Tournament of Managers: Lessons from the Academic Leadership Market.

• Author: Rodrigues, Usha R.

1. INTRODUCTION II. EXECUTIVE COMPENSATION CONTROVERSY AND REFORM A. The Problem With Disclosure: The Market for External CEOs is not Robust III. WHY FAVOR AN INTERNAL CANDIDATE? A. Internal Candidates Perform Better B. Firm-Specific Capital C. Information Asymmetry D. Tournament Theory E. Irrelevance of Leadership Theories IV. THE LACK OF A TOURNAMENT IN ACADEMIA A. Selection Process: President B. Literature V. THE DATA A. Descriptive Statistics B. Methodology 1. Hypothesis One 2. Hypothesis Two 3. Hypothesis Three 4. Hypothesis Four 5. Hypothesis Five C. Apples and Oranges: The Generalizability Question VI. IMPLICATIONS VII. CONCLUSION I. INTRODUCTION

   In Brehm v. Eisner, a defining corporate law case, (1) critics viewed the Disney board's decision to hire Michael Ovitz, and then fire him fourteen months later with a $140 million severance package, as the epitome of the problem of excess executive compensation: compliant, complacent boards who were overly generous with shareholder money. (2)

   Despite the many law review articles focused on the Disney case, and the countless others that cite it, one important lesson remains overlooked by the legal literature. Corporate law scholars, like the Delaware courts, focused on the board's faulty process in negotiations to hire and fire Ovitz. (3) And, catalyzed by Lucian Bebchuk and Jesse Fried's Pay Without Performance: The Unfulfilled Promise of Executive Compensation, they have discussed the topic of executive compensation at great length. (4)

   However, legal scholars start with the firing, all but ignoring the first step in the process: the hiring of Michael Ovitz as the next chief executive of Disney. (5) The responsibility of a public company board is to manage in situations where internal management faces a conflict of interest, and chief among these is the selection, monitoring, and firing of the CEO. (6) Yet, the board's choice of Ovitz--an industry outsider--received relatively little attention.

   Hiring Ovitz, an external candidate, was an anomaly. For-profit corporations generally favor internal candidates, hiring externally only about 22% of the time. (7) The literature is divided as to why boards would favor internal candidates over external ones, (8) but the answer to this question is of paramount importance for evaluating arguments about compensation. Optimal contracting scholars point to the market for managerial talent as a major justification for large pay packages. (9) "If we don't pay our CEO the market rate, we will lose her," or so the argument goes. The compensation committees of public firms typically engage compensation consultants who "benchmark" pay by reporting the pay packages of the chief executives of comparable companies. (10) Critics of benchmarking cite its inexorable tendency to ratchet up pay: if all boards regularly aim for their CEO to exceed the median pay level, pay inevitably rises. (11) But the numbers make clear a more fundamental issue: most CEOs come from inside an organization. (12) This simple fact casts doubt on the existence of a market for managerial talent. And, if there is no market for CEOs, then perhaps firms can safely pay them significantly less.

   A potential objection is that benchmarking is what prevents more of a market for CEOs: by this logic, firms who break with industry practice and pay their CEOs less than market rate risk poaching by other firms. (13) Thus, the question this Article seeks to answer: why do publicly traded firms favor internal CEOs, and what effect does that reason have on determining the appropriate level of executive compensation?

   One potential explanation for favoring internal CEOs is that they perform better. Unlike legal scholars, other literatures focus a great deal of attention on the question of CEO selection. (14) In particular, the focus is on the eternal "make or buy" question: should companies promote internal candidates, or instead hire a replacement from outside the firm? The literature on the topic is vast and not limited in scope to the CEOs of public companies--or, indeed, traditional business settings. There are also studies of internal versus external NCAA basketball coaches and professional baseball managers, among other leadership contexts. (15) Yet, despite the myriad studies, results remain inconclusive--they do not reliably show that internal candidates perform better. (16)

   A second explanation is that internal candidates spend years developing firm-specific capital. (17) As employees rise through the ranks of a company, they learn not only the industry, but also what makes their particular organization run. (18) They build up relational capital, enabling them to navigate the intricacies of the company. (19) They understand the strengths and weaknesses of the organization, and can hit the ground running with relatively little time required to acclimate to their new role.

   Information asymmetry, a third explanation, focuses not on what the internal candidate knows about the firm, but instead on what the board knows about the internal candidate. (20) Internal candidates participate, in a real sense, in a years-long interview process. Boards generally have a sense of a current employee's reputation within the organization, and likely have worked with them and interacted with them socially on a variety of occasions. (21) Boards are at a comparative disadvantage when it comes to external candidates. They generally have relatively little time to interview outsiders and conduct due diligence as to their character, work habits, and knowledge. Typically, the external candidate is interviewing in secret, further limiting the board's ability to investigate an individual's character and attributes. (22)

   The final explanation for public company boards favoring internal candidates is to foster a tournament system that incentivizes lower ranks, and keeps them competing for the ultimate prize of leadership over the organization. (23) When it is hard for employers to monitor output reliably, then it may be preferable to structure a "tournament" that enables the employer to measure, and ultimately rank, employees against one another rather than against an objective output measure. (24) Marc Galanter and Thomas Palay famously applied this principle to promotion at a law firm: associates compete in a tournament against one another, for the prize of making partner. (25)

   Economists Edward P. Lazear and Sherwin Rosen give an excellent example of how their theory plays out in explaining executive compensation. Typically, the "salary of ... the vice-president of a particular corporation is substantially below that of the president of the same corporation." (26) Yet, somehow, "on the day that a given individual is promoted from vice-president to president, his salary may triple. It is difficult to argue that his skills have tripled in that one-day period...." (27) But this compensation structure is not surprising if viewed as a tournament, where high CEO pay reflects not current productivity as president, but rather efforts exerted at lower levels to compete in pursuit of CEO status. (28) The practice of favoring internals reassures current employees that there is a limited pool of competition for the grand prize, and that outsiders will not swoop in to claim the reward an internal aspirant has spent a substantial portion of her career working towards.

   No articles have discussed in depth which of these theories--firm-specific capital, information asymmetry, or tournament theory--best explains the favoring of internal candidates. This Article offers support for the tournament theory, by way of an unexplored source: the academic leadership market.

   Academia offers an ideal setting to test the power of tournament theory for a simple reason: tenured academics do not engage in a leadership tournament. (29) The other two explanations for favoring internal candidates, firm-specific capital and information asymmetry, still apply: internal candidates know more about their university, and the institution knows more about them, as compared to outsiders. (30) Tournament theory, in contrast, does not apply. (31) The typical academic professes something of a horror of administration, sometimes termed moving to the "dark side." (32) Thus, if the tournament theory is the dominant explanation for favoring internal candidates in the for-profit setting, internal candidates should not be as favored in the academic setting. If, in contrast, firm-specific capital or informational asymmetry explain the dominance of internal candidates, then internal candidates should likewise predominate in the academic leadership setting.

   I conduct a study from 1995-2016 of top fifty university presidents, and find support for the tournament theory of CEO selection. Only 24% of the presidents of leading universities are internal. This notable bias in favor of external candidates in a setting with no tournament, but where firm-specific capital and information asymmetry still exist, suggests that tournament theory explains the internal CEO bias. The implications for explaining why public firms favor internal CEOs are profound.

   The Article proceeds as follows: Part II describes the debate surrounding executive pay levels. Notably, the optimal contracting view identifies the market for managerial talent as one justification for high rates of executive pay. However, given the strong tendency to favor internal candidates, such a market may be more fiction than fact. Part III describes the potential reasons for preferring inside candidates: firm-specific capital, information asymmetry, and tournament theory. Part IV moves to the academic setting to test the tournament theory. After establishing the absence of an academic leadership tournament and describing the process of selecting a university president, this Part briefly surveys the scant literature on university presidents. Part V provides the data, a sample of university...