Antitrust class action counsel are in the business of extracting cash from defendants in the form of settlements that are, in effect, a tax on every transaction in the market covered by the case. The bigger the market, the greater the number of the transactions, the bigger the payday for class counsel.
Over the years, plaintiffs' counsel have developed two powerful tools to maximize their profits: the overarching conspiracy allegation and the broad class definition. The first allows plaintiffs to stitch together disparate events – often separated in time and space and involving different actors – into tales of broad market manipulation. This allows them to threaten entire industries with joint and several liability. A broad class definition, in turn, allows plaintiffs to seek damages on behalf of virtually all consumers.
For this strategy to work, however, plaintiffs must get past the pleading stage, at which point the threat of liability starts to become real, and settlement values skyrocket. To date, this hurdle has not been all that hard for plaintiffs. So long as there is some kernel of plausibility to some aspect of their claim, courts have been loath to narrow the scope of the case at the pleading stage. And class certification allegations have been largely immune from attack until discovery has been completed.
But if the recent decision in In re Railway Industry Employee No-Poach Antitrust Litigation, 2019 WL 2542241 (W.D. Pa. 2019), is any indication, the tide may be turning. There, Judge Joy Flowers Conti held that plaintiffs adequately alleged an overarching no-poach conspiracy among the largest railway equipment suppliers, but failed to allege facts that would allow the case to proceed as a class action. Judge Conti, therefore, effectively killed the case. The decision at once shows the trend towards giving overarching conspiracy allegations a light touch at the pleading stage, but also a new boldness in striking class allegations before discovery or a motion for certification.
Plaintiff Successfully Transform Separate Agreements into an Overarching Conspiracy
The three primary defendants – Wabtec, Knorr, and Faiveley (the latter of which had been acquired by Knorr) – collectively employed about 40,000 skilled and unskilled railway workers. In 2018, the Department of Justice entered into consent decrees with the defendants, in which each admitted to separate, bilateral agreements with each other not to solicit or hire each other's employees. Specifically, the DOJ found that, in 2009 Wabtec and Knorr entered the first of the no-poach agreements, followed by an agreement between Knorr and Faiveley in 2011, and an agreement between Wabtec and Faiveley in 2014. Thus, by 2014, each defendant was in a bilateral agreement with the other two.
Those bilateral agreements would normally not be sufficient to make each defendant jointly and severally liable for all the challenged hiring decisions, particularly any harm that may have arisen from an agreement that a defendant was not a party to. So to achieve that end, plaintiffs tacked on an "overarching conspiracy" claim, asserting that each of the three agreements were part of a common, unified scheme. The plaintiffs then alleged the broadest class possible, claiming that each and every employee suffered injury. The table was set to impose maximum settlement pressure on the defendants. Or so plaintiffs thought.
The court began its analysis by resolving a threshold question that a number of courts have touched on but most had sidestepped, namely, whether a no-poach agreement is per se unlawful under the antitrust laws, or whether the more lenient rule-of-reason applies. Backing up the DOJ, the court held that the per se rule applied, and that an agreement among horizontal competitors in the labor market not to compete for employees is a naked restraint of trade.
Having resolved that question, the court next addressed whether each defendant should be jointly and severally liable, a question that turns on whether plaintiffs adequately alleged an overarching conspiracy, or just three separate bilateral agreements. Following courts' normal reluctance to slice and dice the allegations at such an early stage in the case, the court concluded that there were sufficient allegations to support a circumstantial case that, by the time the third agreement was executed, each defendant knew it was acting pursuant to a broader overarching scheme.
Normally, these threshold rulings would have been the end of the pleading phase of the case; the motion to dismiss would be denied and the parties would be sent off into the discovery abyss. But the defendants had one more gambit to play – one that is rarely tried and even more rarely successful. Here, it paid off.
Defendants Knock Out the Class Allegations, Effectively Ending the Case in the District Court
In most antitrust class cases, plaintiffs include in their complaint a conclusory section providing notice of the scope of the alleged class, identifying a number of common issues, and asserting that all the necessary elements for maintaining a class action can be satisfied. But plaintiffs rarely go beyond cursory notice pleading. And so it was here. The Railroad plaintiffs made no effort to allege facts that, if true, would show that class-wide injury could be proven through common evidence. Instead, they banked on the standard practice of addressing the propriety of class certification only after discovery and in connection with their affirmative motion for class certification.
The defendants, however, did not want to wait until after discovery, so they sought to wrest control of the issue earlier by forcing plaintiffs to make their case on the strength of the complaint's allegations. The defendants took two shots. The first failed; the second did not.
Defendants first moved to strike the class allegations under Rule 12(f), which allows a court to "strike from a pleading an insufficient defense or any redundant, immaterial, impertinent, or scandalous matter." But class allegations – no matter how conclusory – are not redundant, immaterial, impertinent, or scandalous. Just the opposite. Class allegations are critical to establishing the right to proceed to discovery on issues relevant to class certification. As such, the court held that Rule 12(f) did not apply.
The defendants last shot was a motion under Rule 23(d)(1)(D), which empowers courts to "require the pleadings be amended to eliminate allegations about representation of absent persons." The provision was not historically viewed as a license to test the adequacy of the complaint at the pleading stage, but rather was designed to conform the pleadings to any decision on class certification in combination with, as the Advisory Committee notes, "a pretrial order under Rule 16." But the court conscripted this rule along with Rule 23(c)(1)(A) – which provides that "[a]t an early practicable time … the court must determine by order whether to certify the action as a class action" – to view its authority more expansively. Read together, the court construed Rules 23(c)(1)(A) and 23(d)(1)(D) to authorize striking class allegations prior to discovery and prior to any motion for class certification.
Rule 23, however, does not offer any standard for evaluating class allegations, and modern case law is clear that class certification must be decided under a "preponderance of the evidence" standard after conducting a "rigorous analysis." The court, however, took a different approach, effectively imparting Twombly's plausibility standard on class action allegations. In so doing, the court imposed on plaintiffs the "burden to set forth factual allegations to advance a prima facie showing" of entitlement to class action treatment "or that at least it is likely that discovery will reveal evidence" to support class action treatment. The court also held that, in making this determination, it was "constrained by the … the allegations" in the complaint, thus, effectively incorporating Rule 12(b)(6)'s non-evidentiary "failure to state a claim" standard on the class action component of the case.
Having concluded that plaintiffs are not automatically entitled to class discovery simply by adequately pleading a conspiracy, the court questioned plaintiffs' broad class definition, which included all defendants' employees, regardless of skill level or job function. Defendants argued that antitrust harm to "the expansive class" is not susceptible to common proof – a requirement under Rule 23 – because the case "will require the Court to consider each employee's contract, salary history, professional qualifications, geographic location and willingness to relocate, and fungibility within labor markets." Defendants' use of the plural – labor markets – was intentional. Accountants and railway engineers, for example, surely compete in different labor markets and, while defendants might control the latter, they certainly lack market power over the former.
The court appeared to be persuaded, noting in a footnote that "plaintiffs' task is more difficult because they seek to represent an expansive class of all defendants' employees, which includes employees highly skilled in the railway equipment supply industry and employees without skills specific to that industry." But the court did not need to reach the question of whether such a broad class definition could be sustained because it found plaintiffs failed to adequately plead it.
Citing Third Circuit law, the court "rejected the notion that antitrust injury in an employee boycott or no hire context can never be proven by common evidence," and instead noted that class certification may be appropriate if there is "evidence showing that compensation of class members was correlated over time." As the court explained, a plaintiff must therefore allege that each defendant set compensation formulaically and that "the compensation structures of the defendants … were so rigid that the compensations of all class members were tethered together," such that a showing of impact on one would mean impact on all.
This notion of a rigid price or wage structure has a long lineage in antitrust cases, dating all the way back to the 1970s, when the existence of such a price structure was first used by plaintiffs as a sword to show common impact. In recent years, it has been used by defendants as a shield to defeat class certification where prices or wages are individually negotiated.
In Railway, the court not only re-affirmed the requirement to show a rigid price structure for class certification, it turned it into a necessary pleading element.
And so, just as the substantive "price structure" requirement has morphed from a plaintiff- to defense-friendly tool, so too has the procedural pendulum swung from favoring the plaintiff to favoring the defense. In days bygone (really, just a few years ago), plaintiffs could get a class certified under what was in effect a notice pleading standard, without much, if any, evidence to support it. Courts had to decide class certification as soon as practicable (which often meant before substantial discovery); they needed to resolve all doubts in favor of class certification; they were barred from considering affirmative evidence presented by defense experts, and they were prohibited from resolving any factual disputes. A series of cases, including the Third Circuit's seminal Hydrogen Peroxide decision, rejected this approach, holding that Rule 23 requires a rigorous analysis of all relevant facts under a preponderance of the evidence standard. This effectively took consideration of class issues beyond the pleading stage.
But while the rigorous evidentiary approach was far better for defendants than the near automatic class certification standards that preceded it, it came with its own set of problems. Defendants now have to go through what is inevitably a long and expensive discovery process before reaching the issue of class certification. Railway potentially solves that problem by allowing defendants to defeat class certification unless plaintiffs meet their Rule 23 burden at the pleading stage.
That is where the Railway plaintiffs fell short. The court held that plaintiffs failed to include allegations that established "the compensation structure for all defendants were so rigid that the compensation of all class members, including employees specific to the railway equipment supply industry and employees with [such] skills …, were tethered together." There was no need to consider whether establishing class impact through common proof was a "clear impossibility," a standard advanced by plaintiffs, or whether discovery would likely yield facts sufficient to establish common impact, a standard espoused by defendants. The simple failure to allege a rigid compensation structure was enough to strike the class allegations from the complaint, sounding the death knell of the case, at least as currently pled.
Implications for the Future
If this decision stands, its implications in future antitrust class actions are profound. As an initial matter, plaintiffs will no longer be able to get away with conclusory class allegations. They will need to include allegations setting forth all the facts that plausibly show that impact will be provable through common evidence. It will also effectively give defendants an opportunity to narrow the scope of the case – at the outset and before any discovery – to classes in which wages (or prices) are formulaically determined via a rigid wage (or price) structure. Markets in which prices are set through individual negotiation would appear to be immune from class certification, or at least subject to attack on a motion to dismiss. And better yet, if either side gets an adverse ruling at the pleading stage, it will be subject to discretionary interlocutory appeal under Rule 23(f), even without the blessing of the District Court.
This decision, therefore, provides a promising new avenue for defendants to combat the crippling threat of a class action at the early stages of a case.