3.1 Introduction

When stock markets are efficient, it becomes harder for management to obtain equity financing for a proposed acquisition as the marginal value of the acquisition decreases.1 In a similar vein, when corporate voting is efficient, it will become harder for management to obtain shareholder approval for a proposed acquisition as the marginal value of the acquisition decreases.2 Thus, both market efficiency and voting efficiency are of critical importance for the efficient allocation of resources in the economy.3 Why, then, have finance and legal scholars devoted such tremendous intellectual efforts to examining market efficiency and so little to examining voting efficiency?

One explanation is that most empirical studies of market efficiency test a straightforward hypothesis: if markets are efficient, it is impossible to consistently outperform the market. The early evidence indicated that this was indeed the case, a remarkable finding that motivated researchers to come up with explanations.4 Among them were Ronald Gilson and Reinier Kraakman, whose seminal paper The Mechanisms of Market Efficiency revealed how the market aggregates information.5 The sub sequent finding that markets are not all that efficient required its own explanations. These were provided by behavioral finance research, which showed that investors have bounded rationality and that there are limits to arbitrage.6 As a result of these scholarly efforts, we now have a nuanced view of market efficiency.

By contrast, there appears to be no equally straightforward hypothesis that can be tested to study voting efficiency. Indeed, the debate has been largely theoretical, and limited in scope. To the extent Milton Friedman's characterization of shareholders as "owners" of the firm lelt room for questioning the efficiency of voting, the debate pretty much seems to have been silenced by Frank Easterbrook and Daniel Fischel's contractarian argument that because shareholders are the residual claimants of the firm they have the appropriate incentives to make discretionary decisions.7 But surely incentives alone do not suffice. Whether shareholders make the right decisions, such as rejecting a proposed merger if the marginal value is too low, is ultimately an empirical question.

The question of whether shareholders make the right decisions has perhaps never been more important. In the wake of the financial crisis, policymakers across the globe are re-thinking the role of shareholders. In the United States, it is felt that shareholders lacked the means to intervene in portfolio companies. Accordingly, their powers have recently been expanded in the area of executive compensation, and may soon be expanded in other areas.8 In the United Kingdom, where shareholders already had broad powers, it is felt they were merely slow to act, and shareholders are called upon to engage with portfolio companies.@@9 The academie debate, meanwhile, is focused on the trade-off between enabling shareholder monitoring to reduce agency costs and preserving managerial discretion to run the business10 Thus, both policymakers and academies are ignoring the preliminary question of whether shareholders are capable of making the right decisions. The aim of this Chapter is to make some progress on this question, taking into account recent advances in law and fmance.

The fundamental insight driving this Chapter is that investment decisions and voting decisions are similar in the sense that both are driven by an investor's belief as to the net present value of an asset. In the case of investment decisions that asset is the share, which represents a pro rata entitlement to the firm's future cash flows. In the case of voting decisions, the asset could be a proposed acquisition, which may be characterized as a real option. In each case, by executing his decision, the investor reveals information underlying his beliefs. In the case of an investment decision, this information is aggregated through the market system, and in the case of a voting decision it is aggregated through the voting system.

The notion that the market is a system for information aggregation can be traced to Friedrich Hayek, who stressed the importance of utilizing knowledge dispersed among people and argued that we "must look at the price system as . . . a mechanism for communicating information if we want to understand its real function."11 Finance scholars, of course, have done precisely this. Sanford Grossman, for one, demonstrated that the competitive system aggregates all the market's information in such a way that the equilibrium price summarizes all the information in the market.12

The notion that voting, too, is a system for information aggregation can be traced to eighteenth century French philosopher Marquis de Condorcet. His Jury Theorem holds that where there are a number of voters who must decide on two alternatives, one of which is correct and the other incorrect, and the probability that any given voter will vote for the correct alternative is greater than 0.5 (i.e., that such voter is more likely to be right than wrong), then the probability that a majority vote will select the correct alternative approaches 1 as the number of voters gets larger.13 Moreover, the majority will be more likely to vote for the correct alternative than any individual voter.14 The Jury Theorem serves as a theoretical foundation for two intriguing recent books, James Surowiecki's The Wisdom of the Crowds and Cass Sunstein 's Infotopia: How Many Minds Produce Knowledge, both of which vividly describe the variety of contexts in which crowds display remarkable wisdom—from football prediction markets to Wikipedia.15

Meanwhile, a growing number of scholars refer to the Jury Theorem as a theoretical foundation for corporate voting.16The basic proposition reads something like this: in a choice between two alternatives (e.g., the firm merges or not), assuming that shareholders vote for the correct option with probability greater than 0.5, then, as the number of shareholders increases, the probability that a majority vote taken at the shareholders' meeting will select the correct (i.e., value maximizing) alternative tends toward certainty.17

This Chapter moves beyond mere references to the Jury Theorem and toward a comprehensive understanding of the determinants of shareholders' ability to make the right decisions as a group. Adrian Vermeule justly notes that the Jury Theorem rests on fragile mechanisms that apply only under narrow conditions.18 As a first step, we need to identify these conditions and determine whether they hold given what we know about how investors make decisions in real life. Fortunately, finance research has already taught us a lot about how investors make investment decisions and how these decisions impact market efficiency. We can make great progress by using these insights to assess how investors make voting decisions and how these decisions impact voting efficiency. As a second step, we need to expand our view by exploring models of crowd wisdom other than the Jury Theorem. Taking these two analytical steps provides us with a taxonomy of what might be referred to as the mechanisms of voting efficiency.

To be sure, even if we look at both the market and voting as systems for information aggregation, differences remain. But a comparative analysis nevertheless yields valuable insights; indeed, this Chapter is not the first to link the two systems. Sunstein offers a "Condorcetian interpretation" of Hayek, arguing that "[p]recisely because many people are making purchasing decisions, their aggregate judgments are highly likely to be correct, at least if most purchasers have relevant inforrnation."19 Niall Ferguson, in his book The Ascent of Money, puts it quite clearly when he states that lijn effect, stock markets hold hourly referendums on the companies whose shares are traded there: on the quality of their management, on the appeal of their products, on the prospects of their principal markets."20

Section 3.2 of this Chapter offers a working definition of voting efficiency, according to which voting is deemed efficient if it leads to an outcome that maximizes shareholder value. Section 3.3 represents the core of the Chapter and identifies and explores four key mechanisms of voting efficiency, or elements that lead to—and limit—voting efficiency.21 The first is informed voting: shareholders need to have at least some information to ensure that they are more likely to be right than wrong. Whereas prior research has focused almost exclusively on this mechanism, it merely forms the starling point of our inquiry. The second is rational voting: the possession of information will only increase the probability that shareholders vote for the correct option if they process such information rationally. As we will see, some of the cognitive biases that have been found to affect shareholders' investment decisions may equally affect their voting decisions. The third is independent voting: to come to a collective judgment that is more accurate than the average individual judgment, each shareholder needs to independently arrive at a judgment on which option maximizes shareholder value by making use of his or her personal cognitive skills. The fourth is sincere voting: shareholders need to vote in accordance with that judgment. When shareholders have heterogeneous preferences and some vote with a view to maximizing their private interests rather than their pro-rata share of the firm's future cash flows, the probability that a majority of the shares is voted for the correct option decreases dramatically.

Even if the initial distribution of information, skills and preferences among shareholders is such that a majority of the shares risks being voted in favor of the incorrect option, arbitrage can reallocate voting power in the hands of shareholders with superior information and skills and with appropriate incentives, thereby increasing the probability that a majority of the shares will be voted in favor of the correct option. Section 3.4 identifies three arbitrage strategies: (1) share trading, (2) proxy solicitation, and (3) vote buying, and analyzes cost constraints and legal constraints to these strategies. The analysis suggests that limits of voting arbitrage are significant and affect voting efficiency much in the same way as limits of securities arbitrage affect market efficiency.

Finally, section 3.5 applies the insights from the Chapter to two issues that are currently being studied by the U.S. Securities Exchange Commission (S.E.C.) and policymakers around the world: voting without a corresponding economie interest (`empty voting'), and the major influence of proxy advisers such as RiskMetrics, which recently acquired Institutional Shareholder Services (ISS). Costs and benefits are brought into sharper focus by showing that each issue involves a trade-off between the various mechanisms of voting efficiency. Several options are then presented to mitigate the costs while fostering the benefits.

The Chapter concludes by summarizing policy implications and formulating hypotheses that can be tested in future empirical research.