How family firms solve intra-family agency problems using interlocking directorates: an extension.

This commentary makes two contributions to a better understanding of interlocking directorates in family firms. First, we compare agency costs in family and nonfamily firms. Second, we present additional propositions on how interlocking directorates address agency issues in family firms.

**********

Using the concept of community-level social capital, Lester and Cannella (2006) discuss how interlocking directorates can address intra-family agency problems as well as some of the characteristics of the interlocking directorates that family firms are likely to employ. To put their paper in context, we use the accounting of agency costs in family and nonfamily firms as presented by Chrisman, Chua, and Litz (2004). We then extend Lester and Cannella's contributions by presenting additional propositions on how interlocking directorates address agency issues in family firms.

The Agency Context

Agency costs consist of the combination of costs incurred by a firm in terms of lower performance through its unwillingness or inability to effectively deal with agency problems, and of costs incurred through the incentives and monitoring activities used to deal with information asymmetries and align the interests or actions of an agent with the interests of a principal (Jensen & Meckling, 1976). Chrisman, Chua, and Sharma (2005) observed that the agency literature covers agency problems that arise between owners and managers, owners and lenders, and majority and minority owners. Schulze, Lubatkin, Dino, and Buchholtz (2001) have added intra-family agency costs to this list. They propose that intra-family agency costs may be caused by asymmetric altruism, and may exist even in the absence of asymmetric information. For example, some agency problems of family firms may be a consequence of an inability or unwillingness to enforce contracts on family members rather than incomplete contracting (Bernheim & Stark, 1988).

Based on the possibility of agency problems between the parties mentioned above, the research question that has begun to receive attention is whether the agency cost of the family form of organization is higher, lower, or equal to those of organizing as a nonfamily firm. As displayed in the following inequality (Chrisman et al., 2004, p. 339), the hypothesis in vogue is that the agency costs are not equal.

(1) ALT + [OM.sub.F] + [OL.sub.F] + [DM.sub.F] [not equal to] [OM.sub.NF] + [OL.sub.NF] + [DM.sub.NF]

where the subscript F stands for family firms, NF for nonfamily firms, and:

ALT = Intra-family agency costs arising from asymmetric altruism and other family interactions.

OM = Agency costs arising from separation of ownership and management.

OL = Agency costs arising from information asymmetries and conflicts of interests between owners and lenders.

DM = Agency costs arising from information asymmetries and conflicts of interests between dominant and minority shareholders.

Observing that recent large sample studies of publicly traded firms (e.g., Anderson & Reeb, 2003) suggest that family firms have higher economic performance than nonfamily firms, Lester and Cannella (2006) asked how this could be, when family firms must incur the costs of dealing with the additional intra-family problems. Since this would result only if the following inequality holds, Lester and Cannella surmised that successful family firms must have developed strategies, structures, or processes to deal with intra-family agency costs efficiently. (1)

(2) ALT < ([OM.sub.NF] + [OL.sub.NF] + [DM.sub.NF]) - ([OM.sub.F] + [OL.sub.F] + [DM.sub.F])

Actually, if the agency costs of nonfamily firms--with respect to the owner-manager, owner-lender, and majority-minority shareholder relationships--are much higher than those of family firms, i.e., the right-hand side of the inequality in equation 1 is as large as some researchers believe (e.g., Fama & Jensen, 1983; Jensen & Meckling, 1976), then intra-family agency costs might not have to be addressed for family firms to obtain higher economic performance. Put differently, dealing with intra-family agency problems, even efficiently, is not a necessary or sufficient condition for family firms to outperform nonfamily firms. On the other hand, dealing efficiently with the sources of intra-family agency costs should be a sufficient condition for family firms that do so to outperform family firms that ignore intra-family agency costs or deal with them inefficiently.

Propositions about Interlocking Directorates

Lester and Cannella (2006) propose interlocking directorships as one means by which successful family firms might solve intra-family agency problems efficiently. They suggest that successful family firms use specific interlocking directorates to build community-level social capital, which then helps reduce or eliminate intra-family agency costs.

Social capital is typically represented as having two key dimensions, a "structural" dimension--characterized by the overall configuration of the network or pattern of connections between actors--and a "relational" dimension--characterized by the actual relationship or bonds between actors that enable them to make claims on one another (Portes, 1998). Lester and Cannella (2006) focus on the structural dimension. Their conceptual contribution focuses on the corporate governance structures that would help family firms overcome intra-family agency costs effectively and efficiently. Their five propositions are:

Proposition 1: Family-controlled corporations will have a higher proportion of board interlocks to other family firms than will nonfamily-controlled corporations.

Proposition 2: Family-controlled corporations will have a higher proportion of interlocks to other firms in the local area and nearby geographic areas than will nonfamilycontrolled corporations.

Proposition 3: Relative to nonfamily corporations, the interlocks of family business boards of directors are more likely to be with firms that share similar strategic orientations.

Proposition 4: The most prestigious and successful family firms are the most likely to be located in the center of the intercorporate network of family-controlled companies.

Proposition 5: For family-controlled corporations, there is a positive relationship between the ratio of board interlocks to other family firms and overall performance.

Graphically, propositions 1-4 may be represented, as in the networking literature, as shown in Figure 1.

[FIGURE 1 OMITTED]

These propositions flow from Lester and Cannella's (2006) arguments about the types of directors who would be most helpful to family firms in solving intra-family agency problems. The comparison is relevant since nonfamily firms do not have intra-family agency costs; therefore family firms should exhibit greater tendencies to have such directors on their boards. The last proposition then concludes that family firms possessing these types of directors on their boards will outperform family firms that do not.

While these propositions add important insights and research directions to the literature, they do not exhaust the propositions that can be derived from Lester and Cannella's (2006) arguments. For example, they do not present intermediate propositions linking the types of directors with the sources of the abilities that might cause them to be used on family business boards. The linkages between the greater abilities of certain types of directors to help family firms deal with the sources of intra-family agency costs and performance could also be specified more precisely. Finally, proposition 5 needs to be clearer about the types of interlocking directorates that lead to superior performance.

Our extensions focus on propositions that deal with these issues. By focusing on intermediate propositions we are arguing that it would be useful to empirically examine the proposed causes of the expected relationships, as well as the relationships themselves. In doing so, we can open the "black box" to test more fully the theoretical arguments underlying the propositions. This is especially important since there may be other explanations for the presence of interlocking directorates in family firms (e.g., serendipity, dominance of family firms in a particular environment, etc.) that have little to do with social capital or reducing intra-family agency costs.

The additional propositions that we present here are based on the arguments made by Lester and Cannella (2006) about which types of directors are most likely to help family firms solve intra-family agency problems. Propositions 1 and 2 are ultimately dependent on their argument related to resource sharing and trust. A director's resources, in terms of knowledge, capabilities, or personal networks, must be shared to help family firms deal with their intra-family agency problems. Resource sharing is more likely to take place if there is trust, and trust is more likely to take place if there are relationships grounded in shared values and experiences. As a result, their arguments about the determinants of trust lead naturally to the following additional propositions:

Proposition 1a: The likelihood of relationships based on shared values and experience is greater between owners of family firms and outside directors from other family firms than between owners of family firms and outside directors from nonfamily firms.

Proposition 1b: There is more trust between owners of family firms and outside directors from other family firms than between owners of family firms and outside directors from nonfamily firms.