the firm probably is not done without a good reason and it is more complex to complete a large spin-off. This could be the reason that CEOs are not punished for their spin-off activities if it is a significant part of the firm, but are punished when it is a less significant portion of the firm.
5.3 Robustness tests
<Insert Table 6>
In Table 6, a robustness test is presented where a different performance variable is used,
ROCE. Colum (3) and Column (4) show that I Spin-Off Year still has a significant negative
effect on CEO Compensation for focusing spin-offs and smaller spin-offs when including
ROCE instead of ROE. Column (1), Column (2) and Column (5) indicate an insignificant
effect of I spin off year on CEO Compensation, just like the main results have shown. Therefore this robustness test is satisfactory.
22
In Table 7, a robustness test is presented where lagged control variables are included instead of unlagged control variables. Column (3) shows significant negative effect of I
Spin-Off Year on CEO Compensation, this is the same as in the main analysis. In Column (4) there
is no significant effect found for small spin-offs, which differs from the main analysis. The coefficient sign (-) remains the same, so although the support for the effect small spin-offs have on CEO Compensation does not increase, the robustness test does not indicate that the conclusion drawn from the main results are incorrect. Column (1), Column (2) and Column (5) indicate an insignificant effect of I spin off year on CEO Compensation, just like the main results have shown. Considering all this, the robustness test is satisfactory.
<Insert Table 8>
In Table 8, a robustness test where an extra control variable L.CEO Compensation is included, following the model of Haynes et al. (2007). L.CEO compensation is significant for all regressions, which indicates that compensation is persistent over time. Colum (3) and Column (4) show that I Spin-Off Year still has a significant negative effect on CEO
Compensation for focusing spin-offs and smaller spin-offs when including this variable.
Column (1), Column (2) and Column (5) indicate an insignificant effect of I spin off year on
CEO Compensation, just like the main results have shown. Therefore this robustness test is
satisfactory.
5.4 Additional tests
To see what type of compensation is affected by spin-offs, I run the same regression model as in the main analysis, but this time I use fixed compensation and variable compensation as the dependant variable respectively. If the same results are given for one type of compensation then a conclusion can be drawn about what type of compensation is actually affected by doing a focusing and a small spin-off.
23
In Table 9, the results are presented when using Salary as the dependant variable.
Salary is the fixed part of total CEO compensation. Column (1) shows no significant effect of I Spin-Off Year on Salary when looking at all spin-offs. Column (2) shows a significant
positive effect of non-focusing spin-offs on Salary. This indicates that fixed compensation actually increases after a spin-off, although the impact on total compensation is likely insignificant, since the main analysis does not give significant results. Column (3), Column (4) and Column (5) do not show significant effects of I Spin-Off Year on Salary for focusing, small and large spin-offs, respectively. This indicates that the results found in the main analysis cannot be explained by an increase in fixed compensation. Next, the same regressions are performed with Variable Compensation as the dependant variable.
<Insert Table 10>
In Table 10, the results are presented when using Variable Compensation as the dependant variable. Column (1) displays the same results as the main analysis for all spin-off firms, I
Spin-Off Years does not have a significant impact on Variable Compensation. Column (2)
also shows the same results for non-focusing spin-offs as the main analysis. When looking at focusing spin-offs in Colum (3), A significant (p<0.05) negative effect is found for I Spin-Off
Year on Variable Compensation. In Column (4), where the impact of small spin-offs on Variable Compensation is examined, the results show a significant (p<0.05) negative impact
of I Spin-Off Year. Column (5) does not show a significant effect for large spin-offs.
All results from this additional test are the same as the main analysis. The analysis therefor clearly indicates which part of compensation is affected. Since the results are the same when using Variable Compensation as the dependant variable, compared to total CEO
Compensation the conclusion can be drawn that the effect found in the main analysis is
24
6. Limitations
Although various robustness tests indicate reliable results, this thesis has its limitations. The sample used in this thesis consists of US listed firms, so no direct conclusions from the results can be drawn for CEOs of firms in other parts of the world. To study the effects of spin-offs on CEO compensation in other countries, a study should be conducted with an international sample, or with samples from other areas separately.
In this thesis the only type of divestiture that is studied are spin-offs. It could be interesting to study what the effect of other type of divestitures have on CEO compensation. The sample also consists only of CEOs. It is also interesting what the effect of spin-offs is on compensation of other executives, like CFOs for example.
Finally, although the sample used in this thesis is of sufficient size, using a bigger sample of spin-offs might give stronger results. This could be combined with a study using an international sample, since this would naturally result in a large sample of spin-offs.
7. Conclusion
In summary, I do not find significant results when looking at all spin-off firms in the sample. But when doing a cross-sectional analysis, I do find interesting results. When looking at focusing spin-offs, the results show a significant negative association between doing a focusing spin-off and CEO compensation. No significant effect was found for non-focusing spin-offs. The same can be said about the cross sectional analysis of the size of spin-off. Although I do not find support for the hypothesis that large spin-offs incrementally affect CEO compensation, the results do in fact show that doing a small spin-off is negatively associated with CEO compensation. The additional test where I examined what type of compensation is affected by spin-offs also gives interesting results. The analysis shows that the effect found for focusing spin-offs and small spin-offs is the result of a decrease in variable compensation and not fixed compensation.
25
A possible explanation for these results are that CEOs are “punished” for simplifying their firm. Literature is clear that firm size has a positive effect on executive compensation. But I only find a significant negative association for focusing spin-offs and not for non-focusing spin-offs. This implicates that it is not firm size directly that is associated with executive compensation, but the complexity that comes with it. This also explains why I only find a significant negative association when looking at small spin-offs and no significant association when looking at large spin-offs. Because if firm size is the main determinant of CEO compensation then a bigger spin-off would result in a decrease in compensation. This is not the case. A possible explanation for this could be that CEOs are punished for simplifying there firm when the spin-off is insignificant. But when doing a spin-off of significant size, which has larger economic consequences, they are not punished.
This thesis has important implications for academia, since the effect of firm size on executive compensation is researched extensively, but the effect of firm complexity to a lesser degree. This holds potential directions for future research. The negative effect of focusing spin-offs on CEO compensation also has important implications for compensation setters, considering the positive effect a spin-off can have on a firm. The results of this thesis shows that CEOs are not awarded for spin-offs even when it increases focus. Compensation setters should take this into account, so that activities that are done in favour of shareholders are not discouraged. Because why would a CEO increase corporate focus through a spin-off, when this would result in a loss of compensation?
26
References
Anderson, M. C., Banker, R. D., & Ravindran, S. (2000). Executive compensation in the information technology industry. Management Science, 46(4), 530–547. https://doi.org/10.1287/mnsc.46.4.530.12055
Best, R. W., Best, R. J., & Agapos, A. M. (1998). Earnings Forecasts and the Information Contained in Spinoff Announcements. The Financial Review, 33(3), 53–68. https://doi.org/10.1111/j.1540-6288.1998.tb01382.x
Chang, W.-J., Hayes, R. M., & Hillegeist, S. A. (2016). Financial distress risk and new CEO compensation. Management Science, 62(2), 479–501.
https://doi.org/10.1287/mnsc.2014.2146
Chemmanur, T. J., & Yan, A. (2004). A theory of corporate spin-offs. Journal of Financial
Economics, 72(2), 259–290. https://doi.org/10.1016/j.jfineco.2003.05.002
Davis, J. H., Schoorman, F. D., & Donaldson, L. (1997). Toward a stewardship theory of management. The Academy of Management Review, 22(1), 20–47.
Desai, H., & Jain, P. C. (1999). Firm performance and focus: Long-run stock market performance following spinoffs. Journal of Financial Economics, 54(1), 75–101. https://doi.org/10.1016/s0304-405x(99)00032-x
Donaldson, L., & Davis, J. H. (1991). Stewardship Theory or Agency Theory: CEO Governance and Shareholder Returns. Australian Journal of Management, 16(1), 49– 64. https://doi.org/10.1177/031289629101600103
Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. The Academy of
Management Review, 14(1), 57-74.
Feldman, E. R. (2016). Corporate spin-offs and capital allocation decisions. Strategy Science,
1(4), 256–271. https://doi.org/10.1287/stsc.2016.0022
Gabaix, X., Landier, A., & Sauvagnat, J. (2014). Ceo pay and firm size: an update after the crisis. Economic Journal, 124(574), 59. https://doi.org/10.1111/ecoj.12084
Habib, M. A., Johnsen, D. B., & Naik, N. Y. (1997). Spinoffs and Information. Journal of
Financial Intermediation, 6(2), 153–176. https://doi.org/10.1006/jfin.1997.0212
Haynes, M., Thompson, S., & Wright, M. (2007). Executive Remuneration and Corporate Divestment: Motivating Managers to Make Unpalatable Decisions. Journal of
Business Finance & Accounting, 34(5–6), 792–818.
https://doi.org/10.1111/j.1468-5957.2007.02008.x
Hite, G. L., & Owers, J. E. (1983). Security price reactions around corporate spin-off announcements. Journal of Financial Economics, 12(4), 409–436. https://doi.org/10.1016/0304-405x(83)90042-9
Hope, O.-K., Porumb, V.-A., Rusanescu, S., & Vyas, D. (2020). Private Information and
Bank-Loan Pricing: The Effect of Upcoming Corporate Spinoffs. Article/Manuscript
in preparation.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. https://doi.org/10.1016/0304-405x(76)90026-x
Jensen, M. C., & Murphy, K. J. (1990). Performance pay and top-management incentives.
Journal of Political Economy, 98(2), 225–264.
John, T. A. (1993). Optimality of spin-offs and allocation of debt. The Journal of Financial
and Quantitative Analysis, 28(1), 139–160.
Keller, W., & Olney, W. W. (2020). Globalization and executive compensation. Journal of
International Economics, 103408, 103408–103408. https://doi.org/10.1016/j.jinteco.
2020.103408
27
corporate dividend policy. Review of Accounting Studies, 22(2), 753–790. Kostiuk, P. F. (1990). Firm size and executive compensation. The Journal of Human
Resources, 25(1), 90–105.
Krishnaswami, S., & Subramaniam, V. (1999). Information asymmetry, valuation, and the corporate spin-off decision. Journal of Financial Economics, 53(1), 73–112. https://doi.org/10.1016/s0304-405x(99)00017-3
Murphy, K. J. (1985). Corporate performance and managerial remuneration:an empirical analysis. Journal of Accounting and Economics, 7(1), 11–42. https://doi.org/10.1016/ 0165-4101(85)90026-6
Sun, F., Wei, X., & Huang, X. (2013). CEO compensation and firm performance. Review of
Accounting and Finance, 12(3), 252–267. https://doi.org/10.1108/raf-jan-2012-0006
Veld, C., & Veld-Merkoulova, Y. V. (2004). Do spin-offs really create value? The European case. Journal of Banking & Finance, 28(5), 1111–1135. https://doi.org/10.1016/s0378-4266(03)00045-1
Wachtell, Lipton, Rosen, & Katz. (2016). Spin-off guide. Retrieved from http://www.wlrk.com/webdocs/wlrknew/WLRKMemos/WLRK/WLRK.25192.16.pdf Wilson, A. H., & Higgins, E. J. (2001). Ceo pay/firm performance sensitivity in the insurance
Appendix
Variable definitions
Test variable
I Spin-Off Year An indicator variable which takes the value of 1 for the three years after the spin-off and 0 for the spin-off year and the two
years prior to the spin-off; Dependant variables
CEO Compensation Total annual executive compensation (form ExecuComp);
Salary Fixed annual compensation of the CEO (from ExecuComp);
Variable Compensation Total annual executive compensation minus fixed annual executive compensation (from ExecuComp);
Control variables
ROE Net income divided by average shareholders’ equity (from CompStat); ROCE EBIT (earnings before interest and tax) divided total assets (from CompStat);
Sales Total sales over the fiscal year in $ millions (from CompStat);
Market-To-Book-Ratio Market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity
(from CompStat);
Cash Flow Cash flow from operating activities in $ millions
Firm Size Natural logarithm of total assets (from CompStat);
29
Leverage Total debt divided total assets (from CompStat);
Loss An indicator variable that equals the value of 1 if net income is negative and 0 otherwise (from CompStat);
Interlock An indicator variable that equals the value of 1 if the CEO is involved in a relationship requiring disclosure in the
"Compensation Committee Interlocks and Insider Participation" and 0 otherwise (from ExecuComp);
Big 4 Auditor An indicator variable that equals the value of 1 if the firm is audited by a Big 4 audit firm and 0 otherwise (from CompStat).
Table 1 Descriptive statistics
Non-spin-off firms Spin-off firms
Variables N Mean Median Std. dev. N Mean Median Std. dev.
CEO Compensation ($ thousands) 43,264 4,503 2,674 5,244 6,421 7,198 4,999 6,794
Total Assets ($ millions) 43,281 9,343 1,656 27,227 6,423 21,175 4,895 45,857
Sales ($ millions) 43,249 4,217 1,106 10,025 6,424 10,181 3,581 17,012
Cash Flow ($ millions) 42,247 548.2 121.2 1,433 6,358 1,325 363.0 2,434
Market-To-Book-Ratio 42,468 2.913 2.060 3.516 6,351 3.019 3.167 3.778
ROE 39,665 0.0798 0.109 0.354 6,055 0.100 0.120 0.358
ROCE 42,703 0.0788 0.0766 0.102 6,418 0.0855 0.0809 0.0797
Leverage 31,244 0.226 0.191 0.211 4,735 0.268 0.241 0.195
Tangibility 41,649 0.258 0.181 0.240 6,216 0.280 0.211 0.231
This table presents the descriptive statistics for both non-spin-off firms and spin-off firms. CEO Compensation is total annual compensation a CEO received in $ thousands. Total Assets is total assets at fiscal year-end in $ millions. Sales is total sales over the fiscal year in $ millions. Cash Flow is cash flow from operating activities in $ millions. Market-To-Book-Ratio is market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity. ROE is net income divided by average shareholders’ equity. ROCE is EBIT (earnings before interest and tax) divided total assets. Leverage is total debt divided total assets. Tangibility is net property, plant and equipment divided by total assets.
31 Table 2 Correlations Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (1) CEO Compensation 1.000 (2) ROE 0.090*** 1.000 (3) ROCE 0.099*** 0.433*** 1.000 (4) Sales 0.466*** 0.094*** 0.056*** 1.000 (5) Market-to-Book-Ratio 0.131*** 0.265*** 0.257*** 0.033*** 1.000 (6) Cash Flow 0.472*** 0.099*** 0.084*** 0.800*** 0.058*** 1.000 (7) Firm Size 0.531*** 0.130*** 0.030*** 0.578*** -0.064*** 0.598*** 1.000 (8) Tangibility -0.061*** -0.009* 0.037*** 0.018*** -0.074*** 0.054*** 0.035*** 1.000 (9) Leverage 0.086*** -0.071*** -0.058*** 0.031*** -0.070*** 0.036*** 0.233*** 0.213*** 1.000 (10) Loss -0.087*** -0.434*** -0.533*** -0.088*** -0.098*** -0.115*** -0.180*** 0.008* 0.115*** 1.000 (11) Interlock -0.069*** 0.006 0.034*** -0.038*** 0.009* -0.039*** -0.064*** 0.027*** -0.039*** -0.021*** 1.000 (12) Big 4 auditor 0.125*** 0.024*** 0.110*** 0.103*** 0.037*** 0.098*** 0.103*** 0.152*** 0.097*** -0.015*** -0.020*** 1.000 This table represents the Pearson pairwise correlations for all observations (spin-off firms and non-spin-off firms). CEO Compensation is total annual compensation a CEO received in $ thousands. ROE is net income divided by average shareholders’ equity. ROCE is EBIT (earnings before interest and tax) divided total assets. Sales is total sales over the fiscal year in $ millions. Market-To-Book-Ratio is market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity. Cash Flow is cash flow from operating activities in $ millions. Firm Size is the natural logarithm of total assets. Tangibility is net property, plant and equipment divided by total assets. Leverage is total debt divided total assets. Loss is an indicator variable that equals the value of 1 if net income is negative and 0 otherwise. Interlock is an indicator variable that equals the value of 1 if the CEO is involved in a relationship requiring disclosure in the "Compensation Committee Interlocks and Insider Participation" and 0 otherwise. Big 4 Auditor is an indicator variable that equals the value of 1 if the firm is audited by a Big 4 audit firm and 0 otherwise. Significance level is indicated by *** (for p<0.01), ** (for p<0.05), * (for p<0.1).
Table 3
The effect of spin-offs
Dependant Variable:
CEO Compensation
Variables Main Regression
I Spin-Off Year -0.123 (0.0771) ROE -0.00551 (0.0232) Sales 0.170 (0.106) Market-To-Book-Ratio 0.0855*** (0.0256) Cash Flow -0.0308 (0.103) Firm Size 0.622*** (0.0777) Tangibility -0.104* (0.0557) Leverage -0.0748 (0.0520) Loss -0.0466 (0.0840) Interlock -0.425* (0.221) Big 4 Auditor 0.107 (0.215) Constant -0.134 (0.263) Observations 987 Number of firms 221 Industry FE YES Year FE YES
This table presents the results for the model which includes all spin-off firms. The dependant variable is CEO Compensation, which consists of all annual compensation of the CEO. The test variable I spin-off year is an indicator variable which takes the value of 1 for the three years after the spin-off and 0 for the spin-off year and the two years prior to the spin-off. The model includes industry fixed effect and year fixed effects. ROE is net income divided by average shareholders’ equity. Sales is total sales over the fiscal year in $ millions. Market-To-Book-Ratio is market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity. Cash Flow is cash flow from operating activities in $ millions. Firm Size is the natural logarithm of total assets. Tangibility is net property, plant and equipment divided by total assets. Leverage is total debt divided total assets. Loss is an indicator variable that equals the value of 1 if net income is negative and 0 otherwise. Interlock is an indicator variable that equals the value of 1 if the CEO is involved in a relationship requiring disclosure in the "Compensation Committee Interlocks and Insider Participation" and 0 otherwise. Big 4 Auditor is an indicator variable that equals the value of 1 if the firm is audited by a Big 4 audit firm and 0 otherwise. Robust standard errors are reported in the brackets. Significance level is indicated by *** (for p<0.01), ** (for p<0.05), * (for p<0.1).
33
Table 4
The effect of focusing spin-offs
Dependant Variable:
CEO Compensation
(1) (2)
Variables Non-focusing Focusing
I Spin-Off Year 0.00170 -0.219** (0.108) (0.108) ROE 0.00303 -0.0165 (0.0269) (0.0417) Sales 0.136 0.133 (0.103) (0.116) Market-To-Book-Ratio 0.0721* 0.100*** (0.0433) (0.0299) Cash Flow -0.0416 0.000551 (0.0903) (0.135) Firm Size 0.655*** 0.687*** (0.123) (0.101) Tangibility -0.131 -0.157*** (0.126) (0.0573) Leverage -0.0705 -0.121** (0.0836) (0.0612) Loss -0.306** 0.195* (0.141) (0.104) Interlock -0.223 -0.305 (0.986) (0.218) Big 4 Auditor 0.581* -0.0276 (0.327) (0.195) Constant -0.431 -0.0441 (0.327) (0.353) Observations 404 583 Number of firms 94 127
Industry FE YES YES
Year FE YES YES
This table presents the results for the model which includes non-focusing spin-offs in Column (1) and focusing spin-offs in Column (2). The dependant variable is CEO Compensation, which consists of all annual compensation of the CEO. The test variable for both models, I spin-off year, is an indicator variable which takes the value of 1 for the three years after the off and 0 for the off year and the two years prior to the spin-off. The model includes industry fixed effect and year fixed effects. ROE is net income divided by average shareholders’ equity. Sales is total sales over the fiscal year in $ millions. Market-To-Book-Ratio is market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity. Cash Flow is cash flow from operating activities in $ millions. Firm Size is the natural logarithm of total assets. Tangibility is net property, plant and equipment divided by total assets. Leverage is total debt divided total assets. Loss is an indicator variable that equals the value of 1 if net income is negative and 0 otherwise. Interlock is an indicator variable that equals the value of 1 if the CEO is involved in a relationship requiring disclosure in the "Compensation Committee Interlocks and Insider Participation" and 0 otherwise. Big 4 Auditor is an indicator variable that equals the value of 1 if the firm is audited by a Big 4 audit firm and 0 otherwise.
34
Robust standard errors are reported in the brackets. Significance level is indicated by *** (for p<0.01), ** (for p<0.05), * (for p<0.1).
35
Table 5
The effect of spin-off size
Dependant Variable:
CEO Compensation
(1) (2)
Variables Small Large
I Spin-Off Year -0.281** 0.0267 (0.121) (0.116) ROE -0.0397 0.00350 (0.0256) (0.0450) Sales 0.326** 0.0218 (0.137) (0.0807) Market-To-Book-Ratio 0.0916*** 0.0822* (0.0321) (0.0432) Cash Flow -0.209** 0.191** (0.0888) (0.0875) Firm Size 0.538*** 0.533*** (0.194) (0.109) Tangibility -0.0786 -0.153** (0.133) (0.0697) Leverage -0.135 -0.0852 (0.0964) (0.0584) Loss -0.193 -0.0299 (0.117) (0.148) Interlock -1.182*** -0.338 (0.228) (0.487) Big 4 Auditor 0.180 0.470 (0.210) (0.356) Constant 0.116 -0.386 (0.399) (0.376) Observations 405 426 Number of firms 87 96
Industry FE YES YES
Year FE YES YES
This table presents the results for the model which includes small spin-off size in Column (1) and large spin-off size in Column (2). The dependant variable is CEO Compensation, which consists of all annual compensation of the CEO. The test variable for both models, I spin-off year, is an indicator variable which takes the value of 1 for the three years after the spin-off and 0 for the spin-off year and the two years prior to the spin-off. The model includes industry fixed effect and year fixed effects. ROE is net income divided by average shareholders’ equity. Sales is total sales over the fiscal year in $ millions. Market-To-Book-Ratio is market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity. Cash Flow is cash flow from operating activities in $ millions. Firm Size is the natural logarithm of total assets. Tangibility is net property, plant and equipment divided by total assets. Leverage is total debt divided total assets. Loss is an indicator variable that equals the value of 1 if net income is negative and 0 otherwise. Interlock is an indicator variable that equals the value of 1 if the CEO is involved in a relationship requiring disclosure in the "Compensation Committee Interlocks and Insider Participation" and 0 otherwise. Big 4 Auditor is an indicator variable that equals the value of 1 if the firm is audited by a Big 4 audit firm and 0 otherwise. Robust standard
36
errors are reported in the brackets. Significance level is indicated by *** (for p<0.01), ** (for p<0.05), * (for p<0.1).
37
Table 6
Robustness test: ROCE instead of ROE
Dependant Variable:
CEO Compensation
(1) (2) (3) (4) (5)
Variables Main Regression Non-focusing Focusing Small Large
I Spin-Off Year -0.120 0.0113 -0.214** -0.271** 0.0180 (0.0750) (0.0997) (0.104) (0.120) (0.115) ROCE 0.0407 0.130 -0.0490 0.0631 0.145* (0.0560) (0.0800) (0.0683) (0.104) (0.0801) Sales 0.177* 0.138 0.142 0.325** 0.0259 (0.107) (0.101) (0.119) (0.135) (0.0783) Market-To-Book-Ratio 0.0786*** 0.0730* 0.102*** 0.0668** 0.0644** (0.0238) (0.0424) (0.0245) (0.0336) (0.0327) Cash Flow -0.0368 -0.0487 0.000427 -0.215** 0.173** (0.104) (0.0867) (0.138) (0.0846) (0.0878) Firm Size 0.622*** 0.686*** 0.672*** 0.558*** 0.561*** (0.0748) (0.119) (0.0953) (0.187) (0.104) Tangibility -0.0973* -0.114 -0.157*** -0.0747 -0.142** (0.0550) (0.119) (0.0570) (0.131) (0.0699) Leverage -0.0788 -0.0813 -0.119* -0.145 -0.109* (0.0507) (0.0787) (0.0610) (0.0949) (0.0580) Loss -0.00732 -0.181 0.152 -0.0973 0.0735 (0.0956) (0.164) (0.115) (0.147) (0.145) Interlock -0.370* -0.314 -0.319 -1.181*** -0.235 (0.221) (0.653) (0.212) (0.245) (0.396) Big 4 Auditor 0.0896 0.542* 0.00925 0.266 0.410 (0.206) (0.301) (0.180) (0.178) (0.377) Constant -0.109 -0.362 -0.0562 -0.0283 -0.320 (0.253) (0.316) (0.344) (0.376) (0.381) Observations 1,003 412 591 407 431 Number of firms 223 95 128 87 96
Industry FE YES YES YES YES YES
Year FE YES YES YES YES YES
This table presents the results for the models which includes ROCE instead of ROE. All spin-off firms are included in Colum (1). Column (2) includes non-focusing spin-offs. Colum (3) includes non-focusing spin-offs. Column (4) includes small spin-offs. Column (5) includes large spin-offs. The dependant variable is CEO Compensation, which consists of all annual compensation of the CEO. The test variable for both models, I spin-off year, is an indicator variable which takes the value of 1 for the three years after the spin-spin-off and 0 for the spin-off year and the two years prior to the spin-off. The model includes industry fixed effect and year fixed effects. ROCE is EBIT (earnings before interest and tax) divided total assets. Sales is total sales over the fiscal year in $ millions. Market-To-Book-Ratio is market capitalization (common shares outstanding divided by share price at fiscal year-end) divided by shareholders’ equity. Cash Flow is cash flow from operating activities in $ millions. Firm Size is the natural logarithm of total assets. Tangibility is net property, plant and equipment divided by total assets. Leverage is total debt divided total assets. Loss is an indicator variable that equals the value of 1 if net income is negative and 0 otherwise. Interlock is an indicator variable that equals the value of 1 if the CEO is involved in a relationship requiring disclosure in the "Compensation Committee Interlocks and
38
Insider Participation" and 0 otherwise. Big 4 Auditor is an indicator variable that equals the value of 1 if the firm is audited by a Big 4 audit firm and 0 otherwise. Robust standard errors are reported in the brackets. Significance level is indicated by *** (for p<0.01), ** (for p<0.05), * (for p<0.1).
