It pays to have friends

https://doi.org/10.1016/j.jfineco.2008.07.005Get rights and content

Abstract

Currently, a director is classified as independent if he or she has neither financial nor familial ties to the CEO or to the firm. We add another dimension: social ties. Using a unique data set, we find that 87% of boards are conventionally independent but that only 62% are conventionally and socially independent. Furthermore, firms whose boards are conventionally and socially independent award a significantly lower level of compensation, exhibit stronger pay-performance sensitivity, and exhibit stronger turnover-performance sensitivity than firms whose boards are only conventionally independent. Our results suggest that social ties do matter and that, consequently, a considerable percentage of the conventionally independent boards are substantively not.

Introduction

Amid corporate scandals and conflicts of interest, increased board independence is an oft prescribed remedy. Many academic studies examine the monitory benefits of independent boards (e.g., Weisbach, 1988; Byrd and Hickman, 1992; Brickley, Coles, and Terry, 1994; Cotter, Shivdasani, and Zenner, 1997; Mayers, Shivdasani, and Smith, 1997; Paul, 2007), and mutual fund investors are calling for more independent directors to oversee fund managers. Moreover, recent corporate-governance reforms issued by the NYSE, Amex, and Nasdaq require that listed firms (with some exceptions) have independent boards. But are these “independent” boards really independent?

Currently, a director is classified as independent if he has neither financial nor familial ties to the chief executive officer (CEO) or to the firm. Absent from these conventional criteria are social ties; that is, the nonfamilial, informal connections. However, given that agents are not driven solely by economic gains (e.g., Mills and Clark, 1982; Silver, 1990; Uzzi, 1996), social ties are a potentially rich source of a director's dependence to the CEO. Board consultants in the popular press broach this issue, saying that when directors debate whether or how to fire a CEO, “they [the directors] typically need the most help in dealing with their attachment to the CEO” (Business Week, 2007). Our purpose is to incorporate these heretofore omitted ties into the definition of board independence and to examine their relevance to the monitory and disciplinary effectiveness of the board.

Drawing from the economics and sociology literatures, we propose mutual alma mater, military service, regional origin, academic discipline, and industry as indications of an informal tie between a director and the CEO. These mutual qualities and experiences, through homophily (i.e., an affinity for similar others), facilitate interactions and thereby foster personal connections. Whether it is conscious or not, actors enjoy an easier mutual understanding and are more comfortable with others who share similar characteristics and experiences (Marsden, 1987; McPherson, Smith-Lovin, and Cook, 2001), and “contact between similar people occurs at a higher rate than among dissimilar people” (McPherson, Smith-Lovin, and Cook, 2001, p. 416).

Using hand-collected data, we focus on the Fortune 100 firms from 1996 to 2005. We find that, under the conventional measure of independence, 87% of the boards in our sample are classified as independent; that is, these boards have a majority composition of conventionally independent directors. Under our new measure, which augments the conventional definition with the proposed social restrictions, this percentage drops to 62%. Moreover, the incidence of socially linked directors increases as a new CEO's tenure at the firm progresses, suggesting that CEOs select directors along these social dimensions.

To illustrate a conventionally independent board that is not conventionally and socially independent, we consider the board of Cardinal Health. In the year 2000, this board had 13 directors, 10 of whom were conventionally independent of the CEO. However, one conventionally independent director was not only from the same hometown, but also graduated from the same university as the CEO (incidentally, this director provided a job, at his own firm, for the CEO's son). Another conventionally independent director graduated from the same university and specialized in the same academic discipline as the CEO. Similarly, three others shared informal ties with the CEO, and ultimately, only five of the 13 directors were conventionally and socially independent of the CEO.

To test the monitory relevance of these social ties, we examine the differential association between board independence and the level of CEO compensation when we replace the conventional measure of board independence (which does not consider social ties) with our new measure. If these social ties do not affect the disciplinary or monitory capacity of directors, then a director who is conventionally independent but socially linked to the CEO is an equally effective monitor as a director who is both conventionally and socially independent. As such, we would expect no differential association between board independence and the level of compensation attributed to this distinction.

We find no significant difference in the CEO's total annual compensation when a conventionally independent board is present. However, when a conventionally and socially independent board is present, the CEO's total compensation decreases, on average, by $3.3 million. This magnitude is not only statistically significant, but also economically meaningful (average annual compensation is $12.8 million), and we make similar observations with respect to the CEO's annual salary plus bonus. In addition, we find a compensation differential within the subsample of firms with conventionally independent boards; those firms with boards that are conventionally independent but not conventionally and socially independent award a significantly higher level of compensation to their CEOs. These results further signify that it is not only the conventional ties but also the social ties that matter. Moreover, the excess compensation attributed to this type of board extends to a negative association with subsequent operating performance. This evidence punctuates the monitory relevance of these social ties because alternative interpretations of this excess component of compensation (e.g., the CEO of a more complex firm could require a higher level of compensation and a friendlier board) cannot explain its negative association with the firm's subsequent performance.

We also examine the role of social ties in other supervisory and disciplinary actions of the board, such as CEO turnover and pay-performance elasticity. We find that, within the subsample of firms with conventionally independent boards, those CEOs whose boards are not conventionally and socially independent exhibit a lower sensitivity of turnover and compensation to performance. We also find that CEOs whose audit committees are conventionally independent but socially linked (to the CEO) receive larger bonuses than otherwise equivalent CEOs whose audit committees are both conventionally and socially independent, suggesting that social ties affect the audit committee's oversight of financial statements.

Overall, our results suggest that social ties affect how directors monitor and discipline the CEO and that, consequently, a considerable percentage of the boards currently classified as independent are substantively not.

This paper is organized as follows. In Section 2, we discuss the significance of social ties, we develop our hypotheses, and we discuss our measures for social ties. In Section 3, we describe our data sources, variables, and summary statistics. In addition, we examine what determines the incidence of socially dependent directors. In Section 4, we examine the monitory relevance of social ties in the level of compensation, pay-performance elasticity, and CEO turnover. Moreover, we explore alternative interpretations of the excess compensation attributed to social ties. In Section 5, we discuss our contribution to the corporate governance literature, and in Section 6, we conclude.

Section snippets

Motivation, hypotheses, and identification of social ties

Given that actors are not driven solely by financial motives, social ties have a potentially large impact on a director's monitory and disciplinary capacity. In particular, when two actors share a social bond, there is a shift in normative expectations, whereby their actions are governed by communal norms, which promote mutual caring and trust, as opposed to exchange-based norms, which promote dispassionate reciprocation (Mills and Clark, 1982; Silver, 1990). Furthermore, a social relationship

Data description

This section discusses our data sources and regression variables. We also explore the determinants of a board's social composition, in particular the hypothesis that CEOs desire directors along our proposed social characteristics.

Empirical results

We now proceed to examine the effect of social ties on executive compensation. In Table 4, we present summary statistics on CEO compensation and various firm characteristics (Appendix C contains a correlation matrix of variables, including the governance variables from Table 2 and our dependent variable, CEO compensation). The overall average salary plus bonus and total compensation are $3.8 million and $12.8 million, respectively (Column 1). In a cross-panel comparison, we observe that CEO

Contribution and discussion

Our paper contributes to the governance literature in the following ways. First, we propose a measure of social ties between directors and their CEOs, and we provide evidence of its practical applicability. In contrast to the survey-based measures generally employed by studies pertaining to social embeddedness (e.g., Uzzi, 1996, Uzzi, 1999; Westphal, 1999; Ingram and Roberts, 2000; McDonald and Westphal, 2003; Westphal, Boivie, and Chng, 2006), our measure is based on several broadly available

Conclusion

Directors are not dispassionate. It is not only financial and familial ties that interfere with their disciplinary and monitory roles; social ties also matter. Here, we propose several observable characteristics that likely connect a director (socially) to the CEO: mutual alma mater, military service, regional origin, discipline, and industry. We augment the conventional definition of board independence with these additional social restrictions and find that the percentage of independent boards

References (69)

  • R.B. Adams et al.

    A theory of friendly boards

    Journal of Finance

    (2007)
  • R.K. Aggarwal et al.

    Executive compensation, strategic competition, and relative performance evaluation: theory and evidence

    Journal of Finance

    (1999)
  • G.P. Baker et al.

    CEO incentives and firm size

    Journal of Labor Economics

    (2004)
  • G.P. Baker et al.

    Compensation and incentives: practice vs. theory

    Journal of Finance

    (1988)
  • R.D. Banker et al.

    Sensitivity, precision, and linear aggregation of signals for performance evaluation

    Journal of Accounting Research

    (1989)
  • L.A. Bebchuk et al.

    The costs of entrenched boards

    Journal of Financial Economics

    (2005)
  • L.A. Bebchuk et al.

    What matters in corporate governance?

    Review of Financial Studies

    (2009)
  • L.A. Bebchuk et al.

    Executive compensation as an agency problem

    Journal of Economic Perspective

    (2003)
  • J.A. Brickley et al.

    Outside directors and the adoption of poison pills

    Journal of Financial Economics

    (1994)
  • J.W. Byrd et al.

    Do outside directors monitor managers?

    Journal of Financial Economics

    (1992)
  • L. Cohen et al.

    The small world of investing: board connections and mutual fund returns

    Journal of Political Economy

    (2008)
  • Cohen, L., Frazzini, A., Malloy, C.J., 2008b. Sell side school ties. Unpublished working paper. Harvard University,...
  • J.L. Coles et al.

    Boards: does one size fit all?

    Journal of Financial Economics

    (2008)
  • J.E. Core et al.

    Corporate governance, chief executive officer compensation, and firm performance

    Journal of Financial Economics

    (1999)
  • J.F. Cotter et al.

    Do independent directors enhance target shareholder wealth during tender offers?

    Journal of Financial Economics

    (1997)
  • Crosse, C., Hocking, B., 2004. Social rehabilitation: what are the issues? In: DVA National Rehabilitation Conference,...
  • R.M. Cyert et al.

    Corporate governance, takeovers, and top-management compensation: theory and evidence

    Management Science

    (2002)
  • P.M. Dechow et al.

    Causes and consequences of earnings manipulation: an analysis of firms subject to enforcement actions by the SEC

    Contemporary Accounting Research

    (1996)
  • J. Elster

    Social norms and economic theory

    Journal of Economic Perspective

    (1989)
  • O. Faleye

    Classified boards, firm value, and managerial entrenchment

    Journal of Financial Economics

    (2007)
  • E.F. Fama et al.

    Industry costs of equity

    Journal of Financial Economics

    (1997)
  • E.F. Fama et al.

    Separation of ownership and control

    Journal of Law and Economics

    (1983)
  • E.M. Fich et al.

    Are busy boards effective monitors?

    Journal of Finance

    (2006)
  • S. Finkelstein et al.

    Chief executive compensation: a study of the intersection of markets and political processes

    Strategic Management Journal

    (1989)
  • M.J. Friedman

    Veterans’ mental health in the wake of war

    New England Journal of Medicine

    (2005)
  • J.J. Gaver et al.

    Additional evidence on the association between the investment opportunity set and corporate financing, dividend, and compensation policies

    Journal of Accounting and Economics

    (1993)
  • P. Gompers et al.

    Corporate governance and equity prices

    Quarterly Journal of Economics

    (2003)
  • M. Granovetter

    The impact of social structure on economic outcomes

    Journal of Economic Perspective

    (2005)
  • B.J. Hall et al.

    Are CEOs really paid like bureaucrats?

    Quarterly Journal of Economics

    (1998)
  • K.F. Hallock

    Reciprocally interlocking boards of directors and executive compensation

    Journal of Financial and Quantitative Analysis

    (1997)
  • P.M. Healy

    The effect of bonus schemes on accounting decisions

    Journal of Accounting and Economics

    (1985)
  • B.E. Hermalin et al.

    The effects of board composition and direct incentives on firm performance

    Financial Management

    (1991)
  • B.E. Hermalin et al.

    Endogenously chosen boards of directors and their monitoring of the CEO

    American Economic Review

    (1998)
  • P. Ingram et al.

    Friendships among competitors in the Sydney hotel industry

    American Journal of Sociology

    (2000)
  • Cited by (602)

    • Shareholder nomination of independent directors: Evidence from China

      2024, International Review of Financial Analysis
    • CEO-director ties and board gender diversity: US evidence

      2023, Journal of Behavioral and Experimental Finance
    • Social ties and related party transactions

      2023, Journal of International Accounting, Auditing and Taxation
    View all citing articles on Scopus

    The authors thank the referee, Tarun Chordia, Kira Fabrizio, Clifton Green, Narasimhan Jegadeesh, Ronald Masulis, Shehzad Mian, Peter Roberts, Belén Villalonga, the brown-bag participants at Emory University, and the PhD seminar participants at Vanderbilt University for very helpful comments. This research was conducted while both authors were doctoral students at Emory University.

    View full text