Julian v. Swift Transportation Incorporated

Opinion

WO

Pamela Julian, No. CV-16-00576-PHX-ROS

Plaintiff, ORDER

v.

Swift Transportation Company Incorporated, et al., Defendants. Swift Transportation Company was entitled to deduct from Plaintiffs’ pay no more than eight hours of time, per day, for time Plaintiffs were logged as “sleeper berth.” In theory, that legal ruling should have allowed the parties to calculate the amount of damages due each Plaintiff. Unfortunately, Swift was not able to produce complete and accurate time records that would allow for a simple calculation of damages. But Swift did produce partial records and those partial records were enough to shift the burden of proof to Swift to prove why the damages calculated under the partial records were inaccurate. Swift has not carried its burden, meaning Plaintiffs are entitled to an award of damages in the amount calculated by their expert. The parties’ other disputes involving the compensability of the first day of orientation and additional uncompensated time in the sleeper berth must proceed to trial. In prior orders, the Court concluded Plaintiffs were entitled to be paid for time they were logged as “sleeper berth” in excess of eight hours per day. After reaching that conclusion, the Court believed Plaintiffs likely could be awarded damages as a matter of law. After all, throughout this litigation Swift has repeatedly argued its time records accurately reflected the amount of time each plaintiff spent logged as “sleeper berth.” (See, e.g., Doc. 157 at 12). But when Plaintiffs filed their motion seeking damages based on Swift’s own records, Swift’s belief in the accuracy of its records evaporated. According to Swift, its records are not, in fact, reliable. Thus, Swift now claims its records are so incomplete and full of errors that they cannot be used to determine Plaintiffs’ damages. While Swift argues the alleged deficiencies in the records means only a jury can determine the appropriate damages, that is not accurate. During discovery, Swift produced two types of records that are relevant for determining damages: 1) “DOT driver logs” and 2) “compensation data.” (Doc. 223 at 9). The “DOT driver logs” refer to the records kept by Plaintiffs to comply with Department of Transportation requirements. Those logs contained the following information for the time period from June 30, 2015 through August 28, 2017: date, driver ID, and a driver’s status throughout each day (i.e., “driving,” “off duty,” “on duty,” and “sleeper berth”). (Doc. 223-1 at 8). As for the “compensation data” Swift produced, those records contained the driver ID, hours worked, amount paid, and a description of the hours worked as either “on duty [driving]” or “on duty not [driving].” (Doc. 234-6 at 39). Swift was not able to produce complete DOT driver logs or “compensation data” for every Plaintiff. Beginning with the DOT drivers logs, Swift produced logs related to 8,238 Plaintiffs. However, 10,210 opt-in consent forms have been filed. (Doc. 237 at 12). The parties agree that some of those opt-in consent forms are duplicates but they disagree on the exact number. It is undisputed, however, that there are at least 9,552 unique opt-in forms, meaning there are at least 9,552 Plaintiffs. Given that number, and the DOT driver logs produced, Swift did not produce any data for 1,314 (9,552-8,238) opt-in plaintiffs. In other words, Swift did not produce DOT driver logs for 13.7% of the collective.1 As for the “compensation data,” Swift was not able to produce anything close to 1 The 13.7% figure assumes the collective is, in fact, limited to 9,552 individuals. comprehensive data. The DOT driver logs covered approximately 52,541 employee workweeks but Swift was able to produce “compensation data” for less than half of those workweeks. (Doc. 237 at 4, 8). Moreover, the “compensation data” did not indicate the amount of time Plaintiffs spent in the sleeper berth. Given that the “compensation data” did not cover over half of the relevant time and contained no information regarding sleeper berth time, the “compensation data” is of very limited use for calculating the damages presently at issue. The DOT driver logs are the only realistic basis for calculating those damages and that is what Plaintiffs’ expert primarily relied upon in reaching his conclusions. Plaintiffs retained an expert, David Breshears, to provide damages calculations.2 Breshears used both the DOT driver logs and the “compensation data” to complete his calculations but the “compensation data” was only used to make a preliminary calculation regarding hourly rates. Starting with the “compensation data,” Breshears calculated “a ‘driving’ pay rate” as the amount paid under the “description of ‘D-On Duty Driv. Hrl’” divided by the hours under that description. Breshears then calculated “a ‘non-driving’ pay rate” as the amount with an income description of “D-On Duty Not Drv” divided by the hours under that description. These two calculations gave Breshears hourly pay rates he then applied to the information contained in the DOT driver logs. Using the DOT driver logs, Breshears calculated each “employee’s weekly total pay” as “(a) the driving hours in the log records . . . multiplied by the related driving pay rate . . . plus (b) the on duty hours in the log records multiplied by the related non-driving pay rate.” (Doc. 223-1 at 10). The sum of those two calculations was an individual’s “weekly total pay.” Breshears then calculated each individual’s “daily sleeper hours more than eight.” Breshears combined those sleeper hours with the “on duty hours” and the “driving hours” to get a total number of hours that should have been compensated. Finally, Breshears divided the “weekly total pay” by the total hours that should have been

2 Breshears is a Certified Public Account who has “consulted on and/or testified in over 250 matters involving wage and hour-related disputes.” (Doc. 223-1 at 7). Swift does not dispute that Breshears is qualified but Swift does attack the substance of Breshears’ calculations. Swift did not depose Breshears. compensated and compared that to the amount the individual would have been paid if he had earned the federal minimum wage for all compensable time. An example using a particular driver illustrates each of Breshears’ steps. Using the “compensation data,” Breshears determined driver D354514 had a “driving pay rate” of $9.50 per hour and a “non-driving pay rate” of $7.25 per hour. Breshears then looked to the DOT driver logs for the week ending July 6, 2014. Those logs indicated D354514 had 50.67 hours for which he was in a status that required compensation. That is, the logs reflected D354514 logged 4.92 hours of “on duty” time and 45.75 hours of “driving time.” Applying the respective hourly rates to these hours (4.92 hours multiplied by $7.25 per hour and 45.75 multiplied by $9.50 per hour) resulted in D354514 being entitled to $470.28 based on the “on duty” and “driving time” reflected in the DOT driver logs. But the DOT driver logs also indicated that during this particular week, D354514 logged “26.83 daily sleeper hours more than eight per day.” Thus, Breshears added the 26.83 hours to the 50.67 hours to arrive at a total of 77.5 hours that should have been compensated. The total compensation D354514 had been entitled to receive under Swift’s calculation was then divided by the total hours ($470.28 divided by 77.5 hours) to determine this individual had been paid an effective hourly rate of only $6.07 per hour. Using the federal minimum wage of $7.25 per hour, D354514 should have been paid $561.88 ($7.25 multiplied by 77.5). Thus, Breshears determined D354514 was entitled to $91.59 in additional wages ($561.88 minus $470.28).3 Swift argues Breshears’ calculations are inaccurate and unreliable. According to Swift, Breshears should have begun his analysis by looking to the “compensation data” to determine how much each driver was, in fact, paid. It is undisputed, however, that Swift was unable to produce “compensation data” for over half the workweeks at issue. Moreover, the “compensation data” did not contain information regarding the amount of sleeper berth hours in excess of eight each day. Thus, Swift’s position that Breshears should have calculated damages by primarily looking to the “compensation data” is 3 There are slight differences in the parties’ calculations due to rounding. misguided. It simply was not possible for Breshears—or anyone else—to calculate damages in the manner suggested by Swift given Swift’s failure to keep complete records. But even beyond attacking Breshears’ basic steps, Swift argues Breshears’ calculations were fatally flawed. Swift points to D354514 and argues the “compensation data” for the week ending July 6, 2014, shows D354514 was paid $568.29. Swift believes Breshears should have started with that total and then, together with the 77.5 hours calculated from the DOT driver logs, assessed whether D354514 was paid a sufficient hourly rate. Here, Swift’s proposed calculation would result in an effective hourly rate of $7.33 per hour ($568.29 divided by 77.5 hours). Swift relies on this calculation to argue driver D354514 is not entitled to any damages as his effective hourly rate calculated in this manner exceeded the $7.25 minimum wage. Rather than identifying an error by Breshears, however, Swift’s calculation suffers from a basic error of its own. According to the “compensation data” for the relevant week, D354514 was paid for 55.11 hours (8.52+8.6+8.72+9.17+9.35+10.75) that were logged as driving hours (i.e., “on duty driv”) and 6.17 hours (0.6+0.75+0.85+1.25+1.32+1.4) that were logged as on duty, not driving (i.e., “on duty not drv”). These two figures sum to 61.28 hours. The DOT driver logs, however, show only 50.67 compensable hours. In other words, the DOT driver logs and the “compensation data” reflect different totals of compensable hours. Using the higher total hours reflected in the “compensation data,” and adding the 26.83 of sleeper berth hours reflected in the DOT driver logs, the total hours for which D354514 should have been compensated becomes 88.11 (61.28+26.83), not the 77.5 that Breshears calculated using the DOT driver logs. And dividing the total pay of $568.29 reflected in the “compensation data,” by the 88.11 hours worked, results in D354514 receiving an effective hourly rate of $6.45 per hour. That is still less than the minimum wage, meaning even under Swift’s approach of beginning with the “compensation data,” D354514 would be entitled to $70.53. Accordingly, while complaining of alleged errors or inaccuracies in Breshears’ work, Swift’s own proposed calculations reflect a failure to include all compensable hours. Under Breshears’ calculations for each of the opt-in Plaintiffs in the DOT driver logs, Swift owes those Plaintiffs a total of $6,758,478 for unpaid sleeper berth hours. This figure does not represent an extrapolation from a limited sample. Rather, this figure reflects Breshears’ actual calculations for each Plaintiff in the DOT driver logs. As for the Plaintiffs not reflected in the DOT driver logs, Breshears recommends extrapolating to cover those missing individuals. To account for the individuals missing from the DOT driver logs, Breshears recommends applying a multiplier equal to the total number of Plaintiffs divided by the number of individuals in the DOT driver logs (9,552 divided by 8,238). That produces a multiplier of 1.16. Applying that multiplier to the previously calculated damages results in a total damages award of $7,839,834.48. To be clear, this extrapolation results in each of the Plaintiffs missing from the DOT driver logs being awarded an amount equal to the average damages. This calculation does not assume every Plaintiff not included in the DOT drivers logs drove for the entire six-week training period nor does it assume the missing Plaintiffs drove an unusual amount. Rather, it assumes every Plaintiff not included in the DOT drivers logs drove for the average amount of time and is entitled to an average amount of damages. Based on Breshears’ calculations, Plaintiffs now seek summary judgment regarding the amount of damages they are entitled to recover for the sleeper berth hours in excess of eight. (Doc. 223 at 1). Swift opposes that motion. According to Swift, it has a “fundamental right” to challenge Breshears “by cross-examination” at trial. (Doc. 234 at 6). As mentioned earlier, Swift also argues Breshears’ calculations suffer from too many errors to be accepted. In addition to the parties’ dispute regarding damages, the parties have also submitted briefs regarding how trial should proceed regarding other issues, such as the compensability of the first day of orientation. I. Planned Cross-Examination Cannot Defeat Summary Judgment Swift’s primary argument against awarding Plaintiffs the sleeper berth damages they seek is that “Swift has a fundamental right to challenge Plaintiffs’ expert by cross- examination at trial as the jury can choose to accept or reject his calculations.” (Doc. 234 at 6). Swift argues it is solely “the province of the jury” whether to accept Breshears’ damages calculations. A hope to cross-examine a witness at trial, however, is not enough to defeat a properly supported motion for summary judgment. Once the party seeking summary judgment has come forward with sufficient evidence, a party cannot avoid summary judgment merely by claiming it wishes to cross- examine a witness. “Neither a desire to cross-examine affiant nor an unspecified hope of undermining his or her credibility suffices to avert summary judgment, unless other evidence about an affiant’s credibility raises a genuine issue of material fact.” Frederick S. Wyle Prof’l Corp. v. Texaco, Inc., 764 F.2d 604, 608 (9th Cir. 1985). To avoid summary judgment regarding damages, Swift was required to point to evidence establishing the existence of a genuine dispute of material fact regarding Breshears’ calculations. Swift’s claim that it is entitled to proceed to trial merely to allow a jury to evaluate a cross- examination of Breshears is incorrect. II. Breshears’ Calculations Shifted the Burden to Swift Plaintiffs’ motion for an award of damages invokes the special framework applicable to claims for unpaid wages under the Fair Labor Standards Act (“FLSA”) when an employer does not have reliable time records regarding the amount of work performed. Because the FLSA requires employers maintain time records, the Supreme Court long ago established a “burden-shifting framework for actions in which the employer has kept inaccurate or inadequate records.” Senne v. Kansas City Royals Baseball Corp., 934 F.3d 918, 939 (9th Cir. 2019) (citing Anderson v. Mt. Clemens, 328 U.S. 680 (1946)). That framework places an initial burden on the employee to prove “he has in fact performed work for which he was improperly compensated.” Mt. Clemens, 328 U.S. at 687. The employee meets this burden by producing “sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference.” Id. Once an employee does so, “[t]he burden then shifts to the employer to come forward with evidence of the precise amount of work performed or with evidence to negative the reasonableness of the inference to be drawn from the employee’s evidence.” Id. at 687-88. If an employer is unable “to produce such evidence, the court may then award damages to the employee, even though the result be only approximate.” Id. An employer in these circumstances is not entitled to complain that “the damages lack the exactness and precision of measurement” it would like because that imprecision is due to the employer’s own misconduct in not keeping accurate records. Id. Thus, there need only be a “reasonable inference as to the extent of the damages.” Id. Applying that framework here, Plaintiffs have come forward with damages calculated by Breshears primarily relying on the DOT driver logs. Those logs are the only available evidence indicating the number of hours Plaintiffs spent driving, on duty not driving, and in the sleeper berth. Those logs were more than sufficient to “show the amount and extent” of Plaintiffs’ work. Id. That is, the logs shifted the burden to Swift to come forward with other evidence that would, at least, cast doubt on the accuracy of the DOT driver logs. Swift did not do so. Instead of a straightforward attack on the reliability of the DOT driver logs, Swift points to the “compensation data” as sometimes indicating differences in the number of hours Plaintiffs were driving or otherwise on duty. It is undisputed, however, that the “compensation data” does not present anything close to a complete picture. The “compensation data” produced by Swift does not cover even half of the workweeks at issue nor does it contain information regarding the number of daily sleeper berth hours each plaintiff recorded. It is not possible to arrive at a more accurate calculation of damages by primarily relying on the “compensation data.” Accordingly, differences between the DOT driver logs and the “compensation data” are not sufficient “to negative the reasonableness of the inference to be drawn from [Plaintiffs’] evidence.” Id. at 688. In these circumstances, there is a “reasonable inference” that the DOT driver logs show Plaintiffs performed work for which they were not compensated. Because Swift did not point to evidence contradicting the DOT driver logs, Plaintiffs are entitled to damages calculated using the DOT driver logs. To be sure, those damages are not exact. But they are not required to be exact. Swift devotes a large portion of its briefing to attacking the precise amount of damages calculated by Breshears. This argument from Swift is, in effect, the argument the Supreme Court rejected in 1946. The fact that the damages calculated by Breshears might “lack the exactness and precision of measurement” Swift would prefer is not enough. Mt. Clemens, 328 U.S. at 687. Swift cannot avoid an award of damages merely because of some uncertainty regarding the exact damages Plaintiffs are entitled to recover. The damages need only be “approximate.” Id. And that is what the damages are in this case.4 In addition to calculating the damages due to the Plaintiffs reflected in the DOT driver logs, Breshears opined it would be reasonable to award damages to the Plaintiffs missing from the DOT driver logs. Breshears recommends calculating damages for those missing Plaintiffs by multiplying the damages calculated from the DOT driver logs by the total number of Plaintiffs divided by the number of drivers for which there are records. (Doc. 237-1 at 10). This would create an “estimate” of the wages due the Plaintiffs not reflected in the DOT driver logs by awarding the average amount of compensation to each missing Plaintiff. (Doc. 223-1 at 22). Swift argues there are genuine disputes of material fact regarding this extrapolation because it assumes all the plaintiffs not reflected in the DOT driver logs worked for the full 6-week training period. (Doc. 234 at 19). It is not clear why Swift interprets the proposed extrapolation in that manner. As noted by 4 Swift complains of numerous relatively minor errors in Breshears’ calculations such as some workweeks where an individual was only logged as “sleeper berth.” In Swift’s view, that was obviously an error as no individual would be confined to a sleeper berth, without driving, for an entire workweek. But even assuming that is accurate, that only occurred in 1.4% of the workweeks at issue. (Doc. 237 at 11). Such a minor alleged error is not sufficient to create a dispute that must be resolved at trial given the Supreme Court’s forgiving standard for calculating damages. Swift also complains that, of the 52,541 workweeks reflected in the DOT driver logs and analyzed by Breshears, 1,141 workweeks contained more than 168 hours. (Doc. 234 at 13). If accurate, that would be errors in approximately 2% of the workweeks. But this alleged inaccuracy has a simple explanation. According to Breshears, “certain log records cross calendar days,” which resulted in some workweeks appearing to contain more than 168 hours. Breshears believes that reallocating the hours following week might simply result in more compensation due for the following week. Thus, the fact that some weeks contain more than 168 hours did not have a large impact on the total damages. Plaintiffs, the extrapolation is based on the average amount of damages owed to individuals reflected in the DOT driver logs. The extrapolation does not assume each plaintiff missing from the DOT driver logs worked for six weeks. Rather, it simply assumes the missing plaintiffs worked the average length of time and are entitled to the average amount of compensation. This is imprecise but Swift has not offered any way to avoid such imprecision given its failure to provide any records regarding these Plaintiffs. It is true that extrapolation based on incomplete records will sometimes present a jury question. See, e.g., Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1049 (2016) (noting “persuasiveness” of representative evidence “is the near-exclusive province of the jury”). But Swift has not pointed to any evidence that would create a genuine dispute of fact regarding the reasonableness or appropriateness of this extrapolation. Assuming this issue were to proceed to trial, Swift would have to present evidence showing the “precise amount of work performed” or at least “evidence to negative the reasonableness of the inference” drawn by Breshears regarding the missing individuals. Mt. Clemens, 328 U.S. at 687-88. To defeat summary judgment, Swift should have pointed to the evidence it would present at trial. Swift did not do so. In fact, Swift chose not to depose Breshears and chose not to retain its own expert to conduct an alternative set of damages calculations. And Swift is not entitled to proceed to trial merely because it hopes to somehow discredit Breshears’ reasonable extrapolation. Therefore, Plaintiffs are entitled to damages for the Plaintiffs in the DOT driver logs and the Plaintiffs not in those logs.5 The total damages are $7,839,834.48. III. Trial Issues The parties have long disagreed on whether this case could proceed to trial on a collective basis. The Court called for Plaintiffs to explain how certain issues would be tried on a collective basis. Plaintiffs explained their plans. The Court makes the following 5 According to Swift, certain records “suggest” 150 Plaintiffs quit after orientation and are not entitled to any additional wages for unpaid sleeper berth time. (Doc. 234 at 18). Swift does not explain why it cannot state with confidence whether the particular individuals did, in fact, quit after orientation. Thus, it appears Swift’s own failure to maintain records is again fatal to a precise calculation of damages owed to each individual. observations to guide the parties’ trial planning. A. Adequacy of Sleeper Berth and Time Plaintiffs plan on arguing at trial that Swift was not entitled to deduct any time for a sleeping period because Swift did not comply with the applicable regulation regarding such deductions. That is, to not pay Plaintiffs for the eight-hour sleeping period, Swift must prove the applicability of 29 C.F.R. § 785.22. See Flores v. City of San Gabriel, 824 F.3d 890, 897 (9th Cir. 2016) (holding burden of proving exemption is on employer). For present purposes, that regulation can be viewed as requiring Swift prove four propositions: 1) Swift and Plaintiffs expressly or impliedly agreed to exclude a sleeping period; 2) the sleeping period was “regularly scheduled”; 3) Swift provided “adequate sleeping facilities”; and 4) Plaintiffs were able to “usually enjoy an uninterrupted night’s sleep” which means “at least 5 hours’ sleep during the scheduled period.” Id. According to Swift, the Court can decide as a matter of law that § 785.22 applies. Swift did not, however, present its arguments in a procedurally proper way, such as through a motion for summary judgment. At present, there are disputes of fact regarding the propositions Swift must prove. In brief, there are disputes about the parties’ alleged agreement to exclude sleeping time, how the sleeping time was scheduled, whether the sleeper berths qualified as “adequate,” and the extent of interruptions Plaintiffs experienced during the scheduled sleeping time. Therefore, whether Swift was entitled to deduct an eight-hour period must proceed to trial.6 B. First Day of Orientation The Court previously ruled the compensability of the first day of orientation must proceed to trial. (Doc. 220 at 6). Swift filed a motion requesting the Court direct Plaintiffs to proffer a “trial plan” of how they will prove the compensability of the first day on a collective basis. (Doc. 222). Plaintiffs opposed that motion but, in doing so, also outlined their “trial plan.” (Doc. 230). While Swift believes the “trial plan” is “vague and inadequate,” there is no need for Plaintiffs to provide additional details.

6 There is also a dispute regarding unpaid studying time while Plaintiffs were in the sleeper berth. Resolution of that issue must also proceed to trial. In prior orders, the Court observed there are two factors that will be especially important to resolving the compensability of the first day of orientation: “expectation of pay” and “job prospects upon completion” of the orientation. (Doc. 220 at 4). Swift complains Plaintiffs’ proposed trial plan does not explain in enough detail how they will address these two factors on a collective basis. Deciding the compensability of the first day of orientation, however, will not be resolved based solely on the two factors. As explained in prior orders, the compensability question does “not depend on isolated factors but rather upon the circumstances of the whole activity.” (Doc. 207 at 10) (quoting Boucher v. Shaw, 572 F.3d 1087, 1091 (9th Cir. 2009)). And the central inquiry is always the “‘economic reality’ of the parties’ relationship.” (Doc. 220 at 4) (quoting Hale v. State of Ariz., 993 F.2d 1387, 1393 (9th Cir. 1993). Thus, while the factors regarding “expectation of pay” and “job prospects upon completion” will be important, they will not be dispositive. The entirety of the parties’ relationship must be assessed. Plaintiffs state they will present evidence that all orientations followed the same structure as well as evidence regarding most participants’ expectations of pay and job prospects. Plaintiffs are entitled to do so. Accordingly, IT IS ORDERED the Motion to Require Plaintiffs Provide Trial Plan (Doc. 222) is DENIED. IT IS FURTHER ORDERED the Motion for Award of Damages (Doc. 223) is IT IS FURTHER ORDERED a status conference is set for January 14, 2020 at 2:00 p.m. / / / / / / / / / / / / / / / 1 IT IS FURTHER ORDERED no later than January 7, 2020, the parties shall file a joint status report listing the matters to be resolved at trial, the expected number of witnesses each side wishes to call, and the anticipated length of trial. Dated this 27th day of December, 2019. fo - ( □

g Senior United States District Judge

-13-