Summary
In Espenscheid, a case alleging off the clock work, the Seventh Circuit rejected a proposal to extrapolate the experiences of forty-two "representative" witnesses to a class of 2,341.
Summary of this case from Deluca v. Farmers Ins. Exch.Opinion
09-cv-625-bbc.
May 23, 2011
OPINION and ORDER
This is a collective action under the Fair Labor Standards Act (FLSA), 29 U.S.C. §§ 201- 219, and a class action under the wage and overtime compensation laws of Wisconsin, Minnesota and Pennsylvania. From the beginning of the case, defendants have insisted that it should not go forward as a collective and class action given the varied employment experiences of the plaintiffs and class members. Nonetheless, I certified a collective and class action because I believed that the main issues in the case were whether defendants' nationwide policies and practices caused violations of federal and state wage law. I divided the case into subclasses, concluding the use of subclasses and a bifurcated trial would address the variety of claims and experiences as well as protect the parties' due process rights.
The case is scheduled for a jury trial on June 6, 2011. However, on May 12, 2011, after reviewing plaintiffs' proposed trial plan and expert reports, I became concerned about the manageability of the case. Plaintiffs' plan did not appear to address the subclasses or differences between the plaintiffs' claims and experiences. In an order dated May 12, 2011, I instructed the parties to advise the court whether this case can remain certified as a collective and class action.
After reviewing the parties' responses to the May 12 order, as well as their proposed trial plans, jury instructions, special verdict forms and the record in this case, I conclude that this case is not fit for adjudication as a collective and class action, at least under the plan proposed by plaintiffs. For the following reasons, I am decertifying this case as a collective and class action and dismissing without prejudice the claims of the opt-in plaintiffs and class members, leaving the original plaintiffs to pursue their claims individually.
PROCEDURAL AND FACTUAL BACKGROUND
Initially, plaintiffs brought this action on behalf of four separate classes of installation technicians employed or formerly employed by defendants. The classes were (1) a nationwide opt-in class for violations of the FLSA; (2) a Wisconsin class for violations of Wisconsin wage and overtime compensation laws; (3) a Minnesota class for violations of Minnesota wage and overtime compensation laws; and (4) a Pennsylvania class for violations of Pennsylvania wage and overtime compensation laws.
Utilizing the two-stage certification approach employed by the majority of courts in determining whether to certify a case as a collective action under § 216(b) of the FLSA, I conditionally certified this matter as collective action on June 4, 2010. Dkt. #129. The parties then sent notices to potential opt-in plaintiffs. In response, roughly 1,000 plaintiffs consented to join in this litigation as opt-in plaintiffs. A little more than four months later on October 22, 2010, plaintiffs moved to certify state law classes under Fed.R.Civ.P. 23 for technicians in Wisconsin, Pennsylvania and Minnesota. On November 12, 2010, defendants moved to decertify the nationwide FLSA class.
Defendants based their motion to decertify on the depositions of several opt-in plaintiffs. Defendants' primary argument for decertification was that the opt-in class members were not similarly situated. The class members testified to varying employment experiences and asserted different theories of liability. Plaintiffs countered with declarations of members of the opt-in class and with depositions of corporate officers describing defendants' uniform policies and practices. After reviewing the evidence presented by both parties, I concluded that "[n]ot all technicians performed all of [the alleged] tasks, nor did they perform them everyday or for a consistent period of time," and explained, for instance, that "some technicians testified that they spent a significant amount of time loading and unloading equipment from their vehicles everyday, while others admitted that they rarely unloaded equipment." Dkt. #387, at 15. However, I concluded that plaintiffs' primary challenge was to defendants' uniform policies and practices and that defendants' concerns could be managed through bifurcation and the creation of the following subclasses: (1) plaintiffs who were denied overtime because they recorded a lunch break that they did not take or otherwise underreported hours they worked between their first and last installation or service job of the day; (2) plaintiffs who were denied overtime because they were not compensated for work performed before their first installation or service job of the day or after their last installation or service job of the day; and (3) plaintiffs whose regular and overtime wages for nonproductive work were calculated improperly. I certified similar subclasses for plaintiffs' state law claims, creating a total of twelve subclasses. I concluded that dividing plaintiffs into these classes alleviated defendants' due process concerns, but I noted that "as the case develops further or reaches the damages phase, it may become unmanageable in its current form." Id. at 54. Defendants sought interlocutory appeal of the certification decision, stressing manageability, typicality and due process concerns. Dkt. #423. In addition, defendants sought leave to file an interlocutory appeal of the denial of decertification. Both attempts at interlocutory appeal were denied.
After certification, the parties continued discovery and filed cross-motions for partial summary judgment. With the exception of concluding that some plaintiffs performed potentially compensable work at home, I denied plaintiffs' motion in large part because plaintiffs had not established who in fact performed the "outside the window" tasks, which tasks those individuals performed or the length and frequency of the work. Dkt. #485 at 46-57.
In connection with the summary judgment opinion, I directed the parties to develop a joint trial plan. The parties submitted separate plans. In plaintiffs' trial plan, plaintiffs revealed that they intended to prove their claims at a non-bifurcated trial and without the use of the subclasses developed by the court. Plaintiffs' proposed damages expert report also failed to address the subclasses. Instead, the report relied on the theory that there was an overall "average" technician representing every technician for every subclaim. A few days later, plaintiffs filed motions to eliminate the subclasses and proceed with a non-bifurcated case. In an order dated May 6, 2011, I denied plaintiffs' motions related to the subclasses and bifurcation, expressing concern with plaintiffs' failure to take into consideration the subclasses and different issues present in the case. Dkt. #549. I directed plaintiffs to explain how their expert report accounted for the subclasses and to file a new trial plan.
In an order dated May 12, 2011, I struck plaintiffs' expert report for its failure to comply with Fed.R.Civ.P. 26. Dkt. #595. I noted that although plaintiffs might be able to cure the problems with their expert report, there were larger problems with plaintiffs' plan for proving the case at trial. In particular, plaintiffs had still failed to articulate how they were planning to address the distinct issues represented by each subclass. I concluded that plaintiffs' inability to address these issues indicated that the case had become too difficult to manage collectively and that I was prepared to decertify the case. I directed the parties to respond to my concerns.
PLAINTIFFS' PROPOSED TRIAL PLANS
Plaintiffs have proposed two trial plans. First, they urge the court to eliminate the twelve subclasses and certify an FLSA national class of "all installation or service technicians who were employed by DirectSat and compensated on a piece-rate basis from three years prior to the date of each individual class members' opt-in until the present date." Dkt. #609 at 15. They propose similar classes for the state law claims. Plaintiffs contend that because the "heart of [their] claims is a set of uniform, national policies that apply to every technician" and that "all technicians are claiming compensation for work that was performed with defendants' knowledge," it is not necessary to create subclasses to account for differences in the effect of defendants' policies. Under this trial plan, plaintiffs would present the testimony of several technicians who were allegedly uncompensated for work performed either at home or during the official work window and the jury would decide whether defendants have failed to pay plaintiffs properly for some time they worked, be that work at home, at meetings, driving between jobs or working on customer jobs.
In the alternative, plaintiffs' second proposed trial plan proposes the use of representative proof to establish liability under the twelve subclasses. For the FLSA subclass related to work performed "outside the window," plaintiffs plan to present 42 technicians who will provide "representative testimony that they performed specific work-related tasks outside the window period and that they were not compensated for these tasks." Plts.' Proposed Trial Plan, dkt. #604, at 4. The same 42 technicians will testify as representatives of the second subclass regarding work they performed within the window period for which they were not compensated. Id. at 5. Finally, for the third subclass, the same technicians will testify that there was no agreement with defendants that payment for productive time was intended to include payment for all time worked, including non-productive time. Id. at 6-7. A subset of the 42 technicians will provide testimony for each of the state law classes.
At the liability stage, the jury will be asked to determine whether (1) plaintiffs and class members worked unrecorded time; (2) defendants knew that plaintiffs worked unrecorded hours; (3) defendant failed to compensate plaintiffs for the unrecorded hours; (4) there was an agreement that pay for productive hours was intended to compensate class members for all non-productive time worked; (5) defendants' method for calculating overtime violated the FLSA and state law; (6) defendants failed to maintain accurate time records for employees; and (7) defendants' violation of federal and state wage and hour law was willful.
Plaintiffs propose to calculate damages by determining the average number of nonproductive and productive hours worked by technicians in a workweek. Thus, the jury will be asked to determine the average number of uncompensated productive and non-productive hours in each work week after hearing the testimony of the 42 representative class members. Plaintiffs contend that they do not need an expert to assist the jurors with this determination. Post-trial, the parties would use the jury's averages to determine the amount of damages owed each class member and opt-in plaintiff. Specifically, the parties would add the average uncompensated hours to the weekly pay totals each class members already received during the relevant time period. This method allows for consideration of training weeks during which a class member did not have any piece-rate earnings as well as work weeks in which a class member may have been on light duty, workers' compensation or leave.
OPINION
A. Certification of Collective Actions under the FLSA and Class Actions under Rule 23
Although it is the eve of trial, it is necessary to reexamine the propriety of collective treatment in this case in light of plaintiffs' proposed trial plan, the nature of the claims and the development of the evidence in this case. District courts are "charged with the duty of monitoring [their] class decisions in light of the evidentiary development of the case. The district judge must define, redefine, subclass, and decertify as appropriate in response to the progression of the case from assertion to facts." Richardson v. Byrd, 709 F.2d 1016, 1019 (5th Cir. 1983); see also Baldridge v. SBC Communications, Inc., 404 F.3d 930, 931 (5th Cir. 2005) (certification of a collective action under § 216(b) of FLSA is "subject to revision before the district court addresses the merits"); In re School Asbestos Litigation, 789 F.2d 996, 1011 (3d Cir. 1986) ("When, and if, the district court is convinced that the litigation cannot be managed, decertification is proper."); Eggleston v. Chicago Journeymen Plumbers' Local Union No. 130, 657 F.2d 890, 896 (7th Cir. 1981); Walker v. Bankers Life Casualty Co., 2008 WL 2883614, *9 (N.D. Ill. July 28, 2008) ("If certification is later deemed improvident at any time during the course of litigation, a court may decertify the class."); Blihovde v. St. Croix County, 219 F.R.D. 607, 614 (W.D. Wis. 2003) (district court "can decertify the class at any time before a decision on the merits") (citing Fed.R.Civ.P. 23(c)(1)).
Under the FLSA, one or more employees may pursue an action in a representative capacity only if they are "similarly situated" to the employees they represent. 29 U.S.C. § 216(b). This standard takes into account (1) whether the factual and employment settings of the individual plaintiffs are similar or disparate; (2) whether defendants may assert various defenses that appear to be individual to each plaintiff; and (3) whether fairness and procedural considerations support proceeding as a collective action. Thiessen v. General Electric Capital Corp., 267 F.3d 1095, 1103 (10th Cir. 2001);Russell v. Illinois Bell Telephone Co., 721 F. Supp. 2d 804, 811 (N.D. Ill. 2010); Mielke v. Laidlaw Transit, Inc., 313 F. Supp. 2d 759, 762 (N.D. Ill. 2004). Similarly, class actions under Fed.R.Civ.P. 23 require that "the questions of law or fact common to class members predominate over any questions affecting only individual members," Rule 23(b)(3), and whether a class action is superior to other methods for the adjudication of the proposed class' claims. Mejdrech v.Met-Coil Systems Corp., 319 F.3d 910, 911 (7th Cir. 2003) ("[C]lass action treatment is appropriate and is permitted by Rule 23 when the judicial economy from consolidation of separate claims outweighs any concern with possible inaccuracies from their being lumped together in a single proceeding for decision by a single judge or jury.").
In many cases, collective and class actions provide efficiency and judicial economy in wage and hour disputes because the a court may resolve common factual and legal questions collectively on basis of evidence that is representative of the whole. However, the more dissimilar plaintiffs' job experiences are from one another and the more individualized an employer's defenses are, the less appropriate the matter is for collective treatment. The problems with collective adjudication of the claims in this case have become more pronounced as it approaches trial.
B. The Claims and Record in this Case
In the May 12, 2011 order, dkt. #595, I noted concern with plaintiffs' proposed trial plan and directed the parties to address whether this case had become unmanageable as a collective and class action. In their response, plaintiffs criticize the subclasses as being unworkable and improper and insist that they should be able to proceed with one class. In the alternative, plaintiffs propose that they be allowed to proceed with the subclasses, based on the assumption that all plaintiffs and class members have claims in every subclass.Neither of plaintiffs' proposals is acceptable. Both are merely attempts to work around the subclasses that I created as an alternative to decertifying the case. Thus, although the subclasses may have problems, the solution is not to ignore them. As I noted in the order creating the subclasses, the opt-in plaintiffs and class members have divergent testimony, distinct theories of liability and unique employment environments and experiences. Dkt. #387 at 39 (concluding that "defendants' concerns are manageable through bifurcation and the creation of subclasses); see also dkt. #549 at 3 (concluding that case could proceed as class action "if plaintiffs' claims were divided into three subclasses for the purposes of summary judgment and trial"); dkt. #595 at 13 (reminding parties that this case cannot be tried with "one class" because the "experiences" and claims of the class members vary too widely").
As the case has developed, I have become more convinced that it cannot proceed as one class. Despite my findings at the certification stage that plaintiffs' claims were grounded primarily in defendants' uniform policies and practices, the evidence in the case suggests that proof of plaintiffs' claims depends on how individual technicians responded to the numerous policies and practices at issue in the case.
However, even if all of the plaintiffs and class members could be lumped together into one class based on a broad theory of liability, plaintiffs have not proposed a trial plan that would lead to a fair result. Under either of their proposed plans, plaintiffs would present the testimony of 42 "representative" technicians, as well as other declarations and documentary evidence, to prove the claims of 2,300 individuals. Plaintiffs have not explained how the 42 technicians are representative of the whole class or how counsel could extrapolate the findings from a small sub-set of individuals to an absent class, particularly without the use of an expert. The idea of representative proof is that plaintiffs could provide testimony from a sample of technicians who can provide detail information regarding their experiences, which can then be extrapolated to the remainder of the group without significant error. Without an expert, it is not clear who would testify as to whether extrapolating from the 42 technicians is scientifically or statistically appropriate.
The single fact that all technicians were subject to the same policies, practices and job descriptions does not permit a conclusion that a small sub-set would automatically be representative of the whole. If it were the case that technicians had relatively uniform experiences, then the reported experiences of representative technicians would closely resemble the experiences of absent technicians who have not provided detailed information. In this case, however, there are many complicating factors that undermine the conclusion that one technicians' testimony is representative of an absent technicians' experiences. Thus, there is too great a chance that the determination by a jury of the "average" number of uncompensated hours worked by class members each week would be unreliable and would result in rewarding some technicians who have already been compensated fully and, simultaneously reducing the awards to which others are entitled.
One of the complicating factors is that the quantity of work available to technicians varied according to time and location. In some offices, technicians report an average of around 30 hours a week while at other offices, the average is closer to 40 or 50. Dkt. #319-20, 21 22. Thus, the amount of nonproductive work, such as the number of customer calls each day, the amount of equipment to load and unload and the number of satellites that must be built at home, would vary depending on time and location. The wide variability in weekly hours worked implies strongly that one technician's experiences may not be a proxy for others.
Another complicating factor is that the technicians used a variety of methods to record the time on their timesheets, including whether they recorded time spent driving, attending meetings and sharing equipment with other technicians. Some technicians reported that they recorded time spent in meetings and driving, while others reported that they did not record such time. Some technicians stated that their supervisors did not allow them to record more than 40 hours a week and required them to record a lunch break regardless whether they actually took one; other technicians stated that they were allowed to record overtime and that they recorded lunch breaks only if they took one. In addition, the time technicians devoted to these activities varied. For example, some declarants stated that it took about five minutes to call customers while others spent more than 30 minutes on the same task. Some declarants stated that it took them 15 minutes to drive to their first customer job while others declared that it took up to two hours. Some declarants state that they spent 30 minutes a week in unpaid meetings while others claim to have spent 90 minutes. These are only a few examples from declarations and deposition testimony of opt-in plaintiffs in the record, but they illustrate the wide differences in employment experiences between individual employees and the problem with using a small subset of technicians to represent the whole. Given the wide variability and inconsistencies in technician testimony, it would be difficult for a jury to determine what experiences are representative of the entire group of technicians at issue in this case without gross margins of error.
Additionally, plaintiffs' proposed trial plan would make it extremely difficult for defendants to assert their various defenses. Defendants contend that their GPS data reveals that many technicians were overreporting the time they worked, or at least, that many technicians were recording and were paid for the time they spent performing nonproductive tasks at home. This defense goes to the credibility of certain plaintiffs' testimony and claims, and would be difficult to assert without the ability to cross-examine individual technicians. Defendants also contend that some technicians would not be entitled to recover for driving time under the continuous workday doctrine because they did not perform their at-home tasks immediately before and after driving.Rutti v. Lojack Corp, 596 F.3d 1046, 1060 (9th Cir. 2010) (citing 29 C.F.R. § 785.16) (continuous workday rule applies only where employee is actually engaged continuously in work or is otherwise not free to use time for his own purposes). Defendants cannot prove this defense by calling a handful of witnesses whose testimony suggests that they waited until after they had been home for a few hours to perform certain tasks, because plaintiffs' witnesses will testify to the contrary. As one court noted, collective treatment is not appropriate where a defendant would be required to "pick the class apart, plaintiff by plaintiff, going into the day-to-day job duties of each of the plaintiffs to prove" their defenses. Johnson v. Big Lots Stores, Inc., 561 F. Supp. 2d 567, 586 (E.D. La. 2008) (decertifying case after court trial on merits because evidence at trial showed class was not similarly situated). "That exercise is tantamount to conducting multiple individual trials on the merits and is the antithesis of a collective action." Id. Defendants cannot be expected to present "representative" proof of its defenses when the plaintiffs and class members are not clearly representative of each other.
In sum, the evidence in this case reveals substantial variations among the opt-in plaintiffs and class members. At a high level of generality, the opt-in plaintiffs and class members perform similar job duties and are subject to the same corporate policies. But in terms of individual experiences, the evidence shows that opt-in plaintiffs and class members have different work experiences and were affected by defendants' policies in different ways. It is unfortunate that this case must be decertified at this stage, on the eve of trial and after the large investment of resources by the parties. However, I cannot allow this case to proceed to a jury trial under plaintiffs' proposed plan. As the court stated in Big Lots, 561 F. Supp. 2d at 588, the "all or nothing posture of this case makes ruling on the merits fundamentally unfair to both sides." Were the jury to rule in plaintiffs' favor under its proposed plan, the jury would have to do so on the basis of proof that is not representative of the whole class, and defendants would be deprived of the opportunity to assert their individualized defenses. On the other hand, a verdict for defendants would extinguish all of plaintiffs' and the class members' claims and would be unfair to those technicians who were denied pay in violation of federal and state law. Under such circumstances, the case must be decertified.
C. Conclusion
After the case is decertified, only the named plaintiffs' claims remain in the case. Plaintiffs must decide whether they wish to proceed to trial on June 6 with their individual claims. Plaintiffs should notify the court of their decision before the final pre-trial conference scheduled for Thursday, May 26. Defendants should also inform the court before the pretrial conference whether they object to the June 6 trial date on plaintiffs' individual claims. In addition, both parties should review their motions in limine and inform the court by midnight on Tuesday, May 24, which of their motions in limine they consider to be relevant to plaintiffs' individual claims.
ORDER
IT IS ORDERED that
1. This case is DECERTIFIED as a class and collective action. The claims of all opt-in plaintiffs and unnamed class members are DISMISSED WITHOUT PREJUDICE, leaving before the court the named plaintiffs who originated these actions.
2. Before the final pretrial conference scheduled for Thursday, May 26, 2011 at 4:00 p.m., plaintiffs should notify the court whether they intend to proceed to trial with the claims of the named plaintiffs. Defendants should also inform the court before the pretrial conference whether they object to the June 6 trial date on plaintiffs' individual claims. In addition, both parties should review their motions in limine and inform the court by midnight on Tuesday, May 24, which of their motions in limine they consider relevant to plaintiffs' individual claims.