About two months ago, this columnist was asked to prepare a short report to the SEC’s Investor Advisory Committee on the then still largely unnoticed trend toward bylaw and charter provisions that imposed some form of a “loser pays” rule on plaintiffs in intracorporate litigation. After a quick and dirty investigation, I reported three interesting facts:

First, between May 29, 2014 and September 29, 2014, some 24 companies had adopted such a provision (always applicable only to plaintiffs and always without the matter being put to a shareholder vote). This was obviously a rapid response to the Delaware Supreme Court’s decision in May, 2014 in ATP Tour Inc. v. Deutscher Tennis Bund,[1] which upheld the facial validity of such a bylaw that had been adopted by a Delaware non-stock corporation. Since then, the number has grown further, and some of the better known law firms in the U.S. appear to be adopting this tactic as part of their standard IPO playbook.

Second, although the board-approved bylaw route used in the ATP Tour case has been followed by others, the more common route is to amend the certificate of incorporation of a firm about to undergo an IPO. The greater use of this technique in the IPO context may reflect the fear that mature, “reporting” companies have of alienating their proxy advisors. If Institutional Shareholder Services (“ISS”) were offended by a board-approved bylaw, they might support a hedge fund activist or a shareholder effort to reverse the bylaw through a precatory shareholder vote.[2] In an era of high hedge fund activism, seasoned issuers do not pick fights with proxy advisors.

Third, most of the firms adopting a one-way “loser pays” bylaw or charter provision did not simply require a losing plaintiff to pay the various defendants’ legal expenses; rather they also required a winning plaintiff to pay if it did not achieve a standard of “complete success,” which they defined to mean that a reimbursement obligation arose to repay “all fees, costs and expenses of every kind and description” of all defendants if the plaintiff (or certain related persons) “does not obtain a judgment on the merits that substantially achieves in substance and amount the full remedy sought.”[3] Drafted very aggressively, these provisions applied not only to the actual plaintiff but to any other person “acting on their behalf” who “joins, offers substantial assistance to or has a direct financial interest in any Claim” filed against the corporation or a director, officer, employee or affiliate thereof.[4] Finally, these provisions cover not only lawsuits, but any “proceeding, whether civil, criminal, administrative or investigative” and any assertion of a claim or counterclaim.[5] Thus, a shareholder whistleblower who advises the SEC or some other agency about alleged misconduct, or a former employee/shareholder who brings a claim for wrongful termination, could be held liable (at least if he or she were a present or former shareholder and were not completely successful).

A familiar pattern is evident here. The aggressive transaction planner tries to draft a provision with the maximum in terrorem impact, so as to chill any prospect of litigation of any kind. This draftsman may feel that his side has nothing to lose from his attempt to intimidate. But this could be a serious (if common) mistake. In pushing the envelope this far, the transaction planner may either invite a federal court to discover a conflict between the fee-shifting provision and the policies underlying the federal securities laws (and thus rule that federal law preempts the provision) or encourage the Delaware Chancery Court to find that the provision had an “improper purpose.”[6] The Delaware courts have a long history of holding that powers legitimately possessed by the corporation may still not be used for an improper purpose. The ATP Tour decision alluded to this line of cases, but also observed that the goal of deterring litigation was not “necessarily improper.”[7] Yet, even if attempting to discourage frivolous litigation seems fair enough, the analysis changes when a bylaw or charter provision demands complete success. Then, it seemingly moves beyond a proper purpose and intentionally seeks to discourage meritorious litigation. That may prove too much for Delaware, which (as later discussed) has little desire to ensure the extinction of intracorporate litigation.

As a result, the overly aggressive, high-testosterone transaction planner who writes the preclusive provision that would shift fees even against a largely successful plaintiff may represent our first category of loser. He trips over his own hubris. But who else is likely to lose?

Let’s turn next to Delaware. It needs to strike a difficult balance. On no other occasion in modern memory have the interests of the Delaware Bar been in as frontal a conflict with the interests of its corporate constituents. Corporate litigation is the leading local industry in Delaware; even DuPont does not have the same impact on the state’s gross domestic product. As a result, the Corporate Law Section of the Delaware Bar Association responded to the ATP Tour decision by quickly drafting a statute that would have largely overturned that decision. But corporate lobbyists howled their protests, and Delaware decided to study the issue further. Sometime early in 2015, a new draft is likely to appear, and already the U.S. Chamber of Commerce has taken to the Op/Ed pages of the Wall Street Journal to warn Delaware of the adverse consequences (and the possibility of a corporate migration out of Delaware) if ATP Tour were legislatively overruled.[8] The level of the rhetoric is likely to rise still further.

These critics of corporate litigation have one valid point. M&A litigation has recently exploded, and depending on the study, an arm’s length corporate merger has something like a 95% to 97% chance of attracting a law suit—even though the transaction is an unconflicted one and is at a substantial premium.[9] This litigation is typically settled on a non-pecuniary basis with the shareholders receiving only additional disclosures, while the plaintiff’s attorneys receive generous attorneys’ fees. These suits are generally not brought by the elite plaintiff’s firms (i.e., the larger ones), but by their fly-by-night competitors. As Delaware decisions have noted,[10] these cases are not truly litigated at all (few motions are made, little discovery occurs); rather, a process of “feigned litigation” follows until a settlement is reached with one or more of these plaintiff firms.

The irony in all this is that these cases tend not to be brought in Delaware or in federal court, but in other state courts where they will be often heard by relatively inexperienced state judges. As a result, defendants face at least some uncertainty about how such a judge might rule. Delaware has sought to address this problem by approving forum selection clauses in bylaws and charter provisions.[11] Law is path dependent, and these decisions on forum selection bylaws laid the foundation for the approval of the “loser pays” bylaw in ATP Tour by articulating the theory that the bylaws set forth a contract binding on all shareholders, including those who acquired their stock before the bylaw’s adoption. Perhaps the remedy of forum-selection bylaws would work over time, but corporate issuers want more (and quicker) relief.

The problem for Delaware is that this theory of shareholder consent to bylaws could lead to extreme possibilities. If a shareholder is deemed to consent to a board-approved bylaw, adopted after the time that shareholder acquires his shares, that imposes costs on him for suing the corporation (even successfully), does this same shareholder also consent to a different bylaw, adopted again by the board at some future date, that does any of the following:

1. requires the shareholder to subscribe to additional shares at a price specified by the board or suffer the forfeiture of the shareholder’s shares;
2. hold the shareholder liable for all expenses experienced by the corporation in a proxy contest brought or supported by the shareholder (unless the shareholder is “completely successful”); or
3. require the shareholder to sell some portion of the shareholder’s shares in excess of a specified level back to the corporation (at a price specified by the board).

If Delaware’s new theory of shareholder consent goes this far, limited liability is potentially lost. Nonetheless, the foregoing bylaws (or charter provisions) are like “loser pays” bylaws (and unlike forum selection bylaws) in that they impose financial liability on the shareholder, and thus they are in tension with the age-old assumption of limited liability for shareholders. For this reason, the 2013 legislation proposed by the Delaware Corporate Law Section would have simply said that no bylaw or charter provision could have imposed financial liability on a non-consenting shareholder.

What alternatives are possible for Delaware? Some possibilities seem obvious. One such option would be to provide that a “loser pays” provision could only be adopted by a shareholder vote taken after the company had become a public corporation (i.e., had qualified as a “reporting” company under Section 12 of the Securities Exchange Act). Still, such a provision would transfer considerable power to ISS and the other proxy advisors and might be little used (which may suggest that shareholders do not really desire such provisions).

Another sensible proposal might be to place some form of ceiling on fee-shifting. After all, it should not take much of a sanction to chill the fly-by-night plaintiff’s firms behind most state court “M&A” litigation. If the goal is to deter the pointless and extortionate M&A class action, a modest ceiling could do that (one can reasonably debate whether it should be $500,000 or $1,000,000, and it could be automatically inflation indexed). But in securities class actions, “loser pays” fee-shifting could threaten a $15 million penalty (or more) in some cases (if the action proceeds deep into the discovery stage). Also, securities class actions are not as easily characterized as frivolous or extortionate because the Private Securities Litigation Reform Act (“PSLRA”) has screened out most (but probably not all) such suits. For all these reasons, a “loser pays” rule, if properly limited by a ceiling, might accomplish some good and do relatively little harm. If the rule were so limited, reputable plaintiff’s counsel could obtain litigation insurance, but the fly-by-night plaintiff’s firms that bring (but never actually litigate) M&A cases could probably not afford such insurance.

Still another way to moderate the bite of a “loser pays” rule might be to limit the fee shifting to costs incurred up to the decision on the motion to dismiss. The initial rationale here would be that a case that survives a motion to dismiss is not clearly frivolous (one can debate this, to be sure). In a federal securities class action based on Rule 10b-5, surviving the motion to dismiss means that the complaint did plead (with particularity) facts giving rise to a strong inference of fraud; that is not a low standard. In a derivative action, surviving the motion to dismiss means that the plaintiff has pleaded facts showing that demand on the board should be excused as futile; that is even less easy.

There is an advantage for both sides in cutting off fee-shifting at the motion to dismiss stage. That advantage comes into focus if we focus on the settlement stage and recognize that settlement will remain the most common disposition of a case (particularly federal securities cases), even under a one-way “loser pays” fee-shifting rule. A “loser pays” rule will probably force the plaintiff to settle more cheaply and to sue less often, but, with respect to those cases that are brought, settlement will remain the most likely outcome (and may become even more likely because fewer high-risk cases will be brought).

If a class action (whether based on federal or state law) is settled in federal court, the court must approve the settlement and certify the class under Rule 23 of the Federal Rules of Civil Procedure. Here, if the action is one in which there is a threat of “loser pays” fee shifting, problems arise under Rule 23(a). That Rule requires the court to make four findings as to: (1) numerosity; (2) commonality; (3) typicality; and (4) adequacy of representation. The key issue will involve whether such a lead plaintiff is either “typical” of the class or capable of fair and adequate representation. As to typicality, the absent class members (who have taken no action and are not subject to fee shifting) will logically compare the proposed settlement to the likely outcome at trial. But the lead plaintiff faces a different calculus: it must also compare the outcome under the proposed settlement against the downside if it lost and faced a possible $10 million (or more) in fee shifting. Facing that choice, the lead plaintiff may consider any settlement to be attractive. But for precisely that reason, it is not “typical” of the rest of the class.

Similarly, a lead plaintiff (or the class counsel) facing fee shifting may be incapable of providing “fair and adequate representation” to the class under Rule 23(a)(4). It is incentivized to accept even a poor financial settlement, because the outcome under fee shifting could be far worse to it. This is a problem for both sides, because if settlements cannot be approved, they will not be entered into (and defendants favor settlements and do not wish to be limited to the outcomes of “win” or “lose” only). Objectors will predictably raise this problem in most class actions subject to fee-shifting.

The class representative may be able to sidestep this problem by pointing to the fact that it has been indemnified against fee shifting by class counsel, but that only shifts the problem back to the class counsel, who is highly conflicted. Large plaintiff’s firms may be able to obtain insurance against fee-shifting (but they may still face some deductible or co-insurance provision). But an even bigger problem looms here: the insurance company will inevitably want some control over the litigation and its settlement.

Predictably, insurance companies will insist on controlling securities litigation if they risk liability for one-way fee shifting. Consider, for example, a case in which the defendants win at trial and offer to waive fee-shifting if the plaintiffs agree to forego their right to appeal. For the insurance company, this choice is a “no-brainer,” and they will insist that plaintiffs accept it (or forego their insurance). Similarly, any cheap settlement is attractive to the insurance carrier because it does not share in the upside. As a result, settlement size in securities class actions could decline drastically under a “loser pays” rule. Indeed, this decline should be greater under a one-way loser pays rule than if both sides risked loss, because a two-sided rule would encourage both sides to moderate their demands. Bottom line: for litigators, fee shifting rules make life more complicated, should erode settlement value, should make certification much more difficult, and may produce new litigation under the obligation to indemnify.

Yet, if the liability for fee shifting ceased at the motion to dismiss stage, life would become simpler. Neither class counsel nor the lead plaintiff would be conflicted at the settlement stage (assuming that the case had earlier survived a motion to dismiss). This implies that halting fee shifting at that stage has attractions for both sides.

The impact of one-way fee shifting should also be considered on one last party: the SEC. To date, it has maintained an ostrich-like profile, refusing to indicate what it would do. Arguably, it could assert (possibly as an amicus curiae) that one-way fee-shifting is at least sometimes preempted by the federal securities laws. Alternatively, it could refuse to accelerate IPO registration statements (as it has done for decades with respect to registration statements containing mandatory arbitration provisions). Or, it could do nothing and hide in its bunker.

The SEC’s continuing passivity adds to the growing sense that it is not the agency it once was. Already, it has fumbled badly on this issue, because in the largest IPO in recent years (the Alibaba Group Holding Limited IPO this fall) the SEC totally missed the fact that Alibaba had adopted in its corporate charter a one-way “loser pays” fee-shifting rule.[12] Not a word was said about this in Alibaba’s registration statement (possibly because Alibaba believed it would not be sued if the offering went forward with such a provision). This is equivalent to the Las Vegas bookies missing that one side in the Super Bowl had invoked a rule under which no penalties could be called against it. If the SEC can miss that, it can miss anything, and it is less surprising that the agency missed Bernie Madoff’s Ponzi scheme for thirty years.

For the immediate future, we are waiting on Delaware, both for new legislation and judicial decisions as to when a “loser pays” provision goes too far and thus demonstrates an “improper purpose.” Conceivably, the SEC could react—but the Sphinx may speak first. If it continues its silence, it begins to lose its relevance, and thus it could be the biggest loser.

Endnotes

[1] 91 A. 3d 554 (Del. 2014).

[2] Although advisory shareholder votes are possible, it is doubtful that shareholders can reverse a board-passed bylaw. See CA Inc. v. AFSCME Employees Pension Plan, 953 A. 2d 227 (Del. 2008) (shareholder-passed bylaw dealing with expense reimbursement must be subject to a “fiduciary out,” meaning that board can refuse to accept the shareholders’ position).

[3] This is the language from Article Sixteenth of the Second Amended and Restated Certificate of Incorporation of Smart & Final Stores, Inc., which seems to have been the first company to adopt such a “loser pays” provision in the wake of the ATP Tour decision. For the text of its provision and others, see John C. Coffee, Jr., Fee-Shifting Bylaw and Charter Provisions: Can They Apply in Federal Court?—The Case for Preemption (available at http://ssrn.com/abstract=2508973) (October 10, 2014).

[4] Id.

[5] Id.

[6] See Schnell v. Chris-Craft Industries, 285 A. 2d 437 (Del. 1971).

[7] 91 A. 3d at 560.

[8] See Lisa A. Rickard, “Delaware Flirts With Encouraging Shareholder Lawsuits,” Wall Street Journal, November 15, 2014 at A-11.

[9] For such studies, see Robert M. Daines and Olga Koumrian, Shareholder Litigation Involving Mergers and Acquisitions (Cornerstone Research 2013); Matthew D. Cain and Steven M. Davidoff, A Great Game: The Dynamics of State Competition and Litigation (Jan. 2013) (available at http://ssrn.com/abstract=1984758).

[10] See in particular, In re Revlon Shareholders Litigation, 964 A. 2d 106 (Del. Ch. 2006).

[11] See, e.g., Boilermakers Local 154 Retirement Fund v. Chevron, 73 A. 3d 934 (2013); City of Providence v. First Citizens Bankshares, Inc., 2014 Del. Ch. LEXIS 168 (Del. Ch. Sept. 8, 2014).

[12] The provision is in 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.

John C. Coffee, Jr. is the Adolf A. Berle Professor of Law at Columbia Law School and Director of its Center on Corporate Governance. A version of this article was originally published in the New York Law Journal on November 20, 2014.