How Not to Write a Class Action “Reform” Bill

It was predictable. Given a solidly Republican Congress and a Republican president, sooner or later, an effort would be made in the Trump administration to curb class actions. Not surprisingly, it has come sooner, with the “Fairness in Class Action Litigation Act of 2017” (H.R. 985). A motley assortment of procedural “reforms”—some good, many bad, and most overbroad—H.R. 985 has been introduced by Representative Bob Goodlatte (R-Va.), chairman of the House Judiciary Committee. Much of this bill is a reincarnation of a similar class action “reform” bill that passed the House in 2015, but died in the Senate (possibly because it was clear that President Obama would veto it). Still, some of it is new, including at least one extraordinary provision that raises constitutional issues. This short overview will examine only the most salient provisions of H.R. 985, explaining why some seem constitutionally suspect and others are clumsily drafted. In fairness, however, a few good and sensible ideas are also set forth in H.R. 985.

The provision that will attract the most controversy in H.R. 985 is proposed Section 1717(b)(2) (“Prohibition of Conflicts”), which bars any federal court from certifying “any class action in which any proposed class representative or named plaintiff is a relative of, is a present or former employee of, is a present or former client of (other than with respect to the class action), or has any contractual relationship with (other than with respect to the class action) class counsel” (emphasis added). Simply put, this means that class counsel cannot have formerly represented the class representative in any matter. Consider next the standard pattern in securities class actions in which a public pension fund acts as a lead plaintiff and retains a major plaintiff’s law firm that it has used before (presumably because it was satisfied with its prior efforts). For example, a CalPers may retain Robbins, Geller, Rudman & Dowd, LLP, and the New York State Pension Fund often retains Bernstein, Litowitz, Berger and Grossmann LLP. Under §1717(b)(2), because the client may not use a firm that it has ever previously retained (apparently for any purpose), the result is to impose a legal regime of “one night stands” on clients and their counsel. Who benefits from this? The only plausible answer is: defendants!

Can this provision be evaded? Conceivably, large plaintiff’s firms could sub-divide themselves or spin off smaller affiliates, which would act as class counsel, but delegate much of the actual litigation work to their parent or affiliate. Or, the lead plaintiff could opt out and sue individually, using its preferred counsel. These attempts will produce much litigation and uncertainty. Thus, the provision, as it stands, seems either a death sentence for the large plaintiff’s firm or the end of large public pension funds serving as lead plaintiff.

Nothing about Section 1717(b)(2) seems carefully drafted. If a law firm represented a client in a small house closing 20 years ago, it could seemingly not today represent such an individual, serving as class representative in a class action. A legal secretary who worked a decade earlier for a law firm could similarly not retain it in a class action.

Can Congress do this? The Sixth Amendment to the U.S. Constitution guarantees the citizen’s right to counsel, but that amendment is largely limited to criminal proceedings. Although there is generally no right to appointed counsel in civil matters, the Due Process Clause of the Fifth Amendment has been read to preclude the state in a civil case from interfering unreasonably with the citizen’s choice of hired counsel. In an illustrative case, Potashnick v. Port City Const. Co.,[1] the Fifth Circuit noted that: “A civil litigant’s right to retain counsel is rooted in fifth amendment notions of due process.”[2] Earlier, in Powell v. Alabama,[3] the Supreme Court wrote broadly: “If in any case, civil or criminal, a state or federal court were arbitrarily to refuse to hear a party by counsel, employed by and appearing for him, it reasonably may not be doubted that such a refusal would be a denial of a hearing, and therefore of due process in the constitutional sense.”[4] U.S. courts of appeals in several circuits have repeatedly held that the Due Process Clause guarantees not simply the client’s right to retain counsel in a civil case, but “the right to choose the lawyer who will provide that representation.”[5]

The logic of extending the right to counsel of the citizen’s choice to civil proceedings is strong. Scholars have pointed out that the Sixth Amendment was limited to criminal cases, because that right was not then protected under English law at the time the Constitution was drafted.[6] At that time, English law did not recognize that the criminal defendant had a right to counsel in the case of felonies (but did recognize the right to use counsel of one’s choice in civil cases). Hence, the Sixth Amendment was drafted to remedy the perceived shortcomings in English law, and the draftsmen saw no need to address civil proceedings, where such a right was already assumed.

To be sure, the right to counsel in civil proceedings can be modified and regulated—but not arbitrarily. Before the client’s choice of counsel can be rejected, the reason has to be “compelling.”[7] This makes sense. Sophisticated persons want to hire professionals—lawyers, doctors, whomever—that they know. Prior association breeds trust, and trust can be efficient. Clients would prefer, if possible, to avoid “blind dates” with new counsel. Nor does the fact that a given counsel had previously been used give rise to any conflict of interest. Indeed, if there is any conflict between the interests of the class and those of class counsel, Rule 23(a)(4) of the Federal Rules of Civil Procedure would already preclude use of a conflicted counsel because it requires as a condition of class certification that class counsel “will fairly and adequately protect the interests of the class.”[8] In short, courts are already well armed by Rule 23 to protect the class from truly conflicted counsel. But Rule 23 does not protect the class from fictitious conflicts, whereas H.R. 985 does this with breathtaking overkill.

The only argument that I anticipate for upholding the disqualification of all major plaintiff’s law firms from (at least) most securities class actions is that institutional investors can still use their preferred counsel if they sue on an individual basis. Indeed, many institutions today do opt out as a matter of course in securities class actions. But the Private Securities Litigation Reform Act deemed the largest investor to be the optimal (and presumptive) lead plaintiff. Also, other class members may also want the large plaintiff’s firm to serve as their class counsel, and they are deprived of their choice of counsel if the large pension fund must opt out. In any event, the justification for interfering with the client’s choice has to be “compelling” under the case law, and that standard is not met when there is no legitimate justification.

Section 1717(b)(2) is overbroad in other respects as well, because it precludes an attorney from serving as class counsel if the attorney “has any contractual relationship” with the class representative or if the latter was ever an employee. Seemingly, the owner of a cleaning service that washes the windows at the attorney’s office or a gardening service that mows his lawn at home would be similarly disqualified, even though the relationship was too immaterial to disqualify the attorney under Rule 23(a)(4).

Other provisions of H.R. 985 are equally objectionable. Section 1718(a) purports to codify the ascertainability doctrine, but goes way beyond that goal by precluding the certification of a class action unless, at the time of certification, class counsel can demonstrate “a reliable and administratively feasible mechanism…for distributing to a substantial majority of class members any monetary relief secured for the class.” At the certification stage, this is generally impossible in consumer class actions involving relatively small, “negative value” claims. In these cases, the claiming rate may be low, and only a minority of the class will eventually seek to file claims once notice of settlement is given. To illustrate, assume a manufacturer sells a defective toaster for $50 through a broad retail and online distribution network, and the class sues for rescission of the sales. The class may consist of over 50,000 purchasers, but it is hard to identify them. Once a settlement is reached or a judgment entered, claims administrators can publicize the result and get much of the class to file claims. But no one in the class will respond until there is monetary relief to be claimed. Under proposed Section 1718, this class cannot be certified, even though liability is clear, because the identities and location of the class members cannot be established at the class certification stage. It makes no sense to deny certification simply because a “substantial majority” of the class cannot be identified at the class certification stage (when it is both costly and infeasible to reach them). Even if only a minority of the class later files claims (probably because the time and effort required to file a claim dissuades them), the fact that a majority remains passive is not a legitimate reason to deny recovery to the minority that does file (although the extent of the class’ participation might well be a factor to be considered in determining the attorney’s fee award).

H.R. 985 will also slow the pace of class actions to a crawl. First, proposed Section 1723 permits appeals of orders granting or denying class certification as a matter of right. Today, such interlocutory appeals are discretionary with the appellate court (and are infrequently granted). The burden on appellate courts will be substantial. Second, discovery is halted if defendant makes any of a variety of motions (see proposed Section 1721). Predictably, such motions will be made one after another, in seriation fashion, to delay discovery.

Possibly the most sweeping provision in H.R. 985 deals not with class actions, but with consolidated personal injury actions in federal court, which have largely replaced class actions in the mass tort context. For example, think of the recent Volkswagen settlement for over $16 billion, which involved consolidated cases, not a class action. H.R. 985 would amend existing law to provide that in any consolidated or coordinated personal injury litigation, the claimants must “receive not less than 80 percent of any monetary recovery obtained in that action by settlement, judgment or otherwise.” This would seem to limit contingent fees to no more than 20 percent (well below the prevailing market rate in personal injury litigation.) But the real impact is even more severe. Assume that in a personal injury case, the settlement is for $1 million, but the plaintiff’s attorney spent $200,000 in costs for expert witnesses, discovery, and depositions. That leaves $800,000, which must go to the claimant under this 80 percent provision, meaning that the attorney gets no fee at all and is out of pocket $200,000. Such a provision will simply drive personal injury cases to state court. Was this provision the result of a deliberate intent to starve plaintiff’s attorneys or simply sloppy drafting? It is hard to tell.

In fairness, there are provisions in H.R. 985 that I applaud. Section 1782(b)(2) would limit the attorney’s fee in class actions to a reasonable percentage of the amount actually distributed to the class. This would help preclude reversionary settlements (in which the undistributed recovery reverts to the defendant) and possibly cy pres settlements as well. A similar provision has long been codified in the Private Securities Litigation Reform Act (see Section 21D(a)(6) of the Securities Exchange Act of 1934), and this provision would simply generalize that provision to apply to all class actions. This section also states that in no event can the plaintiff’s attorneys ever receive a fee greater than the class’ actual recovery. Here I would simply say: Bravo! Another sensible provision (Section 1719(b)) would collect data annually on the actual funds distributed to claimants in class actions. Some studies have suggested that claimants file claims for less than 10 percent of the settlement fund in many consumer class actions. More sunlight here would be desirable.

One of the mysteries surrounding H.R. 985 is why its authors chose the strategies that they did. Attempting to disqualify all the major plaintiff’s firms from securities litigation will elicit a predictable constitutional attack, which I think should be successful. A safer (but less covert) strategy would have been to codify a “loser pays” rule for class actions. Unwise as this might be on policy grounds, it is surely constitutional (and most countries have such a rule). Possibly, the draftsmen of H.R. 985 preferred a sneak attack and did not think that the Senate would pass a full-blown “loser pays” rule (hard to tell, frankly).

Because H.R. 985 was introduced by the chairman of the House Judiciary Committee, it has a good chance of passing the House with only minor changes. The Senate is a different story, but much depends on how much resistance it generates. Judicial resistance seems likely in some areas (appellate courts do not want every class certification motion to be appealable). But judges do not vote in the Senate. Much depends then on whether the bar understands H.R. 985’s implications and speaks out.


[1] 609 F. 2d 1101 (5th Cir. 1980)

[2] Id. at 1118.

[3] 287 U.S. 45 (1932)

[4] Id at 69.

[5] See McGuin v. Texas Power & Light Co., 714 F. 2d 1255, 1257, (5th Cir. 1983); Texas Catastrophe Property Inc. Ass’n v. Morales, 975 F. 2d 1178 1181 (5th Cir 1992); Mosley v. St. Louis Sw. Ry., 643 F. 2d 942, 945-46 (5th Cir. 1981), cert denied, 452 U.S. 906 (1981). Outside the Fifth Circuit, see Gray v. New England Tel & Tel Co., 792 F 2d 251, 257 (1st Circ 1986). Several of the foregoing cases have involved fact patterns where the counsel hired by the client would have forced the judge to recuse himself (because of a family relationship). In McGuin, supra, the Fifth Circuit held that the client’s choice of counsel on these facts had to be respected, unless “the sole or primary motive” for the choice was to disqualify the judge.

[6] See Beaney, THE RIGHT TO COUNSEL IN AMERICAN COURTS, 8-9 (1955); Note The Right to Counsel in Civil Litigation, 66 Colum. L. Rev. 1322, 1327 (1966).

[7] See Bottaro v. Hatten Associates, 680 F. 2d 895, 897 (2d Cir. 1982).

[8] In Eubank v. Pella Corp., 753 F. 3d 718 (7th Cir 2014), class counsel used his father-in-law as class representative. Shocked, Judge Posner, writing for the Seventh Circuit, declared this to be a “palpable” conflict and “even scandalous,” relying on Rule 23(a)(4). Id. at 721.

This post comes to us from Professor John C. Coffee, Jr., the Adolf A. Berle Professor of Law at Columbia University Law School and Director of its Center on Corporate Governance. Professor Coffee wishes to thank Ms. Alexandra Kennedy-Breit, an LL.M. student at Columbia Law School, for her valuable research assistance on this post.

1 Comment



Comments are closed.