Recently, CEO compensation has been a topic of much public discussion. Especially during the years of the financial crisis, when high bonuses for top managers attracted a great deal of attention. As Sarah Anderson, an Institute for Policy Studies analyst who specializes in executive pay stated in a CNBC article (2008), lavish executive pay is something that people are livid about across the political spectrum.
However, the topic of CEO compensation in family businesses has received relatively less attention, while family control is still a common ownership structure. In the United States, for example, one third of the Fortune 500 firms are at least partially controlled by a single family maintaining substantial firm ownership (Anderson & Reeb, 2003). Crystal (1998) showed in a report on CEO compensation at family firm Hilton Hotels that Stephen Bollenbach's compensation far exceeded that of his predecessor, Barron Hilton, the hotel family heir. Does this mean that compensation are higher for outside, non-family CEO's than for family CEO's? In order to be able to answer these kind of questions, this thesis will investigate whether there are differences in compensation between family- and external CEO's. Since in existing literature the main focus is on US family firms, this thesis will take European family firms in consideration as well and compare whether (and to what extent) these potential differences will be similar to those in the USA. With this comparison, a gap in literature will hopefully be closed.
Derived from the problem indication, the central question that will be dealt with is as follows:
What is the impact of the CEO of a family firm being a family- or non-familyÂ member onÂ CEO compensation, compared across the US and Europe?Â
In order to be able to answer the problem statement, this question is divided into two sub questions in which the research problems are defined more specifically.
What is the impact of the CEO of a family firm being a family- or non-familyÂ member on CEO compensation, in the US?
What is the impact of the CEO of a family firm being a family- or non-familyÂ member on CEO compensation, in Europe?
The type of research that will be used in order to be able to answer the problem statement and its research questions will be descriptive using the method of a literature review. This means that there will be relied on secondary sources, thus available literature, found for example via the database of Tilburg University and scientific journals in general.
As the problem statement describes, this thesis will be about whether there are differences in CEO compensations for CEOs with different statuses. Therefore, the concepts that will be important in order to investigate this will be as follows. With 'compensation' in this context is meant the level (relative amount), and the structure of the compensation (composition), the latter being subdivided between the variables 'fixed compensation' or 'variable compensation' (e.g. bonuses). The concept 'CEO statuses' consist of the variables of a CEO being a member of the family of the firm, or not, thus being a professional hired from outside. The relationship between these two concepts with its different variables which will be investigated in this thesis, can be visualized in the following way:
level and structure
family or non-family
The different factors for which this relation will be investigated are whether the family firms are situated in the US or in Europe. This might seem like an unsuitable comparison, since the US is a country and Europe is a continent, but this division is made this way deliberately. Comparing the United States, for example, with the Netherlands seems more logical since this would mean comparing two countries instead of a country and a continent, the size of the investigated areas would not be compatible at all. The United States would bring in much more information than the Netherlands, therefore it has been decided to compare the United States with the European continent. In this way, two better comparable groups emerge.
For the structure of this thesis, the order of the research questions will be followed, since the answers to the research questions will eventually lead to the resolution of the problem statement. Therefore, each chapter will deal with one research question. Before the two research questions will be investigated, however, an extra introductory chapter will first discuss some general theories about the impact of the status of a CEO of a family firm onÂ CEO compensation, containing aspects that hold for both the US and Europe. After this second chapter, the third chapter will discuss the impact of the CEO of a family firm being a family- or non-familyÂ member onÂ CEO compensation in the US. Naturally , the fourth chapter will consider this impact in Europe. The final and fifth chapter will contain the conclusions, discussions and recommendations, thereby answering the problem statement.
Chapter 2: What is the impact of the CEO of a family firm being a family- or non-familyÂ member onÂ CEO compensation in general?
In order to be able to investigate the differences in CEO compensation for family and non-family CEOs and compare this across the United States and Europe, it is of course important to have some knowledge about what has been written about CEO compensations and cultural differences in general.
As already indicated in the first chapter, CEO compensation has been the topic of much public discussion. Let us, therefore, first define what is precisely meant by CEO compensation. As already described in the research methods, with 'compensation' is meant the level (thus the relative amount) , and the structure of the compensation received for the CEOs efforts. This structure is subdivided between the variables 'fixed compensation' or 'variable compensation'. But these variables of the structure of the compensation can be specified further. According to Murphy (1998) 'most executive pay packages contain four basic components: a base salary, an annual bonus tied to accounting performance, stock options, and long-term incentive plans.' Especially in the economic recession, lots of employees lost their jobs, while frequently CEOs still received extreme rewards for their performances, in one of the forms described by Murphy. But are those firm performances even positively influenced by high rewards? McConaughy thinks they are, by stating that agency theory, which will be discussed in the following paragraph, predicts that pay for performance is effective in reducing principle-agent misalignments. Thus the higher the correlation between pay and performance, the better the expected performance is. Jensen and Murphy (1990) detected a low but still significant sensitivity of CEO pay to firm performance as well. However according to Frey and Osterloh (2005), firm performances are not positively influenced by high rewards. They say that managers are given incentives to manipulate performance criteria by pay for performance, and that it motivates them to deceitful accounts that are in conflict with the long term firm interest. Unite et al (2008) state that a positive relation between pay and performance does not hold for firms with a relationship to a family, which indicates that there does seem to be a distinction for family firms within the much discussed topic of CEO compensation.
Fuller and Jensen (2002) suggest that bonus structures makes managers focus more on short term goals instead of long term growth. These structures are however designed to align the managers interests with those of the owners. This brings us to the principle agent theory, to which linking CEO payment to performance might be a solution after all.
Principle agent theory
Agency theory is about the conflict of interest between the principle and the agent of a firm. In this context, the principal is the owner of the firm and the agent is the manager or the CEO. The agency problem deals with the conflict that occurs when cooperating parties have different goals or interests (Eisenhardt, 1989). As stated above, linking CEO compensation to firm performance trough for example bonuses and stock options might be a solution to this problem, since owners and managers interest might be more aligned in this way. But the agency problem for family-influenced firms is different in the first place (Young, Peng, Ahlstrom, Bruton, & Jiang, 2008). When members of an influential family participate in management, the separation and conflicting goals between owners and managers are reduced, compared with non-family firms.
CEO compensation in family firms
When Gomez-Mejia, Larazza- Kintana & Makri (2003) researched CEO compensation in family businesses with more than one family member involved, they concluded that CEOs that belonged to the family received less compensation than CEOs that did not belong to the family. They found three, general applicable, reasons for this. CEOs related to the family will enjoy higher job security than non-family CEOs. As Beehr et al. (1997) pointed out, family CEOs have two roles; a work related role, and a non-work related role in which they 'have' to fulfill family obligations. Because of this duality of roles, a family CEO is rewarded with a more assured job (Gomez-Mejia et al, 2001). Furthermore, when there is an emotional bond between monitors and employees being judged, the evaluator is more likely to give positive performance feedback appraisal. For family firms, this would mean that monitors probably tend to evaluate the family CEOs more positive, and give them the benefit of the doubt more easily. All in all the above suggests that agents, who are according to the agency theory mostly risk averse, in family businesses trade a higher job security for less earnings.
Moreover, because of the emotional ties of belonging to a family, family CEOs are less likely to compete in the external labor market. This means that they are less free to choose the best offer available to them, which decreases the need to compensate family CEOs with the same packages as outside CEOs get, who do need those 'handcuffs' to stay with the company (Gomez-Mejia, Larazza- Kintana & Makri, 2003)
Apart from these reasons found by Gomez-Mejia, Larazza- Kintana & Makri, Block (2008) found a number of aspects in which family CEOs differ from non-family CEOs as well. The first aspect he mentions is that the firm is not just an employer for family CEOs. They are unlikely to go against the interests of the firm, since the firm has a symbolic meaning for them as it symbolizes the heritage and tradition of the family, and therefore is likely to be part of their identity. Going against the interests of the firm would eventually mean going against themselves, which becomes more apparent when family CEOs have the name of the company. Another relevant aspect is that family member mostly want to pass the firm on to the next generation in line (Casson & James, 1999). Logically, this would mean that family CEOs focus on the long-term interest by for example investing in research and development and keeping good relations with the staff of the company (Block, 2008). Following from Block's first two aspects mentioned it appears that the targets of family-CEOs are frequently non-financial (Donckels and Frohlich, 1991; Harris and Martinez, 1994; Tagiuri and Davis, 1992). In conclusion, according to Block (2008), the interests of family-CEOs are frequently powerfully aligned with those of the firm.
The above is confirmed by Jensen & Meckling (1976), who agree with the earlier given statement that the agency problem is less likely to arise in a family business, as a family-CEO probably has other incentives besides pay. In addition McConaughy (2000) agrees with the conclusion made by Gomez-Mejia, Larazza- Kintana & Makri (2003) and suggest that this might mean that family firms do not need to pay family-CEOs as high a compensation as non-family CEOs.
However, a converse perspective on this matter is given by Bebchuck, Fried & Walker (2002), who suggest that since CEOs have the power to influence the board, they can influence their compensation as well and utilize this possibility to engender higher compensations for themselves. They state that family CEOs may have more authority to do this than non-family CEOs. Apart from being able to influence ones fixed pay, CEOs can also influence one of the other forms of executive pay packages cited from Murphy at the beginning of this chapter. Dechow, Hutton & Sloan (1996) indicate that CEOs can be empowered to influence the worth of their stock options as well. This does, however, depend on the extent to which the options are 'visible' (Pollock, Fisher & Wade, 2002). The more family members are on the board and the closer they work with the CEO, the harder it gets to re-price the stock options since they might not be hidden enough.
All the above indicates that there are many aspects of the differences in CEO compensation for family- and non-family CEOs in general, and now that we are aware of some of the most important theories on this the next chapter will focus on those differences specifically for the US.
Chapter 3: What is the impact of the CEO of a family firm being a family- or non-familyÂ member on CEO compensation, in the US?
McConaughy (2000) studied CEO compensation in 82 founding family-firms, from which 47 CEOs are family members and the other 35 are not. The sample of the family-controlled businesses with a family CEO was drawn from the Business Week CEO 1000 and the Business Week Corporate Elite, both containing information on 1000 U.S. firms. The sample of the family-firms with a non-family CEO was drawn from a Loyala University, Chicago, investigation of family businesses in the Fortune 500, which means that all 82 companies were U.S. firms.
' It tests the family incentive alignment hypothesis, which predicts that family CEOs have superior incentives for maximizing firm value and, therefore, need fewer compensation-based incentives. Univariate and multivariate analyses show that family CEOs' compensation levels are lower and that they receive less incentive-based pay-confirming the family incentive alignment hypothesis and suggesting the possible need for family firms to increase CEO compensation when they replace a founding family CEO with a nonfamily-member CEO.'(McConaughey, 2000)
Thus according to these results family CEOs receive less pay, providing less remuneration for their performances. For non-family CEOs this is different, meaning that to attract , hold on to, and stimulate those CEOs, higher pay levels are needed since they lack the incentives that family-CEOs do have. This goes against the view that family CEOs use their position to extract additional compensations, as mentioned in the first chapter in accordance with Bebchuck, Fried & Walker (2002), who suggest that since CEOs have the power to influence the board, they can influence their compensation as well and utilize this possibility to engender higher compensations for themselves.
Family CEOs receiving less pay, however, does not leave out the option that they might receive alternate compensations besides cash payments, in for example one of the forms cited from Murphy earlier.
In conclusion, McConaughey (2002) states that since family CEOs have greater incentives, they do not need as much supplementary incentives trough compensation as non-family CEOs, there by supporting the family control incentive alignment hypothesis. Founding family CEOs receive less pay. These results propose that non-family CEOs have to be paid more to get what family CEOs would do.
Deckop (1988) findings on the subject of CEO compensation are in line with McConaigheys conclusions, since he too stated that CEOs from outside the firm received more compensation than CEOs promoted from within, and that founders of the firm received less pay.
Gomez-Mejia, Larraza-Kintana and Makri (2003) conducted a study on the determinants of CEO compensation in family firms as well. During a four-year period (from 1995-1998) they collected data on 253 family controlled firm, from which 148 (thus 59%) had a family-CEO and 105 (41%) were led by non-family CEOs. This sample came from a randomly selected set of 3000 companies included in the COMPUSTAT database, therefore containing mainly U.S. based firms. The results indicate that family CEOs of family businesses receive a total income that is lower than that of non-family CEOs. This relative disadvantage increases with a rise in the amount invested in research and development and a rise of family ownership concentration, because hereby the family CEOs pay will be depressed. Because of the family ties, family CEOs' pay tends to be more shielded from risk but more sensitive to systematic business risk, which is more uncontrollable (Gomez-Mejia, Larraza-Kintana and Makri, 2003). These results are in line with what has been stated above.
Block (2008) also published a paper on the structure of executive compensation in family and non-family firms. The sample for his investigation was gathered from the Standard & Poor's 500 and their database, resulting in 2578 observations from 393 U.S. firms.
'The findings of this paper support the argument that agency conflicts between owners and management are lower in family firms versus non-family firms. It is found that both family management and the degree of family ownership increase the share of base salary.' (Block, 2008)
At first glance, this seems to be a contradictory statement. Block says that there already is an alignment of interest between the family-CEO and the firm, which suggest that they do not need as much pay as non-family CEOs to stay motivated. However, he states that both family management and the degree of family ownership íncrease the share of base salary, and not decrease as seems more logical in the first place. This seemingly contradiction is clarified elsewhere in his article, were he states that ' in a situation of low information asymmetry between shareholders and
management, the fixed component of pay should be high, and incentive pay should be low' (Block 2008). This statement is confirmed by the mention that:
'In addition to this, a negative relationship between family management and the share of stock option pay is found. There seems to be a strong alignment of interests between a family CEO and the firm for which he or she works. An alternative interpretation is that family managers are intrinsically more strongly motivated to behave in the firm's best interest.' (Block, 2008)
In conclusion, Block (2008) stated that in US family firms the share of stock based pay is lower and the share of base salary is higher than in non-family firms. In a situation of low information asymmetry between shareholders and management, which is the case in family firms since family members often know the firm better, the fixed component of pay should be high and the incentive pay should be low in US family firms, since intrinsic motivation plays an important role for CEOs that belong to the family. Paying a family CEO an extremely high compensation might decrease this intrinsic motivation, and therefore might not be a good idea (Block, 2008).
A more specific investigation concerning CEO compensation in US family businesses has been done by Crystal (1998). He showed in a report on CEO compensation at family firm Hilton Hotels that Stephen Bollenbach's compensation far exceeded that of his predecessor, Barron Hilton, who was the hotel family heir. This indicates that in this case the family CEO received far less pay than the non-family CEO successor.
Chapter 4: What is the impact of the CEO of a family firm being a family- or non-familyÂ member on CEO compensation, in Europe?
Croci, Gonenc & Ozkan (2010) researched the impact of family control and institutional investors on CEO compensation in Continental Europe. Although this chapter deals with CEO compensation in Europe, the investigation still seems a valuable source, since by far the biggest part of Europe is formed by Continental Europe, were only the European islands are left out. For their paper, Croci, Gonenc & Ozkan used a dataset of 915 listed firms with 4045 firm-year observations from 14 countries, during an eight year period from 2001 till 2008. They expected that family control probably affects the level and the structure of the CEOs compensation, an expectation stated in the first chapter as well. Croci, Gonenc & Ozkan (2010) state that in family firms equity-based compensation might be relatively less important than in widely held firms, since family firms do not have to be as concerned about the alignment of interests of CEOs and shareholders and therefore do not need equity based compensation to achieve this alignment of incentives. More generally, since family firms often like to keep control, they might be hesitant to give their CEOs stock options, thereby reducing the part of equity-based compensation (Croci, Gonenc & Ozkan, 2010) Thus, it is expected that a family CEO receives relatively less equity-based pay.
In their findings, Croci, Gonenc & Ozkan (2010) indicate that family control moderates the total level of CEO pay, including cash as well as equity-based compensation. This might be because the controlling family is frequently in an appropriate position to supervise and, if need be, fire the CEO if he/she does not act in accordance with the expectations of the family. The tendency that family control moderates the total level of CEO compensation is even stronger when the CEO is a family-member. Thus, controlling families do not use CEO pay to take away capital from minority shareholders (Croci, Gonenc & Ozkan, 2010).