Largely unnoticed, two parallel actions are now pending in the U.S. District Court for the Northern District of California and the Delaware Court of Chancery that could have greater impact on corporate governance than most recent Supreme Court decisions.1 Both cases challenge bylaws adopted by Chevron and Curtiss-Wright, respectively, that mandate that certain broad categories of shareholder actions, including derivative actions and actions brought to assert a breach of fiduciary duty, may only be brought in the Delaware Court of Chancery. Such an “exclusive forum” provision was recommended in a 2010 Delaware Court of Chancery decision as the best answer to the growing problem of multi-forum litigation under which virtually every merger transaction now attracts a flurry of litigation in state and federal court.2 Both bylaws appear to be a delayed response to that recommendation, and the corporate bar is now anxiously awaiting the outcome. If the bylaws are upheld, many public corporations are likely to adopt similar amendments in the near future.3

This column will argue that the use of charter amendments and bylaws making Delaware the exclusive forum for fiduciary litigation involving Delaware corporations represents the best conceptual answer to a real and growing problem. A variety of factors—including intense competition within a fragmented plaintiffs bar—has produced the perverse phenomenon of “phantom litigation” in which rival plaintiffs firms battle to be named lead counsel, but then do not actively litigate the case following that contest.4 Motions are not made, discovery is not taken, and depositions are not scheduled.

Instead, a largely invisible competition begins to see which plaintiffs firm can settle first with the defendants. Necessarily, this requires the plaintiffs firm to offer to settle cheaply, and the preconditions are thereby satisfied for the commencement of a “reverse auction” under which the winner is generally the low bidder.5 As a result, recent empirical surveys have found that M&A litigation now settles in the vast majority of cases on a “disclosure only” basis with no monetary payment being made by any defendant.6

The evils here are multiple. First, unless one believes that “disclosure only” settlements truly benefit shareholders and justify million dollar fee awards (in which case this author would like to sell you a bridge to Brooklyn at a very cheap price), then such litigation gives off at least a faint odor of collusion. Second, corporations must defend multiple actions, many filed in copycat style in multiple jurisdictions, with shareholders bearing ultimately the costs of both sides. Third, the Delaware Court of Chancery is effectively disabled, both because these cases typically settle outside of Delaware and because no court is given the opportunity to reach any decision of consequence, except for the ultimate decision to bless the settlement.

These problems are intractable because if Delaware takes any step to discourage “phantom litigation” (such as by reducing fee awards or tightening the standards for selection of lead counsel), it will only accelerate the exodus of M&A cases out of Delaware. Indeed, many believe that a few Delaware decisions tightening fee award standards and ending the old “first to file” presumption in selecting lead counsel spurred the migration of cases out of Delaware.7

Although no single explanation can account fully for this rush for the exit from Delaware, the adverse consequences are clear, and only the “exclusive forum” prescription seems realistically capable of addressing these problems. Thus, this column will argue that the case for a forum selection clause, if approved by the shareholders, is compelling. But even if this is the optimal reform, its implementation is tricky. As usual, the devil is in the details. Although Delaware corporate law poses no real obstacle to a forum selection clause (whether in the corporate charter or in a shareholder-approved bylaw),8 federal law arguably may. Both the diversity jurisdiction statute and, more specifically, the Securities Litigation Uniform Standards Act of 1998 (SLUSA) seem to entitle the plaintiff to choose at its option between state and federal court.9 How are these statutes to be read to uphold such a clause? Similarly, some claims (such as Rule 10b-5 claims) can be asserted derivatively, but only in federal court, because state courts lack subject matter jurisdiction over the Securities Exchange Act.10

Trickier still is the appropriate wording of such an “exclusive forum” clause. How ambitious dare counsel be? In the one recent federal case, a forum selection bylaw was invalidated on a variety of grounds, but most clearly because it applied retroactively to already pending litigation and had not been approved by the shareholders.11 Thus, the bylaw was perceived by the disapproving court as a naked instance of self-dealing because the directors who adopted it were voting to terminate litigation brought against themselves.

Yet, if the appearance of self-dealing is to be avoided, can the bylaw give the board of directors the choice as to which forum should hear the case? Strange as it seems, both the bylaws adopted by Chevron and Curtiss-Wright do exactly this. Clearly following the same template, they each allow the corporation to “consent in writing to the selection of an alternative forum.”12 Plaintiffs in both suits attacking these bylaws plausibly assert that the above-quoted language is nothing but a thinly-concealed invitation to the plaintiffs bringing the case in the non-Delaware forum to make a cheap offer to settle on a non-monetary basis (because otherwise the corporation will insist on the case being sent back to Delaware). Clearly, if one wants to restrict collusive settlements, reverse actions, and costly multi-forum litigation, it is no answer to give the corporation the choice of forum after the first action is filed. That invites the worse of both worlds by encouraging collusion while disabling any truly adversarial plaintiff from suing outside of Delaware.

What Explains the Exodus?

Two trends are occurring simultaneously. First, M&A cases are leaving Delaware. A detailed survey by professors John Armour, Bernard Black and Brian Cheffins that will soon appear in the Columbia Business Law Review finds that where formerly over two-thirds of M&A litigation involving Delaware target companies was brought in the Delaware Court of Chancery, this percentage fell by half to 34 percent between 2002 and 2010.13 Most of this decline is accounted for by cases filed not in federal court, but in other state courts, often in the state where the target’s company’s principal place of business is located. An even more recent study by NERA researchers reports that in M&A transactions exceeding $100 million that were challenged in state court between 2006 and 2010, only 20 percent were filed exclusively in Delaware and only 33 percent were filed in Delaware and another state.14 Thus, in 67 percent of these lawsuits, Delaware was simply not involved.

Second, the overall rate of stockholder litigation in state courts is increasing sharply, especially in M&A cases. Where once slightly less than 50 percent of Delaware target companies were sued in M&A litigation, the percentage of M&A transactions now attracting litigation has soared as high as 90 percent in recent years (both for Delaware and non-Delaware companies).15 Apparently, target companies are just too attractive not to sue.

What explains this increase at a time when securities class action litigation is generally subsiding?16 The inconsistency between these two trends is striking. Multiple explanations for this disparity make sense. First, precisely to the extent that the Private Securities Litigation Reform Act of 1995 (PSLRA) discouraged some plaintiff law firms from suing in federal court because of the PSLRA’s enhanced pleading standards, these firms may have turned to state court. Firms deterred by higher pleading standards might consist disproportionately of “bottom fishers” that could not or would not engage in costly discovery, and hence their migration to state court could partly explain the phenomenon of “phantom litigation.”

Second, the increased size of M&A transactions may turn a low cost “disclosure only” settlement into an offer defendants cannot refuse—at least if there is even the slightest risk of either a plaintiff’s victory or a delay in the transaction. If the transaction is for, say, $5 billion, $1 million in attorney fees is only a “rounding error” of one-tenth of 1 percent of the transaction’s cost to risk-averse defendants.

Third, an increased competitiveness may characterize the plaintiffs bar, which is the product in part of the fragmentation of the old plaintiffs bar oligopoly formerly dominated by the Milberg Weiss firm. This new competition tends to imply that every case that might be brought will be brought—at least by one firm.

No one of these theories works perfectly. The PSLRA was passed in 1995, but the rise of multi-forum litigation really dates from sometime after 2000.17 Thus, there is a timing gap. Although the large size of M&A transactions may explain less resistance to a low-cost settlement (particularly when settling is cheaper than defending), the size of M&A transactions peaked several years ago, while the increase in M&A transactions being litigated only peaked more recently in 2009 to 2011.

Probably the best explanation for the increased rate of non-Delaware filings may be that plaintiffs have learned that they gain something by filing outside Delaware. What do they gain? Although the outcome of M&A litigation brought in Delaware is relatively predictable, cases filed outside Delaware are much less predictable. State judges outside of Delaware have often never previously seen an M&A case. Equally important, it may be the largest litigation over which the judge has ever presided, and, unfortunately for defendants, the judge may be fascinated by it (and the possible media attention it attracts). A dozen or more Wall Street attorneys in full dress uniform and limousines parked in front of the courthouse may increase the court’s own sense of self-importance (at the least, the case beats hearing divorces and auto crashes!) In turn, uncertainty about the outcome leads risk-averse defendants to settle (particularly if they can settle on a “disclosure only” basis). Moreover, the less experienced court will take longer to decide, and delay is also problematic for defendants, encouraging them to settle.

Next, there is the strong possibility that Delaware is more parsimonious on legal fees. Given that settlement of the litigation seems predictable in M&A transactions, wherever they are brought, the expected differential in fee awards may be the critical variable motivating the exodus. Lastly, Delaware has moved away from the “first to file” presumption in determining the choice of lead counsel, and, borrowing from the PSLRA, it now looks to other factors, such as the support of large institutional investors.18 Given this, a small firm without institutional clients cannot expect to gain control of the case in Delaware and might as well file elsewhere at the outset.

Once a case is filed in some other state, it will usually attract one or more parallel actions in Delaware and elsewhere, and then the delicate “Kabuki dance” to settlement can begin.19 A recent NERA Economic Consulting study gives a useful illustration: At 7:00 a.m. on Sept. 12, 2011, a merger of two public corporations at a seemingly high 57 percent premium was announced by a press release issued in California.20 The transaction was valued at $3.7 billion and thus can be described as medium-sized. Three hours later, the first plaintiffs law firm (which was based in New York) issued and posted on the Web a press release announcing that it was “investigating” the fairness of the merger. By the end of that day, 11 plaintiffs law firms had posted such notices, and by two days later, the total number had grown to 19 firms. Of course, these notices served a purpose: to attract potential plaintiffs. Relatively quickly, cases were filed in California state court and in Delaware. Within two months, the case settled in California on a “disclosure only” basis (with defendants agreeing not to challenge legal fees of up to $795,000). This is competition—but only over the right to settle and confer preclusion on the defendants. The social value of such an auction for preclusion is dubious at best.

What Reform Is Needed?

The case for forum selection clauses was strongly made by Vice Chancellor J. Travis Laster in his Revlon decision.21 Although it is clearly an answer to Delaware’s problem of a declining market share (with today less than a third of M&A cases being brought in Delaware22), liberals may believe that it is not a good public policy solution because they presume Delaware law to be biased against plaintiffs. Here, they miss the forest for the trees. Unless a single court has control over the case, judicial authority will dissolve in the universal solvent of multi-forum competition. No court will have meaningful authority, and that is the worst outcome for investors. Indeed, the current predominance of “disclosure only” settlements suggests that this point has already been reached. Conceivably, a Delaware forum is not in the best interests of some plaintiffs law firms, but a single forum is in the best interests of shareholders because it enables courts to enforce the rule of law and discourage collusion. Moreover, at least sometimes, plaintiffs attorneys do win in Delaware and do receive high fee awards that are not the product of a settlement.23

But if a single forum is desirable, it is not easily achieved. First, some federal statutes expressly grant the plaintiff the choice between federal and state forums. Perhaps the clearest example is afforded by SLUSA. It preempts class actions (and certain consolidated actions) in state court that raise essentially securities fraud claims, but SLUSA is subject to the “Delaware carve out,” which is set forth in §16(d)(1) of the Securities Act of 1933.24 That provision states that a class action “based upon the statutory or common law of the state in which the issuer is incorporated…may be maintained in a State or Federal court by a private party.”25 On its face, this seemingly gives the plaintiff the option to choose the forum in M&A litigation alleging a fiduciary breach.26

A second problem involves the recent decision in Galaviz v. Berg.27 In this 2011 decision, the district court of the Northern District of California invalidated a bylaw mandating the Delaware Court of Chancery as the exclusive forum for the fiduciary breach action before it. Primarily, it relied upon the facts that (1) the bylaw was retroactive and applied to the already pending case before it, and (2) the bylaw had not been approved by the shareholders, but only by defendant directors. Still, while it acknowledged that the case for the bylaw would be stronger if the bylaw had been adopted by the shareholders, it also found that “the enforceability of a purported venue requirement is a matter of federal common law.”28 This is the key point of law that will be contested in the Chevron and Curtiss Wright cases noted earlier.

Several answers are possible to these claims. First Galaviz probably overstates in claiming that venue is governed by federal common law. In earlier cases (which Galaviz attempts to distinguish), the Supreme Court has upheld forum selection clauses.29 Rather than be a bulwark against state law, federal common law generally should incorporate important state policies, including a state’s preference for enforcing forum selection clauses.30 Nowhere is this clearer than in the case of corporate law, because the court has many times recognized that corporations are creatures of state law.31 In this light, a forum selection clause is less a matter of the procedural law on venue than the substantive law of corporate governance.

The leading case that states this position is Cohen v. Beneficial Industrial Loan,32 which clearly holds that the substantive corporate governance policies of a state must be respected, even if they are expressed in procedural terms that vary from federal procedural norms. Cohen addressed a New Jersey securities-for-expenses statute and whether it applied in federal court, despite the absence of any similar federal procedural rule. The court held that the New Jersey statute did apply precisely because its purpose was substantive (namely, to protect directors and officers from frivolous litigation). Similarly, a properly drafted forum selection clause has an even clearer substantive goal: to allow shareholders to protect themselves from costly litigation and collusive settlements.

The one weakness in this analogy is that Delaware has no formally declared policy favoring forum selection clauses. But it could have such a policy, and, if so, it should be given dispositive effect. Either the Delaware Supreme Court could announce such a policy or, more simply, Delaware could adopt a statute authorizing (but not mandating) such a clause. The latter step would be squarely in the tradition of §102(b)(7) of the Delaware General Corporation Law, which authorized charter amendments exculpating the directors of Delaware corporations from monetary liability for breach of the duty of care. Delaware acted in the wake of Smith v. Van Gorkom,33 which had created a liability crisis for directors and insurers. Here, a new §102(b)(8) could authorize forum selection clauses in the charter or bylaws to the extent they were approved by shareholders and made Delaware the exclusive forum.34 Although state statutes that conflict with federal rules are not always respected, these cases are distinguishable in this context.35

How, then, should a federal court deal with the argument that SLUSA or the diversity jurisdiction statute gave the federal court jurisdiction? The appropriate answer would be that the federal court has subject matter jurisdiction over the action, but it should respect and enforce the forum selection clause. Thus, if the action were a derivative action, brought in federal court, asserting both Rule 10b-5 claims and breach of fiduciary duty claims, the court would retain the Rule 10b-5 claims (which the Delaware state court could not hear), but remand the breach of fiduciary claims, refusing to try to resolve them in federal court.36

Absent such a statute or judicial policy, one can still argue that Delaware has a strong and clearly expressed policy of favoring an enabling body of corporate law under which shareholders can select their own rules.37 In due course, we will see if that works. If not, legislation is the answer.


The corporate bar is waiting for the uncertainty created by Galaviz v. Berg to be clarified. The bar may also be anxious about whether proxy advisers (most notably Institutional Shareholder Services (ISS)) will tell shareholders to support or oppose such clauses. ISS needs to be educated to understand that forum selection clauses that meet three conditions are in the best interests of shareholders. Those conditions are: (1) the clause must be approved by shareholders; (2) it should not be retroactive; and (3) it should not give corporate management or the board discretion to choose the forum after the litigation is brought. Clauses that satisfy these conditions are the best way to curb the shady market that currently sells settlements to the lowest bidder.

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. See Buchansky v. Armacost, No. 4:12–cv–61597 (N.D. Cal. March 29, 2012); Boilermakers Local 154 Retirement Fund v. Curtiss-Wright, No. 7219 (Del. Ch. Feb. 6, 2012). The former suit against Chevron is a derivative action; the latter, a class action.

2. See In re Revlon Shareholder Litig., 990 A.2d 940, 945-46 (Del. Ch. 2010). Vice Chancellor J. Travis Laster eloquently describes the pattern in this case under which “No One Litigates Anything.”

3. In discussions with leading M&A attorneys, I have heard them take the position that such clauses can be inserted today into the charters of IPO companies or spinoffs, but that it is too early to ask for a shareholder vote on them until legal uncertainties are resolved. In part, they are reluctant to seek the approval of proxy advisers, such as ISS, until legal doubts are clarified.

4. Vice Chancellor J. Travis Laster has made this point incisively. See supra note 2. See also In re Cox Communications, 879 A.2d 604, 608 (Del. Ch. 2005) (Strine, V.C.). For academic commentary recognizing this problem as early as 2004, see Elliot Weiss and Lawrence J. White, “File Early, Then Free Ride: How Delaware Law (Mis)Shapes Shareholder Class Actions,” 57 Vand. L. Rev. 1797 (2004).

5. In a “reverse auction,” a defendant seeks to settle with the weakest plaintiff’s team or the plaintiff’s team most distant from trial because that team would recognize that they cannot possibly avoid preclusion by other plaintiffs whose actions would be earlier resolved. The premise is also that those who have invested the least in the action can settle the cheapest. For the original articles coining this term, see John C. Coffee, Jr., “Class Wars: The Dilemma of the Mass Tort Class Action,” 95 Colum. L. Rev. 1343, 1370–73 (1995); John C. Coffee, Jr., “Class Action Accountability: Reconciling Exit, Voice and Loyalty in Representative Litigation,” 100 Colum. L. Rev. 370, 392 (2000). See also Reynolds v. Beneficial Nat’l Bank, 288 F. 3d 277, 282 (7th Cir. 2003) (Posner, J.).

6. See Douglas J. Clark & Marcia Kramer Meyer, “Anatomy of a Merger Litigation” (NERA Economic Consulting, April 2012), at 10. This study was based on a sample of 162 settled merger cases.

7. Although the race to be the “first to file” in Delaware was long motivated by the perception that the first plaintiffs firm to file would win the lead counsel designation, two Delaware decisions after 2000 rejected this standard in favor of preferring the shareholder (and its counsel) with the largest economic interest in the case. See TCW Technology Partnership v. Intermedia Communications, 2000 WL 1654504 at *4 (Del. Ch. Oct. 17, 2000); Hirt v. U.S. Timberlands Service, 2002 WL 155842 (Del. Ch. June 9, 2002). These rules were further refined to avoid a purely quantitative test in Wiehl v. Eon. Labs, 2005 WL 696764 at *3 (Del. Ch. May 22, 2005). This changed standard probably incentivized some plaintiffs lawyers to file elsewhere if they were unlikely to be made lead counsel in Delaware.

8. Delaware law has long rejected any vested rights theory, and the preferences and rights of shares of a Delaware corporation can be amended after their issuance, subject to class voting if a particular class is disfavored. Also, under Delaware’s doctrine of “independent legal significance,” what cannot be done by a charter amendment can be done by a merger. See Federal United v. Havender, 11 A.2d 2d 331 (Del. 1940).

9. Although SLUSA preempts securities fraud class actions in state court, §16(d)(1) of the Securities Act of 1933 preserves Delaware “fiduciary breach” class actions and states that plaintiffs may sue in state or federal court. See 15 U.S.C. §77(p)(d). For an overview, Atkinson v. Morgan Asset Management, 658 F.3d 549 (6th Cir. 2011).

10. State courts can hear cases under the Securities Act of 1933, but not under the Securities Exchange Act of 1934. Of course, a settlement in state court of a fiduciary breach claim can include a release terminating the Rule 10b-5 claim. See Matushita Elec. Indus. v. Epstein, 516 U.S. 367 (1996).

11. See Galaviz v. Berg., 763 F. Supp. 2d 1170, 1171 (N.D. Cal. 2011)

12. The Chevron clause begins:

Unless the Corporation consents in writing to the selection of an alternative forum, the Court of Chancery of The State of Delaware shall be the sole and exclusive forum for (i) any derivative action brought on behalf of the Corporation, (ii) any action asserting a breach of fiduciary duty owed by any director, officer or other employee of the Corporation to the Corporation or the Corporation’s stockholders….

See Buchansky v. Armacost, No. 4:12–cv–61597 (N.D. Cal. March 29, 2012), at para. 37.

13. See John Armour, Bernard Black & Brian Cheffins, “Delaware Corporate Litigation and The Fragmentation of The Plaintiff’s Bar,” 2012 Columbia Business Law Review—(forthcoming in 2012).

14. See Clark & Meyer, supra note 6, at 5. Some of the cases in this study presumably involved non-Delaware corporations; thus, this percentage does not represent the percentage of Delaware incorporated firms being sued out of state.

15. See Armour, Black & Cheffins, supra note 13.

16. For statistics on the decline in securities class action litigation, see Jordan Milev, Robert Patton, Svetlana Starykh & John Montgomery, “Recent Trends in Securities Class Action Litigation: 2011 Year-End Review,” (2011).

17. One cannot date this trend precisely, but the first public signal of it may have been a speech given by the redoubtable Ted Mirvis. See “Anywhere But Chancery: Ted Mirvis Sounds an Alarm and Suggests Some Solutions,” 7 M&A J. 17, at 18 (2007). The pattern was, however, even earlier described in In re Cox Communications, supra note 4.

18. See supra note 7.

19. The phrase “Kabuki dance” is borrowed from Vice Chancellor Laster in In re Revlon Shareholders Litig., 990 A.2d 940, 945 (Del. Ch. 2010).

20. See Clark and Meyer, supra note 6, at 2.

21. See In re Revlon Shareholders Litig., 990 A.2d. at 945-46.

22. See Clark and Meyer, supra note 6, at 5.

23. See, e.g., In re S. Peru Copper Shareholders Litig., 30 A.3d 60 (Del. Ch. 2011). This case eventually resulted in a fee award of $285 million, which has been the subject of much attention on the usual blogs and is seen by some as a signal that Delaware is relaxing its fee award standards to attract back plaintiff law firms. This author considers that a premature prediction.

24. See supra note 9.

25. See 15 U.S.C. §77(p)(d)(1).

26. A close reading of §16(d)(1) of the Securities Act of 1933 suggests that this provision (the so-called “Delaware carve-out”) only intends to permit the suit to be brought in the state of incorporation or in federal court. If so, a Delaware corporation could not be sued under this provision in California state court.

27. See supra note 11.

28. 763 F. Supp. 2d at 1175.

29. See M/S Bremen v. Zapata Off-Shore, 407 U.S. 1 (1972).

30. This point of view is expressed by several justices in Stewart Organization v. Ricoh, 487 U.S. 22 (1988.).

31. This line of cases begins with Santa Fe Industries v. Green, 430 U.S. 462 (1977).

32. 337 U.S. 541, 555-56 (1949).

33. 488 A.2d 858 (1985).

34. Of course, Delaware could permit the board to choose the forum after the fact, which would defeat the purpose of the legislation. Still, my judgment is that proxy advisers, like ISS, would be reluctant to support proposed charter and bylaw amendments that gave management an after-the-fact choice of forum.

35. A brief reference is necessary here to Shady Grove Orthopedic Associates v. Allstate Insurance, 130 S.Ct. 1431 (2010), in which the Court refused to enforce a New York statute that precluded class actions to enforce a penalty, because the New York statute was inconsistent with its reading of Rule 23. This case can be distinguished on a variety of grounds (if space were available), but a critical point is that the Scalia opinion only speaks for a plurality of the justices. The majority of the court—Justice John Stevens and the four dissenters—agreed that the federal courts needed to balance state substantive policies with federal procedural rules. In any event, the same inconsistency is not present here.

36. Some recent bylaw provisions would require that the Rule 10b-5 action be transferred to the federal court in Delaware. This seems consistent with the purposes underlying 28 U.S.C. §1404.

37. The enabling character of Delaware law has been stressed in a number of decisions, perhaps most notably in Stroud v. Grace, 606 A.2d 75 (Del. 1992).