Fact and Fiction in Corporate Law and Governance

In a recent article, entitled Fact and Fiction in Corporate Law and Governance, I evaluate two broad elements of corporate governance scholarship—one conceptual and the other methodological. The conceptual element is the “contractarian” framework within which legal academics have analyzed corporate law since the 1980s. My evaluation is not aimed at the characterization of a corporation as a nexus of voluntary associations—a characterization that I think most of us share—but rather at the belief among some legal scholars that market forces lead to optimal governance arrangements within firms, and that a market dynamic leads states to compete to provide value-maximizing corporate law rules. The methodological element that I address is the use of corporate governance indices in empirical studies of corporate governance, an approach that dates back to the 1990s but that became widespread following Gompers, Ishii and Metrick’s development of the G Index in 2003.

My evaluation of the contractarian theory’s optimality claim is based on the empirical work of many authors, supplemented by a new dataset that I recently collected. I conclude that the theory has little if any empirical support, especially where it is supposedly strongest—at the IPO stage. Market forces do not promote innovative or customized governance arrangements as the contractarian theory posits, and there is no basis on which to conclude that the near-uniformity of plain vanilla governance arrangements, supplemented with staggered boards, is socially optimal. The same is true of the race among states for incorporations. The evidence shows that the race does not exist—to the top or the bottom. On the other hand, where the contractarian perspective was viewed as weak—among companies already trading on public markets and controlled by agents—there is reason to believe the market forces over the past three decades have promoted the emergence of mechanisms that reduce agency costs. These include stock based compensation, the use (and threatened use) of shareholder votes by hedge fund activists and traditional institutional investors, and a heightened exposure of managers to reputational damage at the hands of the press. Thus, if there is a dynamic that leads to optimal (or at least good) corporate governance arrangements, it is more likely a web of self-enforcing implicit promises impinging on already-public firms than explicit charter terms, or legal default rules, that a firm adopts at the IPO stage.

My evaluation of governance indices is based on an understanding of how the mechanisms underlying those indices work. Governance indices are widely misunderstood and misused in the finance literature. Many elements of commonly used indices have no potential to entrench management or otherwise reduce firm value. Poison pills are an example. As is well known among lawyers and legal academics, the presence of a pill is equivalent to the latent ability of a firm to adopt a pill when needed, and all firms have that latent ability. Another misconception embedded in the use of governance indices is that a larger number of takeover defenses amounts to greater protection from takeovers. This of course is factually incorrect. Once a company has a staggered board, additional defenses (other than dual class stock) provide no protection at the margin, and even in companies without staggered boards multiple defenses are generally redundant with one another. These two misconceptions have been pervasive in the empirical finance literature since the 1980s, despite efforts by legal academics to dispel them by explaining how takeover defenses actually work. See, for example, John Coates, “Takeover Defenses on the Shadow of the Pill: A Critique of the Scientific Evidence,” 79 Tex. L. Rev. 271 (2000). Economists, however, have resisted these efforts. The widespread use of corporate governance indices reflects ongoing misconceptions regarding takeover defenses and other elements of these indices.

The full article is available for download here.