JOHN UTNE, Case No. 16-cv-01854-RS Plaintiff, v. ORDER ON MOTIONS TO EXCLUDE HOME DEPOT U.S.A., INC., Defendant.
This action, brought on behalf of a class of Home Depot employees, avers violations of state wage and hour laws at Home Depot stores in California. Background facts have been set forth in prior orders and need not be repeated here. Following class certification, Lead Plaintiff John Utne and Defendant Home Depot U.S.A., Inc. (“Home Depot”) each moved to exclude the testimony and/or reports of three of the opposing party’s proffered expert witnesses.1 This order addresses all six motions. For the reasons discussed below, Plaintiff’s motion to exclude Mr. Crandall is granted in part and denied in part; the motion to exclude Mr. Smith is denied; and the motion to exclude Mr. Trujillo is granted. Defendant’s motion to exclude Dr. Krosnick is granted in part and denied in part; the motion to exclude Mr. Toney is granted in part and denied in part; and the motion to exclude Mr. White is granted. 1 Alfred Pinto was also appointed as an additional class representative. Following briefing and oral argument on the present motions, Plaintiff also filed a Fifth Amended Complaint, which did not A. Plaintiff’s Expert Witnesses 1. Jon A. Krosnick Plaintiff proffers expert evidence from Dr. Jon A. Krosnick, a Stanford University social scientist. Dr. Krosnick first compiled a survey drawn from phone interviews with 322 randomly selected class members. The survey assembled data relating to respondents’ recollections of, among other things, their time spent walking to the back of the store; which entrances they used; their time spent clocking in; and whether they had ever waited to be let out of the store at the end of a shift (and, if so, for how long). Dr. Krosnick then conducted a statistical analysis of the responses to estimate the average amount of time employees spent conducting these activities or waiting to be let out. He concluded, for instance, that the respondents “waited to be let out of the store on 60.61% of closing shifts, on average.” Dkt. 320-2 (“Krosnick First Report”) ¶ 188. His First Report detailing these results was submitted on October 16, 2019. Dr. Krosnick submitted a Second Report on April 30, 2020, prepared as a rebuttal report to the time and motion study performed by Defendant’s expert Robert Crandall. Finally, Dr. Krosnick submitted a Third Report on December 17, 2021, which further critiqued the Crandall study, responded to criticisms by Mr. Crandall of Dr. Krosnick’s survey, and provided an “idiographic” model to estimate classwide damages. 2. James Toney Plaintiff also retained James Toney as a data analysis and damages expert. Working from the data provided by the Krosnick survey, Mr. Toney used a “nomothetic” model to estimate classwide damages (including unpaid wages and penalties) during the class period.2 Toney’s 2 According to Dr. Krosnick, the “nomothetic” model, used by Mr. Toney, “compute[s] averages of key quantities for the survey respondents and then use[s] those averages to calculate damages using the time keeping records and the pay records” for the entire Class Period. Dkt. 320-4 (“Krosnick Third Report”) ¶ 1013. For the “idiographic” model used by Dr. Krosnick, by contrast, each survey respondent’s answers . . . are used in conjunction with his or her time keeping records and pay records to calculate the damages owed to him or her in each of a series of categories (e.g., Opening Report was completed on October 16, 2019, followed by supplemental reports submitted on April 30, 2020; December 17, 2021; and March 18, 2022. 3. Gary White Finally, Plaintiff retained Gary White, a consultant and former retail executive. In his First Report, submitted on October 16, 2019, Mr. White reviewed Home Depot’s policies relating to clocking in and store closing and concluded they were out of step with others in the retail industry. His Second Report, dated January 16, 2020, provided his critiques of portions of the Crandall study commissioned by Defendant. B. Defendant’s Expert Witnesses 1. Robert Crandall Defendant retained Robert Crandall, a partner at the consulting practice Resolution Economics LLC, to perform a “time and motion study” using video recordings taken at thirty randomly selected California Home Depot stores in 2019. Mr. Crandall recorded “approximately 8,200 entrances and 4,400 exits” by employees and subsequently coded their activities. Among other findings, Mr. Crandall’s study concluded that 61.5% of employees experienced no waiting time at all at the end of their shifts and that employees engaged in “personal activities” while walking to the back of the store. The study was submitted on October 16, 2019. 2. Stephen Smith Defendant also retained Stephen Smith, a director at the University of Chicago’s National Opinion Research Center, to rebut the findings contained in the Krosnick survey. Mr. Smith’s report, submitted on April 30, 2020, critiqued the survey for, among other things, containing overly complex questions and being susceptible to “recall error” (that is, misestimations based on
walking time, waiting time). Then, these damages amounts are added up for all survey respondents to yield the total amount of damages owed to the survey respondents. Then this amount is divided by the number of survey respondents . . . to yield an average per person, and this average is multiplied by the number of people in the class . . . to yield the total amount of damages owed to the class. Id. ¶ 1014. inaccurate memories). 3. Timothy Trujillo Finally, Defendant retained Timothy Trujillo, a human resources consultant, to rebut Mr. White’s findings. Mr. Trujillo, for his part, concluded that Home Depot’s policies were in line with industry practices. His report was submitted on April 30, 2020. Rule 702 of the Federal Rules of Evidence requires that a witness proffered as an expert by a party be qualified by “knowledge, skill, experience, training, or education.” Fed. R. Evid. 702. Even if a witness is qualified as an expert in a particular field, any scientific, technical, or specialized testimony is admissible only if it (a) “will help the trier of fact to understand the evidence or to determine a fact in issue,” (b) “is based upon sufficient facts or data,” (c) “is the product of reliable principles and methods,” and (d) “the expert has reliably applied the principles and methods to the facts of the case.” Id. Irrelevant or unreliable testimony is prohibited under Rule 702. Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 589 (1993). Expert opinions are relevant if the knowledge underlying them has a “valid connection to the pertinent inquiry.” United States v. Sandoval-Mendoza, 472 F.3d 645, 654 (9th Cir. 2006) (internal quotation marks and alteration omitted). “Expert testimony which does not relate to any issue in the case is not relevant and, ergo, non-helpful.” Daubert, 509 U.S. at 590. Expert opinion testimony is reliable if such knowledge has a “basis in the knowledge and experience of [the relevant] discipline.” Id. at 592. Under Daubert, courts should consider the following factors when evaluating whether an expert’s proposed testimony is reliable: (1) “whether a theory or technique . . . can be (and has been) tested,” (2) “whether the theory or technique has been subjected to peer review and publication[,]” (3) the known or potential error rate of the particular scientific theory or technique, and (4) the degree to which the scientific technique or theory is accepted in a relevant scientific community. Id. at 593–94. This list is not exhaustive, however, and the standard is flexible. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 151 (1999). The Daubert inquiry “applies not only to testimony based on ‘scientific’ knowledge, but also to testimony based on ‘technical’ and ‘other specialized’ knowledge.” Id. at 141. The court’s task is not to “decid[e] whether the expert is right or wrong, just whether his testimony has substance such that it would be helpful to a jury.” Alaska Rent-A-Car, Inc. v. Avis Budget Grp., Inc., 738 F.3d 960, 969–70 (9th Cir. 2013). Courts may not exclude testimony simply because it is impeachable. Id. at 969. “Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.” Daubert, 509 U.S. at 596. The focus of the inquiry is thus on the principles and methodology employed, not the conclusions reached by the expert. See id. at 595. Ultimately, the purpose of the assessment is to exclude speculative or unreliable testimony to ensure accurate, unbiased decision-making by the trier of fact. “Nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert.” Kumho, 526 U.S. at 157 (internal quotation marks and citation omitted). A. Crandall Study Plaintiff seeks to exclude the Crandall study based on a number of critiques of its methodology and execution. Plaintiff previously moved to exclude the study on nearly identical grounds; that motion was granted only in part and only with respect to the then-pending motion to decertify the class. See Dkts. 238, 272. Plaintiff now rehashes these arguments, adding only the argument that the study should be excluded under Rule 403 of the Federal Rules of Evidence. This argument is unpersuasive, especially in light of the prior conclusion that “Plaintiff’s concerns largely go to weight, not admissibility.” Dkt. 272. Consistent with the prior order, Plaintiff’s motion is granted only with respect to Mr. Crandall’s opinions on the percentage of employees engaged in personal activities during their pre-shift walks to the back of the store or time in the break room before their shifts.3 3 Defendant indicated in its opposition to the motion that, consistent with the Court’s prior order, it will not be seeking to solicit evidence from Mr. Crandall on this topic. See Dkt. 317, at 7 n.5, 16. B. Krosnick’s First Report Defendant moves to exclude Dr. Krosnick’s First Report on grounds broadly akin to Plaintiff’s critiques of the Crandall study. For instance, Defendant argues the Krosnick survey should be excluded because it is inadequately representative, is too small, drew too low a response rate, did not adequately account for differences in respondents’ tenure, and reported findings that were at times inconsistent with Home Depot’s own payroll records. Many of these critiques appear valid — maybe a stratified random sample, for instance, could have better accounted for the unique layouts of all 233 Home Depot stores — but none are fatal. Defendant relies heavily on Jimenez v. Allstate Insurance Co., in which the court granted the defendant’s motion to exclude a similar survey by Dr. Krosnick based on the “substantial indicia of [its] unreliability.” No. LA CV10-08486 JAK, 2019 WL 13088814, at *26 (C.D. Cal. May 13, 2019). These indicia included, among others, the fact that the survey attempted to extrapolate from a non-random sample of the class, id. at *18, and depositions of survey respondents that showed the survey was susceptible to recall bias, id. at *19–20. Besides being noncontrolling, that case is largely inapposite. Dr. Krosnick’s survey here is not susceptible to the critique raised in Jimenez that, by using a “convenience sample” from an attempted census, rather than a random sample, Dr. Krosnick’s results were “neither persuasive nor supported by any identified, scientific principles.” Id. at *18. While Dr. Krosnick appears to have relied on a method of bias control that the Jimenez Court rejected, his First Report offers adequate explanation for why this was rooted in scientific principles. Compare Krosnick First Report ¶ 340, with Shari Seidman Diamond, Reference Guide on Survey Research, in FED. JUD. CTR., REFERENCE MANUAL ON SCIENTIFIC EVIDENCE 359, 384 (3d ed. 2011) (“Researchers have developed a variety of approaches to adjust for nonresponse, including . . . comparing . . . the pattern of responses from easy-to-reach and hard-to-reach responders in telephone surveys . . . .”). Further, Defendant’s argument relating to the lack of access to information about Dr. Krosnick’s cognitive pretesting process has been raised and rejected. See Dkt. 272, at 7–8. Defendant’s objections are better addressed through rigorous cross-examination than full exclusion of the First Report, and the motion is therefore denied with respect to that report. C. Krosnick and Toney Damages Calculations Defendant also moves to strike the idiographic damages calculations contained in Dr. Krosnick’s Third Report and to exclude Mr. Toney’s reports and testimony discussing his nomothetic damages calculations. These calculations are methodologically similar, see supra n.2, but Defendant’s arguments against them are different. Defendant contends Dr. Krosnick’s calculations are procedurally improper: they constitute untimely and improper surrebuttal that denied Home Depot the chance to respond substantively.4 Defendant’s attacks on Mr. Toney’s reports, by contrast, are substantive. For instance, it suggests the reports inaccurately calculate penalties and inappropriately aggregate data across the class, including by failing to account for differences between the injured class members (e.g., by not adjusting for differences in pay between class members). Although they do not raise the issue directly, these two motions necessarily require thinking ahead to how damages could be awarded to injured class members in the event Home Depot is found liable.5 Plaintiff argues the jury should first calculate an aggregate damages award 4 This argument is rooted in the tangled web of discovery this case has woven. In summary, Dr. Krosnick claims that, “[i]n keeping with best practices in survey research,” he did not have access to survey respondents’ identities at the time he submitted his First Report in October 2019. Krosnick Third Report ¶ 1016. After a year of averred back-and-forth with Home Depot and various discovery orders from the assigned magistrate judge, Dr. Krosnick eventually obtained access to the respondents’ identities as well as time and pay records “sufficient to perform an ideographical [sic] damages calculation.” Dkt. 320, at 25. Dr. Krosnick conceded it was his idea to perform the damages analysis once the respondents’ identities had been turned over, and that he had not previously suggested he would perform it if he had access to respondents’ identities. See Dkt. 305-3, at 259–60. Defendant argues these calculations should have been included, if at all, in an earlier report, and that Dr. Krosnick could have obtained the respondents’ identities from SSRS at any time. While the balance of this section of the Order addresses more substantive concerns with Dr. Krosnick’s damages calculations, it is hard to take seriously the idea that Home Depot was unable to respond substantively to the calculations despite the fact that it received them roughly seven months before filing the present Daubert motion. 5 Prior orders have noted only that uninjured class members “may be weeded out at the claims phase.” Dkt. 272, at 14. for each class, and then a claims process can be used to screen out uninjured class members. See Dkt. 319, at 11–12; Dkt. 345, at 18–20. Defendant argues this approach is inconsistent with the law in that it would violate Home Depot’s “due process right not to pay in excess of its liability.” Dkt. 329, at 2 (quoting Victorino v. FCA US LLC, No. 16cv1617-GPC, 2020 WL 2306609, at *3 (S.D. Cal. May 8, 2020)). This is especially true because Krosnick’s data suggest that “certain class members were never subject to the alleged violations that give rise to [P]laintiff’s claims,” and thus an aggregate calculation based on that data would impermissibly sweep in uninjured class members. Dkt. 307, at 9. As such, Defendant contends damages must be determined on an individualized basis. See Dkt. 346, at 13–14. If this is the case, Toney’s calculations must be excluded as irrelevant.6 See Dkt. 329, at 2. Decisions of the Ninth Circuit and the Supreme Court have upheld aggregate damages awards in wage and hour class actions, notwithstanding individual differences among class members’ injuries. In Ridgeway v. Walmart Inc., for instance, plaintiffs proffered an expert who calculated damages for a class of truck drivers, relying on a variety of representative evidence as to the amount of time drivers spent on uncompensated activities, such as rest breaks and vehicle inspections. See Ridgeway v. Wal-Mart Stores Inc., No. 08-cv-05221-SI, 2017 WL 1549329, at *1–2 (N.D. Cal. May 1, 2017). The jury returned an aggregate damages award based on the expert’s calculations, id. at *1, which the Ninth Circuit upheld, see Ridgeway v. Walmart Inc., 946 F.3d 1066, 1089 (9th Cir. 2020). The Ninth Circuit relied in large part on the Supreme Court’s decision in Tyson Foods, Inc. v. Bouaphakeo, another wage and hour class action in which the plaintiff’s expert witnesses calculated average amounts of time employees spent donning and doffing protective equipment based on a video study. See 577 U.S. 442, 450 (2016). The jury “more than halved the damages recommended by” one of the experts, but it nonetheless returned an aggregate award to the class. Id. at 452. 6 Defendant does not raise this same argument with respect to Dr. Krosnick’s idiographic damages calculations, but the logic clearly applies to them as well. At the same time, the Tyson Court recognized that “the question whether uninjured class members may recover” any part of an aggregate damages award was “of great importance.” 577 U.S. at 1050; see also id. at 1053 (Roberts, C.J., concurring) (“[I]f there is no way to ensure that the jury’s damages award goes only to injured class members, that award cannot stand.”). While the Supreme Court has not addressed the issue squarely, the en banc Ninth Circuit recently rejected the argument that a district court cannot permissibly certify a class “that potentially includes more than a de minimis number of uninjured class members.” Olean Wholesale Grocery Cooperative, Inc. v. Bumble Bee Foods LLC, 31 F.4th 651, 669 (9th Cir. 2022) (en banc). This conclusion was in turn supported by the principle that “district court[s are] well situated to winnow out . . . non-injured class members at the damages phase of litigation.” Ruiz Torres v. Mercer Canyons Inc., 835 F.3d 1125, 1137 (9th Cir. 2016). Returning, then, to the immediate question, Plaintiff’s proffered damage calculations cannot thread the needle between these principles. Plaintiff does not contest that Krosnick’s data indicates that some Hourly Class members never walked across the store at the start of their shifts, and that not every Lock-In Class member had to wait to be let out. See Krosnick First Report ¶¶ 178, 188. Both calculations, then, almost certainly include damages for uninjured class members. Unlike the models offered in Tyson and Ridgeway, however, neither model here purports to derive a total award from a sum of individual class members’ awards. Cf. Ridgeway, 946 F.3d at 1088. Were the jury to return an aggregate award, it would “not be possible to know which workers are entitled to share in the award,” or how much each should receive. Tyson, 577 U.S. at 461. At the same time, neither Toney nor Krosnick offer a formula or methodology on which the jury could independently base its own award. See, e.g., Bolding v. Banner Bank, No. C17-0601RSL, 2021 WL 5978231, at *3 (W.D. Wash. Sept. 13, 2021). In the abstract, a classwide damages award may be permissible with respect to either or both classes in this case. Unfortunately, neither Krosnick’s nor Toney’s calculations can provide enough specificity as to the class members to ensure that the Court could efficiently and equitably determine how to distribute any such award and exclude uninjured members.7 Cf. Brown v. DirecTV, LLC, 562 F. Supp. 3d 590, 602 (C.D. Cal. 2021). While they may have value as a heuristic for settlement or to help the Court assess the proper level of penalties, neither calculation should be offered to the jury. The motion to strike Dr. Krosnick’s idiographic calculations from his Third Report is thus granted. The motion to exclude Mr. Toney’s reports is denied except as to his “analysis of the percentage of closing shifts where the time punches were edited upwards,” as this is “relevant to the liability determination for the Lock-In Class.” Dkt. 345, at 17. The motion is otherwise granted.8 D. Smith Testimony Plaintiff moves to strike portions of Stephen Smith’s expert testimony. As noted above, Mr. Smith was hired to provide rebuttal testimony to the Krosnick survey. Plaintiff argues that Smith’s testimony should be stricken because he (1) failed to base his opinions on his own study or investigation of the respondents, (2) failed to provide sufficient methodological foundation for his critiques, (3) needlessly provides evidence cumulative of Mr. Crandall’s reports, and (4) invades the province of the jury by providing opinions a lay person could make. None of these arguments hold water. Mr. Smith’s report is replete with references to competing sources of scientific literature that call the Krosnick survey’s results into doubt. Reasonable minds could disagree as to the veracity of these objections, but they are surely not baseless. Many of Mr. Smith’s opinions are rooted in a comparison between his expectations and
7 Additionally, Plaintiff provides no authority that would support a classwide damages award extrapolated from a representative sample of injured class members, as Krosnick’s idiographic calculation does. In fact, courts have actively disproved of this method in the absence of “extraordinary circumstances.” Gutierrez v. Wells Fargo & Co., No. C 07-05923 WHA, 2009 WL 1247040, at *4 (N.D. Cal. May 5, 2009) (distinguishing Hilao v. Estate of Marcos, 103 F.3d 767 (9th Cir. 1996)); see Spears v. First American eAppraiseIT, No. 5-08-CV-00868-RMW, 2014 WL 4647679, at *21 (N.D. Cal. Sept. 16, 2014); 4 NEWBERG AND RUBENSTEIN ON CLASS ACTIONS §§ 11:4, 12:5 (6th ed. 2022). 8 Both parties have provided helpful supplemental briefing on the question of whether trial should be bifurcated into a liability phase and a damages phase. See Dkts. 344–46. That question ultimately will not be resolved today, but either party may move for bifurcation (or the parties can offer a stipulation to this effect) at any time consistent with this Order’s holdings. the Krosnick survey’s findings; the variances he observes lead him to conclude that the survey was prone to error. Mr. Smith need not have actually interviewed the same survey respondents to draw these conclusions, as Plaintiff suggests; Mr. Smith adequately explains why such inferences can be drawn based on his experience. Furthermore, Mr. Smith does not simply rehash data from the Crandall study to the point that it need be excluded. Finally, while some of Mr. Smith’s conclusions use plain language (such as his critique that the survey questions are too “wordy,” see Dkt. 306-1 Ex. 2, at 14 ¶ 40), this does not ipso facto make them lay opinions. In brief, all of Plaintiff’s critiques, valid though they may be, go to the weight of Mr. Smith’s testimony, not its admissibility. The motion to exclude portions of Stephen Smith’s testimony is therefore denied. E. White Reports and Testimony Defendant finally moves to exclude the reports and testimony of Gary White, primarily on the ground that they include improper legal conclusions and are speculative. In particular, Mr. White opines as to whether, in his opinion, Home Depot employees were under the control of Defendant before clocking in. Defendant also objects to statements by Mr. White that it argues are irrelevant, such as whether being locked in post-shift was “unfair” to employees. Upon review of Mr. White’s report, these arguments are compelling. While an expert’s testimony is “not objectionable just because it embraces an ultimate issue” in the case, Fed. R. Civ. P. 704(a) (emphasis added), it is objectionable where it offers “an opinion on an ultimate issue of law.” Nationwide Transp. Fin. v. Cass Info. Sys., Inc., 523 F.3d 1051, 1058 (9th Cir. 2008). This is precisely what Mr. White does on multiple occasions. See, e.g., Dkt. 303-2, at 16 (“Home Depot established control of their associates by written policy.”); id. at 17 (“Home Depot knew or should have known they were not paying associates for being under the store’s control.”). Meanwhile, Mr. White’s speculation as to Home Depot’s motives in establishing a pre-shift policy are irrelevant to the question of whether employees were “controlled” by Defendant. To the extent Mr. White bases his conclusions about the pervasiveness of post-shift lock-ins on Plaintiff’s depositions, this is an unreliable basis for his opinion that should properly be excluded. The only portion of Mr. White’s reports not subject to this critique is his rebuttal testimony 1 about the veracity of the Crandall study. Specifically, he opines that, based on his experience, it 2 would be “improbable if not impossible” for Mr. Crandall’s cameras to have been installed 3 without employees noticing, and that the cameras would cause employees to be on their “best 4 policy-following behavior.” Dkt. 303-3, at 3. This is not an opinion that requires Mr. White’s 5 “specialized knowledge.” Fed. R. Evid. 702(a). While Plaintiff is free to raise this issue on cross- 6 examination to attack the weight of Mr. Crandall’s study, allowing Mr. White to testify to this fact 7 could mislead the jury. Thus, the motion to exclude Mr. White’s reports and testimony is granted. 8 F. Trujillo Testimony 9 Finally, Plaintiff seeks to exclude the testimony of Timothy Trujillo. As Defendant notes, 10 Mr. Trujillo was retained purely to rebut Mr. White’s testimony. As Mr. White’s testimony will be 11 excluded, Mr. Trujillo’s rebuttal testimony is moot. The motion to exclude is therefore granted. 13 For the reasons discussed above, Plaintiff's motion to exclude the reports and testimony of 14 Robert Crandall is granted with respect to those portions previously excluded under this Court’s 3 15 May 8, 2022, order. The motion is denied in all other respects. Defendant’s motion to exclude Dr. a 16 || Krosnick’s First Report is denied, but its motion to strike the portions of Dr. Krosnick’s Third 3 17 Report concerning the idiographic damages calculations is granted. Defendant’s motion to exclude 18 the testimony of James Toney is denied as to his analysis of the percentage of closing shifts where 19 time punches were edited upwards, and otherwise granted. Plaintiff's motion to exclude the 20 testimony of Stephen Smith is denied. Defendant’s motion to exclude the testimony of Gary White 21 is granted. Finally, Plaintiff's motion to exclude the testimony of Timothy Trujillo is granted. 22 23 ITIS SO ORDERED. 24 25 Dated: November 10, 2022
7 RICHARD SEEBORG _ Chief United States District Judge 28 ORDER ON MOTIONS TO EXCLUDE EXPERT TESTIMONY CASE No. 16-cv-01854-RS