Evans Fruit Co Inc v. United States Department of Labor

1 FILED IN THE U.S. DISTRICT COURT EASTERN DISTRICT OF WASHINGTON 2 Oct 11, 2019 3 UNITED STATES DISTRICT COURT SEAN F. MCAVOY, CLERK EASTERN DISTRICT OF WASHINGTON [4] EVANS FRUIT CO., INC., a No. 1:19-cv-03202-SMJ 5 Washington Corporation; WGE HOLDINGS, LLC, a Washington ORDER DENYING PLAINTIFFS’ 6 Limited Liability Company; MOTION FOR PRELIMINARY McDOUGALL FAMILY FARMING, INJUNCTION 7 INC., a Washington Corporation; McDOUGALL & SONS, INC., a 8 Washington Corporation; DOUBLE S ORCHARDS, LLC, a Washington 9 Limited Liability Company; COLUMBIA FRUIT PACKERS, 10 INC., a Washington Corporation; COLUMBIA ORCHARD 11 MANAGEMENT, INC., a Washington Corporation; and WADE & WADE, 12 LLC, a Washington Limited Liability Company, [13] Plaintiffs, [14] v. [15] UNITED STATES DEPARTMENT OF 16 LABOR; PATRICK PIZZELLA, in his official capacity as Acting United States 17 Secretary of Labor; JOHN P. PALLASCH, in his official capacity as 18 Assistant Secretary of Labor, Employment & Training 19 Administration, United States Department of Labor; CHERYL M. 20 STANTON, in her official capacity as Administrator of the Wage & Hour 1 Division, United States Department of Labor; [2] Defendants, [3] and [4] STATE OF WASHINGTON 5 EMPLOYMENT SECURITY DEPARTMENT; [6] 7 Defendant-Intervenor. 8 On October 9, 2019, the Court held a hearing on Plaintiffs’ Motion for 9 Preliminary Injunction, ECF No. 15. At the conclusion of the hearing, the Court 10 took the matter under submission. For the reasons that follow, the Court now denies 11 Plaintiffs’ motion for a preliminary injunction. 12 BACKGROUND 13 Plaintiffs are eight Washington-based apple growers. ECF No. 1 at 6–8. 14 Plaintiffs assert that their crops “are virtually exclusively picked by hand laborers.” 15 Id. at 15. As the result of a persistent shortage of domestic agricultural laborers, 16 Plaintiffs and other Washington growers hire a significant number of non17 immigrant foreign laborers under the so-called “H-2A program.” Id. at 15. This 18 program authorizes short-term visas for qualifying individuals “having a residence 19 in a foreign country which [they have] no intention of abandoning who [are] coming 20 temporarily to the United States to perform agricultural labor or services.” 8 U.S.C. 1 § 1101(a)(15)(H)(ii)(a); see also Hispanic Affairs Project v. Acosta, 901 F.3d 387 , 2 382 (D.C. Cir. 2018). 3 To ensure that incoming H-2A laborers do not depress the wages of United 4 States citizens, federal law requires employers who hire H-2A laborers to pay the 5 highest of four possible wages: the adverse effect wage rate (AEWR), any 6 collectively-bargained wage, the applicable state or federal minimum wage, or the 7 prevailing hourly or piece wage rate (PWR). 20 C.F.R. § 655.120 (a). 8 A. The PWR 9 The PWR, which is at issue in this case, is intended to reflect the average, i.e. 10 “prevailing,” wage paid to domestic agricultural laborers, who are engaged in a crop 11 activity like apple harvesting, within a given agricultural region. 29 C.F.R. § 12 502.10(a). The United States Department of Labor (“DOL”)—which administers 13 the H-2A program—sets the PWR. The applicable regulations, however, expressly 14 delegate the task of calculating the PWR to “state workforce agencies,” (SWAs) 15 such as Washington’s Employment Security Department (ESD). Id. The SWAs 16 receive grant funds to survey employers and employees, collect wage data, and 17 calculate what they believe to be the PWR. See ECF No. 29 at 3. The SWAs then 18 submit their results to DOL, which reviews the information and “determines 19 whether the survey results may be validated.” Id. at 5 . If so, the PWR is published, 20 and H-2A employers must pay it immediately, even if the change comes mid1 harvest.1 ECF No. 1 at 44. 2 The dispute in this case centers on the process an SWA must follow in 3 conducting the required wage survey. See 29 C.F.R. § 502.10 . The parties agree that 4 a DOL publication known as “Handbook 385,” is the relevant source of authority.2 5 ECF No. 15 at 5; ECF No. 31 at 5. Handbook 385 is not, however, a model of clarity 6 or precision. For example, it requires SWAs to ensure “that the planned [wage] 7 sample will yield data which will be representative of the wage rates paid in the 8 [specified] crop activity.” ECF No. 30-3 at 4. To that end, it refers SWAs to a 9 “general guide” of minimum sampling requirements—for instance, when a crop 10 activity employs 3000 or more workers in a given area, Handbook 385’s “general 11 guide” provides for a 15% minimum sample size. Id. The Handbook also dictates 12 that the wage sample “should include workers of small, medium and large 13 employers” and states that SWAs should employ “probability sampling methods” [14] 15 1 An H-2A employer notified of an increased PWR must pay the new rate immediately. 20 C.F.R. § 655.120 (b). Moreover, besides filing a lawsuit like this 16 one, an aggrieved H-2A employer has no recourse to challenge the PWR. Id. 17 2 The parties agree that Handbook 385 is binding on SWAs in conducting prevailing wage surveys, and this is consistent with DOL’s statements outside this litigation. 18 See 84 Fed. Reg. 36168 (“Currently, the SWAs are required to conduct prevailing wage surveys using standards set forth in Handbook 385.”). They disagree, 19 however, about whether several requirements imposed by Handbook 385 are mandatory or permissive. Compare ECF No. 40 at 9 (characterizing 15% sample20 size threshold as a “mandatory parameter[]”) with ECF No. 31 at 7 (characterizing the same threshold as a “general guide” (quoting ECF No. 30-3 at 4)). 1 which the Handbook does not prescribe. Id. The Handbook also requires wage 2 surveys to be conducted using a “substantial number” of in-person employer 3 interviews, but qualifies that under “certain conditions, employer contacts by mail 4 or by telephone may be made” so long as “the [SWA] assure[s] itself that 5 information gathered in this manner is representative.” Id. at 6 . It is silent, however, 6 as to the “certain conditions” under which this exception may apply. Id. 7 1. PWR for Apples in Washington 8 DOL first set an hourly PWR for apples in Washington in 2017, and only 9 then for honeycrisp apples, for which H-2A employers were required to pay 10 $15.00/hr. ECF No. 1 at 137. In 2018, no hourly PWR for apples was established. [11] Id. at 141 . In 2019, prior to the notice of increase at issue here, the hourly PWR was 12 still only $15.00/hr. for honeycrisp apples, but a $15.03/hr. AEWR3 was in effect. [13] Id. at 13 ; 83 Fed. Reg. 66307 . As such, when Plaintiffs applied to hire H-2A laborers 14 for the 2019 apple harvest, each agreed to pay at least $15.03/hr., and DOL 15 approved their applications on that condition. See, e.g., ECF No. 29-13 at 3. 16 Critically, Plaintiffs agreed that “if the [PWR] is adjusted during a work contract . . . 17 [each] must pay . . . that higher [PWR] upon notice” from DOL. ECF No. 29-12 at [18] 3 The AEWR represents a “wage floor” designed “to ensure that the wages of 19 similarly employed U.S. workers will not be adversely affected,” and is not tied to a particular commodity. 29 C.F.R. § 502.10 ; AFL-CIO v. Dole, 923 F.2d 182 , 184 20 (D.C. Cir. 1991). In this case, it is the increased PWR that is challenged, and so the technical differences between that figure and the AEWR are unimportant. 1 2. On July 24, 2019, DOL sent just such a notice, informing Plaintiffs that the PWR 2 for high-density4 apple harvesting had been set at $16.00/hr. ECF No. 1 at 44–45. 3 The focus of this litigation is the way DOL arrived at that figure. 4 2. The 2019 PWR Survey 5 The 20195 PWR survey, from which the $16.00/hr. PWR was derived, was 6 conducted by ESD using an online survey, telephone calls, and forms sent through 7 the mail.6 ECF No. 30 at 10–11. Before and after the survey was conducted, ESD 8 met with stakeholders in the Washington agricultural community. ECF No. 30 at 6, 9 23–25. At those meetings, ESD previewed the survey form it planned to use, and 10 later reviewed the PWRs it had calculated. Id. Though invited to do so, none in 11 attendance voiced concerns with the form of the survey or the resulting PWRs. Id. 12 at 24–25. 13 Joshua Moll, an economist at ESD, oversaw the 2019 survey and calculated [14] 4 “High density” is defined as apple harvesting in orchards with more than 800 trees [15] per acre. ECF No. 1 at 120. Plaintiffs challenge two PWRs—one for high-density apple harvesting and one for “harvesting-color-picking,” which is defined as [16] “selectively harvesting fruit based on color or maturity.” Id. Because both PWRs were set at $16.00/hr., unless otherwise noted, the Court’s refers to both [17] interchangeably. [18] 5 Although the survey data used to calculate the PWR at issue here was collected between October 2018 and January 2019, the Court refers to this as the “2019 [19] survey” for convenience. ECF No. 30 at 7. 20 6 ESD interviewed farm laborers in-person, but the information it gathered was not used in setting the PWR. ECF No. 30 at 10. 1 the resulting PWRs. Id. at 1. Mr. Moll applied statistical models to first estimate the 2 number of apple growers with high-density orchards and then extrapolated to an 3 estimate that 11,895 laborers were employed in high-density apple harvesting 4 during the season’s most active week. Id. at 19. Responses to ESD’s survey 5 included wage information representing 1745 of those laborers, or 14.67% of the 6 estimated total.7 Id. at 19. Once the survey period closed, ESD used the method 7 prescribed by Handbook 385 to calculate that the new prevailing wage rate for high8 density apple harvesting was $16.00/hr. Id. at 20–21. ESD then reported its findings 9 to DOL. Id. at 24–25. A DOL analyst confirmed that the sample size of ESD’s 10 survey was adequate, confirmed ESD’s calculation of the PWRs from the survey 11 data, and published the results. ECF No. 29 at 8. On July 24, 2019, prior to the start 12 of the apple harvest, DOL informed Plaintiffs that the PWR had increased, and that 13 they would be required to pay the new rate immediately. Id. at 7–8. 14 3. Plaintiffs’ Claims 15 Plaintiffs contend that DOL’s decision to certify the increased PWR was 16 arbitrary and capricious and should therefore be set aside. ECF No. 1 at 20–35. 17 Plaintiffs make three arguments that ESD reached an inaccurate result and failed to 18 conduct the 2019 wage survey in accordance with Handbook 385. ECF No. 15 at [19] 7 Mr. Moll relied on the same methodology to estimate that there were 7728 workers 20 employed in color-picking harvesting, and collected wage data for 1200 of them, or 15.52% of the estimated total. ECF No. 30 at 20–21. 1 8–11. The Court briefly summarizes each of Plaintiffs’ arguments in turn. 2 a. Unrepresentative PWR 3 Plaintiffs first argue that the $16.00/hr. PWR for high-density apple 4 harvesting is unrepresentative of what Washington growers actually pay. ECF No. 5 15 at 8. They attempt to substantiate this claim with declarations from Washington 6 growers—which Plaintiffs assert in the aggregate represent roughly 75% of the 7 state’s production by tonnage—most of which state they do not pay $16.00/hr. See, 8 e.g., ECF No. 16-2 at 3 (“In 2018, the hourly wage rate paid by [Columbia Orchard 9 Management, Inc.] for harvest of apples, excluding Honeycrisp was $14.12 per 10 hour. In 2018, the hourly wage rate paid [] for harvest of Honeycrisp was 11 $15.00/hr.”). Plaintiffs claim that “[t]his alone shows a strong likelihood of success 12 on the merits.” ECF No. 15 at 8. 13 b. Handbook 385 14 Plaintiffs also argue that ESD deviated from the guidelines set out in 15 Handbook 385 in two ways. 16 i. Failure to Sample 15% of Worker Population 17 Plaintiffs contend that the PWR for high-density apple harvesting was 18 calculated based on a sample size “below the 15% threshold” set out by Handbook 19 385. ECF No. 15 at 9. ESD estimated that the 1745 domestic workers for which it 20 collected wage data represented 14.67% of the labor force and rounded up to report 1 a sample size of 15% when it published the survey results. ECF No. 30 at 19; ECF 2 No. 1 at 105. 3 Plaintiffs also contend that ESD’s use of inappropriate statistical models 4 could have resulted in a dramatic underestimate of the domestic workforce and thus 5 that its sample may have represented far less than 15% of the population. ECF No. 6 15 at 9; ECF No. 40 at 7. Plaintiffs argue that ESD should have instead calculated 7 what percentage of the state’s apple crop each responding employer represented, 8 which Plaintiffs claim would have resulted in a more reliable estimate of the sample 9 size. Id. 10 ii. Failure to Conduct In-Person Interviews 11 In their reply brief, Plaintiffs argued for the first time that ESD’s failure to 12 conduct in-person interviews rendered the PWR unreliable and DOL’s certification 13 of it arbitrary and capricious. ECF No. 40 at 10. This issue was not raised in 14 Plaintiffs’ original motion for a preliminary injunction, ECF No. 15. However, the 15 argument does appear in Plaintiffs’ complaint. See ECF No. 1 at 33–35. Thus, 16 notwithstanding the general rule that the Court “need not consider arguments raised 17 for the first time in a reply brief,” the Court will evaluate the merit of this claim. 18 Zamani v. Carnes, 491 F.3d 990, 997 (9th Cir. 2007). 19 // 20 // 1 LEGAL STANDARD 2 A. Standard for Preliminary Relief 3 To obtain preliminary injunctive relief, the moving party must show (1) a 4 likelihood of success on the merits; (2) that irreparable harm is likely if preliminary 5 relief is denied; (3) that the balance of equities tips in its favor; and (4) that an 6 injunction is in the public interest. Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7 7, 20 (2008). In the Ninth Circuit, the movant may also obtain a preliminary 8 injunction by showing that there are “serious questions going to the merits”—a less 9 demanding showing than likelihood of success on the merits—in cases where the 10 balance of equities tips “sharply” in its favor. All. for the Wild Rockies v. Cottrell, [11] 632 F.3d 1127, 1135 (9th Cir. 2011). 12 B. Standard Under the Administrative Procedures Act 13 Under the Administrative Procedures Act (APA), the Court must invalidate 14 “agency action, findings, and conclusions” that are “arbitrary, capricious . . . or 15 otherwise not in accordance with law.” 5 U.S.C. § 706 (2)(A). 16 Review under the arbitrary and capricious standard is narrow, and [the Court does] not substitute [its] judgment 17 for that of the agency. Rather, [it] will reverse a decision as arbitrary and capricious only if the agency relied on 18 factors Congress did not intend it to consider, entirely failed to consider an important aspect of the problem, or 19 offered an explanation that runs counter to the evidence before the agency or is so implausible that it could not be 20 ascribed to a difference in view or the product of agency expertise. 1 League of Wilderness Defs. Blue Mountains Biodiversity Project v. Allen, 615 F.3d 2 1122, 1130 (9th Cir. 2010) (quoting Lands Council v. McNair, 537 F.3d 981 , 987 3 (9th Cir. 2008)). The Court’s deference is greatest when evaluating “an agency’s 4 technical analyses and judgments involving the evaluation of complex scientific 5 data within the agency’s technical expertise.” Id. Moreover, agency action is not 6 arbitrary and capricious simply because it relies on a “dataset [that] was less than 7 perfect.” Dist. Hosp. Partners, L.P. v. Burwell, 786 F.3d 46, 61 (D.C. Cir. 2015). 8 And under the APA, “[w]hen specialists express conflicting views, an agency must 9 have discretion to rely on the reasonable opinions of its own qualified experts even 10 if, as an original matter, a court might find contrary views more persuasive.” Marsh 11 v. Oregon Nat. Res. Council, 490 U.S. 360, 378 (1989). 12 However, “[a]n agency’s unannounced departure in practice from a written 13 regulation is a distinct form of agency action that is challengeable, separate and 14 apart from adoption of the regulation itself.” Acosta, 901 F.3d at 387 . Thus, while 15 a plaintiff cannot prevail simply by showing that the agency reached an imperfect 16 conclusion, if the agency consistently followed a “general policy by which its 17 exercise of discretion will be governed, an irrational departure from that policy,” 18 rather than “an avowed alteration of it” may be arbitrary and capricious independent 19 of the result the agency reached. I.N.S. v. Yueh-Shaio Yang, 519 U.S. 26, 32 (1996). [20] 1 DISCUSSION 2 A. Plaintiffs Are Not Entitled to a Preliminary Injunction 3 “Crafting a preliminary injunction is an exercise of discretion and judgment, 4 often dependent as much on the equities of a given case as the substance of the legal 5 issues it presents.” Trump v. Int’l Refugee Assistance Project, 137 S. Ct. 2080 , 2087 6 (2017) (citing Winter, 555 U.S. at 20, 24 ). This is particularly true in the case at bar, 7 where Plaintiffs marshal only limited evidence in support of their claims. Although 8 this limited showing might be sufficient if Plaintiffs could establish that the equities 9 tip sharply in their favor, the Court finds that they have failed to do so. All. for the 10 Wild Rockies, 632 F.3d at 1132 . Accordingly, because Plaintiffs have failed to 11 establish likelihood of success on the merits, the Court denies the motion for a 12 preliminary injunction. 13 1. Balance of Equities 14 The balance of equities does not tip sharply in Plaintiffs’ favor. The parties 15 devote substantial argument to comparing the facts of this case to those in a related 16 matter in which the Court recently granted preliminary injunctive relief, Zirkle Fruit 17 Co. v. U.S. Dep’t of Labor, et al., Case No. 1:19-cv-03180-SMJ, ECF No. 49. 18 Plaintiff in that matter, a Washington blueberry grower, was notified of a 50% 19 increase in the applicable PWR in the seventh week of a fifteen-week blueberry 20 harvest. Id. at 5. The Plaintiff credibly contended that, were the increased PWR 1 enforced, it would result in $1,400,000 in virtually unrecoverable labor costs to its 2 business alone. Id. at 22. The Court concluded, based on the “remarkable” facts of 3 that case, that the equities tipped “sharply” in the Plaintiff’s favor. Id. at 21. For that 4 reason, and because the Plaintiff had established “serious questions going to the 5 merits,” the Court granted the requested preliminary injunction. Id. at 21, 24. 6 To some extent, the equities of this case are similar to those in the Zirkle 7 matter. For one, Plaintiffs in both cases challenge similar administrative action— 8 DOL’s certification of PWRs calculated by ESD using the same prevailing wage 9 survey. See id. at 6–10; ECF No. 30 at 8–10. Moreover, there is significant overlap 10 between the claims presented in both cases. Id. And like the Plaintiff in the Zirkle 11 matter, Plaintiffs in this case stand little chance of recovering the wages at issue 12 once paid to their non-immigrant foreign employees. Id. at 14. 13 But there are important—indeed, dispositive—differences between the 14 equities of this case and those in the Zirkle matter. First and most significant is the 15 magnitude of the challenged PWR increase. The Plaintiff in the Zirkle litigation 16 challenged a 50% increase in the PWR, while in this case Plaintiffs face only a 6.5% 17 increase. By way of comparison, farmworker wages nationwide increased 7% 18 between 2018 and 2019, and field worker wages grew by 8% during the same 19 period. See Farm Labor, U.S. Dep’t of Agric. Econ. Research Serv. 3 (2019), 20 https://downloads.usda.library.cornell.edu/usda-esmis/files/x920fw89s/ks65hn7 1 6b/7h14b065g/fmla0519.pdf. Although this data cannot disprove Plaintiffs’ 2 substantive claims, it is relevant to the Court’s task of weighing the equities. 3 Additionally, the increased PWR for high-density apples was announced prior to 4 the beginning of the harvest, whereas in the Zirkle litigation, Plaintiff challenged a 5 mid-harvest increase. See Zirkle Fruit Co., Case No. 1:19-cv-03180-SMJ, ECF No. 6 49 at 5; see also ECF No. 15 at 3. Plaintiffs then waited nearly two months to seek 7 preliminary relief. In light of the foregoing, the Court finds that Plaintiffs have not 8 shown the equities tip sharply in their favor. 9 2. Likelihood of Success on the Merits 10 As set out above, the Court concludes that the equities in this matter do not 11 sharply favor Plaintiffs. They must therefore establish a “likelihood of success on 12 the merits” of their arguments that DOL’s certification of the increased PWR for 13 high-density apple harvesting was arbitrary and capricious. All. for the Wild 14 Rockies, 632 F.3d at 1135 . For the reasons that follow, the Court concludes that 15 they have failed to do so. 16 a. Inaccurate PWR 17 Plaintiffs’ first claim, that the PWR for high-density apple harvesting is 18 unrepresentative of the wages actually paid by Washington growers, is insufficient. 19 To substantiate this claim, Plaintiffs provide declarations of numerous Washington 20 apple growers, the lion’s share of which assert they do not pay $16.00/hr. for high1 density apple harvesting. See, e.g., ECF No. 16-2 at 63 (“At no point in 2018 did 2 [Columbia Orchard Management, Inc.] pay $16.00 (or more) per hour as an hourly 3 wage for the harvest of apples.”). Plaintiffs assert that “[t]his alone shows a strong 4 likelihood of success on the merits.” ECF No. 15 at 8. 5 Under the APA, the Court may not set aside the PWR simply because DOL 6 or ESD “relied on a “dataset [that] was less than perfect.” Burwell, 786 F.3d at 61 . 7 Rather, Plaintiffs bear the burden of showing that DOL’s certification of the PWR 8 “r[an] counter to the evidence before the agency or [was] so implausible that it could 9 not be ascribed to a difference in view or the product of agency expertise.” Allen, 10 615 F.3d at 1130 (quoting McNair, 537 F.3d at 987 ). The Court also notes that most 11 of the declarations on which Plaintiffs rely state that in 2018, growers paid 12 $15.00/hr. for the honeycrisp harvest, and $14.12/hr. for all other apple harvesting. 13 See, e.g., ECF No. 16-2 at 3. These figures are—the Court presumes not 14 coincidentally—the PWR and AEWR in effect during the 2018 harvest when 15 Plaintiffs received clearance orders to hire H-2A laborers. See ECF No. 1 at 137; [16] 82 Fed. Reg. 60629 . Thus, this evidence is most easily interpreted to mean that 17 Plaintiffs agreed to pay as little as legally permissible during the 2018 harvest, and 18 not, as Plaintiffs seem to suggest, that these rates were reflective of what the labor 19 market would bear. ECF No. 15 at 7. Thus, while this evidence may help to inform 20 the Court’s ultimate conclusion that DOL acted in an arbitrary or capricious manner, 1 standing alone it is inadequate to show that Plaintiffs are likely to succeed on the 2 merits. 3 b. Failure to Sample 15% of the Workforce 4 Neither of Plaintiffs’ arguments that ESD relied on an inadequate sample of 5 workers engaged in high-density apple harvesting establishes likelihood of success 6 on the merits. Plaintiffs first argue that ESD’s 14.67% sample size did not conform 7 with the requirements of Handbook 385. ECF No. 30 at 14, 19. Handbook 385 8 dictates that when more than 3000 domestic workers are engaged in a given crop 9 activity, as a “general guide,” a prevailing wage survey should include data 10 representing at least 15% of that population. ECF No. 30-3 at 4. The Court reads 11 this provision to mean that the 15% threshold prescribed by Handbook 385 is not 12 an absolute floor, anything less than which results in an invalid PWR, but rather an 13 approximate guideline. And even if the 15% threshold was an absolute minimum, 14 Plaintiffs have failed to establish that DOL and ESD’s practice of rounding to the 15 nearest whole number—in this case, from 14.67% to 15%—is an arbitrary or 16 capricious deviation from the Handbook’s provisions. See ECF No. 30 at 19. 17 Plaintiffs also argue that the wage information collected by ESD could have 18 represented far less than 15% of the relevant labor force because ESD 19 underestimated the total population—in other words, that the denominator in ESD’s 20 sample size calculation was wrong. ECF No. 40 at 7. In support of this argument, 1 Plaintiffs provide declarations of Stephen Bronars, Ph.D, an economist. See ECF 2 Nos. 17, 42. Dr. Bronars opines that ESD’s calculation of the relevant labor force 3 is a “lower bound” estimate, meaning that “the survey covers no more than 14.67 4 to 15.53% of the relevant population of workers.” ECF No. 17 at 6. Dr. Bronars 5 further opines that the statistical models employed by ESD were unreliable and 6 could have resulted in a substantial underestimation of the workforce. Id. at 5–13. 7 Plaintiffs cannot prevail simply by showing that ESD relied on imperfect data 8 or used an imperfect approach in evaluating the adequacy of its sample size. Dist. 9 Hosp. Partners, L.P., 786 F.3d at 61 . Plaintiffs do not contend that ESD failed 10 entirely to apply a “probability sampling method[]” in conducting the wage survey, 11 but instead argue that its chosen methods were unreliable. ECF No. 40 at 8; ECF 12 No. 30-3 at 4. Notably, although Dr. Bronars devotes significant argument to 13 criticizing ESD’s approach, he nowhere proposes an alternative. See generally ECF 14 Nos. 17, 42. And for his part, Mr. Moll asserts that the models he applied were 15 suitable to the task of estimating the size of the workforce. ECF No. 30 at 17–20. 16 Plaintiffs independently assert that relying on tonnage would be “a mechanism to 17 obtain a more reliable estimate of the relevant workforce than the unreliable 18 estimation methodologies used by ESD,” but cite no authority for this proposition, 19 which they simply label “intuitive.” ECF No. 40 at 7. Nor do Plaintiffs make any 20 attempt to tie this tonnage-based approach to the text of Handbook 385, which 1 directs SWAs to employ “probability sampling methods.” ECF No. 30-3 at 4. 2 The role of judicial review under the APA is not to search for a superior 3 alternative to agency action. Instead, the Court gives great deference to “an 4 agency’s technical analyses and judgments involving the evaluation of complex 5 scientific data within the agency’s technical expertise,” such as ESD’s selection of 6 the statistical models it applied. Allen, 615 F.3d at 1130 (quoting McNair, 537 F.3d 7 at 987). At this stage, Plaintiffs have done no more than present a “specialist[] [who] 8 express[es] conflicting views.” Marsh, 490 U.S. 378 . DOL has discretion, however, 9 “to rely on the reasonable opinions of its own qualified experts,” such as Mr. Moll, 10 “even if, as an original matter, a court might find contrary views more persuasive.” [11] Id. Mindful of the principle that the Court does not “substitute [its] judgment for 12 that of the agency,” the record at this stage is insufficient to find that Plaintiffs are 13 likely to prevail on the merits of this argument. Allen, 615 F.3d at 1130 (quoting 14 McNair, 537 F.3d at 987 ). 15 c. Failure to Conduct In-Person Interviews 16 Plaintiffs’ final argument, that ESD failed to conduct in-person interviews, is 17 also insufficient to warrant preliminary relief. The parties agree that ESD conducted 18 zero in-person interviews in calculating the PWR. ECF No. 30 at 8–9. DOL and 19 ESD argue that Handbook 385 affords SWAs freedom to deviate from the in-person 20 interview requirement, and that ESD’s limited manpower made it necessary to do 1 so. ECF No. 31 at 21–22. Handbook 385 dictates that under “certain conditions, 2 employer contacts by mail or by telephone may be made” if “the [SWA] assure[s] 3 itself that information gathered in this manner is representative.” ECF No. 30-3 at 4 6. But the Handbook does not suggest under what conditions this exception would 5 apply, and neither DOL nor ESD make a compelling attempt to interpret this 6 provision for the Court. The agencies also rely on guidance distributed to SWAs 7 which can be read to suggest that an SWA forced to “prioritize[] its limited 8 resources” may rely on alternative survey means such as telephone contacts as long 9 as doing so “will yield statistically valid wage findings.” ECF No. 30-2 at 20. 10 Neither of these arguments convincingly rebuts Plaintiffs’ argument that the failure 11 to conduct in-person interviews could render the resulting PWR arbitrary or 12 capricious. See Acosta, 901 F.3d at 387 (“An agency’s unannounced departure in 13 practice from a written regulation is a distinct form of agency action that is 14 challengeable.”); see also Yueh-Shaio Yang, 519 U.S. at 32 . 15 However, Plaintiffs have not established how the failure to conduct in-person 16 interviews resulted in an artificially inflated PWR—in short, they lack evidence of 17 causation. Plaintiffs’ complaint merely states that had ESD conducted in-person 18 interviews, “the resulting PWR . . . would have been substantially lower than 19 $16.00 per hour.” ECF No. 1 at 35. But Plaintiffs make no attempt to substantiate 20 this claim. In the Zirkle litigation, the Plaintiff argued that conducting in-person 1 interviews would have resulted in ESD making contact with a more representative 2 sample of blueberry growers, and the Court concluded that this argument was 3 “plausible.” Case No. 1:19-cv-03180-SMJ, ECF No. 49 at 18. But in that case the 4 Court only required evidence of “serious questions going to the merits,” and here 5 Plaintiffs face a more onerous burden. Id. Because they have failed to make a 6 persuasive showing of causation, the Court concludes that they have failed to make 7 the required showing of likelihood of success on the merits of this claim. 8 3. Irreparable Injury & Public Interest 9 Because the Court finds that Plaintiffs have failed to make the required 10 showing that they are likely to succeed on the merits, a preliminary injunction is not 11 warranted. Thus, the Court does not evaluate the remaining two Winter factors. 12 CONCLUSION 13 For the foregoing reasons, the Court finds that Plaintiffs have failed to 14 establish a likelihood of success on the merits of any of their claims that DOL’s 15 certification of the PWR calculated by ESD was arbitrary or capricious. The 16 requested preliminary injunction is therefore unwarranted. 17 Accordingly, IT IS HEREBY ORDERED: 18 1. Plaintiffs’ Motion for Preliminary Injunction, ECF No. 15, is 19 DENIED. [20] 1 IT IS SO ORDERED. The Clerk’s Office is directed to enter this Order and 2 || provide copies to all counsel. 3 DATED this 11th day of October 2019. 4 eee SALVADOR MENDEIz/A, JR. 5 United States District Judge [20] ORDER DENYING PLAINTIFFS’ MOTION FOR PRET TIMINARY INTITINGTION .%1