Tyson Foods v. Bouaphakeo: Not All Representative Evidence is Made Equal

Edward Nasser

Edward Nasser is a student at Harvard Law School.

In Tyson Foods v.  Bouaphakeo, decided on March 22nd, the Supreme Court held favorably for thousands of workers in an opinion that limits the sweeping Wal-Mart Stores v.  Dukes decision of 2011.  As noted recently on this blog, courts have become increasingly hostile to collective action, but this case, along with Campbell-Ewold Company v.  Gomez, stand out against the modern trend.  This post will examine the differences between Wal-Mart and Tyson and explain why the Court was correct in Tyson even given the binding effect of Wal-Mart.

Tyson Recap

The plaintiffs in Tyson, 3,334 workers in the kill, cut and retrim departments of an Iowa meat processing facility, won $2.9 million in damages following a jury trial for violations of the Fair Labor Standards Act and the Iowa Wage Payment Collection Law; their primary claim was that they should have received overtime pay for time spent donning and doffing the protective gear required to carry out their jobs.  Tyson compensated some, but not all workers, for time spent donning and doffing this gear, but did not keep track of how long it took for each worker or department.

The District Court allowed class certification for the Iowa law claim under Rule 23 because the common questions, e.g., whether time spent donning and doffing protective gear should have been compensated under the FLSA, were “susceptible to classwide resolution even if not all of the workers wore the same gear.” Tyson Foods, Inc.  v.  Bouaphakeo, No.  14-1146, 2016 WL 1092414, at *1 (U.S.  Mar.  22, 2016).  The FLSA claim was certified as a “collective action” under 29 U.S.C.  § 216, which permits employees to sue on behalf of “themselves and other employees similarly situated.” § 216 (b).

To recover lost overtime wages, the employees had to show they each worked more than 40 hours a week, but because Tyson failed to keep records of the time workers spent donning and doffing, plaintiffs relied on a study purporting to establish representative proof.  An industrial relations expert analyzed how long various donning and doffing activities took by watching videotape of the procedures.  He then averaged the time taken by individuals in this observed sample to estimate that cut and retrim workers spent 18 minutes a day and kill department workers 21.25 minutes on donning and doffing.  These estimates were added to the timesheets of each employee to determine which class members worked more than 40 hours a week and to assess the overall value of the recovery.  The Eighth Circuit affirmed the judgment and the $2.9 million verdict.

On appeal, the parties agreed the most significant question common to the class was whether time spent donning and doffing protective gear should be compensated under the FLSA.  Tyson argued that individual inquiries into the time each worker spent were paramount to answering this question, so class certification was improper because plaintiffs’ primary method of proving injury wrongly assumed each employee spent the same amount of time donning and doffing the protective gear.  The workers argued that individual inquiries were unnecessary because the average time established by observation could be assumed to extend to all employees.  The Court held that such representative proof from a sample could be used to show predominance of questions of law or fact common to class members over those affecting only individuals under FRCP 23(b)(3).

SCOTUS Distinguishes Wal-Mart

In his majority decision, Justice Kennedy makes clear he does not think this case is all that similar to Wal-Mart, where the Supreme Court overturned certification of a class of 1.5 million current and former female employees who asserted a Title VII case against their retail employer.  Wal-Mart Stores, Inc.  v.  Dukes, 564 U.S.  338, 367 (2011).  That case was decided under the more preliminary requirement of 23(a), which ensures that the named plaintiffs are appropriate representatives of the class whose claims they are litigating and that the class share a common question of fact or law.  Id.  at 349.

Kennedy examines at length the factual reasons why Tyson and Wal-Mart should be distinguished.  There, plaintiffs alleged that local supervisors exercised their discretion over pay and promotion in a discriminatory way against women and sought injunctive and declaratory relief along with an award of backpay.  To show a common question of fact, plaintiffs relied on three forms of proof: statistical evidence showing disparities between men and women, testimony from about 120 female Wal-Mart employees, and the testimony of a sociologist who concluded Wal-Mart was vulnerable to gender discrimination after conducting a social framework analysis of Wal-Mart’s culture and personnel practices.  Id.  at 346.

The Wal-Mart Court rejected each of these methods.  It held that the regression analysis, which showed a statistically significant disparity in the aggregate between men and women at Wal-Mart, was insufficient to infer that discriminatory treatment is typical of the retailer’s employment practices at individual stores.  Id.  at 357.  Furthermore, and more fundamentally, the Court held that the statistical evidence failed to identify and challenge a specific employment practice as is necessary to prevail on a Title VII suit.  It similarly rejected the anecdotal testimony as suffering from the same defects.  Id.  at 358.  Finally, it dismissed the sociological study because its author conceded that he “could not calculate whether 0.5 percent or 95 percent” of the employment decisions were determined in a biased manner—and the question of whether it was closer to 0.5 or 95 was precisely that which the plaintiffs’ theory of commonality depended.

The Court also pointed out the apparent contradiction in plaintiffs’ allegations, noting that the fact each manager had discretion to make pay and promotion choices hedged against there being a uniform employment practice that could provide the necessary common question to certify a class.  Wal-Mart Stores, 564 U.S.  at 355.  In contrast, the plaintiffs in Tyson , “worked in the same facility, did similar work, and were paid under the same policy.” Tyson Foods, Inc., 2016 WL 1092414, at *2.  The representative evidence did not create the commonality of the class, as it did in Wal-Mart, but was used only as a means of establishing the extent of the injury suffered by that already existing class.

The difference boils down to the finding that the Wal-Mart plaintiffs “did not provide significant proof of a common policy of discrimination to which each employee was subject,” and sought to use representative evidence to overcome this flaw.  Id.  Kennedy explained that although the underlying question of whether the sample at issue could be used to establish liability in an individual action by any member of the class was the same, the facts were different enough to warrant the opposite result.

Thomas’ dissent argues that the majority erred in distinguishing from Wal-Mart.  He writes that in the same way that statistical evidence failed to show common discrimination where managers made discretionary decisions in Wal-Mart, so too should statistical evidence fail where workers spend different amounts of time donning and doffing defeat the plaintiffs in Tyson.  This critique overlooks the theories the statistical evidence was meant to illustrate.  The Wal-Mart plaintiffs asserted systematic discrimination disparately impacting women across hundreds of stores.  The Tyson plaintiffs asserted that thousands of workers in the same factory were not being paid overtime for time spent on the same kind of work activity.  While these workers did undisputedly spend different amounts of time donning and doffing, the claim that they all should be compensated for that time is precisely the kind of common question intended for a class action dispute.


The takeaway seems to be that the court will not allow class certification based on a statistical showing of commonality where the individual impact of injury cannot be proven.  Plaintiffs may use representative evidence to establish harm where they already show sufficient commonality to be certified as a class.  On this reading, Tyson is a victory for class-action plaintiffs insofar as it clarifies that Wal-Mart was not the sweeping restriction it might be read as.

Enjoy OnLabor’s fresh takes on the day’s labor news, right in your inbox.