Master Thesis
The effects of family ownership on executive compensation:
Evidence from the United States
Name: Joris Meens Student ID: 6118887 MSc: Business Economics Specialization: Finance Supervisor: dr. F.S. Peters Number of Credits: 15 EC Abstract
This study investigates executive remuneration within family owned businesses. The main focus of this research is to compare the structure of CEO remuneration packages between family and non-family firms. Therefore a sample of 623 unique firms during the years 2006 - 2010 is used. Evidence is found that family firms use less equity based compensation. It is also found that family firms where the founder is active as either CEO or board member reduce the use of equity compensation even further.
2 Statement of Originality
This document is written by Student Joris Meens who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Table of content
Introduction ... 4
Literature Review ... 6
Introduction to remuneration ... 6
Introduction to Family ownership ... 8
Research aim and expectations ... 11
Sample and Methodology ... 11
Family variables ... 11 Methodology ... 12 Descriptive statistics ... 13 Hypotheses ... 17 Empirical Analysis ... 17 Robustness... 22
Robustness remuneration structure ... 22
Conclusion and discussion ... 23
4
1. Introduction
This research aims at providing new insights on CEO compensation within family owned firms. Existing academic literature extensively suggests that family owned businesses perform relatively well on their financials compared to nonfamily firms (Maury, 2006). The most used explanation for this outcome is that family firms typically experience less agency problems between shareholders and management. Family firms often have an active role for family members on their higher management positions. These family managers are commonly associated with a strong connection towards the company’s heritage. This is believed to strengthen the balance between self- and company interest towards the company, reducing agency conflicts (Anderson & Reeb 2003, Barontini & Caprio 2006). The second most used explanation is that family owned firms are often subscribed to have an increased focus on long term performance and risk management. Due to the stronger emotional bond, more attention is paid to the company’s continuity. This reduces the likeliness of excessive risk taking behavior and improves the attention for long-term stability. (Anderson & Reeb, 2003 and Barontini & Caprio 2006).
Over the recent years, the influence of CEO compensation on company performance has become a heavily debated topic. The connection between remuneration and firm
performance was first pointed out by Jensen and Murphy (1990), who famously stated that what matters in executive compensation is not how much you pay, but how you pay it. Today executive compensation is well recognized as one of the key issues in corporate governance. Compensation packages are seen as one of the primary tools for aligning managers and shareholders’ interests. The attention for this topic grew even more during the recent financial crisis, where CEO remuneration became a fierce topic in public debate (Kirkpatrick, 2009).
Academic literature suggests that managerial behavior can to some extend be
controlled by remuneration. The ultimate goal in designing CEO compensation packages is to reduce agency problems between shareholders and managers, and to discourage managerial short-termism and excessive risk taking behavior (Anderson & Reeb, 2003). This raises the question to what extent the appointed reduced agency problems and long-term focus for family firms is reflected in their remuneration strategy.
Various papers investigate the effects of firm characteristics on executive
compensation. Most of these researches include inputs on firm risk characteristics, size, financials and ownership structures. Research within this topic that pays special attention to family ownership is scarce though, despite the significant role of family owned firms on the wider economy. Approximately one-third of all S&P 500 companies is family owned
5
(Anderson & Reeb, 2003 & Mauray, 2006). Additional research that specifically investigates executive remuneration within family owned firms could lead to new insights and a better understanding of the use of both different compensation strategies and family firms in general. The central research question of this research will therefore be:
“What is the influence of family ownership on executive compensation? Evidence from the United States.”
To create a complete answer to the above stated question, the research will be executed in two parts. The first part is concerned with investigating whether the expected reduced classical agency problems in family run businesses indeed is reflected in the structure of the
remuneration packages. In order to do so, the structure of family firm remuneration packages will be analyzed and compared to remuneration packages of nonfamily firms. The second part of the empirical tests will take a closer look at family firms and investigate if
remuneration is further affected when the CEO is a member of the shareholding family or when the founder of the firm is still present.
The empirical results do report a difference in remuneration structure. A positive and significant effect was found for family ownership on the use of both base salary and cash bonus. Family ownership has a negative significant effect on the use of restricted stock. No significant relation was found for stock options. However, when equity based compensation was taken collectively, a significant negative relation was reported, further proving the greater role for cash based compensation within family firms. No further significant relation was found for the remuneration structure when the CEO belongs to the shareholding family. The presence of the founder as either CEO or board member further reduces the use of equity based compensation.
This research is organized in the following way. Section II discusses existing academic literature on executive compensation and family firm ownership. Section III
extensively describes the empirical methodology, the used samples and the relevant summary statistics. In section IV the empirical results are presented. In section V the robustness of the found results is tested, followed by the conclusion and discussion in section VI.
6 2. Literature Review
Introduction to remuneration
CEO remuneration is a widely discussed and investigated topic. Early debate regarding CEO remuneration primarily focused on levels rather than methods of payment. This changed after Jensen and Murphy’s (1990) innovative article that gave new insights in the effects of
different payment methods on company performance. The main conclusion of their research was that the effectiveness of CEO compensation packages is determined by the method of payment rather than by how much is paid. Since the appearances of this article academic research has primarily focused on the effects of different remuneration packages on company performance. CEO compensation became recognized as a tool for aligning managers’ and shareholders’ interests and was therefore valuable for shareholders in their goal to maximize company value (DeFusco et al, (1990)). The awareness that CEO compensation could be used as a device to somewhat control a managers actions also linked payment methods to corporate governance (Core, Holthausen and Larcker, 2000). One of the highly recognized primary goals of CEO compensation is to prevent managers’ excessive self-interested behavior. This puts an even greater weight on the relevance of CEO compensation in companies with lower governance standards.
Although a lot has been written about the effects of certain remuneration packages on company performance and corporate governance, less has been written about what company characteristics are key in determining the optimal remuneration structure. The following paragraphs briefly explain the most used payment methods and describe what existing literature claims to be the most relevant company characteristics regarding CEO compensation.
Payment methods
In order to align managers’ and shareholders’ interests, companies typically use four payment methods. These most commonly considered forms of executive compensation are base salary, cash bonus, restricted stock awards and stock options (Jensen and Murphy’s (1990), Gopolan et al. (2014)). The variety of payment methods is designed to give managers’ different
incentives, which helps companies to create a balance between shareholders’ interests and managers risk exposure. Base salary is the cash payment that a manager receives regardless of any performance measure. Cash bonus is payed if certain predetermined performance goals are met. Cash bonuses are mostly used as an incentive for achieving short-term goals. Restricted stock and stock options are forms of equity based compensation that get awarded
7
when certain performance goals have been reached. Restricted stocks are stocks awarded to the manager and directly link a manager’s wealth to the company’s market value. Awarded stock options give managers the option to purchase stock or liquidate the option values at some time in the future at a predetermined price. The option therefore only has value to the manager if the stock price at time T exceeds the exercise price.
Company characteristics and remuneration
In 1995, Mehran published a paper which describes the relationship between compensation package structures, ownership types and board structures. The paper states that CEO
monitoring options are one of the primary concerns in CEO compensation. The ease by which a CEO’s actions can be monitored has a large influence on the chance of CEOs acting in a way that does not match shareholders’ interests. According to Mehran, the primary goal in designing a CEO compensation package is the alignment of shareholders- and directors interests. Therefore, remuneration should be used in such a way that the likelihood of managerial actions that do not meet shareholders’ interests is reduced to a minimum.
Mehran states that the board of directors is one of the primary organs that is concerned with CEO monitoring. A directors board typically consists of both company in- and outsiders. Because company insiders are often already involved in the company’s daily process, it is believed that outsiders are beneficial for improving both impartial monitoring and supporting external investor interests. It is therefore stated that the composition of this board should be reflected in the CEO compensation structure.
Mehran’s (1995) empirical results indeed show that firms with a higher percentage of outsiders in their board of directors use more equity based compensation. The explanation given for this outcome is that compensation structures are determined collectively within the board. A larger number of outsiders is therefore thought to extent the pressure of external shareholders in their desire to use more equity based compensation.
These results are largely in line with Ryan and Wiggins (2001), who performed a more extensive research regarding the same topic. In this paper, the aforementioned most
commonly used payment methods are analyzed and regressed on a large variety of company and CEO characteristics. The included set of company characteristics is much more extensive compared to Mehran (1995) and is used to test pre-stated potential agency problems between shareholders and managers and their relation with the structure of compensation packages.
8
alternative for monitoring. In their article, a positive and highly significant relation is found between the weight of restricted stock awards and the percentage of outside directors.
Apart from monitoring difficulties, special attention is paid to what is known as the
‘investment opportunity hypothesis’. This describes the relation between firm specific risk
and the use of incentives. Firms that experience high growth opportunities base a large part of their market value on pending investments. Since the actual outcome of these investments is highly uncertain, monitoring managers’ actions becomes difficult. This also counts for firms that operate in risky markets and experience large operating cash flow volatilities. It is therefore claimed that firms with higher company specific risks should pay their managers using long-term equity-based awards. The empirical results confirm this statement by showing a significant positive relation between both R&D intensity and operating cash flow volatility and stock options and a negative relation between R&D intensity/operating cash flow volatility and both restricted stock awards and cash bonus.
The explanation for this outcome is based on the different nature of the used equity awards. Restricted stock awards are company stocks awarded to the CEO. They represent a value that matches the company’s current stock price. By awarding stock, a manager’s wealth is directly connected to the company’s market value. This residual risk could make mangers reluctant to engage in risky investments. Stock options on the other hand give managers the right to purchase stocks at some future date at a predetermined price. This only creates wealth for a manager if the stock prices exceeds the option’s exercise price. Therefore both
investment intensive companies and companies with high operating income fluctuations are thought to encourage their managers to invest by awarding options rather than restricted stock and cash bonus.
Introduction to Family ownership
Family firms play a significant role in the corporate landscape. As mentioned before, roughly one third of all S&P 500 firms is claimed to be family owned (Anderson & Reeb (2003), Mauray, (2005)). Academic literature extensively suggests that family owned firms typically have different standards on several company characteristics. Examples of such differences can be found in firm financials and governance standards. The following paragraphs will describe the most important differences according to existing literature.
9
Family firm performance
Anderson & Reeb (2003) studied the performance of US family owned firms. Their results indicate that family owned firms on average outperform non-family firms. This is explained by the assumption that the corresponding families are highly invested in their own firms and therefore their own wealth is directly connected to the company’s value. This boosts
incentives to maximize company value but also reduces the likeliness of freeriding. Similar results are found by Maury (2006), who also claims that family firms typically face less agency costs.
Family firm corporate governance
It is often stated that part of the supreme performance is due to differences in governance structures that commonly characterize family firms. Klein et al (2005) confirms that the alignment between managers and shareholders’ interests is better within family owned firms. As mentioned before, family firms often have an active role for family members on the higher management positions (Anderson & Reeb 2003, Barontini & Caprio 2006). It is believed that the stronger bond with the company’s heritage and the direct contact line between the family shareholders and family managers serves as an alternative for monitoring and goal alignment, and reduces the likeliness of excessive risk taking behavior.
The same emotional bond is used as an explanation for the long-term focus that is often subscribed to family firms. Although literature extensively suggests that this long-term focus indeed exists, empirical evidence is scarce. James (1999) investigates the expected long-term horizon by comparing return on investments between family and non-family firms. Empirical results show that family firms indeed have a better return on investment. The explanation used is that families who possess a firm are often engaged in a long-term commitment with the company. This leads to an extended investment horizon which further urges the relevance for efficient long term investment strategies. Anderson and Reeb (2003) state that long-term vision within family owned firms can be proven by their supreme stock performance and lower costs of long-term debt. Chen et al (2008) tried to prove long-term vision for family firms by looking at stock return volatilities. Lower volatilities would indicate that a company is less vulnerable for sudden operational or market changes,
indicating a more steady and long-term focused company climate. Although they find a lower volatility for family firms, the difference with non-family firms is not significantly different.
10
Founder presence
Literature is fairly consistent in stating that the proposed performance and governance characteristics are only valuable if a family member is present as either the CEO and/or as member of the board of directors. The opinion about what is the most valuable role for founder families differs among papers. Burkart et al (2003) & He (2008) found that family ownership creates value only if the founder or a descendant is present as CEO and/or chairman. The combination of block ownership and a high managerial position within the company is required to fully enforce the advantages that come with the aforementioned respect to the company’s heritage. Villalonga and Amit (2006) even sharpen this conclusion by stating that family ownership only creates value if the founder of the firm executes a top position within the firm. The founder being present reduces the likeliness of heirs getting into conflict over company decisions and works as a powerful extra monitoring device over other managers. Andres (2008) findings lay somewhere in the middle, by stating that family
ownership is only valuable if founder family is present and that the highest results are reached if the founder himself is present as either CEO or board member.
Family ownership and CEO compensation
There are many reasons to believe that CEO compensation packages could differ between family and non-family firms. As mentioned before, literature is nearly unanimous in stating that incentive alignment, monitoring possibilities and governance standards play a key role in designing an effective compensation package. Literature also claims that family firms
typically differ in these subjects compared to non-family firms. Despite this, academic research that specifically investigates CEO remuneration within family firms is scarce.
Gomez-Mejia et al. (2003) did some research regarding the level of CEO
compensation in public family firms. One of their conclusions is that the total CEO payment in a family firm is less when the CEO belongs to the family. This is less the case when the CEO is an outsider. The given explanation is that family insiders benefit job protection, which reduces personal risk. Outsiders experience far less job protection and therefore demand a higher compensation.
McConaughy (2004) did some basic research on CEO compensation in family firms and found that CEOs that are member of the founding family earn less incentive based salary compared to companies that are family owned but have an outside CEO. Again, the reduced agency problem is used as an explanation for this outcome.
11
Gomez et al (1987), explain this however by the idea that because the CEO is a member of the company owning family, his personal wealth is already likely to be connected to the company’s market value. This would make the family CEO less sensitive to additional capital compared to family outsider CEOs, who don’t possess a comparable stake.
Research aim and expectations
As presented by the literature study above, family owned firms tend to differ in areas that are referred to as key issues in remuneration research. This paper attempts to investigate whether the claimed differences in reality are reflected in the remuneration strategies. Family
shareholders are often said to be highly concerned with the company’s well-being. Their involvement is thought to have a monitoring function over managers and reduces the
likeliness of agency conflicts. It is also claimed that family firms are typically operated with greater risk aversion and an increased long-term interest. Based on the literature that appoints the relevance of equity based compensation as an alternative to monitoring and to align optimal investment horizons, it is expected that family firms use less equity compensation. It is also expected that the presence of the founder and that CEO family membership of the further reduces the use of equity based compensation. In the following section the used data and methodology will be described.
3. Sample and Methodology
This section will give an extensive description on the collected data and the methodology used to conduct the empirical tests. The empirical tests for this research can be divided into two parts. In the first part, the influence of family ownership on the structure of executive compensation will be examined. The second part will investigate the effects of the founder being present and the CEO belonging to the shareholder family.
Family variables
In this research, the same definition for family ownership will be used as presented by Anderson et al. (2012). A firm is considered family owned if a single family possess a stake of at least 5%. To be able to identify family firms, the dataset from Anderson’s website (2015) is used. This dataset contains the 2000 largest U.S. companies in terms of total assets from 2001 until 2010, including a dummy variable that indicates whether a firm is family owned. Financial companies are not included in the set due to their complex company profile.
12
To test more specific effects of family ownership on remuneration, a number of family related variables will be added to the dataset. The selected variables are chosen based on their
predicted relation to the compensation subject according to literature. The chosen family variables are: (1) family owns stake of at least 5%, (2) CEO is family of founder, (3) founder is present as either CEO or board member. The first mentioned variable is identified using Anderson’s dataset, the latter are extracted from ISS directors database or manually collected from SEC filings and company websites if missing.
Dependent variables
The weighted payment methods as presented in the literature part serve as the dependent variables for the regressions. All information regarding compensation is extracted from Compustat’s Execucomp database. Due to the in 2006 introduced FASB123 reporting standards, Execucomp changed its calculation methods for determining executive
compensation levels. In order to create the most recent insights, remuneration data from 2006 onwards will be used for the empirical tests.
Control variables
The regression models will include a limited number of control variables. The set of variables is a combination of the controls used by Harley and Wiggans (2001) and Mehran (1995), of which the most important ones are already discussed in the literature review. The control variables can be divided into company financial measures and management specifics. The used financial controls are Market-to-book value of assets, R&D intensity, Capital
expenditure intensity, Firm size, Business risk, long term debt ratio and convertible debt ratio. All the financial control variables are extracted from the Compustat Core Financial database.
The management specific control variables are percentage of outsiders on the board of directors, CEO duality, CEO age and CEO stock ownership. The first two are collected from the ISS Directors database, the latter from Execucomp. A more specific definition of the control variables and their expected relation to the payment methods can be found in the appendix.
Methodology
In order to analyze the compensation structure of family firms six compensation forms are used: (1) percentage of total compensation in base salary, (2) percentage of total
13
(4) percentage of total compensation in awarded restricted stock, (5) percentage of total compensation paid in cash (Base Salary + Cash Bonus) and (6) percentage of total compensation that is equity based (Restricted Stock + Stock Option awards). Total compensation is defined as the dollar value sum of base salary, cash bonus, stock option awards, restricted stock awards and all other compensation as reported in Execucomp. Other compensation will not be treated as a separate compensation method in the regressions since the nature, rationale and methods of these payments are very difficult to trace and compare. To create a first insight on the effects of family ownership on executive compensation a univariate two group mean t-test will be performed, comparing the mean values of the remuneration portions between family- and non-family firms. This will be followed by regressions of the payment portions on the above described variables. Because the dependent variables are fractions that are truncated between 0 and 1 (or 0% and 100%), the regressions will be conducted using the Tobit method for truncated data corrected for robust standard errors. The regression is expressed in the equation presented below:
% Payment method i = β0 + β2 CEO is family and block shareholder + β3 Founder is present + β4 Control variables
To further specify the effects of family ownership on remuneration, similar regressions will be performed using only the family owned firms. This makes it possible to test if the stated effects of the presence of the company’s founder and whether the CEO is a family member indeed exist. This regression will have the following form:
% Payment method i = β0 + β1 CEO is (family of ) founder + β2 Founder is present + β3 Control variables
Descriptive statistics
Table 1 displays the descriptive statistics of the final sample used for this research. 623 unique firms are included in the dataset, of which 141 (22.6%) are family owned. The total dataset contains 2465 year observations and runs from 2006 until 2010.
The sample’s total compensation averages US$ 5.15 million dollar and ranges from US$ 0 to US$ 128.21 million. Bonus and other compensation are on average the smallest fractions of total compensation with a weight of 3.69% and 5.77% and a value of US$ 195.069 and US$ 303.925. Salary, restricted stock and stock options on average all account
14
for roughly 30% of total compensation. Restricted stock has the highest average dollar value which amounts US$ 2.013.920. The average CEO age is 55.59 years. More details can be found in the table below.
Tabel 1: Descriptive statistics on compensation data for full sample including both family and non-family firms. The period ranges from 2006 until 2010.
VARIABLES N Mean Stand. Dev. Min Max
Panel A: Dollar Value
Salary 2,465 $865.945 $360.652 $0 $3.300.000 Bonus $195.069 $662.912 $0 $8.624.545 Stock options $1.773.881 $3.457.073 $0 $78.421.032 Restricted stock $2.013.920 $2.876.727 $0 $26.397.500 Other compensation $303.925 $2.130.411 $0 $100.202.100 Total compensation $5.152.925 $2.876.727 $0 $128.208.404 Panel B: Percentages % Salary 28.54% 20.32% 0% 100% % Bonus 3.69% 10.33% 0% 82.41% % Restricted stock 32.09% 25.52% 0% 95.89% % Stock options 29.89% 24.20% 0% 100% % Other Compensation 5.77% 10.98% 0% 100%
Panel C: Age and Family
CEO age 55.59 6.97 32 94
Total firms 623
Family owned firms 141 (22.63%)
CEO Duality 65.49%
Table 2 displays the firm characteristics that are thought to explain the CEO remuneration structure. The firms’ average total assets amounts US$11.3 billion dollar with a median of US$2.4 billion. The market to book value averages 1.539 while the median lays slightly lower at 1.308. Capital expenditures average 3.09% of total assets, compared to 2.32% of R&D expenditures. Long-term debt ratio and convertible debt ratio average 0.267 and 0.0376 respectively. The percentage of outside directors is added to the regression as a proxy for CEO monitoring. The firms on average have a board of directors that consists for 80% out of company outsiders. Further details are presented in table 2.
15
Table 2: Descriptive statistics firm characteristics of full sample running from 2006 until 2010. Standard
Median Mean deviation Min Max
Cashfl. Vol. 108.389 450.819 1.261.827 7.294 14.438.970 Total Assets $2.409.921 $11.344.260 $42.463.930 $57.442 $797.769.000 Market-Book 1.308 1.539 0.904 0.201 8.212 R&D/assets 2.32% 4.43% 5.84% 0% 67.99% CapEx/assets 3.09% 4.02% 3.21% 0.00% 25.16% LTD ratio 0.129 0.267 0.624 0 16.50 Conv. Debt. 0 0.0376 0.133 0 2.767 Outside Direc. 0.8 0.783 0.117 0.143 0.941
The dollar value of firm size should be multiplied by 1000 to report the actual value. Operating cash flow volatility is calculated using 10 years of monthly operating cash flow data as reported in Compustat. This information is used to determine the standard deviation, which will be used in the regressions as the estimator for operating cash flow volatility.
In order to find out if the presence of the founder and whether the CEO belongs to the family has any effects on remuneration, a subsample including only the family owned firms will be analyzed. CEO and firm characteristics of this subsample are presented in table 3. In the subsample an average total compensation of US$4.451 million was paid which, coherent to literature, is lower than the average of US$5.515 million from the full dataset. Both cash payment components lay slightly above the full-sample average, with 36.38% for base salary and 5.61% for cash bonus. Restricted stock and stock options both represent around 25% of total compensation with, similar to the full dataset, a slightly higher weight for stock options. 48.03% of the family owned firms has a CEO in position that belongs to the founding family. The average CEO age is 57.45 years which is fairly similar to full sample’s value of 55.59. 55.22% of the family firms still has its founder active as either CEO or board member.
Of the family firm CEO’s, 47.18% has a double function as chairman. This is considerably lower compared to the full sample in which 65.49% has a double function. The percentage of outsiders on the board of directors is also lower with an average of 70%, compared to 78% for the non-family firms. With a median total assets of US$ 1.515 dollar, the subsample consist of smaller firms. Both R&D- and capital expenditure intensity and debt ratios are comparable to the full company set. More details can be found in table 3.
A correlation matrix is presented in the appendix. The matrix reports no high
correlations between the variables. All variables are included in the table, except for the year fixed effects.
16
Table 3: Descriptive statistics family firm subset, Descriptive statistics on compensation data for family firm subset. The sample ranges from 2006 until 2010.
VARIABLES N Mean Stand. Dev. Min Max
Panel A: Dollar Value
Salary 498 $800.736 $349.742 $0 $2.500.000 Bonus $236.025 $723.857 $0 $5.613.503 Stock Options $1.864.772 $6.112.805 $0 $78.421.000 Restricted Stock $1.304.767 $2.480.838 $0 $21.540.500 Other Compensation $244.878 $542.973 $0 $5.737.161 Total Compensation $4.451.346 $7.157.518 $0 $80.914.950 Panel B: Percentages % Salary 36.38% 24.69% 0% 99.99% % Bonus 5.61% 14.09% 0% 82.41% % Stock Options 27.58% 25.77% 0% 100% % Restricted Stock 23.17% 24.41% 0% 91.52% % Other Compensation 7.24% 13.91% 0% 100% Panel C: Family CEO age 57.45 8.89 32 94 CEO family 48.03% Founder present 55.22% CEO duality 47.18% Panel D: Specifics
Median Mean Stand. Dev. Min Max
Cashfl. Vol. 70.02 365.94 1.063 7.35 7.113
Total Assets (US$) 1.514.98 8.716.537 28.566.530 171.9 279.264.000
Market-Book 1.390 1.606 0.913 0.201 6.157 R&D/assets 1.62% 4.01% 6.39% 0% 53.8% CapEx/assets 3.52% 4.27% 3.04% 2.36% 18.42% LTD ratio 7.14% 26.35% 98.49% 0% 164.4% Conv. Debt. 0% 1.89% 6.69% 0% 55.20% Outside direc. 0.71 0.698 0.132 0.143 0.938
17
Hypotheses
In order to create clear answers to the stated problem sets two hypotheses will be tested. The hypothesis are designed based on what literature claims to be possible determinants of differences in family firm compensation. The hypotheses are presented below.
Hypothesis 1: Family owned firms pay their CEO using relatively less equity based compensation
Hypothesis 2: Non family CEO’s operating a family firm receive a larger part of their compensation in equity form
4. Empirical Analysis
In this section the outcomes of the empirical models will be presented. Table 4 displays the results of a two group mean t-test, comparing the mean values of the remuneration portions between family- and non-family firms. For all four payment methods the null hypotheses are rejected, suggesting that the compensation structure differs between family and non-family firms. The found means for salary and bonus are larger for the family firm group, appointing that family firm CEOs receive a larger portion of their compensation in salary and cash bonus while non-family firms use relatively more restricted stock and stock options.
Table 4 Two group mean t-test comparing family and non-family firms
Non-family-owned Family-owned
VARIABLES Mean Mean Difference t-value Critical t-value
% Salary 0.2656 (0.0042) 0.3638 (0.0111) -0.0983 -8.307 (T<t) 0.0000 % Bonus 0.0321 (0.0020) 0.0561 (0.0063) -0.0240 -3.617 (T<t) 0.0002 % Rest. Stock. 0.3435 (0.0057) 0.2317 (0.0109) 0.1118 9.0601 (T>t) 0.0000 % Stock Opt. 0.3048 (0.0053) 0.2759 (0.0115) 0.0289 2.2744 (T>t) 0.0116
Table 5 reports the results of the multivariate Tobit regression of the compensation structure on CEO and firm characteristics. For this regression the full 2006 till 2010 sample is used, including both family and non-family firms. Year dummies are added to control for year fixed effects. Consistent with the univariate t-test, the regression shows a positive significant (1% level) relation between base salary and family ownership, indicating that after being
controlled for the included company characteristics, family firm CEOs on average have a (4.7%) higher portion of their total compensation paid in base salary. Similar results were found for the cash bonus regression, with a positive and significant (0.0414 at 10% level) coefficient for the family firm dummy. In line with these findings, family ownership has a
18
positive and highly significant (0.0629 at the 1 percent level) coefficient in the total fraction paid in cash regression, confirming that family firm CEOs indeed receive a larger portion of their compensation in cash payments.
In line with the expectation that family firms used less equity based compensation, family ownership has a negative effect on restricted stock, that is significant at the 1% level. Although a negative coefficient is also reported on the fraction paid in stock options, it is not significant at the 10% level. Nevertheless, in the regression where the total portion of equity compensation is used as the dependent variable, the resulting coefficient is negative (-0.0911) and significant (at 1%). These results support the hypothesis that family firms use relatively less equity based compensation.
A further inspection of the table shows no unexpected outcomes on the control variables. An increased amount of outsiders on the board of directors increases the weight of equity based compensation. Increased company risk through higher R&D intensity, higher capital expenditures and a larger amount of total assets lead to higher equity based
compensation, which is in line with Ryan & Wiggins (2001) and Mehran (1995).
As discussed in the literature review, the reduced equity based compensation within family firms might be explained by the idea that family firm CEOs on average already possess a relatively large stake of the company. This could make them insensitive to the incentive effects that are normally subscribed to equity based compensation. This statement is tested by additional regressions including a variable for the company stake possessed by the CEO, as presented in table 6. A univariate regression of CEO stake on family ownership confirms that family firm CEOs on average possess a larger stake. However, the multivariate regressions show very similar results compared to the above displayed regressions in which CEO stake is not included. The family ownership variable is significant and has the same sign for all regressions with the exception of stock options, where again no significant relation was found.
Table 7 displays the regressions of the compensation structure on the family firm subsample. These regressions are conducted to investigate if the expected effects of the founder being present and the CEO belonging to the family has any effects on the
compensation structure. The outcome of the regressions suggest that, against expectations, being a family CEO has no significant effect on either the total proportion paid in equity or cash. Looking at the regressions where the salary components are taken separately the only significant effect is found for restricted stock awards. The negative relation of -0.104, which is significant at the 1% level, confirms the expectation that CEOs belonging to the family-
19 Tabel 5 Tobit regression of compensation structure on company and CEO characteristics
N = 2465 % Salary % Bonus % Restr. Stock % Stock options % Equity % Cash
Family firm 0.0467*** 0.0414* -0.102*** -0.0206 -0.0911*** 0.0629*** (4.79) (1.83) (-6.07) (-1.26) (-7.08) (5.82) Market-to-book -0.00820* 0.00770 -0.0652*** 0.0553*** 0.00277 -0.00406 (-1.72) (0.67) (-7.73) (6.98) (0.44) (-0.77) R&D intensity -0.427*** 0.0572 -0.367*** 0.897*** 0.576*** -0.466*** (-6.05) (0.34) (-2.88) (7.74) (6.20) (-5.95) CapEx intensity -0.221* -0.220 -0.163 0.728*** 0.531*** -0.332** (-1.85) (-0.78) (-0.79) (3.69) (3.38) (-2.51)
Log total assets -0.0632*** 0.0266 0.0200 0.0675*** 0.0766*** -0.0572***
(-6.71) (1.15) (1.25) (4.31) (6.17) (-5.48)
Oper. Inc. Volatility 0.00867 -0.0186 0.0255 -0.0437*** -0.0201 0.00450
(0.89) (-0.78) (1.55) (-2.71) (-1.57) (0.42)
Long-term debt ratio 0.00961 0.00339 -0.0183* 0.00151 -0.0167** 0.0102
(1.52) (0.22) (-1.74) (0.15) (-2.02) (1.45) Conv.-debt ratio -0.0240 -0.254** 0.0976** -0.0214 0.0577 -0.0454 (-0.84) (-2.35) (2.01) (-0.45) (1.53) (-1.43) CEO duality 0.00936 -0.0205 -0.0227* 0.00180 -0.0154 0.00363 (1.19) (-1.07) (-1.69) (0.14) (-1.49) (0.42) % Outside directors -0.230*** -0.319*** 0.286*** 0.136** 0.281*** -0.285*** (-6.56) (-3.90) (4.72) (2.33) (6.08) (-7.32) CEO age 0.00260*** 0.00345*** -0.00173* -0.00361*** -0.00366*** 0.00356*** (4.75) (2.68) (-1.83) (-3.91) (-5.03) (5.86) Intercept 0.797*** -0.191 -0.0266 -0.0779 0.0437 0.838*** (14.69) (-1.49) (-0.28) (-0.86) (0.61) (13.91)
Year fixed effects Yes Yes Yes Yes Yes Yes
t statistics in parentheses
20
Tabel 6 Tobit regression of compensation structure on company and CEO characteristics including CEO ownership stake
N=1864 % Salary % Bonus % Restr. Stock % Stock options % Equity % Cash
Family firm 0.0326*** 0.0427* -0.0557*** -0.0182 -0.0670*** 0.0480*** (2.81) (1.68) (-2.73) (-0.94) (-4.42) (3.73) CEO stake 0.156* -0.283 -1.098*** -0.116 -0.457*** 0.131 (1.80) (-1.36) (-4.66) (-0.78) (-3.86) (1.36) Market-to-book -0.0110** 0.00423 -0.0673*** 0.0610*** 0.00531 -0.00748 (-2.06) (0.35) (-7.11) (6.88) (0.76) (-1.26) R&D intensity -0.427*** 0.221 -0.334** 0.803*** 0.549*** -0.412*** (-5.38) (1.29) (-2.30) (6.16) (5.31) (-4.68) CapEx intensity -0.284** 0.210 -0.569** 1.026*** 0.484*** -0.293** (-2.13) (0.72) (-2.44) (4.66) (2.79) (-1.99)
Log total assets -0.0653*** 0.00399 0.00767 0.0878*** 0.0846*** -0.0614***
(-6.01) (0.16) (0.41) (4.86) (5.97) (-5.10)
Oper. Inc. Volatility 0.0133 0.00153 0.0375** -0.0706*** -0.0320** 0.0117
(1.20) (0.06) (1.98) (-3.83) (-2.21) (0.95)
Long-term debt ratio 0.00620 0.0134 -0.0220** 0.00694 -0.0163* 0.00855
(0.94) (0.91) (-2.01) (0.65) (-1.92) (1.17) Conv.-debt ratio -0.0205 -0.407*** 0.0478 0.0283 0.0667 -0.0592 (-0.57) (-3.10) (0.77) (0.48) (1.42) (-1.48) CEO duality 0.000663 -0.0270 -0.0164 0.0160 -0.00346 -0.00740 (0.07) (-1.30) (-1.05) (1.05) (-0.29) (-0.73) % Outside directors -0.243*** -0.289*** 0.228*** 0.177** 0.276*** -0.301*** (-5.89) (-3.17) (3.21) (2.57) (5.13) (-6.60) CEO age 0.00232*** 0.00216 -0.000661 -0.00390*** -0.00298*** 0.00291*** (3.76) (1.60) (-0.62) (-3.74) (-3.68) (4.27) Intercept 0.827*** -0.0651 0.0248 -0.147 0.00715 0.884*** (13.16) (-0.47) (0.23) (-1.41) (0.09) (12.70) t statistics in parentheses * p<0.10, ** p<0.05, *** p<0.010
21
Tabel 7 Tobit regression of compensation structure on family firm characteristics
N=477 % Salary % Bonus % Restr. Stock % Stock options % Equity % Cash
CEO family of founder -0.00548 -0.0222 -0.104*** 0.0365 -0.0242 -0.00762
(-0.22) (-0.39) (-2.70) (0.90) (-0.66) (-0.27) Founder present 0.0892*** -0.0197 -0.129*** -0.0703* -0.142*** 0.0892*** (3.60) (-0.36) (-3.37) (-1.76) (-3.93) (3.25) Market-to-book -0.00959 0.0128 -0.115*** 0.0362* -0.0350* -0.00196 (-0.72) (0.42) (-5.00) (1.68) (-1.78) (-0.13) R&D intensity -0.671*** 0.773** -0.279 1.095*** 0.969*** -0.587*** (-3.77) (2.14) (-0.88) (3.90) (3.74) (-2.97) CapEx intensity -0.690** -0.597 0.477 0.848 1.175** -0.750* (-2.00) (-0.76) (0.87) (1.54) (2.35) (-1.96)
Log total assets -0.0822*** 0.0764 0.0319 0.0640 0.0754* -0.0683**
(-3.04) (1.31) (0.76) (1.47) (1.91) (-2.28)
Oper. Inc. Volatility 0.0280 -0.0488 -0.00324 -0.0368 -0.0355 0.0186
(1.01) (-0.81) (-0.07) (-0.82) (-0.88) (0.60)
Long-term debt ratio 0.00879 0.0113 -0.0421** 0.00782 -0.0259 0.0132
(0.71) (0.41) (-2.33) (0.41) (-1.47) (0.96) Conv.-debt ratio -0.333* -0.553 0.827*** -0.0928 0.565** -0.484** (-1.90) (-1.33) (3.18) (-0.33) (2.24) (-2.49) CEO duality -0.0165 -0.0611 -0.0154 -0.00723 -0.0234 -0.0344 (-0.67) (-1.08) (-0.40) (-0.18) (-0.65) (-1.26) % Outside directors -0.252*** -0.408** 0.297** 0.0939 0.287** -0.311*** (-3.06) (-2.18) (2.26) (0.71) (2.39) (-3.40) CEO age 0.00147 0.00817*** -0.00254 -0.000854 -0.00139 0.00364** (1.11) (2.72) (-1.19) (-0.39) (-0.71) (2.49) _cons 0.995*** -0.776** 0.137 -0.245 0.00864 0.904*** (7.24) (-2.56) (0.63) (-1.11) (0.04) (5.93) t statistics in parentheses * p<0.10, ** p<0.05, *** p<0.010
22
receive smaller fractions of their salary paid in restricted stock. Because of the
aforementioned insignificance of the family CEO variable in the total equity regression, the second hypothesis will not hold, rejecting the expectation that non-family CEOs receive a larger portion of their compensation in equity.
The second independent variable of interest presented in table 6 is the dummy that indicates whether the founder still is active within the company as either CEO or board member. For all payment methods, except cash bonus, a significant relation is found. Both restricted stock and stock options have a negative coefficient (-0.129 and -0.0703) and are significant (at 1% and 10%). The relation between cash bonus and the founder variable is not significant if the payment methods are taken separately. This changes in the regression where cash based compensation is taken as a whole. Conform expectations the presence of the CEO increases the weight of cash based compensation.
5. Robustness
In this section an additional test is presented in order to check the robustness of the results discussed above.
Robustness remuneration structure
The data used for the remuneration structure is conform the FASB123 reporting standards, introduced in 2006. As a consequence the used sample for this research ranges from 2006 until 2010. During this period, the corporate landscape was marked by a financial crisis that started in 2007 (Conyon et al., 2011). Based on the heavily declining primary US stock markets, almost all companies were financially affected. The incentive part of executive compensation is often related to performance measures that include current stock price returns. It is therefore near certain that the financial crisis affected executive remuneration to some extent. In order to check if similar results are found in a period that is less affected by the financial crisis, an additional regression will be executed on a sample that runs from 2001 until 2010. For the pre-2006 period, compensation levels as computed by Compustat will be added to the dataset. The regression results are presented in table 6.
Family ownership has a comparable and significant effect in both the total equity and total cash regression, further endorsing that family owned firms put more emphasis on cash based compensation. This strengthens the first hypothesis that family firms use less equity based compensation compared to non-family firms.
23
Although the total cash- and equity compensation measures have similar effects, differences arise in the regressions with the separate compensation components as the
dependent variable. In the original 2006-2010 regressions, family ownership was positive and significantly correlated to cash bonus while no significant relation was found for stock
options. The 2001-2010 regression shows an opposite result, in which there is no significant relation between family ownership and cash bonus, while there is a significant relation with stock options. This relation is in accordance with the original expectations that non-family CEOs receive less equity based compensation and therefore less stock options.
The change in significance between stock options and cash bonus could partially be explained by the declining role of employee stock options. Literature extensively reports a vast reduction in the use of stock options between 2001 and 2010 and describes two main causes for this change. Coles, Daniel and Naveen (2006) devote the reduced role for employee stock options to new insights in the incentive effects of stock options. In this research a positive relation is found between the use of stock options and riskier investment patterns and the implementation of more aggressive debt policies, suggesting that stock options encourage managers to engage in riskier behavior. Hayes et al (2012) add to this explanation that after the adaption to FAS 123R accounting standards in 2006, many
accounting benefits for the use of employee stock options were removed. This further reduced the attractiveness for companies to use this compensation type.
The resulting coefficients of salary and restricted stock remain fairly consistent with comparable coefficient values (0.0467 vs. 0.0624, -0.0762 vs. -0.1020 respectively) that all are significant at the 1% level . No large changes occur with control variable coefficients, further indicating that the results between the two different time samples are properly comparable. More information regarding the robustness test is presented in the table below.
6. Conclusion and discussion
This study investigates CEO remuneration within family owned firms. Literature
comprehensively states that family owned firms typically have different standards on various company characteristics such as firm financial stability and governance standards. Many of these characteristics are believed to be key issues for determining efficient remuneration strategies. This research therefore investigates whether the claimed differences in company characteristics in reality are reflected in the used remuneration strategies. Tests are conducted that compare structures of compensation packages between family and non-family firms.
24
N=4731 % Salary % Bonus % Restr. Stock % Stock options % Equity % Cash
Family firm 0.0624*** 0.0129 -0.0762*** -0.0650*** -0.102*** 0.0720*** (8.81) (1.41) (-5.46) (-5.47) (-10.09) (8.47) Market-to-book -0.0178*** -0.000766 -0.0626*** 0.0523*** 0.0221*** -0.0200*** (-5.98) (-0.20) (-9.55) (10.59) (5.19) (-5.58) R&D intensity -0.287*** -0.423*** -0.645*** 1.048*** 0.702*** -0.578*** (-5.41) (-5.89) (-5.77) (11.97) (9.29) (-9.07) CapEx intensity -0.0376 -0.329*** -0.108 0.524*** 0.411*** -0.254** (-0.46) (-3.09) (-0.66) (3.82) (3.49) (-2.56)
Log total assets -0.0581*** -0.0112 0.0230* 0.0687*** 0.0819*** -0.0685***
(-9.05) (-1.35) (1.84) (6.40) (8.92) (-8.87)
Oper. Inc. Volatility 0.00490 0.0232*** 0.0202 -0.0417*** -0.0322*** 0.0214***
(0.75) (2.75) (1.58) (-3.83) (-3.45) (2.73)
Long-term debt ratio 0.0207*** -0.0118** -0.0205** -0.00199 -0.0208*** 0.0162***
(5.29) (-2.17) (-2.57) (-0.31) (-3.74) (3.45) Conv.-debt ratio -0.0156 -0.131*** 0.0150 0.115*** 0.0514* -0.0447* (-0.74) (-3.46) (0.36) (3.31) (1.71) (-1.76) CEO duality 0.0116* -0.0123 -0.0197* -0.00621 -0.0177** 0.00671 (1.93) (-1.56) (-1.73) (-0.62) (-2.07) (0.93) % Outside directors -0.170*** -0.0492* 0.219*** 0.188*** 0.238*** -0.194*** (-7.54) (-1.73) (4.83) (4.97) (7.36) (-7.17) CEO age 0.00252*** 0.00129** -0.00312*** -0.00328*** -0.00407*** 0.00330*** (6.26) (2.47) (-3.90) (-4.83) (-7.03) (6.82) _cons 0.680*** 0.0767 -0.291*** 0.0542 0.0882* 0.812*** (18.38) (1.62) (-3.92) (0.88) (1.67) (18.27)
Fixed year effects Yes Yes Yes Yes Yes Yes
t statistics in parentheses
25
Based on existing literature, two proposed relations between family firm characteristics and corporate remuneration will be tested. The primary research questions are:
(1) Do family owned firms use less equity based compensation?
(2) Do non-family CEOs operating a family firm receive a larger part of their compensation in equity form?
To find out if differences occur between the structure of compensation packages, several regressions are executed. In these regressions, the percentage compensation
components are regressed on a set of variables, including a dummy that indicates whether the firm is family owned. Conform expectations, the family ownership variable has a positive and significant effect on the total fraction paid in cash. Furthermore, a negative significant relation is found between equity based compensation and family ownership. This confirms the
expectations that family owned firms use less equity based compensations. These results remain valid when the same regressions is run over an extend time sample and with the inclusion of a variable that represents the stake possessed by the CEO. This further strengthens the validity of the outcomes. Ryan and Wiggans (2001) find that companies facing larger potential agency conflicts use more equity based compensation. The results found in this research therefore tend to subscribe both Anderson & Reeb (2003) Barontini & Caprio (2006) in their statement that family firms experience less agency problems.
To create a deeper insight in the family firms, a similar regression is executed
including only the family owned firms. Existing literature states that the potential benefits of family ownership is dependent on the presence of the founder as CEO or board member and whether the CEO belongs to the family. To check the potential effects of these relations, the family ownership dummy is replaced by dummies reflecting if the CEO is part of the family and whether the founder is present as either CEO or board member.
In this regression no significant effect for the CEO belonging to family on either total cash or total equity based compensation is found. Expectations were that family CEO’s should receive relatively less equity based compensation. This is based on the idea that the family CEO’s personal wealth often already is connected to the company through the family share capital. This could make him less sensitive to the incentive effects of additional equity compensation. The regressions however display no significant relations that confirm or reject this statement. Therefore no general statement about the effects of CEOs belonging to the family shareholders can be made.
A possible explanation for this outcome is presented by McConaughy (2004) who states that CEOs that are member of the controling families typically engage in long-term
26
ownership commitments. This reduces the ease and frequency by which family CEOs normally trade the company shares, which is often considered family capital. These
limitations might be far less severe with new awarded stocks or stock options. The payoffs to equity pay could therefore still make a considerable impact on the disposability of the total compensation. This means that there is no empirical evidence suggesting that non-family CEOs receive relatively more or less equity based compensation.
The regressions further show a significant positive effect for the presence of the founder on total cash based compensation and a negative significant effect on total equity based compensation, which is according to expectations. This result has a two sided explanation. Literature states that in case the founder serves as CEO generally less equity based compensation is used (He, 2008). The founders often already posses a significant stake and are believed to have greater intrinsic motivation, reducing the relevance of incentive based compensation. The relatively large weight for base salary is further underlinded by the regressions where payment components are taken separatly. For stock options and restricted stock the founder dummy has a highly significant and negative coefficient. No significant effect was found for founder on cash bonus while a positive and highly significant relation was found in the base salary regression.
Another explanation for this arises when the founder is not active as CEO but does hold a board position. Both Burkart et al (2003) and Villalonga & Amit (2006) highlight the special status founders typically have. This status is thought to work as a strong manager monitoring device and to reduce the chance of managerial conflicts. During this research no distinction is made for the position of the founder, either than being present or not. Additional research that does make this separation could further clarify the effects of the position of the founder on CEO compensation.
This paper contributes to existing literature by giving new insights in remuneration strategies used by family owned firms. This helps to get a better understanding of both the governance choices of family firms and the use of remuneration in general. This paper
however has certain limitations. Ownership structures are recognized as an important factor in the use of compensation, since one of the goals of remuneration is to align managers and shareholders’ interests. External shareholders could play an important role in the decision making process of family firms, especially when then the stake possessed by each group is comparable. However, collecting information on ownership structures is difficult and
therefore not included in this research. Research that does include more extensive ownership information could therefore be an useful addition.
27
Literature
Anderson, R.C., Reeb, D.M., (2003). Founding-family ownership and firm performance: evidence from the S&P500. Journal of Finance (no. 58), pp. 1301– 1328.
Anderson, R.C., Duru, A., and Reeb, D. M. (2009). Founders, heirs, and corporate opacity in the United States. Journal of Financial Economics, Vol. 92(2), pp. 205–222.
Anderson, R.C., (2015). Family Data Distribute, www.ronandersonprofessionalpage.net/data-sets.html, visit 02-04-2015
Andres, C. (2008). Large shareholders and firm performance. An empirical examination of founding-family ownership. Journal of Corporate Finance, Vol. 14(4), pp. 431–445.
Barontini, R., & Caprio, L. (2006). The effect of family control on firm value and performance. Evidence from Continental Europe. European Financial Management, 12(5), pp. 689–723.
Burkart, M., Panuzi, F., Shleifer, A., (2003) Family Firms, Journal of Finance (vol 58, issue 5), pp. 2167-2022.
Bryan, S., Hwang, L., Lilien, S., 2000. CEO stock-based compensation: An empirical analysis of incentive-intensity, relative mix, and economic determinants. Journal of Business (no.73), pp. 661–693.
Chen, S., Chen, X., Cheng, Q., (2008). Do family firms provide more or less voluntary -disclosure? Journal of Accounting Research (no. 46), pp. 499–536.
Conyon, M., Judge, W.Q., Useem, M., (2011) Corporate governance and the 2008-09 financial crisis. Corporate governance, An International Review, Vol. 19(5), pp. 399-404.
Coles, J., Daniel, N., Naveen, L., (2006). Managerial incentives and risk taking. Journal of Financial Economics, (no.790), pp, 431–468
Core, J.E., Holthausen, R.W., Larcker, D.F., (1999). Corporate governance, chief executive officer compensation, and firm performance. Journal of Financial Economics (no. 51), pp. 371-406.
DeFusco, R., Johnson, R., Zorn, T., (1990). The effect of executive stock option plans on stockholders and bondholders. Journal of Finance (no. 45), pp. 617-627.
Gomez-Mejia, L.R., Larraza-Kintana, M., Makri, M. (2003). The determinants of executive compensation in family-controlled public corporations. Academy of Management Journal,(no. 46), pp. 226-237
28
Gopolan, R., Milbourn, T., Song, F., Thakor, A., (2014). Duration of Executive Compensation. Journal of finance, (no. 6).
He, L. (2008). Do founders matter? A study of executive compensation, governance structure and firm performance. Journal of Business Venturing (No. 3), pp. 257-279.
James, H.S., (1999). Extended Horizons and the Family Firm. International Journal of the Economics of Business (Vol. 6), pp. 41-55
Jensen, M.C., Murphy, K.J., (1990). CEO incentives: It’s not how much you pay, but how. Harvard Business Review (no. 68), pp. 138-149.Klein, P., Shapiro, D., Young, J. (2005).
Corporate Governance, Family Ownership and Firm Value: The Canadian Evidence. Corporate governance, An International Review (vol. 13), pp. 769-784
Kirkpatrick, Grant (2009), “The corporate governance lessons from the financial crisis”, OECD Journal: Financial Market Trends, (no. 1).
Maury, B. (2006). Family ownership and firm performance: Empirical evidence from Western European corporations. Journal of Corporate Finance, (Vol12(2), pp. 321–341.
McConaughy, D.L.,(2004). Family CEOs vs. Nonfamily CEOs in the family controllerd firm. Family business review,(no. 13).
Mehran, H., (1995). Executive compensation structure, ownership and firm performance. Journal of finance (no. 5)
Ryan, H.E.,Wiggins, R.A., (2000). The influence of firm- and manager-specific characteristics on the structure of executive compensation. Journal of corporate finance (no. 7), pp. 101-123
Villalonga, B., Amit, R., (2006). How Do Family Ownership, Control and Management Affect Firm Value? Journal of financial economics (Vol. 80), pp. 385-417
Worldbank Financial Database (2015). http://data.worldbank.org/indicator/NY.GDP MKTP.CD/countries/US?page=1&display=default. Visited 02-06-2015
29 Appendix 1: Definition and source of used variables
Variables Definition Source
Family variables
Family firm Single family posses at least 5% stake Anderson website
CEO is family CEO belongs to family Company websites
Founder present Founder is still active as either CEO or board member
ISS/ SEC fillings/ Company website
Investment strategies
Capital Expenditure intensity Capital expenditures divided by book value total assets Compustat
R&D intensity R&D expenditures divided by book value total assets Compustat
Market-to-book value of assets (Market value equity + book value debt)/ book value of assets Compustat
Firm financials
Firm size logartihm of total assets Compustat
Operating cash flow volatility Coefficient of variation company’s operating cash flow Compustat
Long-term debt ratio Book value of long-term debt/ market value of equity Compustat
Convertible debt ratio Book value of convertible debt/ market value of equity Compustat
Governance and CEO specifics
CEO duality CEO is also chairman ISS risk metrics
% oustide directors Percentage of company outsider on board of directors ISS risk metrics
CEO age CEO Age Execucomp
30 Tabel Appendix 2: Correlation matrix
% Salary % Bonus % Restr. Stock % Stock options Family firm Market-to-book
R&D
intensity CapEx intensity
% Salary 1,00 % Bonus 1,00 % Restr. Stock 1,00 % Stock options 1,00 Family firm 0.1997 0,10 -0,51 -0,05 1,00 Market-to-book -0.0744 0,06 -0,18 0,24 0,04 1,00 R&D intensity -0.0811 -0,02 -0,19 0,21 -0,04 0,35 1 CapEx intensity -0.0178 0,00 -0,11 0,06 0,04 0,07 -0,14 1
Log total assets -0.4077 0,01 -0,03 0,11 -0,11 -0,01 -0,16 0,03
Oper. Inc. Volatility -0.4019 0,01 0,23 0,13 -0,10 0,10 -0,08 0,08
Long-term debt ratio -0.0104 -0,02 0,21 -0,05 0,00 -0,26 0,14 0,02
Conv.-debt ratio -0.0041 -0,04 0,05 -0,01 -0,07 -0,12 0,14 -0,01
CEO duality -0.0361 0,00 0,03 -0,01 -0,05 -0,03 -0,15 0,05
% Outside directors 0.2484 -0,13 0,21 0,04 -0,37 -0,08 -0,03 -0,07
CEO age 0.1191 0,07 -0,04 -0,99 0,13 -0,07 -0,11 0,01
Log total assets Oper. Inc. Volatility Long-term debt ratio Conv.-debt ratio
CEO duality
% oustide
direct Age
CapEx intensity
Log total assets 1
Oper. Inc. Volatility 0,26 1
Long-term debt ratio 0,16 0,11 1
Conv.-debt ratio -0,06 -0,07 0,27 1
CEO duality 0,20 0,19 0,02 -0,06 1
% Outside directors 0,23 0,22 0,01 0,0057 0,1566 1
