«Henry T.C. Hu* Bernard Black** Most U.S. public companies have a one-share, one-vote capital structure, in which voting power is proportional to ...»
Empty Voting and Hidden Ownership:
Taxonomy, Implications, and Reforms
Henry T.C. Hu*
Most U.S. public companies have a one-share, one-vote capital structure, in which voting power
is proportional to economic ownership. This ensures that shareholders have economic
incentives to vote to increase firm value and helps to legitimate managers' exercise of authority
over property the managers do not own. Berle-Means’ “separation of ownership and control” suggests that shareholders have limited say over firm actions. Even so, mechanisms rooted in the shareholder vote, including proxy fights and takeover bids, constrain managers not to stray too far from shareholder wealth maximization.
The derivatives revolution and other capital markets developments threaten this familiar pattern.
Both outside investors and insiders can readily decouple economic ownership of shares from voting rights to those shares. This decoupling -- which we call "the new vote buying" -- is often hidden from public view and is largely untouched by current regulation. Hedge funds have been especially creative in decoupling voting rights from economic ownership. Sometimes they hold more votes than economic ownership: a pattern we call “empty voting.” Sometimes they hold more economic ownership than votes, though often with de facto ability to acquire the votes if needed -- a situation we call "hidden ownership" because the economic ownership is often not disclosed. Insiders also often use empty voting techniques.
This article offers a taxonomy of the new vote buying that unpacks its functional elements. We discuss the implications of decoupling for control contests and other shareholder oversight. We also propose a disclosure-based regulatory response. Our disclosure proposal would integrate and simplify five existing, inconsistent share ownership disclosure regimes, and is worth considering independent of its value with respect to decoupling. In the longer term, substantive responses to empty voting may be needed; we sketch some possible responses.
* Professor Hu is Allan Shivers Chair in the Law of Banking and Finance, University of Texas Law School ( 512-232-1373, email@example.com).Professor Black is Hayden W. Head Regents Chair for Faculty Excellence, University of Texas Law School, and Professor of Finance, University of Texas, McCombs School of Business.
(512-471-4632, firstname.lastname@example.org). The authors thank workshop participants at the American Law and Economics Association annual meeting (May 2005), the Conference on Boundaries of SEC Regulation at the Financial Economics Institute of Claremont McKenna College (Feb. 2006), Iman Anabtawi, Larry Harris, Bruce Johnsen, Harold J. Mulherin, Jeff Netter, Edward Rock, David Skeel, Janet Kiholm Smith, and [to come] for comments and suggestions.
IV. A Proposal for Integrated Ownership Disclosure A. General Considerations Table 4. Proposed Integrated Ownership Disclosure B. Large Shareholder Disclosure (Schedules 13D and 13G) C. Institutional Money Managers and Mutual Funds D. Disclosure of Record Date Capture (Expansion of Form 13F) E. Summary V. Toward Substantive Responses to Empty Voting A. General Considerations B. Strategies Focused on Voting Rights
1. Direct Limits on Voting Rights: The Martin-Partnoy Proposal
2. Voting By Record Owners: Extension to Equity Swaps
The governance structure of the modern corporation emerged in a period characterized by two central concepts. One was the “separation of ownership and control,” which has been central to thinking about corporate governance since the 1930s. In stylized form, as conceived by Adolph Berle and Gardiner Means,1 and converted to modern language, the professional managers of publicly held corporations hold few shares yet have substantial control over their firms. Shareholder-owners face large collective action problems in overseeing managers.
Constraining managerial autonomy and reducing the divergence between managerial and shareholder interests are thus core goals for an overall governance system. State corporate law rules and federal disclosure rules are largely evaluated against these goals. So too are marketcentered constraints, including institutional investor activism and the market for corporate control. These goals and the legal and market tools that foster these goals together comprise the corporate governance paradigm.
The second core concept was that a “shareholder” has economic ownership coupled with voting power. For much of the period in which modern corporate governance scholarship developed, one-share, one-vote structures were nearly universal, enforced by New York Stock Exchange listing standards. Even when dual-class structures became possible, around 1990, they remained uncommon. Our views of the ways in which institutional investor oversights and the market for corporate control can discipline wayward management depend on the coupling of economic interest and voting power. Terminology reflects this coupling: we use a single term, "ownership," to refer to possession of both the economic return on shares and corresponding
shareholders have an economic interest in increasing share value and will vote to further that interest. Beyond the instrumental role of voting, Delaware courts treat the concept of shareholder-as-owner-and-voter as the core ideological basis for managerial exercise of authority
Commission (SEC) “large shareholder” disclosure rules also largely assume the coupling of economic ownership and voting power.3 The assumption that votes are tightly linked to economic interest is, however, increasingly untenable. The derivatives revolution in finance, especially the growth in equity swaps and other privately negotiated ("over the counter" or "OTC") equity derivatives, and related growth in the stock lending market, are making it ever easier and cheaper to decouple economic ownership from voting power.4 Both company insiders and outside investors (especially hedge funds) are taking advantage of this new opportunity. Sometimes they hold more votes than shares -- a pattern we call "empty voting" because the votes have been emptied of an accompanying economic interest. Persons with more votes than economic interest are said to be "long the vote." In an extreme case, an investor can be long the vote while holding a "net short" economic position, which gives the investor an incentive to vote in ways that reduce company value.
2 See, e.g., Blasius Industries, Inc. v. Atlas Corp., 564 A.2d 651 (Del. Ch. 1988); cf. MM Companies Inc.
v. Liquid Audio, Inc., 813 A.2d 1118 (Del. 2003).
3 We provide citations to SEC rules and discuss these rules in Part III.
4 The complexities raised by the “new vote buying” are new examples of the challenges the derivatives revolution and modern financial innovation generally pose to corporate governance principles. For discussions of the overarching issue, see, e.g., Henry T. C. Hu, New Financial Products, the Modern Process of Financial Innovation, and the Puzzle of Shareholder Welfare, 69 TEX. L. REV. 1273 (1991); Henry T. C. Hu, Behind the Corporate Hedge: Information and the Limits of “Shareholder Wealth Maximization,” J. APPLIED CORP. FIN., Fall 1996, at 39-51.
rights, combined with informal access to the corresponding voting rights. This ownership is typically not disclosed under current large shareholder disclosure rules, which focus on voting power rather than economic interest, and do not clearly require disclosure of the informal voting power that often exists. Investors can often engage in what we call “vote morphing,” moving at will between informal, undisclosed voting power and actual voting rights. We use the term “hidden ownership” to refer to the combination of undisclosed economic ownership plus probable informal voting power.
We refer to empty voting and hidden ownership together as "the new vote buying" or simply as "decoupling." In the past several years, this decoupling has affected takeover battles and control of public companies in (at least) the U.S., the U.K., Germany, Japan, Australia, and New Zealand. Its full extent is unknown. Policymakers abroad are beginning to confront the new vote buying, and requiring additional disclosure. Policymakers in the U.S. have barely begun to address it, but will soon need to.5 There are a number of ways to decouple votes from economic ownership. One common method relies on the stock lending market, which lets one investor “borrow” shares from another.
Under standard lending arrangements, the stock borrower has voting rights but no economic ownership, while the stock lender has economic ownership without voting rights. A second approach employs an equity swap, in which the person with the long equity side (the “equity leg”) of the swap acquires economic ownership of shares without voting rights, while the short side (the "interest leg") often hedges its economic risk by holding shares, thus ending up with 5 The only public sector recognition in the United States that we know of is a July 2005 speech by Vice Chancellor Leo Strine, where he stated that what we term “empty voting” and “related factors” are “making it difficult for corporate law makers to avoid a fundamental look” at corporate law. See David Marcus, Thinking Big Thoughts, CORPORATE CONTROL ALERT, Aug.-Sept. 2005, at 6.
relying on put and call options or, where they exist, on single-stock futures.6 Surprising results can flow from empty voting. A recent public U.S. instance illustrates the potential risks. Perry Corp., a hedge fund, owned seven million shares of King Pharmaceuticals. Mylan Laboratories agreed in late 2004 to buy King in a stock-for-stock merger at a substantial premium. However, Mylan's shares dropped sharply when the deal was announced. To help Mylan obtain shareholder approval for the merger, Perry bought 9.9% of Mylan – becoming Mylan’s largest shareholder -- but fully hedged the market risk associated with the Mylan shares. Perry thus had 9.9% voting ownership of Mylan but zero economic ownership. Including its position in King, Perry's overall economic interest in Mylan was negative. The more Mylan (over)paid for King, the more Perry stood to profit.7 A second, potentially beneficial use of empty voting involves outside shareholders magnifying an existing long ownership position. Other things equal, this can reduce shareholder collective action problems. For example, an activist hedge fund can borrow shares just before the record date for a shareholder vote, then reverse the transaction afterwards. The first publicly reported instance of this "record date capture" strategy occurred in the U.K. in 2002. Laxey Partners, a hedge fund, held about 1% of the shares of British Land, a major U.K. property company. At the annual general meeting, Laxey emerged with voting power over 9% of British Land's shares, the better to support a proposal to dismember British Land. Just before the record date, Laxey had borrowed 42 million shares. Hedge funds may see this record date capture as a 6 The corresponding U.K. instrument is known as a "contract for differences" or CFD. In this article, we use the term "equity swap" to refer to both instruments.
7 Part II provides further details on this and other new vote buying examples. Table 2 provides source citations.
Empty voting by institutions is a close cousin to hedging techniques widely used by insiders (zero-cost collars, variable prepaid forward contracts, and the like) by which managers and controlling shareholders retain formal ownership of shares, while shedding some or most of their economic ownership.8 In the U.S., these strategies have typically been driven by managers' desire to shed risk while deferring taxes, rather than by vote buying motives. But insiders can easily also use empty voting techniques to cement their control. As we discuss in Part II, they are doing so in other countries.
Conversely, investors can have greater economic ownership than voting rights -- but with the de facto ability to acquire those rights quickly when they are needed. Under existing disclosure rules, the lack of formal voting rights may let the investor's position remain hidden.
Perry's stake in a New Zealand company, Rubicon Ltd., which came to light in 2003, illustrates this. Perry used equity swaps to hold an undisclosed 16% economic stake, despite New Zealand’s large shareholder disclosure rules, which are similar to U.S. Section 13(d) and require disclosure by 5% shareholders. When an election came along, Perry in effect exchanged its swap rights for shares held by the swap counterparties, thus morphing its de facto voting rights into actual voting rights. Its failure to disclose the swaps was held not to violate New Zealand law.
8 See, e.g., J. Carr Bettis, John M. Bizjak & Michael L. Lemmon, Managerial Ownership, Incentive Contracting, and the Use of Zero-Cost Collars and Equity Swaps by Corporate Insiders, 36 J. FIN. & QUANT.
ANALYSIS 345 (2001); Rachel Emma Silverman & Jane J. Kim, IRS Targets Strategy for Wealthy Executives, WALL ST. J., Feb. 9, 2006 (describing variable prepaid forwards).