The U.S. Supreme Court has held that statistical sampling may be proper in some contexts in its long-awaited decision in Tyson Foods, Inc. v. Bouaphakeo. The case involves  the use of statistical sampling by plaintiffs in support of class certification and determining class-wide liability in wage and hour lawsuits brought under the Fair Labor Standards Act and similar state laws. 

In a 6-2 decision on March 22, 2016, the Court clarified that its ruling in Wal-Mart Stores, Inc. v. Dukes did not categorically prohibit the use of statistical sampling. Tyson represents the second victory for class action plaintiffs this term (the first being Campbell-Ewald Co. v. Gomez, which held that unaccepted Rule 68 offers do not moot class actions), and the outcome was a disappointment to those who hoped the Court would render an opinion making class certification standards more onerous. However, the case was decided on narrow grounds and it should have limited application.

In Tyson, plaintiffs worked in a pork processing plant which required them to wear protective equipment which varied from job to job. In violation of federal law, Tyson failed to keep records of how long it took each employee to don and doff this protective gear each day. Instead, Tyson paid the employees a flat rate based on its estimate of the time required to don and doff each job's protective gear. 

The plaintiffs alleged that Tyson's policy of not paying for the actual time spent donning and doffing protective gear denied them overtime compensation in violation of the FLSA. The plaintiffs moved to certify a class pursuant to Rule 23 for violation of Iowa state law, and to certify a ''collective action'' pursuant to the FLSA on these grounds.

The district court certified both classes over Tyson's objection that resolution of the employees' claims on a class-wide basis was improper due to the variability in time each employee spent donning and doffing job-specific gear. Because Tyson did not keep records, the trial court allowed plaintiffs to introduce testimony of an expert witness who estimated the time spent donning and doffing gear for each job using observation, samplings, and other evidence. The expert concluded that depending on the gear, employees were entitled to 18 to 21.25 minutes of additional pay per day on average. Using these estimates in conjunction with other time records kept by Tyson, a different expert opined that some class members, but not all, worked more than 40 hours per week and were entitled to overtime compensation.

The jury awarded $2.9 million in unpaid wages to plaintiffs. The district court denied Tyson's motion to set aside the verdict on the basis that the classes should not have been certified due to individual variance in donning and doffing time. A divided panel of the Eighth Circuit affirmed.

Agreeing with the Eighth Circuit, the Supreme Court held that use of ''representative evidence'' was an appropriate vehicle for determining class-wide liability on these facts. The Court rejected calls by Tyson and supporting amici to categorically disallow the use of statistical or representative evidence in class actions. Instead, wrote Justice Kennedy, such evidence is sometimes appropriate for determining liability, but its propriety depends on the particular cause of action and facts of the case. According to the majority, the use of a statistical sample was appropriate here in a wage and hour action to ''fill an evidentiary gap created by [Tyson's] failure to keep adequate records'' and because there were ''no alternative means for the employees to establish their hours worked . . . ''

The Court also rejected the argument that class certification was improper under its previous decision in Wal-Mart. Justice Kennedy clarified that Wal-Mart did not ''stand for the broad proposition that a representative sample is an impermissible means of establishing class-wide liability.'' Wal-Mart was distinguishable, according to Justice Kennedy, because the plaintiffs there sought to use statistical sampling to show the existence of a policy to discriminate.  Here, however, Tyson's policy of paying something for time spent donning and doffing gear, but not keeping records, was not in dispute. In the absence of such evidence, even plaintiffs bringing individual actions would need to rely on the expert's estimates to determine if they were entitled to overtime compensation. Thus, a statistical sample was appropriate, because plaintiffs ''worked in the same facility, did similar work, and [were] paid under the same policy . . . the experiences of a subset of employees [could] be probative as to the experiences of all of them.''

Though the Court distinguished Wal-Mart and held that the use of statistical sampling was permissible on these facts, it was also quick to mention that sampling is not always appropriate and would ''depend on the purpose for which the sample is being introduced and on the underlying cause of action.'' The Court declined to issue a broad ruling that statistical sampling was always suitable, stating that ''[t]he fairness and utility of statistical methods in contexts other than those presented here will depend on facts and circumstances particular to those cases.'' 

The Court also rejected Tyson's argument that the jury improperly awarded compensatory damages on a class-wide basis, despite the fact that plaintiffs' own expert admitted that some class members were not entitled to any overtime pay at all. Instead, the Court remanded to the district court to determine how to prevent an allocation of damages to uninjured class members.

Chief Justice Roberts, in a concurring opinion, expressed doubts that the district court could fashion a way to apportion damages between injured and uninjured class members. Given that ''Article III does not give federal courts the power to order relief to any uninjured plaintiff, class action or not,'' he wrote, ''if there is no way to ensure that the jury's damages award goes only to injured class members, that award cannot stand.'' This statement may shed light on the Chief Justice's thinking in another important class action pending before the Court regarding a plaintiff's entitlement to statutory damages in the absence of injury, Spokeo, Inc. v. Robins.

Given the Court's reluctance to broadly condemn or endorse the use of statistical or representative sampling in class actions, it remains to be seen what effect this case will have on class actions in the future.

Ballard Spahr's Consumer Financial Services Group advises on regulatory matters, assists in the design and documentation of credit products, and represents clients in class actions, regulatory enforcement proceedings, and other lawsuits nationwide. Ballard Spahr's Labor and Employment Group has experience in counseling and litigating an array of labor—including FLSA, employment, and ERISA matters in the public, private, profit, and nonprofit sectors and in representing a large variety of industries.

Copyright © 2016 by Ballard Spahr LLP.
(No claim to original U.S. government material.)

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, including electronic, mechanical, photocopying, recording, or otherwise, without prior written permission of the author and publisher.

This alert is a periodic publication of Ballard Spahr LLP and is intended to notify recipients of new developments in the law. It should not be construed as legal advice or legal opinion on any specific facts or circumstances. The contents are intended for general informational purposes only, and you are urged to consult your own attorney concerning your situation and specific legal questions you have.