Fairer, quicker elections are held in Zimbabwe. The bitter CSX proxy fight, which was scheduled to have been resolved at CSX’s June 25 shareholders’ meeting, may finally reach a definitive outcome on July 25, when that shareholders’ meeting is reconvened. 1 Then again, it may not. CSX may prefer to keep the outcome unresolved until the Second Circuit hears its appeal in August of Judge Lewis Kaplan’s compromise decision, which satisfied neither side.2 So far, excess has been countered with excess, and just about everyone has embarrassed themselves, including, most of all, the SEC, which has waffled and sadly demonstrated that it is woefully behind other major securities regulators. The one exception to this generalization is U.S. District Judge Lewis Kaplan, who once again has shown himself to be not only exceptionally able, but virtually fearless. In successive years, he has faced down the Southern District U.S. Attorney’s Office and now the SEC’s staff.

Nonetheless, Judge Kaplan’s decision is troubling, both in the uncertain reach of its very fact-specific holding and in the potentially universal scope of its broader suggestion that all equity swaps confer beneficial ownership of the underlying, referenced shares upon the “long” side of the swaps contract. This column will explore what intermediate positions are available – both to the Second Circuit and the SEC when it at last awakes to the fact that it must commence rulemaking to address synthetic ownership through equity derivatives.

Judge Kaplan’s Decision

By now, most are familiar with the facts of the CSX case, and they will thus be only briefly summarized. Between October 2006 and October 2007, The Children’s Investment Fund Management (UK) LLP (“TCI”) entered into total return equity swaps with eight different counterparties, all major swap dealers, pursuant to which it acquired the “economic” ownership of 8.8 percent of CSX’s outstanding shares. Predictably, each of the swaps dealers hedged its short position by buying an equivalent number of CSX shares. The court found that “TCI carefully distributed its swaps among eight counterparties so as to prevent any one of them from acquiring greater than 5 percent of CSX’s shares and thus having to disclose its swap agreements with TCI.”3 The court’s view is that TCI’s goal was to avoid any of its dealers being required to file a Schedule 13D. But this assumption overlooks the likelihood that each of these dealers could have crossed Section 13(d)’s 5 percent threshold without filing a Schedule 13D in reliance on Rule 13d-1(b)(1), which would permit the dealer to file instead a Schedule 13G within 45 days after the end of the calendar year.4

What then could explain TCI’s clear insistence on a broad distribution? In my judgment, TCI may have more feared that a swaps dealer relying on Schedule 13G would have been reluctant to vote with TCI for fear that such a vote would demonstrate that it was “a participant in any transaction” having the “purpose or effect” of “changing or influencing the control of the issuer.”5 In this light, a broad distribution made it easier for TCI to assure itself that its swap dealers would accommodate it by voting their shares in its favor. The difference is important because it suggests a different and potentially stronger rationale for the court’s decision.

Nevertheless, in late 2007, TCI largely moved its swaps to Deutsche Bank and Citigroup, contracting with them for an aggregate exposure equal to approximately 9 percent of CSX’s outstanding shares. The court found that at least one motive for TCI’s decision to shift was that Deutsche Bank ran an in-house hedge fund, Austin Friars, which had its own position in CSX stock.6 Thus, unlike the other swap dealers that were fully hedged, Deutsche Bank had its own reasons for voting with TCI (and these apparently motivated TCI to select it as its primary counterparty).

The strongest evidence in the case for CSX’s side involves the behavior of Deutsche Bank as the record dates for voting the CSX stock approached. Not surprisingly, Deutsche Bank lent the shares it purchased to hedge its position to short sellers in order to earn interest on its stake. This is both normal and logical because a fully hedged swaps dealer has no rational interest in voting its shares because it can neither gain nor lose. By lending the shares to short sellers, they pass on their vote for a fee to others, who do have a rational interest in voting. Yet, as the Feb. 27, 2008 record date for the CSX annual meeting approached, Deutsche Bank “recalled the loans so that it would own the shares on the record date and thus be entitled to vote them.”7 Thereafter, it again reloaned its CSX shares immediately after the record date. Moreover, when CSX adjourned the record date, “a similar influx and outflow of shares took place around the adjourned date.”8 Deutsche Bank’s repeated recalling of its loaned shares cost it the interest payments thereby foregone from short sellers, and suggests that it was operating as the agent of TCI.9

Judge Kaplan, however, did not decide the case on this basis. Initially he implies that TCI should be deemed the beneficial owner of the shares that its swap dealers held because SEC Rule 13d-3(a) deems a person to be a beneficial owner if such person has “(1) voting power which includes the power to vote, or to direct the voting of, such security; and/or (2) investment power which includes the power to dispose, or to direct the disposition of such security.”10 With regard to the latter “investment power” prong of this rule, he reasons that TCI knew that its counterparties would buy CSX shares corresponding to the shares referenced in their swap agreement with TCI and that they would sell these shares the moment TCI unwound its swap. Therefore, he suggests that TCI had “investment power” because it held and exercised the de facto ability to cause its counterparties to buy and sell CSX shares. The problem with this argument is that it proves too much and does not distinguish Deutsche Bank from any other professional swaps dealer, which all behave similarly. Moreover, if the decision were rested on this overbroad rationale, it could be easily avoided by predictable strategic adaptations. Swaps dealers could, for example, instead hedge, not by buying shares, but by using single stock futures or combinations of puts and calls, or, most likely, by taking the long side of an equity swap with another swap dealer. All these techniques would presumably be more costly, but they would negate “investment power.” Because it would achieve little, this is the one argument that the Second Circuit should not adopt.

The “voting power” prong has, however, greater promise. Most swaps dealers do not vote the shares they buy to hedge their position for a variety of reasons: because they have no economic motive to vote, because they fear losing Schedule 13G eligibility, or because it is more profitable to lend their shares to short sellers. Those that do vote often divide their votes, voting both ways according to some formula or according to the recommendations of a proxy advisory firm. Thus, the repeated act of recalling shares just prior to the record date and relending the shares immediately afterwards does distinguish Deutsche Bank in this case from the vast majority of swaps dealers. A narrower decision, written on this rationale, could find a violation without generally deeming the “long” side of an equity swap to be the beneficial owner of the referenced shares held by the dealer.

Good reason exists to rest any decision about the existing rule on narrow grounds. Any decision deeming “beneficial ownership” to be held by the “long” side of an equity swap of the shares held by the “short” side could trigger any number of existing shareholder rights plans (or “poison pills”), which are often written in terms of the subject person acquiring the beneficial ownership of a defined percentage of the issuer’s shares (usually 15 percent, but sometimes as low as 10 percent).

Ultimately, Judge Kaplan did not actually find that TCI had beneficial ownership under Rule 13d-3(a)(1) of Deutsche Bank’s CSX shares. Rather, he observed only that “the question whether there was an agreement – explicit or implicit – between Deutsche Bank and TCI with respect to the voting of the shares is a close one,”11 and then he proceeded to rest his holding on Rule 13d-3(b). That latter rule deems a person to have beneficial ownership of shares when the person enters into a contract, arrangement or device “with the purpose or effect of divesting such person of beneficial ownership of a security or preventing the vesting of such beneficial ownership as part of a plan or scheme to evade the reporting requirements or Section 13(d) or 13(g) of the Act . . . .” No earlier decision had ever seriously sought to construe this rule, and the SEC has also been largely silent on its scope. Writing on a blank slate, Judge Kaplan found this rule to be applicable when the actor “enters into a transaction with the intent to create a false appearance that there is no large accumulation of securities that might have a potential for shifting corporate control by evading the disclosure requirements of Section 13(d) or (g) through preventing the vesting of beneficial ownership in the actor.”12 In so doing, Judge Kaplan politely disagreed with the SEC’s Division of Corporation Finance’s view (which it borrowed uncritically from an expert witness for TGI) that Rule 13d-3 “is narrower in coverage than the statute.”13 These differing views will be a major issue when the case is argued before the Second Circuit. If one assumes that an attempt to hide a potential control block by using a collection of swaps dealers who will each own less than five percent violates Rule 13d-3(b), then the court clearly had sufficient evidence to find such an attempt.

But what if the Second Circuit adopts a narrower interpretation of Rule 13d-3(b)? It may fear that Judge Kaplan’s interpretation reaches too broadly because, under it, most any significant accumulation of equity swaps with respect to a single issuer would violate Rule 13d-3(b) – in part because the swaps dealer could rely on Schedule 13G and delay disclosing its accumulation of shares. If the Second Circuit wants to write narrowly, it may be necessary to return to the issue that Judge Kaplan called a “close call” but that his complete opinion suggests would have been easy for him to decide: was there an implicit agreement as to voting between Deutsche Bank and TCI? Here, TCI’s original decision to distribute its swaps among eight dealers constitutes better evidence of its desire to secure the voting support of these dealers than of an intent to avoid their filing Schedule 13Ds (because each should have been filed only under Schedule 13G). More importantly, Deutsche Bank’s decision twice to recall its lent shares back from short sellers as the two record dates approached cost it money and evidences a desire to accommodate TCI (or, alternatively, to vote in favor of its own in-house hedge fund, in which case there was arguably an undisclosed “group”). Taken together, these facts do suggest an implicit voting agreement, and such a narrower decision should not cause shock and consternation within the swap dealers community, but a remand for further findings may be necessary.

The Need for SEC Rulemaking

The most that can be said for the SEC staff’s performance in the CSX saga is that they understandably did not want to make law retroactively, or to announce policy changes, in the context of a litigated case. In fairness, the securities bar did believe that the shares underlying equity swaps were normally not beneficially owned by the “long” side of the swap (although this may say more about the automatic tendency of the securities bar to converge around the most deregulatory position). But, shortly after Judge Kaplan’s decision, the Financial Services Authority (“FSA”), the U.K. securities regulator, announced a new policy requiring disclosure of derivative contracts, on an aggregated basis with ownership of common stock, at the 3 percent level.14 Australia took a similar position in 2007.

In contrast, the SEC seems in a state of regulatory paralysis, unwilling or unable to challenge the powerful lobbying forces of the swaps dealers. The Division of Corporation’s Finance’s letter to Judge Kaplan seemed something out of the last century. Meanwhile, the FSA estimates that approximately one third of the trading on the London Stock Exchange is attributable to equity swaps.15 In ignoring swaps, the SEC is wearing blinders.

But is any realistic compromise possible? Some academics, including Professors Black and Hu (who lined up on opposite sides in this case to their mutual advantage), have called for very broad disclosure of significant accumulations of all functional equivalents to common stock – single stock futures, put and call option combinations, and all equity derivatives.16 But this is not what the FSA recently required. Its proposed rules focus on voting and would give an exemption to swap dealers where the “long” side of the swaps contract could not compel or influence the swaps dealer to vote its underlying shares. Suppose then that the SEC, after rule-making, were to require disclosure by the “long” side of an equity swap of the shares held by the “short” side (i.e., the swap dealer’s hedged position) – unless the swaps dealer had publicly taken the position that it would not vote its referenced shares in any election or consent solicitation. This would not be a hard or costly position for swap dealers to adopt, and it would still leave them free to lend their shares to unaffiliated short sellers (who could vote them). Such a proposed rule could also provide that if a swaps dealer did not want to make a public statement that it would never vote its underlying shares, it could still also provide in the swaps contract that it would not vote (and thereby avoid disclosure). This variation might require some tinkering with the current SEC rules because it would be a contract between the insurgent and the dealer with respect to the issuer’s shares (which would sometimes need to be disclosed under the current rules).17 Little doubt exists that the SEC has authority to define “beneficial ownership” more broadly than it has done to date.18

The key point is that the position here outlined would only require disclosure by the “long” side of the shares that could be voted and not of equity positions that simply gave the “long” side increased economic incentive to wage a proxy fight, but no increased voting power. Thus, it would tolerate some “uncoupling” of economic ownership and voting rights, but bar the hidden acquisition of voting power by implicit agreements.

Stay tuned. The saga of the CSX train wreck will continue, the Second Circuit must rule, and at some point the SEC may awake from its slumber. But at present, it is apparent that Wall Street believes that transparency is only a norm by which the other guy should abide.

John C. Coffee, Jr. is the Adolf A. Berle professor of law at Columbia University Law School and director of its Center on Corporate Governance.


1. The two dissident hedge funds claim to have elected four out of the five candidates that they ran for CSX’s 12 person board. The one insurgent candidate not elected – Gary Wilson, a former board chair of Northwest Airlines – was not supported by RiskMetrics, the proxy advisory firm, apparently because he serves as a director of Yahoo!, where RiskMetrics is currently considering its position. See John D. Boyd, “Fund Claims Four CSX Seats,” Traffic World, July 7, 2008 at p. 26.

2. See CSX Corporation v. The Children’s Investment Fund Management (UK) LLP, 2008 U.S. Dist. LEXIS 46039 (S.D.N.Y. June 11, 2008).

3. Id. at *36.

4. Of course, a Schedule 13G filing, which would have been necessary by Feb. 14, 2008 (i.e., 45 days after year end), would also give the market belated notice that swap dealers were assembling CSX’s stock.

5. See Rule 13d-1(b)(1)(i), 17 C.F.R. §240.13d-1(b)(1)(i).

6. CSX Corporation v. The Children’s Investment Fund Management (UK) LLP, supra note 2, at *75.

7. Id. at *76.

8. Id. at *77.

9. Alternatively, one can infer that Deutsche Bank was voting to assist Austin Friars, its hedge fund, but this interpretation is factually open to challenge and would in any event lead only to the conclusion that TCI, Deutsche Bank and Austin Friars were an undisclosed §13(d) “group.”

10. See Rule 13d-3(a), 17 C.F.R. §240.13d-3(a).

11. CSX Corporation v. The Children’s Investment Fund Management (UK) LLP, supra note 2, at *78.

12. Id. at *95.

13. Id. at *99.

14. See James Mackintosh and Jennifer Hughes, “FSA imposes disclosure to block covert sales,” Financial Times, July 2, 2008.

15. Id. (“The FSA estimates that CFDs account for almost one third of equity trading.”)

16. For their position, see Henry T.C. Hu and Bernard Black, “Equity and Debt Decoupling and Empty Voting II: Importance and Extensions,” 156 U. of Pa. L. Rev. 625 (2008), and Henry T.C. Hu and Bernard Black, “The New Vote Buying: Empty Voting and Hidden Ownership,” 79 Southern California L. Rev. 811 (2006).

17. See Item 6 to Schedule 13D (“Contracts, Arrangements, Understandings, or Relationships With Respect to Securities of the Issuer”).

18. Indeed, the SEC has gone much further in relying on §14(e) of the Exchange Act to enable it to broadly extend its substantive tender offer rules so that they go well beyond the statutory text of §14(d). See Polaroid Corp. v. Disney, 862 F.2d 987, 994-995 (3d Cir. 1998) (upholding All Holders Rule); see also Exchange Act Release 34-23421 (July 18, 1986). Section 23(a) of the Exchange Act gives the SEC broad “power to make such rules and regulations as may be necessary or appropriate to implement the provisions of this chapter,” including presumably its anti-fraud provisions. Finally, Rule 14e-3, which addresses a related context, goes well beyond the normal scope of insider trading liability and was upheld in United States v. O’Hagan, 521 U.S. 642, 669 (1997).