Fee-Shifting and the SEC: Does It Still Believe in Private Enforcement?


Corporate law normally moves at a glacial pace, but sometimes there are periods of rapid change, much of it invisible to the ordinary observer.  2014 may be witnessing such a period of rapid, low-visibility change.  Between May 29 and September 29, 2014, some 24 public companies adopted either bylaws or charter provisions mandating that an “unsuccessful” plaintiff in shareholder litigation (whether in state or federal court or arbitration) must pay the fees and expenses of all defendants.  This list includes high profile examples, such as the recent Alibaba IPO.[1]  Generally, “reporting” companies accomplish this result through a board-adopted bylaw, while IPO companies place this provision in their charter before they go public.  The fee-shifting thereby mandated is considerably tougher than the British “loser pays” rule because many of these provisions deem a plaintiff who is not entirely successful on all claims and requested forms of relief to have been unsuccessful and thus required to reimburse.[2]  Also, unlike the British rule, this approach is one-sided; that is, a defendant who loses does not pay the successful plaintiff’s fees and expenses.  Although 24 companies is not a large number, the trend is accelerating, and it resembles the first trickle of water through a leak in a dam.  Soon the dam breaks, and a cascade descends upon those below. By the end of September, adoption of fee-shifting provisions was occurring on a virtually daily basis. Moreover, many of the most prestigious law firms in the country were helping their clients clear registration statements containing such provisions with the SEC.  This could quickly become part of the standard IPO game plan.

The timing of this sudden burst of new bylaws is not surprising.  They all follow a decision in May 2014—ATP Tour Inc. v. Deutscher Tennis Bund[3]—in which the Delaware Supreme Court held that a very punitive fee-shifting bylaw was “facially valid” and could be enforced against shareholder/plaintiffs who acquired their shares both before and after its adoption—unless the shareholder could show that the bylaw was adopted for an “improper purpose.”[4]  But here the Delaware Supreme Court added that “the intent to deter litigation…is not invariably an improper purpose.”[5]

One cloud on the horizon remains the attitude of Delaware.  Last June, the Corporate Law Section of the Delaware State Bar Association proposed legislation to overturn ATP Tour by prohibiting charter or bylaw provisions from containing any provision that would impose monetary liability on a shareholder.  That legislation was blocked, but is being refashioned.  At hearings before the SEC Investor Advisory Committee last week, Professor Lawrence Hamermesh, who is involved in that process, predicted that something might emerge in early 2015, but offered no clues on the likely shape of new legislation.  Delaware is uniquely conflicted on this issue, because ATP Tour could imply a significant decline in Delaware-based litigation, but such a decline would greatly benefit management and directors of Delaware corporations.  Never before have the interests of the Delaware bar and its clients clashed so directly.  Still, even if Delaware were to act, the issue would still not disappear for three distinct reasons: (1) Delaware might only modestly limit the use of such bylaws, still permitting a substantial chill on securities class actions; (2) Corporations incorporated in other jurisdictions may adopt similar provisions (and the prestige of the Delaware Supreme Court may lead other courts to accept its ruling, even if the Delaware legislature were to reverse or amend it); and (3) Corrective action by Delaware might fuel an interjurisdictional competition, as other, more conservative states (think, Texas) might seek to lure companies to reincorporate there to exploit their tolerance for such provisions.

Still, the even larger question then is what will the SEC do—if anything at all.  To date, it has been standing passively on the sidelines.  Indeed, in the Alibaba IPO, its staff missed the forest for the trees, requiring no disclosure of the impact or scope of Alibaba’s charter provision shifting fees to unsuccessful plaintiffs.

I. Fee-Shifting Provisions in Federal Court

What then is the status of such a board-adopted bylaw in federal court? Here, it is important to note that the ATP Tour case was in fact brought in federal court in Delaware.  The plaintiff lost at trial,[6] and the defendant moved for its costs pursuant to the bylaw.  The District Court denied this motion, effectively ruling that federal law preempts the enforcement of fee-shifting agreements when antitrust claims are involved.[7]  But the Third Circuit reversed the District Court’s order, ruling that before addressing federal preemption issues, the District Court should have first determined whether the bylaw was enforceable under state law.[8]  In fact, the Third Circuit expressed skepticism that the bylaw was enforceable under Delaware law.[9]  On remand, the District Court certified this question of the bylaw’s enforceability under Delaware law to the Delaware Supreme Court.

With that initial question thus resolved, the focus now shifts back to the preemption issue.  In general, federal courts have refused to enforce state law penalties intended to deter frivolous litigation.  In a leading case, Burlington Northern Railroad Company v. Woods,[10] a tort action was removed by the defendant from Alabama state court to federal court, where the defendant lost at trial and then appealed, posting a mandatory bond to stay the judgment.  On affirmance of the decision below, the Eleventh Circuit granted plaintiff’s motion for a mandatory penalty of 10% of the judgment amount, based on an Alabama statute that sought to deter frivolous appeals by imposing an automatic 10% penalty.  The Supreme Court reversed, finding that the Alabama statute had no application to a case in federal court (even when removed to federal court based on diversity jurisdiction).  Following its leading precedent of Hanna v. Plumer,[11] the Supreme Court held that federal procedural rules applied in federal court.  Further, it noted a conflict between the Federal Rules of Civil Procedure (which permit a court in its discretion to impose a penalty for a frivolous appeal) and the Alabama statute (which made the penalty mandatory, whether the appeal was frivolous or not).

If we assume Burlington Northern to govern, the same conflict arguably exists between mandatory fee-shifting under a bylaw and the discretionary sanctions authorized for frivolous litigation by Federal Rule of Civil Procedure 11.  Yet, other decisions have applied state law requirements that amount in substance to fee-shifting in federal court.[12]  In any event, before one concludes that federal law should preempt state law on this issue, one must focus on a critical difference between Burlington Northern and the case of a fee-shifting bylaw.  Fee-shifting under ATP Tour is not based on a state statute, but on a contract among the shareholders.  Indeed, the Delaware statute authorizing broad bylaws is no different from that of any other state jurisdiction.  Rather, under standard “black letter” corporate law, bylaws set forth a contract among the shareholders.  From this perspective, the fee-shifting bylaw is the same as an indemnification provision in which one party by contract agreed to pay the other’s legal expenses (at least under certain circumstances).  In fact, the ATP Tour decision emphasizes that contractual agreements are an exception to the usual American rule on fee-shifting (under which each side bears its own legal expenses) that Delaware normally follows.

Hence, the Supreme Court’s standard position that federal procedural rules apply in federal court (and state procedural rules do not) might be sidestepped here if we view the bylaw as simply a contract among the parties.  But again one cannot stop at this point either—for at least two reasons.  First, federal courts have held that some contracts for indemnification are unenforceable because they conflict with the policies underlying specific federal statutes.  The leading such decision is Globus v. Law Research Service, Inc.,[13] in which the Second Circuit denied indemnification to an underwriter who had knowledge of the misstatement because, it said, such indemnification would be contrary to the policies underlying the federal securities laws.  Second, in Atlantic Marine Construction Co., Inc. v. United States District Court for the Western District of Texas,[14] the Supreme Court held in 2013 that contractual provisions among the parties do not supercede the Federal Rules of Civil Procedure but must be interpreted in a manner consistent with them.  In Atlantic Marine, the parties had agreed to a forum-selection clause providing that all disputes would be litigated in Virginia.  Yet, when a dispute arose, the plaintiff sued in Texas, and the defendant sought to have the case either dismissed or, in the alternative, transferred to federal court in Virginia.  Both the district court and the Fifth Circuit refused to do either, ruling that 28 U.S.C. §1404(a) was the exclusive mechanism for enforcing a forum-selection clause.  Both further concluded that the district court had to undertake a balancing-of-interests analysis.  On appeal, the Supreme Court partially disagreed.  Although it found that Section 1404 (and not the contract, itself) governed, it held that the District Court’s analysis improperly placed the burden on the defendant to prove that the requested transfer was appropriate.  Instead, it said the burden was on the plaintiff, as the party “flouting” its contractual obligations, to show that public-interest factors overwhelmingly disfavored the requested transfer to Virginia:

“Only under extraordinary circumstances unrelated to the convenience of the parties should a §1404(a) motion be denied.”[15]

Because there was no preemption issue in Atlantic Marine, its relevance is limited, but it does have two implications: (1) contracts governing litigation are not necessarily enforced as written but must be interpreted through the prism of the Federal Rules of Civil Procedure; and (2) public policy questions may retain some modest relevance, even when there is no federal statute involved.  On this basis, a forum selection bylaw requiring federal securities class actions to be brought in a preferred federal forum (for example, federal district court in Delaware) is likely enforceable, but will have to be implemented by means of Section 1404.

II. Is One-Sided Fee-Shifting Inconsistent With the Policies Underlying the Federal Securities Laws?

To ask this question is not to answer it.  Indeed, the likely answer is probably: sometimes yes, sometimes no.  The specific provision needs to be considered to determine the extent of the burden it imposes.  Most such provisions will presumably require a losing shareholder who sued the company unsuccessfully to pay all the defendants’ expenses, but many go even further and require the plaintiff to be completely successful on all its claims and obtain most of the relief sought—or  face fee-shifting.[16]  Thus, if a plaintiff sued for $100 million and obtained only $40 million (a notable victory by any realistic standard), it may not have been sufficiently successful to escape fee-shifting.  Most (if not all) such bylaws will not pay the successful plaintiff’s fees or expenses.[17]  Thus, without more, such provisions have two key faults: (1) they are one-sided in that they reimburse successful defendants, but not successful plaintiffs (thus, they are unlike the English Rule which shifts fees both ways evenly); and (2) they require fee-shifting even in cases that were reasonable or even meritorious (but lost on a technical legal defense or were largely, but not entirely, successful).

But there is even more objectionable about many of these provisions.  Some of these provisions are drafted so broadly that they expressly apply to “investigations” as well as to legal actions, and some also purport to require anyone who assists a plaintiff in such litigation to also share liability for fee shifting.  Thus, a shareholder/whistleblower could be arguably held liable for the corporation’s fees and expenses in defending a civil or criminal investigation by regulators—at least if not all the charges raised by the whistleblower were fully confirmed.  In a given case, this could be inconsistent with federal whistleblower protections.  Also, efforts to cover those who assist the shareholder plaintiff might even apply to expert witnesses and attorneys who assist the litigation (at least if they own shares).

Since at least J. I. Case Co. v. Borak in 1964,[18] federal decisions and the SEC have asserted that private enforcement of law is a “necessary supplement” to public enforcement by the SEC and the Department of Justice.[19]  Although the Supreme Court’s attitude may be more equivocal today on this point (and clearly it will no longer imply a federal private cause of action, absent clear legislative direction), the Court has still shown itself unwilling to dismantle Rule 10b-5 class actions.[20]  The SEC has not formally retreated from its support for private enforcement, but it is currently on the sidelines, and this issue will put its resolve to the test.

The incentives created by an automatic “loser pays” rule seem particularly perverse.  A reckless or incompetent attorney who files a half-baked complaint will face only a modest to moderate sanction when the action is dismissed on a motion to dismiss (because the defendant will not yet have incurred substantial legal expenses).  But if the action is more meritorious and survives the motion to dismiss, it will proceed into the discovery stage.  Now, the expenses really begin to mount (and can easily exceed several million dollars a year).  The harder and longer the plaintiff’s attorney works to prepare the case, the greater the potential sanction he faces.  As a result, the incentive effect here is to encourage early (and probably premature) settlement before the facts are really developed.  If the plaintiff’s attorney loses (at trial, summary judgment, or whenever), the plaintiff’s attorney has a strong incentive to negotiate a deal under which the attorney waives the right to appeal in return for defendants’ waiver of fee-shifting.  As a result, little appellate law may be made.

These outcomes seem inconsistent with Congress’s attitude towards fee-shifting in securities class actions.  Preemption seems especially justified here because Congress struck a special balance in this area.  That balance recognizes that fee-shifting against the losing side may sometimes be appropriate, but it is conditioned on judicial oversight and applies to both sides.  Under Section 21D(c)(3) of the Securities Exchange Act of 1934 (which provision was added by the Private Securities Litigation Reform Act of 1995), a presumption in favor of fee-shifting is created if any motion or pleading fails to comply with Rule 11(b) of the Federal Rules of Civil Procedure.  This two-sided approach is equally punitive to both sides in making full fee-shifting (rather than a lesser financial sanction) the presumptive penalty, but it requires the court to find a violation of Rule 11.  Because this approach is tougher and more punitive than the normal approach under Rule 11 of the Federal Rules (which would typically involve lesser financial sanctions), it represents a carefully balanced federal policy, but one that is in sharp conflict with automatic and one-sided fee-shifting without any role for judicial discretion.  Effectively, board-adopted bylaws can turn a Congressionally-mandated system of two-sided fee shifting that is dependent on judicial discretion into an automatic system of one-way fee-shifting.

What standard for preemption does the caselaw suggest?  On a number of occasions, federal courts have heard challenges by takeover bidders asserting that state anti-takeover provisions were so preclusive as to be preempted by the Williams Act.  Most of these cases have converged on a test known as the “meaningful opportunity for success” standard.[21]  Under this test, if a takeover is wholly precluded by the state statute, the statute is preempted.  Correspondingly, under an analogous test applicable to securities class actions, it would be possible to impose some penalty on an unsuccessful litigant, but not one so punitive as to be preclusive.

III. What Should the SEC Do?

The SEC could take a number of steps, all consistent with past practice.  First, the SEC could assert the case for preemption selectively as an amicus curiae in cases where no violation of Rule 11 seemed present.  This should require some careful case analysis by the SEC and should not be an automatic response.  Alternatively, the SEC might assert that automatic fee-shifting is always in violation of the Securities Exchange Act, unless it is predicated on a judicial finding that Rule 11 was violated.  This would be a riskier approach, and it might force the SEC to defend a less-than-attractive plaintiff’s attorney.

Second, the SEC has in a closely related area refused to accelerate registration statements where the company’s certificate of incorporation or bylaws contained a mandatory arbitration clause.  This threat seems to have been effective, and companies that have considered challenging the SEC have ultimately backed down.[22]  Yet, the SEC has not held up registration statements with fee-shifting provisions.  This is inconsistent.  Functionally, the two cases are equivalent, because both provisions effectively bar private enforcement.

Third, the SEC could require registrants to state in their registration statements that they understand that the SEC believes that the federal securities laws are inconsistent with fee-shifting bylaws.  Such a statement is already specified in Forms S-1 and S-3 with respect to indemnification provisions.[23]  This at least imposes an embarrassment cost on the issuer and alerts courts to the SEC’s views without the need for an SEC amicus position.

Fourth, the SEC could focus disclosure on such provisions, thereby raising the “embarrassment cost” to the issuer.  In the case of IPOs that involve such provisions, the SEC could require these terms to be disclosed up front as a major “risk factor.”  In contrast, in the Alibaba IPO, no disclosure focused on the impact of its fee-shifting charter provision.  Although the SEC’s staff appears to have missed the forest for the trees here, it must be stressed that enhanced disclosure alone is not an adequate remedy.  Corporations will still use such provisions, even if they modestly impact the IPO price.  Investors cannot adequately price the impact of a provision denying them the ability to enforce their legal rights because they do not know how likely it is that the corporate insiders will breach their duties.

Finally, the SEC is uniquely positioned to gather relevant data.  In assessing the impact of fee-shifting bylaws, it would be useful to know what the average costs are that defendant firms incur on such litigation and that they would seek to shift.  The impact on the typical plaintiff’s firm could also be evaluated empirically.  Similarly, the SEC could assess whether insurance could alleviate this problem (if it were available) and at what cost.  If Delaware were to impose partial curbs (permitting only some limited fee shifting), the SEC could assess the likely empirical impact of such a modified fee-shifting bylaw.


The impact of fee-shifting provisions could be decisive on the future of private enforcement of the securities laws.  The defendant’s expenses in a securities class action can easily exceed $10 million, and this amount would bankrupt many smaller plaintiff’s law firms.  It is questionable (and certainly unresolved) whether plaintiff’s law firms could obtain liability insurance to cover these amounts.  Even if a bold plaintiff’s law firm did sue, it would likely have to agree to indemnify the class representative from fee-shifting, and some class representatives might decline the position, fearing that the plaintiff’s firm could not fully protect them.

As the case proceeded, the defendant’s expenses will progressively mount, increasing the potential penalty.  This will predictably force cheaper settlements, thereby injuring the class.  If fee-shifting bylaws are upheld, defendant issuers should logically regard them as a riskless move that has little downside.  Probably, proxy advisors would object to such board-adopted bylaws,[24] but this is not the kind of board action that could easily fuel a proxy contest or be easily overturned by a shareholder vote.[25]  As a result, such provisions, unless challenged by the SEC, will predictably become prevalent.

The full impact of bylaw and charter provisions are potentially even more far-reaching.  Potentially, not only litigation, but proxy contests could be deterred.  If a bylaw can require shareholder plaintiffs to reimburse the other side, it is arguable that it could also require proxy insurgents to do so, unless they were “completely” successful.

Bottom Line: For the short-term, the ball is still in Delaware’s court while its legislature considers possible curbs.  That process will not likely be resolved until 2015.  Although the SEC need not oppose all fee-shifting provisions adopted through board or shareholder action, it must be prepared to take on open-ended and more sweeping provisions—or concede the decline of private enforcement.  At present, the SEC seems to be ducking this issue.  At last week’s hearings, SEC staffers observed that although mandatory arbitration provisions in corporate charters would not be accelerated, they were not yet certain that fee-shifting provisions were equally preclusive.  Continued irresolution will only further injure the SEC’s already damaged reputation.

Final Thought:  If Delaware does act to restrain fee-shifting through bylaws, the potential for a “race to the bottom” arises.  Other states of a more conservative bent (consider, for example, Texas) might accept or even endorse fee-shifting provisions.  At this point, some corporate lawyers will predictably advise their clients to reincorporate in Texas, and many IPO issuers might prefer to incorporate in Texas initially.  Even if small changes in corporate law will not produce a migration into or away from Delaware, the permissibility of automatic fee-shifting is a major difference that will fuel interjurisdictional competition because it protects corporate managers and directors from potential personal liability.  In this light, an SEC announcement that it will challenge onerous fee-shifting provisions would chill interstate charter competition (and might even be welcomed in Delaware).  Delaware alone cannot solve this problem.


[1] For the Alibaba example, see Section 173 (“Claims Against the Corporation”) of the Amended and Restated Memorandum and Articles of Association of Alibaba Group Holding Limited, which is incorporated in the Cayman Islands (available at http://www.sec.gov/Archives/edgar/data/1577552/000119312514333674/d709111dex32.htm).

[2] For an example of how sweeping some of the bylaws can be, see Article Sixteenth of the Second Amended and Restated Certificate of Incorporation of Smart & Final Stores, Inc., which may have been the first company to adopt this approach in its IPO (available at http://www.sec.gov/Archives/edgar/data/1563407/000104746914007436/a2221270zex-3_1.htm). The company is a West Coast food and supply chain, which is incorporated in Delaware. Article Sixteenth reads as follows:

Notwithstanding anything in this Certificate of Incorporation to the contrary, to the fullest extent permitted by law, in the event that (i) any current or prior stockholder or anyone on their behalf (a “Claiming Party”) initiates any action, suit or proceeding, whether civil, criminal, administrative, or investigative, or asserts any claim or counterclaim (each, a “Claim”) or joins, offers substantial assistance to or has a direct financial interest in any Claim against the Corporation (including any Claim purportedly filed on behalf of any other stockholder) and/or any director, officer, employee or affiliate thereof (each, a “Company Party”), and (ii) the Claiming Party (or the third party that received substantial assistance from the Claiming Party or in whose Claim the Claiming Party had a direct financial interest) does not obtain a judgment on the merits that substantially achieves, in substance and amount, the full remedy sought, then each Claiming Party shall be obligated jointly and severally to reimburse the applicable Company Party for all fees, costs, and expenses of every kind and description (including, but not limited to, all reasonable attorneys’ fees and other litigation expenses) that the applicable Company Party may incur in connection with such Claim. If any provision (or any part thereof) of this Article SIXTEENTH shall be held to be invalid, illegal or unenforceable facially or as applied to any circumstance for any reason whatsoever: (1) the validity, legality and enforceability of such provision (or part thereof) in any other circumstance and of the remaining provisions of this Article SIXTEENTH (including, without limitation, each portion of any subsection of this Article SIXTEENTH containing any such provision (or part thereof) held to be invalid, illegal or unenforceable that is not itself held to be invalid, illegal or unenforceable) shall not in any way be affected or impaired thereby, and (2) to the fullest extent permitted by law, the provisions of this Article SIXTEENTH (including, without limitation, each such portion containing any such provisions (or part thereof) held to be invalid, illegal or unenforceable) shall be construed for the benefit of the Corporation to the fullest extent permitted by law so as to (a) give effect to the intent manifested by the provision (or part thereof) held invalid, illegal, or unenforceable, and (b) permit the Corporation to protect its directors, officers, employees and agents from personal liability in respect of their good faith service. Any person or entity purchasing or otherwise acquiring any interest in the shares of capital stock of the Corporation shall be deemed to have notice of and consented to the provisions of this Article SIXTEENTH.

Obviously, this provision goes beyond a “loser pays” rule and is in effect “a-less-than-100%-successful-plaintiff-pays” rule.

[3] 91 A. 3d 554 (2014). This decision is fully consistent with other recent Delaware decisions upholding board-adopted bylaws containing forum selection clauses requiring intracorporate litigation to be brought in Delaware. See Boilermakers Local 154 Retirement Fund v. Chevron Corp., 73 A. 3d 934 (2013) and City of Providence v. First Citizens Bankshares, Inc., 2014 Del. Ch. LEXIS 168 (Del. Ch. Sept. 8, 2014). State court decisions in Louisiana, New York, Illinois and elsewhere have recently upheld and enforced Delaware forum selection clauses.

[4] This “improper purpose” requirement dates back to Schnell v. Chris-Craft Industries, 285 A. 2d 437 (Del. 1971), which essentially holds that powers legitimately possessed may not be used for an inequitable purpose.

[5] 91 A.3d at 560.

[6] For the treatment of the substantive claims, see Deutscher Tennis Bund v. ATP Tour, Inc., 610 F. 3d 820 (3d Cir. 2010).

[7] See Deutscher Tennis Bund v. ATP Tour, Inc., 2009 U.S. Dist. LEXIS 97851, 2009 WL 3367041 at *4 (D. Del. Oct. 19, 2009).

[8] See Deutscher Tennis Bund v. ATP Tour, Inc., 480 Fed. Appx. 124, 126 (3d Cir. 2012).

[9] Id. at 127 (“Indeed, we have doubts that Delaware courts would conclude that Article 23.3 imposes a legally enforceable burden on Deutscher and Qatar.”).  Apparently, it guessed wrong.

[10] 480 U.S. 1 (1987).

[11] 380 U.S. 460 (1965).

[12] The case most favorable to enforcing a fee-shifting bylaw is probably Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541 (1949), which upheld the application of a state mandatory security-for-expense bond requirement in federal court. The Supreme Court concluded that this statute was not truly procedural because its real intent was to deter frivolous litigation, not regulate procedure. Such an argument could also be made for fee-shifting bylaws, which similarly levy costs against the losing plaintiff. But Cohen was enforcing a substantive state policy intended to regulate corporate governance (which is traditionally left to state law). In this sense, Cohen is easily distinguishable. First, Delaware has no substantive policy favoring fee-shifting (but rather generally follows the American Rule and leaves departures from that rule to private ordering at present). Second, while the security-for-expense bond in Cohen applied only to derivative actions (which are principally brought in state court), the fee-shifting bylaw here at issue will apply to federal securities class actions (which cannot be brought in state court, at least in the case of Rule 10b-5, because federal courts have exclusive subject matter jurisdiction). Hence, it is arguable that Delaware lacks any legitimate interest in regulating actions that can exclusively be filed in federal court.

[13] 418 F. 2d 1276 (2d Cir. 1969). Globus does not stand alone and has been widely followed. See, e.g., Heizer Corp. v. Ross, 601 F. 2d 330, 334 (7th Cir. 1979); DeHaas v. Empire Petroleum Co., 286 F. Supp. 809, 815 (D. Colo. 1968), aff’d in relevant part, 435 F. 2d 1223 (10th Cir. 1970).

[14] 134 S. Ct. 568 (2013).

[15] 134 S. Ct. at 581.

[16] For example, the actual bylaw in the ATP Tour case was even more sweeping (and was still upheld by the Delaware Supreme Court) in that it required the plaintiff to reimburse the other side’s expenses if it did “not obtain a judgment on the merits that substantially achieves, in substance and in amount, the full remedy sought.” 91 A. 3d at 555. Wow! Plaintiffs rarely win on everything and so could achieve a 95% victory and still face fee-shifting. See also the bylaw set forth supra in Note 2.

[17] The successful plaintiff’s attorney can seek a court-awarded fee from the class’s recovery, but under the American rule defendants are not usually liable for the plaintiff’s attorney fees, even if the facts were overwhelmingly in the plaintiff’s favor. Thus, the plaintiff’s fees comes from its successful clients’ recovery, not from the defendants.

[18] 377 U.S. 426 (1964).

[19] Id at 432.

[20] See Halliburton Co. v. Erica P. John Fund, Inc., 134 S. Ct. 2398 (2014) (upholding “fraud on the market” doctrine).

[21] See City Capital Associates v. Interco, Inc., 696 F. Supp. 1551 (D. Del. 1988); West Point-Pepperell v. Farley, Inc., 711 F. Supp 1996, 1102 (N.D. Ga. 1989); RP Acquisition Corp. v. Staley Cont’l Inc., 686 F. Supp. 476, 482 (D. Del. 1988); BNS Inc. v. Koppers Co., 683 F. Supp. 458, 469 (D. Del. 1988).

[22] In 2012, the Carlyle Group inserted a mandatory arbitration provision into its governance documents as it was preparing its IPO, but quickly dropped it under regulatory and investor pressure.  See Note, In a Bind: Mandatory Arbitration Clauses in the Corporate Derivative Context, 28 Ohio St. J. on Disp. Resol. 737, 745 (2013).

[23] See Form S-3 at Item 13 (“Disclosure for Securities Act Liabilities”). This Item requires the registrant to provide the information required by Item 510 of at Regulation S-K (17 C.F.R. 229.510), which requires the registrant  (if indemnification is authorized) to state that it has “been informed that in the opinion of the Securities and Exchange Commission such indemnification is against public policy as expressed by the Act and is therefore unenforceable.”

[24] Indeed, Institutional Shareholder Services has also objected to board-adopted forum selection bylaws and supported an effort to repeal Chevron’s bylaw to this effect.

[25] Activists could seek passage of an advisory shareholder vote recommending the repeal of the bylaw, but such a vote (presumably pursuant to SEC Rule 14a-8) will likely have to be precatory only. See CA, Inc. v. AFSCME Employees Pension Plan, 953 A. 2d 227 (Del. 2008) (holding that a bylaw mandating reimbursement of expenses in proxy contests must be subject to a “fiduciary out,” meaning that the board can refuse to accept the shareholders’ position). In administering Rule 14a-8, the SEC will generally permit issuers to exclude shareholder resolutions that are not simply precatory. With respect to the possibility of a proxy contest, the passage of a fee-shifting bylaw will hardly reduce the corporation’s value and thus will not make it a candidate for hedge fund activism (which usually is aimed at increasing share value over the short-term). No one else is likely to undertake the high costs of a proxy contest.

John C. Coffee, Jr. is the Adolf A. Berle Professor of Law at Columbia University and Director of its Center on Corporate Governance.