Revisiting Executive Pay in Family-Controlled Firms

In our paper, Revisiting Executive Pay in Family-Controlled Firms, which was recently made publicly available on SSRN, we reexamine executive pay in family-controlled firms and challenge the findings in the existing literature.

According to the prior literature, family executives of family-controlled firms receive lower compensation than non-family executives. Using 82 family-controlled firms in the U.S. in 1988, McConaughy (2000) report that family CEOs are paid lower compensation than non-family CEOs. Likewise, Gomez-Mejia, Larraza-Kintana, and Makri (2003) show similar findings using a sample of 253 family-controlled firms in the U.S. during 1995-98.

Why is it the case? Explanations in the existing literature can be summarized into three. First, not all family members are involved in management. Some are directly involved, while others are not. These second group of family members, however, are not just bystanders. They play an important role of monitoring and disciplining family-executives, and because of their multidimensional and long-term relationship with family-executives, they are quite good at playing their role (Fama and Jensen, 1983; McConaughy, 2000). Thus, the compensation level of family-executives is kept at its necessary minimum.

Second, family executives enjoy benefits that cannot be enjoyed by non-family executives. They receive dividends from their ownership stakes, and enjoy higher job security (Gomez-Mejia, Larraza-Kintana, and Makri, 2003). Family-executives may trade these benefits for lower compensation. Third, family-executives are handcuffed. Emotional attachment to the firm makes them unlikely to compete in the external job market, and take more lucrative outside offers (Gomez-Mejia, Larraza-Kintana, and Makri, 2003). This lessens the need to reward family-executives with pay packages typically paid to lure or retain professional executives.

These explanations, however, are likely to be viable when the family-controlled firm is a stand-alone firm and not a part of a business group. With only one firm under family control, only a limited number of family members can take managerial positions. A large fraction of family members, with no managerial positions, will simply hold shares as outsiders. With their welfare heavily dependent on the prosperity of the firm, however, these outside family owners will have a strong incentive to carry out their monitoring role and make sure that family-executives do not overpay themselves. Availability of dividend income, job security, and absence of outside job offers will serve as justifications to demand a low compensation level to family-executives.

In this paper, we revisit executive compensation in family-controlled firms, and empirically show that family discount often found in U.S. stand-alone firms cannot be generalized into other types of family firms. Our prediction is that family discount disappears or even switches into family premium if a firm is a part of a large business group.

Which features of a business group may drive such an outcome? First, we predict that the existence of multiple firms within a business group matters. With multiple firms under family control, almost all the family members can be involved in management. With so many managerial positions available for family members, it becomes hard to find family members simply holding shares as outsiders. In their absence, their monitoring and disciplining roles also disappear.

Second, family-controlled business groups are typically formed by pyramiding, circular shareholdings, or a combination of the two, which allows families to control the whole business group without holding large fraction of shares in each individual member firm. In this setting, family-executives in one member firm is not likely to closely monitor those in other member firms as they have little incentive to pay for the costs of monitoring. Moreover, they have control rights that are often high enough to effectively entrench themselves from outside monitoring or disciplining—such as takeover threats—by non-family shareholders.

Third, non-family executives in a business group are typically hired internally. They may move across firms within the same group that forms an internal labor market, but seldom between two different groups (Khanna and Yafeh, 2007). Since they do not have job opportunities outside the group, firms do not compete to retain them with lucrative pay packages. Consequently, their pay level tends to be lower than otherwise. Presence of powerful family executives in the board may lower their pay level even further.

We test our predictions using family-controlled firms in Korea. We believe Korea provides an ideal laboratory setting for a number of reasons. First, it is dominated by a wide range of family-controlled firms: Small stand-alone firms at one extreme and large business groups, also known as chaebol groups like Samsung and Hyundai Motors, at the other extreme. This feature allows us to investigate how family premium varies with the size of business group.

Second, chaebol groups have a large number of member firms. As of April 2014, Samsung, Hyundai Motors, SK, LG, and Lotte—also known as the Big Five—respectively have 74, 56, 80, 61, and 74 member firms. Also, the extensive use of pyramiding and circular shareholding by chaebol groups renders family executives to have control rights high enough to entrench themselves from outside shareholders and to have cash flow rights low enough to lose interest in monitoring other family executives. These features exactly match with the circumstances that lead to an absence of monitoring and disciplining, not only among family members, but also by outside non-family shareholders.

Using 2014 compensation data of 564 executives in 368 family-controlled firms in Korea, we find evidence supporting our predictions. First, we find that family-executives are in general paid more than non-family executives (by 27% more, on average). Second, we find that this family premium is pronounced in firms affiliated to large business groups. Third, we find that pay to family-executives rises with the drop in outside family members’ influence (measured by their aggregate ownership in the firm in excess of the ownership held by the firm’s family executive). Fourth, we find that family premium rises for group chairs, who are least likely to be monitored or disciplined by other family members.

The full paper is available for download here.