The Uneasy Case for Favoring Long-Term Shareholders
abstract. This Article challenges a persistent and pervasive view in corporate law and corporate governance: that a firm’s managers should favor long-term shareholders over short-term shareholders, and maximize long-term shareholders’ returns rather than the short-term stock price. Underlying this view is a strongly held intuition that taking steps to increase long-term shareholder returns will generate a larger economic pie over time. I show, however, that this intuition is flawed. Long-term shareholders, like short-term shareholders, can benefit from managers’ destroying value—even when the firm’s only residual claimants are its shareholders. Indeed, managers serving long-term shareholders may well destroy more value than managers serving short-term shareholders. Favoring the interests of long-term shareholders could thus reduce, rather than increase, the value generated by a firm over time.
author. Dane Professor of Law, Harvard Law School. For helpful comments, I am grateful to Jesus Alfaro, Adi Ayal, Lucian Bebchuk, Ilan Benshalom, Bill Bratton, John Coates, Paul Edelman, Einer Elhauge, Luca Enriques, Allen Ferrell, Jill Fisch, Mireia Giné, Assaf Hamdani, Sharon Hannes, Ron Harris, Ehud Kamar, Louis Kaplow, Reinier Kraakman, Adam Levitin, Yoram Margalioth, Alan Miller, Jacob Nussim, Gideon Parchomovsky, Ariel Porat, Mark Ramseyer, Morgan Ricks, Maribel Saez, Steve Shavell, Doron Teichman, Randall Thomas, Michael Wachter, Yesha Yadav, Omri Yadlin, and participants at the Harvard Law and Economics Seminar and faculty workshops and conferences at Bar-Ilan University, CERGE-EI, Haifa University, Harvard Law School, Hebrew University, IDC-Herzliya, Tel Aviv University, Universidad Carlos III de Madrid, U.C. Berkeley, the University of Pennsylvania Law School, and Vanderbilt Law School. Special thanks to José Marín, Mark Roe, and Eric Talley for detailed suggestions on earlier drafts. For financial assistance, I am grateful to the Harvard Program on Corporate Governance and the John M. Olin Center for Law, Economics, and Business at Harvard Law School. Elaine Choi, Dennis Courtney, Edward Dumoulin, Ledina Gocaj, Matt Hutchins, June Hwang, David Kirk, Audrey Lee, Da Lin, Amanda Tuninetti, and Brandon Une provided excellent research assistance.
This Article questions a persistent and pervasive view about the proper objective of corporate governance: that managers should favor long-term shareholders over short-term shareholders and aim to increase long-term shareholder value rather than the short-term stock price. This view is widely shared by leading academics, executives, corporate lawyers, and judges.1 It is also at the heart of recent reform proposals—in the United States, the United Kingdom, and elsewhere—to give long-term shareholders more power over public companies.2
The persistence of this view derives from a widely held and appealing intuition: because managers serving short-term shareholders may destroy economic value to boost the short-term stock price, managers serving long-term shareholders will necessarily generate more economic value over time (a bigger “pie”). The problem with this intuition is that it is wrong, at least for the typical U.S. firm that transacts in (buys and sells) a large volume of its own shares.3 Yes, managers serving short-term shareholders might destroy economic value to boost the short-term stock price. But in a transacting firm, managers serving long-term shareholders might also destroy economic value to boost the long-term stock price. In fact, long-term shareholders may well benefit more from value destruction than will short-term shareholders. Therefore, favoring long-term shareholders in the typical firm could, paradoxically, reduce the size of the pie created by the firm over time.
A firm’s directors and CEO (collectively, its “managers”) have at least some incentive to serve shareholders’ interests, even if they are not completely faithful agents of the firm’s shareholders. How managers respond to this incentive will depend, in part, on shareholders’ time horizons. If short-term investors exert greater influence on managers than do long-term shareholders, then managers can be expected to focus on increasing the short-term stock price rather than long-term shareholder value. If long-term shareholders are more powerful than short-term shareholders, then managers can be expected to focus less on the short-term stock price and more on increasing long-term shareholder returns.4
Much attention has been focused on the potential problems that can arise when a firm’s investor base consists largely of short-term shareholders.5 In particular, managers seeking to serve short-term shareholders may engage in “short-termism”: taking steps that boost the short-term stock price but reduce the economic value created by the firm over the long term. The cost of short-termism is borne by other parties, including long-term shareholders (if any) and future shareholders who purchase shares in the short term.6
Short-termism has long been considered a major problem for publicly traded U.S. firms. For decades, legal academics,7 business school professors,8 executives,9 and corporate lawyers10 have decried the potentially perverse interests of short-term shareholders. The recent financial crisis, which many blame on the influence of short-term shareholders, has renewed and intensified criticism of these investors.11
While short-term shareholder interests are roundly criticized, the interests of long-term shareholders are generally all but put on a pedestal. Legal academics12 and business school professors13 urge managers to ignore the short-term stock price and focus on maximizing value for long-term shareholders.14 Henry Hansmann and Reinier Kraakman have gone so far as to say a decade ago that “[t]here is no longer any serious competitor to the view that corporate law should principally strive to increase long-term shareholder value.”15 Managers, in turn, appear to have accepted the norm of maximizing long-term shareholder value.16
However, even managers who wish to serve long-term shareholders may believe that short-term shareholder pressure prevents them from doing so. Consequently, policymakers are considering various types of proposals to increase the power of long-term shareholders in public companies relative to the power of short-term shareholders.17 One set of proposals aims to give long-term shareholders more voting rights in the firm.18 Another set of proposals seeks to increase the numberof long-term shareholders by rewarding them with additional dividends or other cash-flow rights.19 A third set of proposals seeks to increase the number of long-term shareholders by revamping the income tax system to make long-term stock ownership relatively more attractive.20
Yet the norm of favoring long-term shareholders over short-term shareholders, and efforts to boost the number and power of long-term shareholders in public companies, are driven by a flawed intuition: that managers serving long-term shareholders will necessarily generate more value over time than managers serving short-term stockholders. I show that in a typical U.S. firm—that is, a firm that transacts heavily in its own shares—managers serving long-term shareholders will not necessarily generate more value over time than managers serving short-term shareholders, and may well generate less. All of the recent efforts to favor long-term shareholders may thus, perversely, reduce the value generated by firms over the long term.21
For most of this Article, I focus on a firm in which the only residual claimants on the value created by the firm are the firm’s current and future shareholders: the investors who own or will own shares between now and “the long term” (by which I mean the relevant end period, however that period is determined). In other words, the firm’s current and future shareholders capture all of the value generated by the firm over time. I will call this a “shareholder-only” firm.
I begin by considering a “non-transacting” shareholder-only firm: one that does not repurchase its own shares or issue additional shares before the long term arrives. I show that, in this type of firm, the conventional view is correct: managers serving long-term shareholders will generate more economic value over time than managers serving short-term shareholders. In particular, long-term shareholders will want managers to maximize the economic pie. Short-term shareholders, on the other hand, may benefit when managers engage in what I call “costly price-boosting manipulation”—actions that boost the short-term stock price at the expense of the pie generated over the long term. Therefore, in such a firm, it is better for the economy, and for investors in the aggregate, if managers seek to maximize long-term shareholder value instead of doing all they can to boost the short-term stock price, regardless of the consequences for the size of the pie.
Most U.S. firms, however, are “transacting.” They buy and sell large volumes of their own shares each year: approximately $1 trillion worth market-wide.22 The magnitude is staggering, not only in absolute terms, but also relative to firms’ market capitalization. Over any given five-year period, U.S. firms buy and sell stock equivalent in value to approximately 30% of their aggregate market capitalization.23 Thus, for example, a company with a market capitalization of $10 billion today can be expected to buy and sell $3 billion of its own shares over the next five years.
I show that, in a transacting firm, managers can boost long-term shareholder payoffs by taking value-destroying steps in the short term. I first consider a “repurchasing firm”—a firm that buys back its own shares before the long term arrives. In a repurchasing firm, long-term shareholder payoffs depend, in part, on the price and quantity of previously repurchased shares. Therefore, long-term shareholders benefit when managers conduct “bargain” repurchases—buybacks at a price below the stock’s actual value.
Bargain repurchases need not destroy economic value. In principle, they may merely redistribute a slice of the pie from short-term shareholders to long-term shareholders without shrinking the pie itself. If bargain repurchases are merely value-shifting and not value-destroying, then managers serving long-term shareholders will seek to enrich them either by creating economic value (as in the non-transacting firm) or by shifting value, but never by destroying economic value. Thus, as in a non-transacting firm, it would be better for managers to serve long-term shareholders rather than short-term shareholders.
However, I show that managers seeking to boost long-term shareholder payoffs in a repurchasing firm may take two kinds of steps that do destroy value and diminish the economic pie. First, managers may engage in “costly contraction”: the undertaking of economically excessive repurchases that divert funds from valuable projects inside the firm to buy back sharply discounted shares. For example, suppose that buying $100 of a firm’s temporarily cheap stock would yield a 40% return to long-term shareholders, a return that represents a transfer of value from other shareholders. And suppose that investing that same $100 in a firm project would yield a 30% return ($30) for all the firm’s shareholders. Finally, suppose that if the $100 were distributed to selling shareholders in a repurchase, those selling shareholders would earn a 10% return ($10) on the $100 through various investments outside the firm. Thus, if the $100 is distributed in a repurchase, the total amount of value created for shareholders in the aggregate will decline by $20, from $30 to $10. But managers serving long-term shareholders will distribute that $100 through a repurchase rather than invest in the higher-value internal project yielding 30%, because the repurchase diverts enough value from other shareholders to provide long-term shareholders with a 40% return. There is evidence suggesting that firms engaging in share repurchases cut back on valuable activities inside the firm.24
Second, managers serving long-term shareholders may engage in costly price-depressing manipulation either to make a bargain repurchase possible or, if the stock is already underpriced, to increase the extent of the bargain. Once a firm decides to repurchase shares, long-term shareholders can benefit if managers engage in value-destroying manipulation to lower the price further before the repurchase occurs. As with costly contraction, such manipulation increases the amount of value transferred to long-term shareholders while shrinking the overall pie. Indeed, firms conducting repurchases engage in such manipulation with the intention of boosting the long-term stock price.25
I then turn to consider the case in which a transacting firm issues additional equity before the long term. Here, long-term shareholders’ payoffs depend on the price that future shareholders pay for the stock and the amount of shares sold to future shareholders. Managers can serve the interests of long-term shareholders by conducting inflated-price issuances, and this benefit increases with the amount of shares sold.
Inflated-price issuances, like bargain repurchases, need not be value-destroying. In theory, they may merely redistribute value from future shareholders to long-term shareholders. But managers conducting inflated-price equity issuances purely for the benefit of long-term shareholders can be expected to engage in two types of actions that destroy value.
First, when the stock price is high, managers seeking to boost the long-term stock price may cause the firm to issue shares to acquire assets even if the economic value of those assets declines when they are absorbed into the firm. America Online’s (AOL) acquisition of Time Warner in 2000, for $162 billion of stock, is a well-known example of long-term shareholders’ benefitting ex post from an issuance that destroyed economic value.26 The acquisition destroyed so much economic value that AOL and Time Warner were forced to part ways nine years later. Nevertheless, from an ex post perspective, AOL’s long-term shareholders undeniably benefitted from the transaction; it enabled them to buy Time Warner’s valuable assets at an extremely cheap price. In 2009, their combined stakes in AOL and Time Warner were worth approximately 400% more than the AOL stake they would have held absent AOL’s acquisition of Time Warner.27
Second, managers conducting inflated-price issuances can further benefit long-term shareholders by engaging in costly price-boosting manipulation (such as earnings manipulation). Such manipulation can transfer even more value to long-term shareholders. As a result, when a firm sells its own shares at an inflated price, the very same pie-reducing strategies that benefit short-term shareholders can serve the interests of long-term shareholders. Indeed, AOL engaged in such value-destroying manipulation when it issued stock to Time Warner shareholders, benefitting AOL’s current long-term shareholders to the detriment of its future shareholders.28 AOL is not alone. There is evidence that firms issuing equity to acquire the assets of other companies systematically engage in value-destroying manipulation to boost the apparent value of the consideration being given to target shareholders.29
Given the volume of repurchases and equity issuances undertaken by the typical U.S. firm, the amount of value that managers can transfer to long-term shareholders by exploiting mispricing in firms’ securities and engaging in these types of value-destroying manipulations is likely to be substantial. Indeed, a recent study by Richard Sloan and Haifeng You confirms that there already has been a large-scale transfer of value to long-term shareholders in publicly traded U.S. firms via equity transactions. The study finds that over the last forty years, an aggregate of over $2.3 trillion has been transferred to long-term investors through bargain repurchases and inflated-price equity issuances.30 So managers have been annually transferring an average of $50 billion in value to long-term shareholders. Across the market, aggregate value transfer exceeds 20% of aggregate net income, suggesting that almost 20% of the wealth created by publicly traded companies for their long-term shareholders is generated via transfers from other shareholders; for smaller firms, the percentage is much higher, around 50%.31 To the extent these transfers involve actions such as stock price manipulation or costly contraction, the cost to other shareholders is likely to exceed $50 billion per year. These amounts are likely to increase if, as is proposed, long-term shareholders are given even more power in widely held firms.
My purpose in this Article is not to argue that managers focused on serving long-term shareholders necessarily generate less economic value than managers focused on serving short-term shareholders. Rather, my objective is to show that neither managers serving long-term shareholder interests nor managers serving short-term shareholder interests will seek to maximize the economic value created by the firm over time. Managers faithfully serving either type of shareholder at the expense of the other can be expected to take steps that shrink the pie. Consequently, the case for favoring long-term shareholders is substantially weaker than it might appear.
The fundamental problem with focusing on either short-term or long-term shareholder interests is that neither type of shareholder interest reflects the value flowing to both the firm’s current and future shareholders. The short-term stock price does not reflect the value flowing to long-term shareholders and future shareholders. Long-term shareholder value does not fully incorporate the value flowing to short-term shareholders and future shareholders. In sum, both short-term shareholder and long-term shareholder payoffs can be enhanced by managers inefficiently transferring value from other shareholders in the firm, to the detriment of the overall economic pie.
So should long-term shareholders be favored? Ultimately, the desirability of favoring long-term shareholders depends not only on the analysis of the financial interests of short-term and long-term shareholders that I have offered, but also on two additional considerations. These two additional considerations are the interests of non-shareholder “stakeholders” and managerial agency costs. Although a complete analysis of these two considerations is beyond the scope of this project, I address them briefly at the end of the Article.
I first turn to stakeholders. Margaret Blair and Lynn Stout have long pointed out—correctly, in my view—that non-shareholder stakeholders also have residual claims on the corporation.32 One might believe that these stakeholders are likely to be better off if managers run the firm for the benefit of long-term shareholders rather than short-term shareholders. However, I explain that as a matter of economic theory, the effect of managers’ time horizons (that is, whether managers serve short-term or long-term shareholders) on stakeholder welfare is actually indeterminate.33 Indeed, if markets are as inefficient as many believe, managers faithfully serving long-term shareholders may sometimes seek to squeeze more value from other stakeholders than managers serving short-term shareholders. For example, the long-term shareholders controlling Wal-Mart did not build their fortunes by overpaying employees.34 Thus, the potential existence of non-shareholder residual claimants may or may not strengthen the case for favoring long-term shareholders.
I next turn to managerial agency costs, which have long been considered one of the most significant problems in the corporate governance of widely held firms.35 The desirability of giving long-term shareholders relatively more power in the corporation (relative to short-term shareholders) will depend on whether long-term shareholders are better or worse at controlling managerial agency costs. But here, as with stakeholders, it is unclear which type of shareholder is preferable. On the one hand, long-term shareholders’ horizons give them a greater interest in controlling managerial agency costs. On the other hand, to the extent that short-term shareholders are willing to accumulate larger positions than long-term shareholders, they may have greater incentives and ability to discipline managers than do long-term shareholders. If the latter effect dominates, favoring long-term shareholders by impeding short-term shareholders may lead to higher managerial agency costs.
All in all, it is unclear whether long-term shareholders’ alignment with stakeholder interests and their ability to control managerial agency costs are better or worse than those of short-term shareholders. Therefore, adding stakeholder interests and the control of managerial agency costs to the mix is unlikely to strengthen or weaken the case for favoring long-term shareholders. That case remains much weaker than it might otherwise appear, and it certainly is far from compelling.
Before proceeding, a word about controlling shareholders is in order. The purpose of this Article is to reexamine the desirability of favoring long-term shareholders in a widely held firm where law, regulation, and private ordering can be used to shift power away from short-term shareholders to long-term shareholders. But my analysis also has implications for the large number of firms in which long-term shareholders already dominate: firms with controlling shareholders.36 While it is well understood that a controlling shareholder may seek to engage in inefficient self-dealing transactions in order to directly transfer value from the corporation to the controlling shareholder,37 my analysis suggests another problem: controlling shareholders, qua long-term shareholders, may also have their firms engage in a variety of other inefficient transactions that transfer value from other shareholders to themselves—namely, costly contraction, costly expansion, and share-price manipulation around repurchases and equity issuances.38 As far as I can tell, this problem has been largely overlooked.39 Therefore, the economic costs associated with controlling shareholders may be even higher than is widely believed.
This Article is organized as follows. Part I describes the conventional wisdom about short-term and long-term shareholders. Part II lays out my positive and normative assumptions for the analysis. Part III shows that in a non-transacting firm (a firm that does not repurchase or issue any shares), the intuition that managers serving long-term shareholders will generate more value than managers serving short-term shareholders is correct. Part IV shows that in a repurchasing firm, long-term shareholder interests do not align with the maximization of economic value. Part V identifies the various ways in which managers in a repurchasing firm may destroy value to benefit long-term shareholders. Part VI shows that in a transacting firm that issues shares, long-term shareholder interests do not align with economic value maximization. Part VII details the various ways that managers in an issuing firm may destroy value to benefit long-term shareholders. Part VIII describes the circumstances in which managers serving long-term shareholders are more likely to destroy value than managers serving short-term shareholders. Part IX addresses the implications of stakeholders and managerial agency costs for the desirability of favoring long-term shareholders. A conclusion follows.
This Part describes the conventional view about short-term shareholders (Part I.A) and long-term shareholders (Part I.B). It then surveys recent proposals designed to increase the proportion and power of long-term shareholders in public companies (Part I.C).
Even the staunchest proponent of shareholder empowerment must be prepared to accept the following proposition: short-term shareholder interests do not completely coincide with the goal of maximizing the economic value created by the firm over time. In particular, managers seeking to serve short-term shareholders may engage in “short-termism”: taking steps that boost the short-term stock but reduce the size of the pie.40 The economic cost of any short-termism is borne (at least in the first instance) by other parties with residual claims on the value created by the firm. These residual claimants include both long-term shareholders and those future shareholders who buy shares at an inflated price in the short term.41
The question, then, is not whether short-termism can exist, but rather: how bad is it? Some argue that short-termism has been and continues to be a large problem. For several decades, Martin Lipton and his colleagues at Wachtell, Lipton, Rosen & Katz have attacked short-term shareholders for having objectives “not in accordance with the long-term interests of other shareholders and other constituencies.”42 During the 1980s and early 1990s, business thought leaders routinely blamed short-term shareholders for the poor performance of U.S. firms relative to those in Germany and Japan.43 The recent financial crisis has renewed and intensified criticism of short-term shareholders from legal academics,44 business school professors,45 and leading business figures.46
While the interests of short-term shareholders are denigrated, the interests of long-term shareholders are exalted. Legal academics of a variety of persuasions have long believed that managers should ignore the short-term stock price and focus on maximizing long-term shareholder value.47 Even Stephen Bainbridge, who has long argued against shareholder empowerment and for a “director primacy” view of corporate governance, has written that directors should be “obliged to make decisions based solely on the basis of long-term shareholder gain.”48 The wide acceptance of the long-term shareholder value norm is illustrated by the earlier-quoted statement by Henry Hansmann and Reinier Kraakman, that “[t]here is no longer any serious competitor to the view that corporate law should principally strive to increase long-term shareholder value.”49
America’s leading business academics share this view. For example, Harvard Business School’s Michael Porter has written that “long-term shareholder value should be identified as the explicit corporate goal. The burden of proof should shift so that managers must explain any decision that is not consistent with long-term shareholder value.”50 Kellogg School of Management’s Alfred Rappaport has argued that “management’s primary responsibility is to maximize long-term shareholder value,” which means “that management’s primary commitment is to continuing shareholders rather than to day traders, momentum investors, and other short-term-oriented market players.”51
Significantly, the Delaware Supreme Court has emphasized the importance of serving long-term shareholders over short-term shareholders.52 Of course, directors owe fiduciary duties to all (current) shareholders.53 But some shareholders, apparently, are more equal than others. For example, in Gantler v. Stephens, the Delaware Supreme Court described “enhancing the corporation’s long term share value” as a “distinctively corporate concern.”54 And a former Delaware Supreme Court Justice, Norman Veasey, wrote a law review article that distinguished between short-term and long-term shareholders, describing only the latter as the firm’s “underlying investors.”55
With legal academics, business school professors, and judges all humming the same tune, it is not surprising that managers have fully internalized the norm of maximizing long-term shareholder value. For example, the Business Roundtable, a collection of CEOs from major U.S. corporations that seeks to influence public policy, referred to “the paramount duty to optimize long-term shareholder value.”56 In 2006, the Business Roundtable’s Institute for Corporate Ethics and the Chartered Financial Analyst Institute’s Centre for Financial Market Integrity produced a report recommending changes in corporate governance to align managers’ actions with long-term shareholder interests.57 Furthermore, the U.S. Chamber of Commerce Center for Capital Markets Competitiveness reports that it advocates policies for the “positive promotion of long-term shareholder value.”58 Firms routinely report in their proxy statements that their primary purpose is to build “long-term shareholder value.”59
Because of the perceived undesirability of short-term shareholder interests and the perceived desirability of long-term shareholder interests, lawmakers, regulators, and firms are considering and (in some cases) implementing a variety of measures to increase the power of long-term shareholders in public companies relative to the power of short-term shareholders.
As discussed in more detail below, these measures fall into three categories: (1) enhanced voting and control rights for long-term shareholders; (2) enhanced cash flow rights for long-term shareholders, to encourage long-term shareholding; and (3) changes in the tax system designed to increase the relative number of long-term shareholders.
A number of prominent business commentators in the United States have suggested that long-term shareholders should receive more voting rights in the firm in order to increase the relative power of each long-term shareholder. For example, Justin Fox and Jay Lorsch have argued for “giving a favored role to long-term shareholders” by increasing voting power with the length of share ownership or simply “restrict[ing] voting in corporate elections of any kind to those who have owned their shares for at least a year.”60 Similar proposals have been floated by academics and regulators in the UK61 and the EU.62
The Aspen Institute’s Business and Society Program, a collection of leading executives and corporate governance specialists, has also recommended consideration of such arrangements.63 The director of this program, Judith Samuelson, has written that corporations should be permitted to overweight the votes of long-term equity-holders so that “only true ‘owners’ of stock, and not transient ‘renters,’ . . . have a substantive voice in a company.”64 Similarly, an American Bar Association corporate governance task force has expressed support for exploring the use of such arrangements.65
Federal regulators have also shown sympathy for the view that long-term shareholders should receive special privileges. When the Securities and Exchange Commission was considering its proxy access rule, the Business Roundtable and others sought to restrict access to long-term shareholders.66 The SEC’s final version of the rule in 2010 (which was later invalidated by the D.C. Circuit) responded to these concerns by limiting proxy access to shareholders that had held their shares continuously for at least three years.67
Delaware judges, both in their decisions and in other fora, have indicated an acceptance of discrimination in voting rights against short-term shareholders and in favor of long-term shareholders. For example, in Williams v. Geier, the Delaware Supreme Court upheld the validity of a charter provision granting superior voting rights to shareholders who held their shares for a continuous three-year period.68 One of the firm’s justifications for this arrangement, which the court did not dispute, was to “[m]aintain [the] ability to maximize long-term value for shareholders.”69
More informal and indirect approaches to increasing the influence of long-term shareholders have also been proposed. For example, the U.K.’s 2012 Kay Commission report recommended that firms consult “major long-term investors” regarding board appointments.70 Delaware Supreme Court Chief Justice Leo Strine has similarly urged firms to consult long-term investors when making board appointments.71
Boosting the number and proportion of long-term shareholders would also increase their power. One set of proposals seeks to convert short-term shareholders into long-term shareholders by rewarding long-term shareholders with additional dividends or other cash-flow rights. For example, economists Patrick Bolton and Frédéric Samama have suggested that long-term shareholders receive “L-shares”—shares entitling them to additional stock in the firm.72 Under the Bolton and Samama proposal, buy-and-hold investors would get a free call option, or warrant, if they held their shares for a pre-specified period of time (three years, for example).73
Several European firms have already modified their corporate arrangements to give long-term shareholders more cash-flow rights vis-à-vis short-term shareholders. For example, the French firms L’Oreal, Electricité de France, Air Liquide, and Credit Agricole have all offered or will offer extra dividends to long-term shareholders.74 Air Liquide is also giving long-term shareholders extra shares.75 Similar arrangements have been proposed in the Netherlands and the U.S.76
A third set of proposals seeks to increase the number of long-term shareholders by revamping the income tax system to make long-term stock ownership relatively more attractive. For example, the Aspen Institute has proposed a graduated long-term capital gains tax rate, with the lowest rate available to shareholders who own their stock for the longest period of time.77 The Aspen Institute has also proposed increasing the deductibility of long-term capital losses.78Similarly, Vanguard’s John Bogle has suggested eliminating the tax deductibility of short-term capital losses and increasing the tax rate on ordinary income generated by stock trading.79
The Aspen Institute is trying to rally support for a tax on transactions in the stock market.80 Such a tax was first proposed by John Maynard Keynes in the 1930s81 and later endorsed by economists Joseph Stiglitz and Larry Summers in the late 1980s.82 One of the main purposes of such a securities tax is to make short-term stock ownership less attractive and thereby increase the proportion of long-term shareholders in public firms.83
The conventional view is simple. Short-term shareholders have undesirable interests: their payoffs do not align with economic value creation over time. Accordingly, these shareholders pressure managers to act in ways that destroy economic value. The solution to this problem is that managers should run the firm for the benefit of long-term shareholders and should seek to maximize long-term shareholder value. Indeed, it might well be desirable to give long-term shareholders more power in the corporation so that managers focus less on maximizing the short-term stock price and more on maximizing long-term shareholder value.
As we will see, long-term shareholder interests are better (that is, better aligned with economic value creation) than short-term shareholder interests in a particular firm: one that does not transact in its own shares. However, it is far from clear that long-term shareholder interests are better than short-term shareholder interests in the typical U.S. firm, which heavily buys and sells its own shares. In such a firm, long-term shareholders may have worse interests. Therefore, shifting power to long-term shareholders might actually reduce the value generated by the firm over time.
This Part provides two critical building blocks for my analysis. Part II.A outlines what I take to be the policy objective of corporate governance regulation: maximizing “economic value”—the (net) economic output generated by the firm, from today through the long term. Part II.B describes my assumptions about short-term and long-term shareholders’ objectives.
I take it as given that the regulation of public companies should be designed to maximize the economic value created by a typical firm over time.84 In particular, I assume that it is desirable to maximize the net economic output of the firm from today until “the long term”—the relevant end point, however that end point is determined.85 The net economic output of the firm is simply the value of assets distributed or retained by the firm less the value of assets contributed to the firm. I call this maximand “economic value,” or, more figuratively, “the pie.”
To focus the analysis, I will for now consider a “shareholder-only” firm. That is, I assume that the only residual claimants on the economic pie are the firm’s current shareholders (who own shares now) and future shareholders (who will buy its shares in the future, but before the long term arrives).86 As a result, economic value is equivalent to the net amount of value flowing to current and future shareholders through the long term—cash they receive from the firm less cash (or other assets) they transfer to the firm.87
In effect, I treat current and future shareholders as if they collectively were a “sole owner” of the corporation. Such a hypothetical sole owner would wish to maximize the net amount flowing to it over time—the amount withdrawn from the corporation (via dividends and repurchases) less the amount invested in the firm (through the purchase of equity from the firm).
To be sure, the premise that it is desirable to maximize the net value flowing to all shareholders of the firm (both current and future), rather than current shareholder value, might be questioned. In the U.S., directors generally owe a fiduciary duty to the firm and its current shareholders; future shareholders are not owed a fiduciary duty until after they have acquired stock in the firm.88 One might therefore believe that a firm should be run to maximize the value flowing solely to current shareholders.
From an economic perspective, however, a dollar flowing to a current shareholder is no more or less valuable than a dollar flowing to a future shareholder (ignoring, of course, the time value of money). Consequently, there is no economic reason for policymakers or analysts, in assessing policy proposals (including proposals designed to shift power to long-term shareholders), to weigh these dollars differently.89 Accordingly, I assign the same weight to every dollar flowing to or from a firm’s shareholders, whether the dollar flows to a current or future shareholder before the long term arrives.90
My focus in this Article is on the objectives of short-term and long-term shareholders—that is, the outcomes they want the firm’s managers to produce. I assume that short-term shareholders will sell their entire equity interest in the short term, and that long-term shareholders will sell their entire equity in the long term, and not engage in any equity transactions in the short term. In other words, I assume “ideal types.”91 I will also assume that shareholders seek the highest possible financial return from their investment in the firm, given their holding period. Therefore, they will want managers to maximize the stock price in the period when they will be selling their shares: short-term shareholders will seek a higher short-term stock price, while long-term shareholders will seek a higher long-term stock price.92
In most of my analysis, I abstract from short-term and long-term shareholders’ abilities to achieve their objectives. That is, I put aside the problem of managerial agency costs: namely, that managers, who directly control the firm, will pursue their own interests rather than those of either short-term shareholders or long-term shareholders. In Part IX, I will briefly consider the possibility that short-term and long-term shareholders may differ in their abilities to reduce managerial agency costs. Until then, however, my focus is solely on the objectives of short-term shareholders and long-term shareholders.
The conventional view is that managers serving long-term shareholders will generate more value over time than managers serving short-term shareholders. This Part shows that the conventional view is correct in a shareholder-only firm that does not transact in its own shares.
Part III.A introduces a simple analytical framework for examining the relationship between shareholders’ returns and economic value in a non-transacting shareholder-only firm. Part III.B describes long-term and short-term shareholders’ returns in this setting. In such a firm, long-term shareholder returns are purely a function of the size of the economic pie. In contrast, short-term shareholder returns are not.
Consider a hypothetical non-transacting shareholder-only firm, ABC Corporation, in a three-period setting: (1) today; (2) the short term; and (3) the long term. The long term is the relevant end period. The short term is a future point in time, occurring before the long term.
ABC Corporation’s situation in the three periods is as follows:
· Today: ABC has two shares outstanding. One share is held by short-term shareholders (denoted “ST”). One share is held by long-term shareholders (denoted “LT”).
· Short term: Short-term shareholders sell their one share to future shareholders (denoted F).93 The sale price is $P per share. The price $P may or may not reflect the share’s actual (full-information) value.
· Long term: ABC’s assets are sold for $V in cash, which reflects their actual value.94 A total of $V is distributed to long-term shareholders and future shareholders. Because each type of shareholder holds one share, long-term shareholders receive $V/2 and future shareholders receive $V/2.
The sequence of events is illustrated in Figure 1 below.
abc as a non-transacting firm
Between today and the long term, the only cash flowing between ABC and its shareholders is the payment of $V made by ABC to long-term shareholders and future shareholders when ABC’s assets are sold in the long term. Accordingly, economic value—the net amount of value flowing from ABC to ABC’s shareholders over time—is $V.95 Economic value (the pie), and the net amounts flowing to short-term shareholders and long-term shareholders, are summarized in Table 1 below.
shareholder payoffs and the pie in a non-transacting firm
In a non-transacting shareholder-only firm, short-term shareholder interests may not align with pie maximization, but long-term shareholder interests always will.
Short-term shareholders will want managers to maximize the short-term stock price ($P) at which they will sell shares to future shareholders. In a rational market, $P would reflect the best possible estimate (based on public information) of $V. And in a rational market with full information, $P would equal $V. Accordingly, in a full-information rational market, managers serving short-term shareholders would strive to maximize the pie.
But in the real world, future shareholders do not have full information about the value of a firm’s stock.96 Future shareholders must rely on information provided by managers. Managers, in turn, can engage in “price-boosting manipulation”—providing information to make the firm appear more valuable than it really is, thereby boosting the price that future shareholders will pay for the stock. Indeed, short-term shareholders will want managers to engage in price-boosting manipulation because it will increase their returns.
If price-boosting manipulation were always economically costless—that is, if it would not reduce the size of the pie—then short-term shareholders’ interests would be consistent with pie maximization. Short-term shareholders would want managers to boost the short-term stock price through manipulation, and managers serving short-term shareholders would do so. But no value would be destroyed in the process. Instead, value would merely be transferred from one type of shareholder (future shareholders) to another (short-term shareholders) without any reduction in the size of the pie as a whole.97
However, short-term shareholders can also benefit from, and therefore will want managers to engage in, costly price-boosting manipulation—manipulation that boosts the short-term stock price but destroys economic value.98 Because short-term shareholders care only about the short-term stock price, any corporate action that boosts the short-term stock price serves their interest, even if it destroys value.
Using the analytical framework introduced in Part III.A, suppose that managers could engage in an action X that boosts the short-term stock price $P by $1 and reduces $V by $0.25. X would destroy $0.25 of economic value. But short-term shareholders would want managers to engage in X because X boosts $P, the price at which short-term shareholders unload their shares.
Managers already engage in a variety of practices that constitute costly price-boosting manipulation. One practice is earnings manipulation: reporting earnings different from the “correct” amount of earnings given the firm’s actual business activity and cash flows.99 Another practice is real earnings management: the postponing of desirable transactions or premature acceleration of transactions to boost short-term accounting results and the short-term stock price at the expense of long-term economic value.100 Each of these strategies shrinks the economic pie but makes the short-term stock price higher than it would otherwise be.101
While managers certainly engage in costly price-boosting manipulation, they may do so to enrich themselves rather than to enrich short-term shareholders.102 Indeed, only a few published studies have found evidence of a link between short-term shareholders and costly price-boosting manipulation.103 Therefore, many legal academics are skeptical that pressure from short-term shareholders causes managers to engage in value-destroying activities.104 But whatever one’s view on this empirical question, there is no escaping the conclusion that short-term shareholder interests are not perfectly aligned with pie maximization.105
Now consider long-term shareholders. Long-term shareholder interests are straightforward. As Table 1 makes clear, long-term shareholders’ payoff ($V/2) rises and falls with economic value ($V).106 In a non-transacting shareholder-only firm, long-term shareholder interests are aligned with maximizing the economic pie.
Accordingly, managers loyally serving long-term shareholders will seek to generate more value than managers loyally serving short-term shareholders. The conventional view and intuition about long-term shareholder interests is correct—at least for a non-transacting, shareholder-only firm.
Most firms, however, aggressively transact in their own shares. Indeed, publicly traded firms in the U.S. buy and sell, in aggregate, approximately $1 trillion of their own shares each year.107 As we will see in the next four Parts, when firms buy and sell their own shares, long-term shareholder returns become decoupled from pie maximization and can be boosted by steps that actually destroy value. In such firms, neither short-term shareholder interests nor long-term shareholder interests align with pie maximization, and it becomes an open question as to which shareholders’ interests are better aligned with pie maximization.
In this Part, I examine the interests of long-term shareholders in a repurchasing firm. Part IV.A describes the widespread use of repurchases by U.S. firms. Part IV.B modifies the analytical framework presented in Part III to explain how stock buybacks change the relationship between long-term shareholders’ interests and economic value. Part IV.C shows that long-term shareholders benefit when managers buy back stock at a cheap price; it also provides evidence that managers currently engage in such “bargain repurchases.”108
Publicly traded U.S. companies increasingly distribute cash through repurchases rather than through dividends.109 Over 90% of U.S. public firms that distribute cash to stockholders do so through share repurchases.110 In 2007, S&P 500 firms distributed almost $600 billion through repurchases,111 and total repurchases market-wide reportedly reached $1 trillion.112 While repurchase volumes declined during the financial crisis, they have since returned to pre-crisis levels: S&P 500 firms repurchased $475 billion of shares in 2013 and were expected to repurchase nearly $600 billion of shares in 2014.113
The volume of these share repurchases is substantial relative to firms’ market capitalizations. Goldman Sachs reports that over the past decade, including the financial-crisis years, S&P 500 firms completed buybacks totaling 3% of their market capitalization per year on average.114 In other words, over a five-year period, firms can be expected to distribute through share repurchases approximately 15% of their market capitalization. While many firms distribute more, and many less, the typical firm is likely to repurchase a substantial amount of its own stock.
The overwhelming majority of share repurchases take the form of “open market repurchases” (OMRs).115 In an OMR, the firm repurchases its shares in the open market through a broker. The transactions are anonymous: shareholders are unaware that the firm is buying shares as the repurchases are occurring. Investors learn about the transactions only after the end of the quarter, typically one to four months after the transactions occur, when the firm reports the prior quarter’s monthly share repurchases.116
To be sure, investors are aware that an OMR might be occurring. A firm cannot conduct an OMR unless its board has previously announced that it has authorized an OMR.117 However, such an authorization is not binding on the firm. Firms do not commit—and are not obligated—to buy back any stock.118 In fact, one study found that almost 30% of firms announcing repurchases do not buy back a single share within the fiscal year in which the repurchase announcement occurs, and about 15% do not buy back any shares within four fiscal years of the announcement year.119 Therefore, investors will not know with certainty that the firm is repurchasing shares until months after the company starts buying back shares.120
OMRs can generate some benefits for shareholders, such as providing a more tax-efficient payout mechanism than dividends.121 But, as we will see below, OMRs also have a dark side. Managers may use OMRs to transfer value from short-term shareholders to long-term shareholders in ways that shrink the pie.
To see how share repurchases decouple long-term shareholder interests from economic value, we will modify the analytical framework introduced in Part III to consider the scenario in which ABC Corporation repurchases its own equity in the short term for $P. For now, I assume that the repurchase does not increase or decrease the size of the pie.
The three periods are as follows:
· Today: ABC has two shares outstanding. One share is held by short-term shareholders (denoted “ST”). One share is held by long-term shareholders (denoted “LT”).
· Short term: Short-term shareholders sell their one share to ABC. The sale price is $P per share. The price $P may or may not reflect the share’s actual (full-information) value.
· Long term: ABC’s assets are sold for $(V-P) in cash, which reflects their actual value. The amount $(V-P) is distributed to long-term shareholders, who in the long term hold 100% of ABC’s equity.
Because short-term shareholders sell their equity for $P per share to ABC rather than to future shareholders, there are no future shareholders in this scenario. The sequence of events is illustrated in Figure 2 below.
abc as a repurchasing firm
Although ABC’s value in the long term is different from what it was in the non-transacting-firm scenario ($(V-P) instead of $V), the pie—the amount of value flowing to shareholders over time—is the same: $V. The amount $(V-P) flows to long-term shareholders in the long term and the amount $P flows to short-term shareholders in the short term.122
The pie and payoffs to shareholders are summarized in Table 2 below.
shareholder payoffs and the pie in a repurchasing firm
Unlike in the non-transacting-firm scenario, in a repurchasing firm there is now a disconnect between long-term shareholder returns and economic value. As $P falls, long-term shareholders’ payoff increases even though the pie remains unchanged.
The analysis in Part IV.B suggests that managers can conduct “bargain repurchases”—OMRs at a cheap price—to transfer value from short-term shareholders to long-term shareholders. And there is considerable evidence that they do just that.
As I have shown elsewhere, a share repurchase has the same distributional consequences as a transaction in which the shareholders whose shares are repurchased directly sell their stock to the remaining shareholders at the repurchase price.123 As a result, a repurchase at a low price (that is, a price lower than the no-transaction intrinsic value of the stock) transfers value from selling shareholders to non-selling shareholders.124
We can see this in terms of our ABC example. In non-transacting ABC (where there is no repurchase), the long-term shareholder will receive $V/2 for its equity. In the event of a repurchase, it will receive $(V-P). The repurchase payout will exceed the no-repurchase payout if and only if $P < $V/2. Thus, when the short-term stock price is lower than the no-transaction intrinsic value of the firm’s shares, long-term shareholders will be better off if the repurchase occurs than if it does not.125
Warren Buffett, in explaining why Berkshire Hathaway seeks to buy its own stock at a low price, puts it much more colorfully:
[T]here is no surer way [of making money for continuing shareholders] than by buying an asset—our own stock—that we know to be worth at least x for less than that—for .9x, .8x or even lower. (As one of our directors says, it’s like shooting fish in a barrel, after the barrel has been drained and the fish have quit flopping.) . . . And the more and the cheaper we buy, the greater the gain for continuing shareholders.126
Warren Buffett is not the only CEO who has figured out that bargain repurchases can generate long-term shareholder value. Evidence that managers engage in bargain repurchases has been steadily accumulating. This evidence includes (a) executives’ own statements and behavior and (b) post-repurchase stock returns, which are discussed in turn below.
When one wishes to understand why executives engage in a particular kind of transaction (such as share repurchases), it is often helpful to ask them. This is exactly what several economists did; they used a confidential survey to ask executives about their firms’ payout policies. According to Alon Brav and his colleagues, “The most popular response for all repurchase questions on the entire survey is that firms repurchase when their stock is a good value, relative to its true value: 86.4% of all firms agree or strongly agree with this supposition.”127 Importantly, these economists reported, “executives tell us that they accelerate (or initiate) share repurchases when their company’s stock price is low.”128
But there was probably no need to use a confidential survey to elicit this information. Executives should be willing to say the same things publicly. After all, firms frequently and openly describe repurchase programs as designed to buy shares at favorable prices for long-term shareholders.For example, in a 2013 press release, UnitedHealth Group announced that it increased its share repurchase program to buy stock at low prices for the benefit of long-term shareholders.129
UnitedHealth Group’s willingness to acknowledge openly that it uses inside information to stealthily acquire shares at a low price from short-term shareholders to transfer value to long-term shareholders would be shocking in a world in which managers believed that they owed the same duty to all shareholders. But managers have been repeatedly told—by leading legal academics, business school professors, and (perhaps most importantly) Delaware judges—that they should maximize long-term shareholder value, even at the expense of short-term shareholders. There is simply no reason for managers to be circumspect.
Managers do, in fact, use inside information to time repurchases. One recent study found that firms systematically buy stock at low prices in each quarter, often transferring large amounts of value to long-term shareholders.130 In one firm, 7.76% of the total market capitalization was shifted from selling shareholders to long-term shareholders in this manner.131 Unsurprisingly, this study also found that managers’ tendency to exploit non-public information in timing repurchases increases with insider equity ownership—that is, the extent to which their interests are aligned with those of long-term shareholders.132 Other recent studies have reported similar results.133
The movement of stock prices following repurchases also suggests that many repurchases are driven by the desire to buy stock at a low price. Researchers have repeatedly found that companies announcing OMRs experience, on average, cumulative abnormal (market-adjusted) returns of approximately 25% over the next four years.134 These abnormal returns suggest that firms announcing OMRs were, on average, 20% undervalued at the time of the OMR announcement.
However, as I noted earlier, many firms announcing OMRs do not actually buy back any stock.135 We would expect firms that announce OMRs and then actually repurchase shares to be more undervalued, on average, than all firms announcing OMRs. Indeed, one study found that “value” firms (those with a high book-to-market ratio) that had announced repurchases and subsequently repurchased more than 4% of their shares in the following year experienced average four-year post-announcement abnormal returns of 57%.136 These subsequent returns provide additional evidence that managers currently use repurchases to shift value from selling shareholders to long-term shareholders.137 Indeed, such studies have led economists to conclude that repurchasing stock at a low price has become a widespread practice.138
Not surprisingly, there is evidence of a link between managerial stock ownership (which aligns managers with long-term shareholder interests) and the propensity to undertake bargain repurchases. One study found that abnormal returns following repurchase announcements, which are associated with pre-repurchase underpricing, are positively correlated with pre-buyback executive stock ownership.139 Another study compared executive stock ownership in two types of repurchasing firms: (1) relatively infrequent repurchasers, which are likely to engage in opportunistic bargain repurchasing when the stock price is low; and (2) relatively steady repurchasers, which are likely to repurchase shares to acquire stock for employee-option programs and to distribute cash rather than to buy back stock at a low price. It found that relatively infrequent repurchasers tend to have higher levels of executive ownership.140 Both of these studies suggest that executives are more likely to engage in bargain-price repurchases when their interests are more aligned with those of long-term shareholders.
Part IV explained that managers can and do use bargain repurchases to shift value from selling shareholders to long-term shareholders. If these bargain repurchases were economically costless, they would merely shift value among different types of shareholders without reducing the size of the pie. Managers could boost long-term shareholder returns without destroying value. Unfortunately, however, the use of bargain repurchases can give rise to economic costs that shrink the pie.
This Part describes two such costs. Part V.A explains that the use of bargain share repurchases to boost long-term shareholder returns can destroy economic value by inefficiently shrinking the firm. Part V.B explains that managers can, and do, benefit long-term shareholders by engaging in costly price-depressing manipulation around share repurchases.
The use of bargain repurchases to benefit long-term shareholders can lead to “costly contraction”: managers seeking to buy back stock at a low price may give up economically valuable projects to fund the repurchase, reducing the total amount of value available to all the firm’s shareholders over time.
From an economic perspective, a firm should distribute cash to shareholders via a repurchase (or dividend) if, and only if, total economic value will be increased.141 For example, suppose that the firm is considering distributing $100 in cash. If an outside project would yield a 15% return and an inside project would yield 10%, the cash should be distributed. But if the best outside project available to shareholders would yield a 10% return and an inside project would yield a 15% return, then the cash should not be distributed.
Importantly, from this perspective, the firm’s stock price should not be a relevant consideration in determining whether the firm should distribute cash via a repurchase. The stock price affects only the distribution of value among different types of shareholders. The only relevant consideration is whether the total economic pie will be bigger or smaller as a result of the repurchase.
However, as we saw in Part IV, managers make payout decisions based on the stock price: when the stock price is low, they initiate or accelerate repurchases. When managers use an extraneous factor such as the stock price to determine the timing of payout, payout policy can become distorted from an economic perspective.
Consider our simple analytical framework involving ABC Corporation. Recall that, absent a repurchase (or equity issuance), the payoff to long-term shareholders is $V/2. If there is a repurchase, and no value is created or destroyed, then the payoff to long-term shareholders is $(V-P). Therefore, long-term shareholders benefit (assuming no value is created or destroyed) from a repurchase whenever $P < $V/2. If, however, $X of economic value must be destroyed to effect the repurchase, long-term shareholders will still be better off with a repurchase as long as $(V-P-X) > $V/2, that is, whenever $P < $(V/2-X).
It might be helpful to offer a numerical example. Suppose that ABC has $110 in fixed assets and $80 in cash that, if invested in a firm project, will yield a return of 12.5% (or $10). Assume, for simplicity, that any cash distributed by ABC will generate a 0% return outside the firm. If ABC does not repurchase a share in the short term, $V will equal $200 ($110+$80+$10), and the long-term shareholder payout ($V/2) will equal $100. Suppose that, in the short term, ABC’s shares trade at $80. If ABC repurchases a single share for $80, $V will equal $190 ($110+$80), and the long-term shareholder payout $(V-P) will equal $110. The long-term shareholder payout is higher if managers distribute cash that would generate $10 more economic value inside the firm.142
Who loses if ABC’s managers distribute cash that could generate a larger return inside the firm? Not the short-term shareholder, who would have sold its share for $80 in any event (but to a future shareholder, rather than to ABC). Indeed, if we relax the example’s assumptions a little bit and consider the possibility that the repurchase might boost the short-term stock price (by, among other things, increasing competition for the short-term shareholder’s share), then the repurchase might actually benefit the short-term shareholder. The loser is the future shareholder, which would have purchased a share for $80 (assuming, as I do in the example, that the repurchase does not boost the short-term stock price) that is actually worth $100.
From a current-shareholder perspective, the outcome of costly contraction might not be objectionable. The long-term shareholder and the short-term shareholder come out ahead, jointly and perhaps even individually. But from an economic perspective, which takes into account the size of the total pie produced by the firm over time for all of its shareholders, current and future, destroying $10 of economic value to boost the payouts for current shareholders is undesirable.
One might argue that a firm repurchasing its stock should be able to pursue both the valuable project and buy back stock that trades at a low price. In a world of perfectly efficient capital markets, this objection would have considerable force: the firm should be able to borrow cheaply enough to buy its own stock and invest in valuable new projects.143
However, as I have explained elsewhere,144 a firm may not be able to borrow enough money to fund the desirable project while also buying back stock at a low price. First, as economists have long understood, the information asymmetry between lenders and the firm may make it difficult for the firm to borrow money cost-effectively.145 From their perches inside the firm, managers may understand that the firm’s prospects are good. But lenders outside the firm, who have less information than the managers, may not be as confident. They may insist on loan terms that make the costs of financing the desirable project higher than the benefits.
Second, the firm’s contractual arrangements with other lenders might impede additional debt financing. Even if Lender A were willing to lend to the firm on cost-effective terms, loan covenants with Lenders B, C, and D might bar the firm from borrowing additional funds (from Lender A or any other creditor).146 While renegotiation is possible in theory, it might be difficult in practice, particularly if the borrower must simultaneously renegotiate with multiple creditors to obtain the modifications needed to facilitate the new investment.
In short, firms may need to choose between engaging in a bargain price repurchase and funding desirable projects. Indeed, empirical evidence suggests that repurchases often divert cash that would otherwise be used for research and development and other productive investments in the firm.147 For example, one study found that repurchases appear to have a significantly negative effect on a firm’s short-term investments and R&D, with a doubling of repurchases leading to an 8% reduction in R&D expenditures.148
We saw in Part III that managers serving short-term shareholders may engage in costly price-boosting manipulation to lift the short-term stock price. We will now see that managers serving long-term shareholders may engage in costly price-depressing manipulation to reduce the short-term stock price around bargain repurchases; indeed, there is evidence that such costly price manipulation already occurs around repurchases.
As Part IV explained, long-term shareholder returns in a repurchasing firm depend on the price at which the firm buys its own shares. Long-term shareholders benefit when the repurchase price is low (relative to the no-transaction value of the stock); the lower the price, the better off long-term shareholders will be. Managers repurchasing cheap shares in the short term can therefore benefit long-term shareholders by further depressing the short-term stock price.149
Importantly, managers can help long-term shareholders by manipulating the stock price around repurchases even when some economic value is sacrificed. In particular, assuming that long-term shareholders’ losses from value destruction are lower than the benefit of the reduced repurchase price, long-term shareholders will prefer that managers engage in costly price-depressing manipulation.150
In fact, there is evidence that managers manipulate prices before and during repurchases, deliberately driving earnings and stock prices down to increase the amount of value transferred to long-term shareholders. One study examined 1720 OMR announcements between 1984 and 2002 that were followed by actual repurchases during the quarter of the announcement or the following quarter.151 The study found significant negative earnings manipulation among firms that announced and conducted OMRs, but not among firms that announced OMRs but did not conduct them.152
Not surprisingly, downward earnings manipulation was more aggressive in firms in which the equity ownership of the CEO was higher—that is, when the CEO’s interests were more aligned with the interests of long-term shareholders.153 This finding strongly suggests that, the more the board and management focus on maximizing long-term shareholder value, the more likely managers will be to engage in costly price-depressing manipulation.
To be sure, long-term shareholders will not always benefit from costly price manipulation. If costly price manipulation destroys too much of the pie, then long-term shareholders will be made worse off. But the important point is that, just as short-term shareholders can benefit from costly price-boosting manipulation that lifts the short-term stock price, long-term shareholders can benefit from costly price-depressing manipulation of a kind that reduces the short-term stock price when the firm is repurchasing shares.
It is also worth noting that unlike costly contraction, which can benefit short-term shareholders and hurts (would-be) future shareholders, costly price-depressing manipulation hurts short-term shareholders and benefits future shareholders by causing the former to sell, and enabling the latter to buy, at a lower price. Because this manipulation is value-destroying, the benefit to long-term shareholders and future shareholders will, by definition, be less than the cost to short-term shareholders.
We have seen that managers seeking to serve long-term shareholders may cut back on economically desirable investments to buy stock at a low price and, when repurchasing shares, waste resources trying to drive the stock price down. We should not be surprised that managers in fact engage in these value-wasting activities. They have been told by legal academics, business school professors, and the courts not only that it is permissible to shift value from short-term shareholders to long-term shareholders, but also that it may well be the managers’ duty to do so. And, of course, managers have compensation arrangements tied to long-term shareholder value that incentivize them to act in this manner.
We now turn to the “mirror image” of repurchases: equity issuances. Like share repurchases, equity issuances decouple long-term shareholders’ interests from pie maximization. As we will see in Part VII, long-term shareholders in an issuing firm, like long-term shareholders in a repurchasing firm, can benefit when managers take steps that reduce the economic pie.
Part VI.A describes the widespread use of equity issuances. Part VI.B modifies the analytical framework presented in Part III to explain how equity issuances change the relationship between long-term shareholders’ interests and economic value. Part VI.C shows that long-term shareholders benefit when managers sell stock at an inflated price, and it provides evidence that managers currently engage in such “inflated-price” issuances.154
The typical publicly traded firm not only repurchases shares, but also issues a considerable amount of shares over its life as a public firm. Indeed, issuances typically exceed repurchases. For example, during each of the years in the period from 1993 to 2002, an average of 66.5% of large firms made net stock issues (issuances less repurchases).155 Strikingly, these net stock issuances averaged 7.5% of assets, which is on the same order of magnitude as net debt issuances.156 The fact that net issuances are so common and so large suggests that the dollar volume of share issuances is the same order of magnitude as the dollar volume of repurchases. Interestingly, the aggregate share count for S&P 500 firms has remained relatively stable over the last decade,157 suggesting that the number of shares issued is in the same ballpark as the number of shares repurchased. The dollar volume of repurchases averages approximately 15% of a firm’s market capitalization over five years.158 It appears that equity issuances are of a similar magnitude, and thus that the total volume of equity issuances and repurchases are approximately 30% of a firm’s market capitalization over five years.
While almost all repurchases take the form of OMRs, equity issuances come in a variety of flavors. I will focus on two of the most important: (1) acquisition-related issuances and (2) seasoned equity offerings.159
Acquirers often issue equity to provide currency for purchasing the shares of a target company, in part because the use of equity rather than cash can provide a tax benefit to the target shareholders.160 An example of such an acquisition (and one to which we will return in Part VII.A) is AOL’s acquisition of Time Warner in 2000 for $162 billion in equity.
Seasoned equity offerings (SEOs) raise cash to fund operations and strategic investments, or to pay down debt.161 Many firms engage in SEOs, and the amount of stock sold is substantial.162 SEOs come in two forms: “firm commitment” and “at-the-market” (ATM).
Until relatively recently, SEOs were almost always “firm-commitment”: the firm arranges to sell a specified number of shares at a fixed price through an underwriter that guarantees to sell the shares at that price and then offers the shares to investors.163 When the market learns of a firm-commitment SEO, the stock price tends to fall.164 The market’s reaction to SEO announcements is not surprising, as an issuance may signal that the stock is overvalued.
In part due to the adverse effect of firm-commitment SEOs on the stock price, firms have taken advantage of recent regulatory changes to issue stock via so-called “at-the-market” SEOs.165 In an ATM, the firm sells shares directly (and quietly) on the market through a sales agent.166 A firm need not—and typically does not—announce these sales as they are occurring, much as firms do not announce OMR transactions as they take place.
Indeed, ATMs are marketed as a way for firms to issue shares quickly when the price appears favorable without alerting the market to the issuance in real time167 (which might cause the stock price to fall). As a recent article by several securities lawyers put it, an ATM enables “the issuer [to] opportunistically take advantage of stock price movements.”168
To be sure, investors know that an ATM might occur. Before conducting an ATM, the firm must have an effective shelf registration statement (including a prospectus) on file with the SEC.169 In these disclosures, the firm must indicate either the maximum number of shares to be sold or the maximum proceeds from the sales, and the firm must identify the sales agent.170
However, these pre-transaction disclosures do not provide much useful information to future shareholders, for two reasons. First, these disclosures can be updated at any time to increase the ceiling on the number of shares to be sold. As a result, investors do not know the maximum number or value of shares that will actually be sold. Second, the filing of these disclosures does not compel the firm to enter into a single transaction. Like an OMR announcement, an ATM filing gives a firm the option, but not the obligation, to trade in its shares on the open market. Investors will not learn of a sale until months after it takes place.171 Consequently, managers have considerable ability to transfer value from future shareholders to long-term shareholders through stealth sales of shares on the open market.
To see how equity issuances decouple long-term shareholders’ interests from the pie, we modify the analytical framework introduced in Part II to consider the scenario in which ABC Corporation issues a third share in the short term for $P. I assume that the issuance does not create or destroy economic value.
The periods are as follows:
· Today: ABC has two shares outstanding. One share is held by short-term shareholders (denoted “ST”). One share is held by long-term shareholders (denoted “LT”).
· Short term: Short-term shareholders sell their one share to future shareholders (denoted F). ABC also sells an additional share to future shareholders. As a result, future shareholders acquire two shares. In both transactions, the sale price is $P per share. The price $P may or may not reflect the share’s actual (full-information) value.
· Long term: ABC’s assets are sold for $(V+P) in cash, which reflects their actual value. The amount $(V+P) is distributed to long-term shareholders and future shareholders. There are a total of three shares outstanding, so the holder of a share receives $(V+P)/3. Long-term shareholders receive $(V+P)/3. Each future shareholder also receives $(V+P)/3, for a total to future shareholders of $2(V+P)/3.
The sequence of events is illustrated in Figure 3 below.
abc as an issuing firm
Although ABC’s value in the long term is different from what it was in the non-transacting case ($(V+P) instead of $V), economic value—the amount of value flowing to shareholders over time—is the same: $V. The amount $(V+P) flows to long-term shareholders and future shareholders in the long term, and $P flows from future shareholders to the firm in the short term.
The economic pie and payoffs to the different types of shareholders are summarized in Table 3 below.
shareholder payoffs and the pie in an issuing firm
Unlike in the non-transacting case, but as in the repurchasing case, there is a disconnect between long-term shareholder returns and the pie. As $P increases, long-term shareholders’ payoff rises, but the pie remains the same. Put another way, the equity issuance converts a long-term shareholder into a partial short-term shareholder (one whose payoff is also linked to $P, the short-term stock price).
The analysis offered in Part VI.B suggests that managers can transfer value from future shareholders to long-term shareholders by conducting “inflated-price issuances”—equity offerings at a price higher than the stock’s actual value. Indeed, they frequently do so.
An equity issuance has analogous distributional effects to a share repurchase. A share repurchase transfers value from short-term shareholders to long-term shareholders when the stock price is lower than its actual value. A stock issuance transfers value from future shareholders to long-term shareholders when the stock price is higher than its actual value. Hence, managers can benefit long-term shareholders by selling stock at an inflated price.172
We can see this in terms of our ABC example. When ABC does not issue equity (or repurchase stock), long-term shareholders will receive $V/2 for their equity. But if ABC issues one share for $P, long-term shareholders will receive $(V+P)/3 for their equity. If $P > $V/2, long-term shareholders are better off.
Who loses from the sale of overpriced equity? In our example, the losers are future shareholders, who buy overpriced shares. But stepping outside this framework for a moment, we should note that a firm’s sale of stock might not increase the number of shares purchased by future shareholders; rather, it might decrease the shares sold by short-term shareholders, who are partially displaced by the firm’s selling activity. If future shareholders buy the same number of shares, and short-term shareholders sell fewer shares, then short-term shareholders will lose from the sale of overpriced equity.
Of course, in the real world, a firm’s sale of stock can be expected to have at least some negative effect on the short-term stock price. This price effect, on which I do not focus, would also need to be taken into account to fully determine which shareholders win and which shareholders lose, and by how much. But the precise identity of the losing shareholders and the extent of their loss are both irrelevant to the point that I wish to emphasize here—namely, that long-term shareholders are better off when the firm sells overpriced stock.
We saw in Part III that many managers acknowledge that their firms repurchase shares when the price is low.173 Similarly, and not surprisingly, many managers also acknowledge that they issue shares when they believe the stock price is “high.”174 Moreover, just as empirical studies repeatedly find that managers tend to conduct repurchases at low prices, there is considerable evidence that managers conduct equity issuances—either to acquire other firms or to raise cash—when the stock is overpriced.
Turning first to acquisition-related issuances, firms tend to use overpriced stock as consideration in acquisitions.175 Such acquisitions benefit the long-term shareholders of the acquiring firms by enabling the acquiring firms to purchase assets with overvalued currency, reducing the effective purchase price paid.176
Equity issuances for cash show a similar pattern. There is evidence, going back decades and from around the world, that traditional SEOs are, on average, overpriced.177 A recent paper examining 2600 SEOs between 1992 and 2010 suggests the magnitude of this benefit. It finds that firms timing traditional SEOs boost average returns to long-term shareholders by approximately 3% over the subsequent three years.178Importantly, and not surprisingly, the propensity to engage in inflated-price issuances appears to increase with insider equity ownership—that is, as managers’ interests become more aligned with those of long-term shareholders.179
Although ATMs are relatively new, the evidence now emerging is consistent with insiders’ using ATMs to sell overvalued equity. For example, one recent study finds that ATMs are announced after significant stock price run-ups and that the market reacts negatively to their announcement.180 This study also finds that actual sales under ATM programs are executed after the stock price has recently risen and market conditions are relatively favorable.181 These findings should not be surprising, since ATMs are marketed to firms as a method of enabling managers to issue shares quickly when the price appears favorable without alerting the market to the issuance and causing the stock price to fall.182We can expect ATMs, like traditional SEOs, to be used to transfer value from future shareholders to long-term shareholders.183
It is worth pausing here to reflect on the fact that corporate lawyers in the United States feel comfortable advising firms to adopt arrangements that allow managers to transfer value from future shareholders to current shareholders. In a regime such as the United Kingdom’s, in which managers are considered to owe a fiduciary duty to both current and future shareholders,184 lawyers might think twice before advising firms to use inside information to secretly sell shares to future shareholders at a price that benefits current shareholders. But in the United States, where it is considered self-evident that managers owe a duty only to current shareholders185 and therefore can and should use any legal means to transfer value from future shareholders to current shareholders, advising firms to do so might seem not only natural but also desirable.
Part VI explained that managers can and do use inflated-price issuances to shift value from future shareholders to long-term shareholders. If these issuances were economically costless, then they would merely shift value among different types of shareholders without reducing the size of the pie. Unfortunately, however, managers serving long-term shareholders may destroy economic value through inflated-price issuances to enrich long-term shareholders.
This Part examines two types of potential economic costs associated with inflated-price issuances. Part VII.A explains that managers using inflated-price equity issuances to benefit long-term shareholders may engage in “costly expansion”—inefficiently moving assets from outside to inside the firm. Part VII.B explains that managers engaged in inflated-price issuances can further benefit long-term shareholders by engaging in costly price-boosting manipulation around the issuances.
Managers serving long-term shareholders may increase the size of the firm through the sale of overpriced equity, even though the expansion may destroy economic value.
We saw in Part V that a repurchase can reduce economic value by distributing cash that, from the perspective of all the firm’s current and future shareholders, could generate higher returns if invested in the firm’s own projects. Analogously, an equity issuance can reduce economic value if the equity issuance causes the firm to absorb assets that would generate more value outside the firm. Furthermore, just as long-term shareholders can benefit from managers’ sacrificing valuable in-firm projects to buy back stock at a low price, they can benefit from managers’ acquiring assets at a discount through the use of overpriced stock that would generate more economic value outside the firm.186
Do real-life long-term shareholders actually benefit from the sale of overpriced equity to finance a value-destroying acquisition? From an ex post perspective, one can certainly find many examples in which long-term shareholders of acquirers benefit from value-wasting acquisitions financed with inflated stock. If there were a poster child for such a transaction, it would likely be AOL’s acquisition of Time Warner in 2000.187
AOL used $162 billion of stock to acquire Time Warner.188 The companies had roughly equivalent market capitalizations before the merger. A hypothetical AOL shareholder owning 2% of AOL before the merger thus would have ended up with approximately 1% of the combined firm.
This transaction is now widely seen as a disaster.189 There can be little doubt, from an ex post perspective, that the acquisition destroyed economic value. The expected synergy benefits failed to materialize. In fact, AOL and Time Warner parted ways nine years later,190 suggesting that synergy effects were actually negative. The economic costs of this failed marriage included the transaction costs associated with combining and then splitting the businesses, as well as the negative synergy costs incurred while keeping the two firms stapled together.
Nevertheless, AOL’s long-term shareholders appear to have benefitted from the transaction. When AOL and Time Warner separated in 2009, AOL was worth $2.5 billion, while Time Warner was valued at about $36 billion,191 for a combined value of about $39 billion. Assuming AOL would have been worth the same ($2.5 billion) in 2009 had it not acquired Time Warner in 2000, our hypothetical 2% AOL shareholder would (absent the merger) have owned shares worth $50 million. Instead, as a result of the merger, that shareholder would have owned 1% of AOL (an interest worth $25 million) and 1% of Time Warner (an interest worth $360 million), for a total value of approximately $385 million—more than seven times the value of her hypothetical no-transaction stake in AOL.
To be clear, I am not claiming that the AOL-Time Warner deal was driven by AOL managers seeking to serve long-term shareholders through costly expansion. AOL’s managers may or may not have believed that AOL was overpriced, and they may or may not have believed that the merger would destroy economic value. I describe AOL’s acquisition of Time Warner simply to offer a concrete example of how long-term shareholders can benefit ex post from a transaction that destroyed economic value.
We have seen that long-term shareholders can be made better off if their firm issues overpriced equity for value-wasting acquisitions (rather than refrains from doing so). But long-term shareholders would be even better off if the firm could use inflated-price issuances to acquire assets that do not decrease in value when brought into the firm. In particular, long-term shareholders would prefer managers of firms with overpriced stock to either (a) acquire assets, via an acquisition-related issuance, that do not decrease in value when brought into the firm; or (b) engage in an SEO and invest the cash in a way that enhanced (or at least did not waste) economic value.
However, in many situations, these two alternatives might either be unavailable or deliver less value to long-term shareholders than costly expansion. Consider first the possibility of conducting an acquisition-related issuance aimed at bringing “good” assets (assets that do not lose value when acquired) into the firm. To begin, the firm may be at its optimal scale and scope, so that expanding the firm can only shrink the economic pie. In addition, even if the firm could be expanded in ways that did not reduce long-term economic value, it may not always be possible to find and acquire good assets during the window when the acquirer’s stock is overpriced. If good assets are unavailable during this window, long-term shareholders might be better off if managers engage in a value-destroying acquisition rather than no acquisition at all.192
Next, consider the possibility of conducting an SEO (and not mis-investing the cash). In theory, long-term shareholders would be better off if managers did not engage in a value-wasting acquisition, but instead sold overpriced equity for cash and then either kept or distributed the cash to shareholders, avoiding any shrinkage of the pie. But a firm conducting an SEO, whether traditional or ATM, must inform its old and new investors of the purpose of the financing.193 If the firm announces that it will take all of the funds raised and hold them in cash or distribute them to shareholders, then investors are likely to infer that the firm is issuing stock merely to exploit the fact that it is overpriced.194 Investors might then lower their valuations of the firm, making it more difficult for the firm to sell overpriced equity through the SEO. Thus, managers can be expected to indicate that the money raised will be used for investment.
Not surprisingly, there is evidence that managers conducting SEOs to sell overpriced stock normally accumulate excessive capital rather than distribute the cash to shareholders. One study, by Robert Chirinko and Huntley Schaller, examined publicly traded U.S. firms during the period from 1980 to 2004 and concluded that high-priced firms with poor investment opportunities accumulated between 15% and 45% of excessive capital while they were overpriced.195 Another study of firms undertaking SEOs between 1980 and 1999 found that these firms dramatically increased investment rather than retiring debt or increasing working capital, and that this spike in investment tended to reduce returns on assets by an economically and statistically significant amount.196
In sum, the “first best” outcome for long-term shareholders of a firm with overvalued stock that cannot efficiently absorb new assets would be for managers to issue overpriced shares and distribute the cash to the firm’s shareholders. But managers must disclose the purpose of the equity issuance, and disclosing that managers will distribute the issuance proceeds to shareholders will clearly signal that the firm is issuing shares solely because they are overpriced, causing the stock price to fall. Therefore, the “second best” outcome for long-term shareholders may well be for managers to engage in costly expansion, rather than refrain from issuing the equity.
Just as managers serving long-term shareholders may engage in costly price-depressing manipulation around bargain repurchases to further depress the stock price, these managers may engage in costly price-boosting manipulation around inflated-price issuances to further boost the stock price. Indeed, there is evidence that managers routinely manipulate the stock price around equity issuances.197
We saw in Part VI that long-term shareholder payoffs depend on the price at which the firm issues additional shares. Long-term shareholders benefit when the issuance price is high (relative to the no-transaction value of the stock). The higher the stock price, the greater the benefit. Thus, managers issuing shares at an inflated price can boost long-term shareholder returns by raising the issuance price.198
Importantly, managers can serve the interests of long-term shareholders by manipulating the stock price around issuances even if some economic value must be sacrificed to do so. In particular, as long as long-term shareholders’ losses from the value destruction are lower than their benefit from an issuance at a higher price, long-term shareholders will prefer that managers engage in costly price-boosting manipulation.199
To be sure, long-term shareholders will not always benefit from costly price manipulation. If costly price manipulation destroys too much economic value, then long-term shareholders will be made worse off. But the critical point is that even long-term shareholders, just like short-term shareholders, may benefit from costly price-boosting manipulation.200
Managers already engage in costly price-boosting manipulation around equity offerings—both acquisition-related issuances and SEOs. Turning first to acquisition-related issuances, managers engage in earnings manipulation when issuing stock to acquire another company.201 One study of mergers announced between January 1992 and December 2000 found that, in acquisitions in which acquirer-firm stock is used as consideration, acquiring firms show significant positive accruals in the quarter before the announcement.202
AOL again offers a useful illustration. During the period in which AOL acquired Time Warner, AOL’s managers engaged in aggressive, costly price-boosting manipulation: they massively inflated advertising revenues.203 The combined entity was later sued by the SEC, the Justice Department, and shareholders, and forced to pay $2.5 billion to investors and the government.204 (The settlement somewhat reduced the gains accruing to AOL’s long-term shareholders from the costly price-boosting manipulation that occurred before and during the acquisition.)205
Next, consider SEOs. Firms conducting traditional SEOs may attempt to boost their stock prices by engaging in real earnings management,206 earnings manipulation,207 or a combination of the two. One study, examining 1511 completed traditional SEOs during the period from 1987 to 2006, found that firms conducting these SEOs engaged in both accruals management and real earnings management.208
Given that long-term shareholders are the investors most likely to benefit from costly price-boosting manipulation around equity issuances, it is not surprising that firms with large shareholders (which tend to be long-term shareholders) are more likely to engage in earnings manipulation around equity offerings than firms without such blockholders. For example, one recent examination of 1372 firm-commitment SEOs found that firms’ accruals increase around the SEOs when there are large outside shareholders that own more than 5% of the firm’s stock, but found that there is no increase in the absence of such shareholders.209 This study suggests that costly price-boosting manipulation around equity issuances may, in fact, be designed to serve the interests of long-term shareholders.
Parts IV through VII have demonstrated that, when a firm buys or sells its own shares, managers seeking to boost long-term shareholder returns may well take steps that destroy economic value. Neither short-term shareholder interests nor long-term shareholder interests align fully with pie maximization. Rather, each type of shareholder will want managers to maximize the payout to her own group, even if those steps may reduce the size of the pie—the value flowing to all shareholders—short-term, long-term, and future—over time.
One cannot be confident, a priori, that long-term shareholder interests are necessarily better- or worse-aligned with pie maximization than are short-term shareholder interests. Short-term shareholders benefit from costly price-boosting manipulation. Long-term shareholders benefit from costly contraction, costly expansion, and costly price manipulation (price-boosting around equity issuances, price-depressing around repurchases). Each type of shareholder, therefore, is characterized by its own set of “vices.”210
While it is difficult to make sweeping statements about whose interests are better aligned with all the firm’s shareholders over time, it is possible to identify factors that will tend to make long-term shareholders worse representatives than short-term shareholders, at least in the shareholder-only firm (in which shareholders are the only residual claimants) that I have been considering so far.
As we saw in Part III, managers serving long-term shareholder interests in a non-transacting firm will seek to maximize the pie. However, as we saw in Parts IV through VII, share repurchases and equity transactions decouple the interests of long-term shareholders from pie maximization and may well encourage managers serving long-term shareholders to destroy value (that is, to pursue long-term shareholder interests at the expense of the pie).
In the U.S., publicly traded firms buy and sell approximately $1 trillion of their own shares each year.211 A typical U.S. firm buys and sells 30% of its market capitalization over a five-year period.212 Obviously, some firms buy and sell more than 30%; others, less. The higher the volume of repurchases and equity issuances, the more likely it is that long-term shareholder interests will be less aligned with pie-maximization than are short-term shareholder interests.
By contrast, the lower the volume of repurchases and equity issuances, the more likely it is that long-term shareholder interests will be better (that is, more closely aligned with pie maximization) than short-term shareholder interests. Consider, for example, a firm that rarely transacts in its own shares—that is, an (essentially) non-transacting firm. In such a firm, long-term shareholder interests will almost inevitably be better aligned with pie maximization than short-term shareholder interests. This raises two policy-related questions explored below: (1) Can rarely transacting firms be identified in advance? (2) Should firms be prohibited from transacting in their own shares?
Is it possible to identify firms today that, in the future, will only rarely transact in their own shares? One might think that long-term shareholders in firms that have rarely transacted in their own shares in the past are more likely to be better representatives of shareholder interests going forward. Unfortunately, however, it is difficult to know whether a firm that has rarely transacted in the past will continue to rarely transact in the future. To begin, the degree to which a firm has transacted in its own shares in the past can be expected to depend, in part, on the historic divergence between share price and share value: the smaller the historic divergence, the less likely the firm will have transacted in its own shares. But the fact that the stock price did not historically diverge much from share value does not necessarily mean there will be little divergence in the future.
Second, the degree to which a firm has until now transacted in its own shares may depend on long-term shareholders’ historic influence on management. If long-term shareholder influence has been weak, then the firm may not have fully exploited divergences between the share price and share value to buy stock at a cheap price and sell stock at an expensive price. But if long-term shareholders gain more power in that firm, we can expect that firm to transact more in the future than it has in the past, everything else remaining equal. So shifting power to long-term shareholders in rarely transacting firms may well turn these firms into frequently transacting firms.
In short, we can confidently conclude that, in a rarely transacting firm, managers serving long-term shareholder interests are likely to generate more value than managers serving short-term shareholder interests. But since the magnitude of market mispricing and the degree of long-term shareholder influence may change in a firm over time, we have little idea as to which firms will rarely transact in the future.
When presenting this work, I have often been asked whether firms should simply be prohibited from transacting in their own shares. Such a prohibition would eliminate all of the distortions I have identified in this Article. It would also tie long-term shareholders’ payoffs to the value flowing to all of the firm’s shareholders over time. This, in turn, would substantially strengthen the case for favoring long-term shareholders.213
In my view, from the perspective of society as a whole, there is likely to be substantial benefit to sharply limiting, if not eliminating, public firms’ ability to repurchase their own shares. As I have explained elsewhere, the marginal social benefit of enabling a firm to engage in share repurchases, given its ability to distribute cash through a regular dividend or a special dividend, is fairly limited.214 On the other hand, the social costs of repurchases, which go beyond those described in this Article, are likely to be considerable.215
Equity issuances, however, are indispensable. While a firm with excess cash that cannot repurchase its own shares can always distribute the excess cash through a dividend, there is no alternative to an equity issuance for a firm that needs additional equity capital. So even if repurchases were prohibited, which is unlikely to happen any time soon, firms would (and should) still be permitted to issue shares. In short, we are likely always to have transacting firms and, therefore, a disconnect between long-term shareholder returns and the size of the pie.
The desirability of having managers pursue long-term shareholder interests will depend, in part, on managers’ ability to buy shares at a cheap price and sell shares at an inflated price. This ability, in turn, will depend on the degree of mispricing in markets, the efficiency of markets in absorbing information about repurchases and equity issuances, and the degree to which managers can hide from investors information about the firm’s ongoing share repurchases and equity issuances. The greater the degree of mispricing and inefficiency, and the looser the disclosure regulations for repurchases and equity issuances, the more likely it is that managers serving long-term shareholders will destroy more value than managers serving short-term shareholders.
For example, and as discussed earlier, U.S. firms can purchase shares secretly through open market repurchases (OMRs) and sell shares secretly through at-the-market offerings (ATMs); this enables firms more easily to exploit differences between the share price and share value. Elsewhere, I have argued that firms trading in their own shares should be subject to the same disclosure requirements that apply to the firms’ insiders trading in those same shares.216 Imposing such disclosure requirements would make it more difficult for managers to use inside information in share repurchases and equity issuances, and it would therefore better align long-term shareholder interests with the creation of economic value.
In short, managers serving long-term shareholders are more likely to destroy value if, for any given volume of repurchases and equity issuances, managers can easily exploit informational asymmetry when having the firm buy and sell its own shares.
Both short-term shareholders and long-term shareholders can benefit from costly price manipulation. But the costly price manipulation that benefits long-term shareholders is likely to destroy considerably less value than the costly price-boosting manipulation that benefits short-term shareholders.
First, costly price manipulation can benefit long-term shareholders only if the firm is repurchasing or issuing shares. By contrast, costly price-boosting manipulation can benefit short-term shareholders whether or not the firm is transacting in its own shares. Because firms may not constantly repurchase or issue shares, at any given time managers serving long-term shareholders are less likely to engage in price-boosting manipulation than managers serving short-term shareholders.
Second, because they have continuing interests in the firm, long-term shareholders are hurt ex post by costly price manipulation that destroys too much economic value; by contrast, short-term shareholders are not. Consequently, when managers engage in costly price manipulation for long-term shareholders, it is likely to be less destructive than when managers engage in costly price manipulation for short-term shareholders.
Therefore, everything else being equal, the harder it is for managers to engage in costly price manipulation, the more likely it is that short-term shareholder interests will be better (more closely aligned with pie maximization) than long-term shareholder interests. For example, consider a world in which costly price manipulation is impossible, but in which long-term shareholders can benefit from costly contraction and costly expansion. In such a world, long-term shareholder interests will be less aligned with pie maximization than short-term shareholder interests.
My purpose has been to show that managers serving long-term shareholder interests, like managers serving short-term shareholder interests, may act in ways that destroy economic value. Managers serving long-term shareholders may not generate more value over time than managers serving short-term shareholders. Indeed, they may well generate less. Proposals to increase the number and power of long-term shareholders could thus lead to a smaller pie, not a larger one.
My analysis has focused on a highly stylized setting with two important simplifications. First, I have focused on a shareholder-only firm, where shareholders are the only residual claimants on the value generated by the firm. In other words, I have ignored other stakeholders. Second, I have assumed that managers seek to advance the interests of either short-term or long-term shareholders rather than to benefit themselves at shareholders’ expense; in other words, I have abstracted from managerial agency costs. In most firms, however, there are likely non-shareholder residual claimants and, more importantly, managerial agency costs. Ultimately, the desirability of favoring long-term shareholders will depend, at least in part, on the effects of increasing the relative power of long-term shareholders on non-shareholder residual claimants and managerial agency costs.
In this Part, I briefly touch on these two issues. Part IX.A discusses the degree to which favoring long-term shareholders is likely to benefit or hurt non-shareholder residual claimants. Part IX.B addresses the degree to which favoring long-term shareholders is likely to increase or decrease managerial agency costs. In both of these discussions, I implicitly assume that the corporation is a simple non-transacting firm; accordingly, the financial interests of long-term shareholders are more closely aligned with shareholder pie maximization than with the interests of short-term shareholders. My goal here is not to provide a complete analysis of these issues. Rather, it is to explain why it is far from clear that considerations of stakeholders and managerial agency costs strengthen the case for favoring long-term shareholders.
My analysis so far has considered a firm in which the only residual claimants on the value generated by the firm are its current and future shareholders. However, while shareholders might be the most important residual claimants on the economic pie created by the firm, they are clearly not the only residual claimants. Other parties, such as employees and creditors, may also be affected by managers’ decision making. Indeed, the existence of such non-shareholder residual claimants has led prominent legal academics such as Margaret Blair and Lynn Stout to argue for a “stakeholder approach” to corporate governance: corporations should be run for the benefit of all stakeholders with residual claims on the corporation, not just shareholders.217
How would the presence of non-shareholder residual claimants affect the desirability of favoring long-term shareholders? It is believed that managers serving short-term shareholders will take steps that harm other stakeholders to boost the short-term stock price.218 On the other hand, it is claimed that managers seeking to benefit long-term shareholders will (or should) directly serve these constituencies as a means to that end.219 The idea that stakeholders are better off if managers seek to serve long-term shareholders rather than short-term shareholders is intuitively appealing.
However, intuition can lead one astray. Neither short-term nor long-term shareholder interests can be counted on to align with the interests of non-shareholder parties. For example, managers serving either short-term shareholders or long-term shareholders may have an incentive to transfer value to shareholders and away from long-term stakeholders or creditors.220
Indeed, one of the very problems associated with “short-termism” may lead managers serving long-term shareholders to squeeze more value from these non-shareholder stakeholders than managers serving short-term shareholders. Critics of short-termism typically argue that the stock market cannot properly value investments with a long-term payoff.221 The market sees only the short-term costs associated with these investments, not the long-term benefits. As a result, managers serving short-term shareholders will not undertake such investments.
But if the stock market does not properly value investments with a long-term payoff, it is unlikely to properly reflect the present value of future savings from cost-minimizing or cost-reducing steps at the expense of non-shareholder residual claimants. Thus, short-term shareholders may well derive less benefit than long-term shareholders from “investing” in strategies that shift value from non-shareholder constituencies to shareholders over time.
Suppose, for example, that in XYZ Corporation the only residual claimants on the value generated by the firm are shareholders and employees. A union is seeking to organize XYZ employees. If the union succeeds, wages will rise. The present value of the extra wages, which represents a transfer from shareholders to employees, is $500 million. XYZ’s managers can thwart the union and prevent a wage increase by inefficiently idling a factory for a month, reducing current earnings by $300 million. Assume that, from an economic perspective, idling the factory is undesirable: it destroys $300 million of value. Assume further that all of this information is public. However, the market has difficulty properly valuing the extra cash that will flow to shareholders over time if wages do not increase.
If managers serve long-term shareholders, then they will idle the factory because the move will save long-term shareholders $200 million ($500 million less $300 million). But if managers serve short-term shareholders, they may not shut down the factory. To the extent the market impounds the $300 million short-term loss, but has difficulty impounding the $500 million long-term gain, such a move could lower the short-term stock price, thereby hurting short-term shareholders. If managers serving short-term shareholders believe that idling the factory will depress the short-term stock price, then they may refrain from doing so.
Wal-Mart may well provide an instructive illustration. The firm is well known for keeping employee wages extremely low, even as compared to other large discount retailers.222 Indeed, some analysts have suggested that its workers are so poorly paid that they are forced to make much greater use of public health and welfare programs than the employees of similar firms,223 costing taxpayers upwards of $3000 per employee annually in public benefit programs.224 These low wages are reportedly maintained via aggressive anti-union tactics.225 These tactics are employed not to benefit short-term shareholders, but rather to serve the interests of the long-term shareholders who control the firm: members of the Walton family.226
I don’t bring up Wal-Mart to criticize or defend the firm’s labor practices. Rather, I use the Wal-Mart example to illustrate the simple point that managers serving long-term shareholders cannot automatically be expected to treat non-shareholder constituencies better than managers serving short-term shareholders. To the extent that markets are not efficient, they might treat them even worse.
Of course, even if managers serving long-term shareholders do treat non-shareholder residual claimants better than managers serving short-term shareholders, the question remains as to whether managers loyally serving long-term shareholders will generate a bigger economic pie than managers loyally serving short-term shareholders. Even if long-term shareholder interests are more closely aligned with those of non-shareholder constituencies, they could still be less aligned with overall pie maximization than are short-term shareholder interests.
My analysis has considered a firm in which managers loyally serve either short-term shareholders or long-term shareholders. In the real world, there are managerial agency costs: managers will tend to act in ways that benefit themselves at the expense of shareholders.227 Indeed, shareholders’ ability to minimize managerial agency costs is one of the most important challenges in the corporate governance of widely held firms. Therefore, the desirability of steps to favor long-term shareholders will depend in large part on how these steps affect shareholders’ ability to reduce managerial agency costs.
On the one hand, it is possible that managerial agency costs would be lower if long-term shareholders had more power. If markets are not efficient, then the benefit of reducing managerial agency costs may not be fully reflected in the short-term stock price. In such a scenario, long-term shareholders may have a greater interest in reducing managerial agency costs than short-term shareholders because they hold their shares for a longer time.228 Long-term shareholders may find it easier to evaluate managerial performance because they are more familiar with it. Moreover, managers may be more willing to bend to the demands of long-term shareholders than the demands of short-term shareholders, knowing that the long-term shareholders are there to stay.
On the other hand, favoring long-term shareholders at short-term shareholders’ expense could also increase managerial agency costs. Many long-term shareholders might (everything else being equal) not be willing to hold as concentrated a position in the stock as short-term shareholders, reducing their incentiveto monitor managers. There may also be certain types of shareholders, such as activist hedge funds, that are particularly capable of monitoring and disciplining managers but cannot or will not commit to holding stock for the very long term.229 Their ability to induce desirable change in firms will decline as long-term shareholders acquire more power. Therefore, shifting power away from short-term shareholders to long-term shareholders could lead to either lower or higher managerial agency costs.
In Part VIII, I described the factors affecting whether managers serving long-term shareholders will generate more or less long-term economic value than managers serving short-term shareholders—assuming a world where shareholders are the only residual claimants and there are no managerial agency costs. In short, in this simple setting, it is far from clear that favoring long-term shareholders will increase the size of the pie when the firm transacts in its own shares. Indeed, favoring long-term shareholders may well decrease the size of the pie in a transacting firm.
In this Part, I have briefly described two other considerations: (1) the possibility that non-shareholder “stakeholders” are also residual claimants on the pie; and (2) managerial agency costs. We have seen that, in the context of any firm (including a non-transacting firm), favoring long-term shareholders may benefit or hurt non-shareholder stakeholders and may reduce or increase managerial agency costs. Thus, when these two considerations are added to the mix, the case for favoring long-term shareholders at any firm could become weaker or stronger. In short, even after accounting for these two factors, the case for favoring long-term shareholders is not compelling.
The power of short-term shareholders, it has been argued, leads to “short-termism”: managers feel pressured to boost the short-term stock price at the expense of maximizing the size of the economic pie created by the firm over time. To counter short-termism, academics and policymakers, as well as leading business executives and corporate lawyers, have proposed various reforms aimed at increasing both the number and power of long-term shareholders relative to short-term shareholders. These proposals appear to reflect the strongly held intuition that managers serving long-term shareholders are likely to generate more economic value over time than managers serving short-term shareholders.
In this Article, I have shown that this intuition is flawed, at least for the typical U.S. firm—one that heavily buys and sells its own shares. On average, U.S. firms trade 30% of their own shares over a five-year period.230 In transacting firms, managers can benefit long-term shareholders merely by transferring wealth from other shareholders through bargain repurchases and inflated-price issuances. Indeed, almost 20% of the wealth created for long-term shareholders today is generated by managers using repurchases and equity issuances to transfer wealth from other shareholders.231 In a transacting firm, long-term shareholders, like short-term shareholders, can benefit from managers’ deliberately destroying value. In particular, long-term shareholders can benefit when managers underinvest in firm projects to buy back stock at a low price, overinvest in projects to sell stock at a high price, manipulate the stock price downward around the time of repurchases, and manipulate the stock price upward around the time of equity issuances. There is evidence suggesting that managers seeking to increase long-term shareholder value engage in these very activities,232 which may well destroy more value than short-termism. As a result, reforms that further favor long-term shareholders in the typical firm could actually reduce the size of the pie created by the firm over time.
My analysis indicates that it is ultimately an empirical question as to which shareholders—short-term or long-term—have interests that are better aligned with a firm’s creation of economic value over time. One of my purposes in writing this Article is to encourage academics to take up this question.233 Determining whether managers serving long-term shareholders are likely to generate more value than managers serving short-term shareholders is crucial for properly evaluating proposals for regulatory intervention, as well as new private ordering arrangements. It is not enough to assume, as many do, that it will be desirable to strengthen the hand of long-term shareholders in public companies simply because they hold their shares for a longer time than short-term shareholders.