Do Boards Have Style? Evidence from Director Style Divergence and Board Turnover

A board of directors performs essential strategic and oversight roles that maximize the value of the shareholders’ residual claim. However, despite careful selection of board members, too often boards neither reach their full potential nor perform their necessary governance obligations. The role of structural characteristics such as firm size and composition have been thoroughly explored. However, evidence tying traditional board variables to board and firm performance remains contradictory or incomplete. As a result, there has been an increasing interest in the role of behavioral and cognitive traits to explain how organizations work. Beginning with the seminal work of Bertrand and Schoar (2003),[1] scholars have used manager fixed-effects to test whether firms are subject to the influence of an individual’s managerial style. A style is a mechanism by which individuals express their self-definition and values through various policies and outcomes inside and outside the firm.

We extend Bertrand and Schoar’s work to the role of board-member style. The study of a style of an individual board member or a collective board of directors is a fertile source of understanding. Unlike managers who operate individually and hierarchically, boards typically function collaboratively and co-equally. As a result, board members display group behaviors such as coalitions, social isolation, and management of dissent. Also, unlike CEOs who are limited to leading one firm, boards of directors can be populated by “busy board members” who serve on numerous boards at once. These additional traits can be exploited to understand more precisely how style works in top management teams. Apparently due to the difficulty of acquiring evidence of board-member style, few authors have explored this subject, leaving a gap in the literature.

We close this gap by assembling a firm-director matched panel dataset that makes it possible to track individual members of a board of directors across firms and over time. We begin with directors who have been on boards of at least two firms and remain on boards for at least three years. Applying the methodology developed by Abowd, Kramarz and Margolis (1999),[2] we leverage it to distinguish between director fixed-effects and firm fixed-effects on corporate policies related to investment, financial policy, operations and performance, and compensation and governance. This methodology also has the advantage of enabling us to estimate the fixed effects of multi-board directors and the fixed effects of a single-board director or firm who sits on a board with a multi-board director. This results in a sample size that is significantly larger than prior employer-employee datasets and is susceptible to less sample selection bias.

We find robust evidence of firm style influencing corporate decisions related to investment, financial policy, operations and performance, and compensation and governance. For all investment policy proxies, the estimates of director style are significant at the 1 percent level, with Investment and Goodwill R-square, a measure of the overall explained variation in this corporate policy variable, increasing by over 4 percent when director fixed effects are included. For financial policy proxies, we find meaningful firm and director fixed effects for cash holdings and leverage variables, with the Leverage adjusted R-square increasing by 7.51 percent when including director fixed-effects. For operations and performance measures, SG&A (6.85 percent), Advertising (5.72 percent), and R&D (7.57 percent) results show adjusted R-square increases as well. Finally, compensation and governance proxies find improvements in R-square due to the inclusion of director fixed-effects for all three compensation metrics of CEO Total Compensation (5.85 percent), CEO Salary plus Bonus (7.11 percent), and CEO Stock Compensation (4.57 percent), as well as CEO duality (at least 4.5 percent) as a governance proxy. Furthermore, the average of all board member fixed-effects in a given year also affects future performance. Firms with high average board style fixed-effects for board size have lower abnormal returns, consistent with prior work on the relationship between board size and firm value. Conversely, firms with high average board style fixed-effects for advertising expense ratios have higher abnormal returns, consistent with prior evidence showing the value of advertising to the firm.

We also find significant evidence that diverging director style relative to that of boards affects board composition. Directors whose style diverges from that of the board as a whole tend to leave. Style differences regarding CEO Total Compensation, Advertising, and Acquisition on PP&E stand out as particularly influential. Notably, style differences over a CEO’s Total Compensation amplify the likelihood of style-divergent director departure when the CEO’s power (measured by total compensation) is high, while style-divergent director departure is muted when firm performance (measured by firm ROA) is high. Finally, we find robust evidence that survival on key board committees is sensitive to the style-divergence of board members for a variety of variables. Notably, the greater a board member’s style differs regarding the CEO’s Total Compensation or Salary and Bonus from other members, the more likely that the style-divergent board members will leave the compensation committee.

Our results differ from the neoclassical account that firm leaders are homogenous inputs of production with little if any role for idiosyncratic influence on the firm. Instead, we provide evidence that economically significant decisions may be influenced by individual board members and their respective board styles. Board members may interpose their personal attributes and attitudes when making decisions and engaging in interpersonal behavior on the board. We thus further open the ”black box” of the board of directors in order to better understand how internal board dynamics work.

Our results also extend and reinforce various literatures. First, we provide evidence that the presence and influence of board management style is both present and robust in an environment that operates substantially differently than typical executive decision-making. Second, we also provide evidence supporting the importance of board stability, a subject that has been associated with a variety of conditions, including poorer initial performance among IPO firms, organizational crises, and bankruptcy filings. Third, we help understand the role of dissent on boards. We report evidence that boards displaying undesirable levels of style divergence have mechanisms that will encourage differing board members to leave. At the same time, boards with higher style differences for R&D and cash ratios outperform those with lower board-level style divergences. Thus, while divergent style can be suboptimal, some difference in style can generate better firm performance.

We also provide evidence that board style has a significant influence on key committee composition across a variety of financial and compositional variables.  We notably find that CEO compensation variables are significantly tied to style-divergent director departure from the compensation committee. This finding supports research showing that CEOs may use their influence to sway board members toward a pro-CEO agenda. Finally, our work contributes to understanding group dynamics and idiosyncratic socio-cognitive factors on board interactions.  With increasing interest in the cognitive traits of key organizational actors, our research on board style paves the way for further understanding of how individual board members, and potentially board coalitions, shape and reinforce board norms and practices. Insights from style research also further our understanding of board members’ “construed reality” and how that reality affects their field of vision (where board members look and listen), selective perception (what they actually experience), and interpretation (the meaning attached to what they experience).

ENDNOTES

[1] Bertrand, M. and Schoar, A., 2003. Managing with style: The effect of managers on firm policies. The Quarterly Journal of Economics, 118(4), pp. 1169-1208.

[2] Abowd, J.M., Kramarz, F. and Margolis, D.N., 1999. High wage workers and high wage firms. Econometrica, 67(2), pp.251-333.

This post comes to us from professors Robert C. Bird, Paul Borochin, and John D. Knopf at the University of Connecticut, and from Luchun Ma, who is a PhD student at the university. It is based on their recent article, “Do Boards Have Style? Evidence from Director Style Divergence and Board Turnover,” available here.