United Farm Workers v. The United States Department of Labor

Opinion

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 1 of 39

1 2 3 4 5 6 7 8 UNITED STATES DISTRICT COURT 9 FOR THE EASTERN DISTRICT OF CALIFORNIA 10

11 UNITED FARM WORKERS, et al., No. 1:20-cv-01690-DAD-JLT 12 Plaintiffs, 13 v. ORDER GRANTING PLAINTIFFS’ MOTION FOR A PRELIMINARY INJUNCTION 14 THE UNITED STATES DEPARTMENT OF LABOR, et al., (Doc. No. 5)

15 Defendants.

16

17 18 This matter came before the court on December 14, 2020 for hearing on the motion for a 19 preliminary injunction on behalf of plaintiffs United Farm Workers and UFW Foundation 20 (collectively, “plaintiffs”). (Doc. No. 5.) Attorneys Mark Selwyn, Derek Woodman, Nicholas 21 Werle, Bruce Goldstein, Trent Taylor, and Gabriela Hybel appeared by video for plaintiffs, and 22 United States Department of Justice Trial Attorney Michael Gaffney appeared by video for 23 defendants the United States Department of Labor (“DOL”) and Eugene Scalia, the Secretary of 24 Labor (collectively, “defendants”). For the reasons explained below, the court will grant 25 plaintiffs’ motion for a preliminary injunction. 26 BACKGROUND 27 In their complaint, plaintiffs allege the following. Of the two to three million 28 farmworkers currently in the United States, over 200,000 are H-2A foreign guestworkers. (Doc.

2

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 2 of 39

1 No. 1 (“Compl.”) at ¶ 24.) The H-2A agricultural guestworker program permits agricultural 2 employers to hire foreign workers on a temporary basis under certain circumstances. (Id. at ¶ 17.) 3 The H-2A program is rooted in the Immigration and Nationality Act of 1952 (“INA”), which 4 created a broad class of non-immigrant “H” visas for temporary admission of foreign workers to 5 provide temporary or seasonal labor in sectors of the economy where there are shortages of U.S. 6 workers. (Id. at ¶ 32.) The INA was later amended to establish the separate H-2A visa 7 classification for agricultural labor. (Id.) As amended, the INA prohibits the United States 8 Department of Homeland Security from issuing an H-2A visa unless the employer seeking to hire 9 foreign guestworkers has applied for and received a certification from the DOL that: (a) “there 10 are not sufficient workers who are able, willing, and qualified” and available to perform the 11 sought for services, and (b) the foreign workers’ temporary employment “will not adversely 12 affect the wages and working conditions of workers in the United States similarly employed.” 13 (Id. at ¶ 32.) This certification requirement furthers the INA’s purpose of protecting U.S. workers 14 from the potential adverse effects of an influx of guestworkers in that certification prohibits 15 agricultural employers from hiring foreign guestworkers unless they have shown that the U.S. 16 labor market cannot supply the required workers, and then requires that this supplemental, foreign 17 labor supply not harm U.S. farmworkers’ wages and working conditions. (Id. at ¶ 33.) 18 To prevent adverse effects on U.S. workers, the DOL’s regulations require that employers 19 utilizing the H-2A program pay a wage that is the highest of either: (1) the Adverse Effect Wage 20 Rate (“AEWR”), (2) the prevailing wage rate, (3) an agreed-upon collective bargaining wage, or 21 (4) the federal or state minimum wage. (Id. at ¶ 35.) Under those regulations, the DOL relies 22 primarily on a two-pronged approach based on the AEWR and prevailing wage rate to guard 23 against wage depression that would otherwise result from the hiring of high numbers of foreign 24 agricultural workers. (Id.) The prevailing wage rate protects local wages paid, while the AEWR 25 sets a state-wide wage floor to prevent wage disparities over larger geographic areas. (Id. at 26 ¶ 36.) The AEWR, however, is the primary wage rate under the H-2A program because it is 27 higher than the other wages in most circumstances. (Id.) As a result, the AEWR determines the 28 /////

3

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 3 of 39

1 wages of approximately 92 percent of the farmworkers working for H-2A program employers. 2 (Id.) 3 Prior to December 21, 2020, the DOL’s regulations required the DOL to use the United 4 States Department of Agriculture’s (“USDA”) Agricultural Labor Survey, commonly referred to 5 as the Farm Labor Survey (“FLS”), in order to calculate the AEWR. (Id. at ¶ 37.) The USDA 6 has conducted the FLS since 1910. (Id. at ¶ 52.) The FLS collects information from farm 7 employers to obtain data on farm employment, hours worked, wages paid, and other statistics. 8 (Id.) The National Agricultural Statistics Service (“NASS”)—the USDA’s statistical branch— 9 publishes FLS data semiannually in May and November in the Farm Labor Report (“FLR”). The 10 May report includes employment and wage estimates based on January and April reference 11 weeks, and the November report includes estimates based on July and October reference weeks. 12 (Id. at ¶ 53.) The November report also provides annual data based on quarterly estimates. (Id.) 13 Aside from a brief two-year period starting in 2008, the DOL’s regulations required it to 14 use the FLS to calculate the AEWR for the H-2A program since the program’s inception in 1986. 15 (Id. at ¶ 37.) The DOL had also used FLS data for the H-2A’s predecessor program since 1953. 16 (Id.) Because of the DOL’s longstanding reliance on the FLS, the USDA has conducted the FLS 17 in cooperation with the DOL, and the DOL has funded the FLS since July 2011 pursuant to a 18 memorandum of understanding between the two agencies. (Id.) Indeed, in a December 2019 19 memorandum of understanding between the DOL and the USDA, the DOL agreed to continue 20 funding the FLS through December 31, 2022. (Id. at ¶ 56.) 21 Under the DOL’s prior regulations, which were adopted in 2010, the DOL sets an AEWR 22 for each state or multi-state region using “[t]he annual weighted average hourly wage for field 23 and livestock workers (combined) . . . as published annually by the U.S. Department of 24 Agriculture . . . based on its quarterly wage survey,” the FLS. (Id. at ¶ 38); see also Temporary 25 Agricultural Employment of H-2A Aliens in the United States, 75 Fed. Reg. 6,884 (Feb. 12, 2010) 26 (“the 2010 Rule”). That 2010 Rule explained that the AEWR seeks to approximate the market 27 wages that would exist absent an influx of foreign workers, thus “put[ting] incumbent farm 28 workers in the position they would have been in but for the H-2A program.” (Compl. at ¶ 39.)

4

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 4 of 39

1 The DOL elucidated that the AEWR was premised on the idea that “an increase of workers under 2 the H-2A program” would prevent wages from “increas[ing] by an amount sufficient to attract 3 more [U.S.] workers until supply and demand were met in equilibrium.” (Id.) In other words, 4 “the AEWR avoids adverse effects on currently employed workers by preventing wages from 5 stagnating at the local prevailing wage rate when they would have otherwise risen to a higher 6 equilibrium level over time.” (Id.) The DOL has recognized that without the protections afforded 7 by AEWRs set at regional or state-wide market rates, farmworkers “would be adversely affected 8 by lowered wages as a result of an influx of temporary foreign farm workers.” (Id.) 9 In adopting the 2010 Rule, the DOL also concluded that the FLS was the best available 10 data source for establishing AEWRs. (Id. at ¶ 40.) The DOL explained at that time that “[t]he 11 FLS is the only annually available data source that actually uses information sourced directly 12 from [farm employers],” which “is a strong advantage of the FLS as the AEWR data source 13 compared to all other alternatives.” (Id.) Additionally, the FLS’s “broader geographic scope 14 makes the FLS more consistent with both the nature of agricultural employment and the statutory 15 intent of the H-2A program.” (Id.) In short, FLS data was best suited to protect against adverse 16 effects because it allowed the DOL to establish AEWRs at regional market rates. (Id.) 17 Conversely, the DOL recognized that using data other than the FLS to calculate AEWRs—in that 18 case, the DOL’s Occupational Employment Statistics (“OES”) survey data—“entails a significant 19 risk that U.S. workers may in the future experience wage depression as a result of unchecked 20 expansion of the demand for foreign workers.” (Id.) 21 On July 26, 2019, the DOL published a Notice of Proposed Rulemaking (“NPRM”) 22 proposing to continue its reliance on FLS data to establish AEWRs under the H-2A program. (Id. 23 at ¶ 43); see also Temporary Agricultural Employment of H-2A Nonimmigrants in the United 24 States, 84 Fed. Reg. 36,168 (July 26, 2019) (“the NPRM”). Specifically, the NPRM proposed to 25 establish separate AEWRs for distinct agricultural occupations within each state or region relying 26 on FLS data. (Id. at ¶ 44.) If the FLS did not report a wage for a specific occupation in a given 27 state or region, the AEWR would instead be based on OES data. (Id.) Further, if OES data did 28 not include a statewide annual average hourly wage for a standard occupational classification,

5

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 5 of 39

1 then the AEWR would be based on the national wage for that occupational classification as 2 determined by the FLS or OES. (Id.) 3 However, on September 30, 2020, the USDA abruptly announced that it had suspended 4 data collection for the October 2020 FLS and canceled the November 2020 publication of the 5 biannual FLR. (Id. at ¶ 55); see also Notice of Revision to the Agricultural Labor Survey and 6 Farm Labor Reports by Suspending Data Collection for October 2020, 85 Fed. Reg. 61,719 7 (Sept. 30, 2020) (“FLS Suspension Notice”). On October 13, 2020, plaintiffs sued the USDA 8 seeking a temporary restraining order and preliminary injunction preventing the USDA from 9 implementing the FLS Suspension Notice. (Compl. at ¶ 57); see also United Farm Workers v. 10 Perdue, No. 1:20-cv-1452-DAD-JLT, 2020 WL 6318432 (E.D. Cal. Oct. 28, 2020). In that case, 11 plaintiffs argued that the USDA’s decision was arbitrary and capricious—largely because it failed 12 to consider the DOL’s reliance on the FLS data—and that it failed to comply with the APA’s 13 notice-and-comment requirements. (Compl. at ¶ 57.) Plaintiffs also argued that H-2A foreign 14 guestworkers and U.S. farmworkers would be irreparably harmed if FLS data from 2020 and the 15 November 2020 FLR were not available for the DOL to use in establishing the 2021 AEWRs. 16 (Id.) On October 28, 2020, this court granted plaintiffs’ motion and enjoined the USDA from 17 canceling the October 2020 FLS and ceasing publication of the November 2020 FLR. (Id. at 18 ¶ 58); see also Perdue, 2020 WL 6318432. 19 On November 5, 2020, the DOL published a final rule in the Federal Register announcing 20 changes to its methodology for setting AEWRs under the H-2A program. (Compl. at ¶ 4); see 21 also Adverse Effect Wage Rate Methodology for the Temporary Employment of H-2A 22 Nonimmigrants in Non-Range Occupations in the United States, 85 Fed. Reg. 70,445 (Nov. 5, 23 2020) (“the Final Rule”). The Final Rule became effective on December 21, 2020. (Compl. at 24 ¶ 4.) In that Final Rule, the DOL elected to freeze the current 2020 AEWRs for two years. (Id. at 25 ¶ 61.) The 2020 AEWRs are based on FLS data reflecting what farmworkers were paid in 2019. 26 (Id.) The Final Rule maintains these 2019 wage rates as the AEWRs for most agricultural jobs in 27 2021 and 2022. (Id.) Then, beginning in 2023, the DOL would adjust the AEWRs annually 28 using the Employment Cost Index (“ECI”)—an index that measures the change in the cost of

6

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 6 of 39

1 labor by surveying various private industries, but notably excluding farms and agricultural 2 workers. (Id.) Under the Final Rule, the DOL will also establish AEWRs each year—including 3 in 2021 and 2022—for a smaller set of “higher-skilled” agricultural jobs using the annual 4 statewide average hourly gross wage for the occupation based on the OES survey—a program 5 administered by the DOL’s Bureau of Labor Statistics that excludes farms from its data 6 collection. (Id.) 7 On November 30, 2020, plaintiffs filed their complaint against defendants in this action 8 seeking declaratory and injunctive relief and asserting the following three claims: (1) a violation 9 of the Administrative Procedure Act (“APA”), 5 U.S.C. § 706, because the Final Rule 10 contravenes the governing statute; (2) a violation of the APA, 5 U.S.C. § 706, because the Final 11 Rule lacks adequate justification and analysis of the economic effects it will have on U.S. 12 workers; and (3) a violation of the APA, 5 U.S.C. § 553, because defendants failed to comply 13 with the requirements of notice-and-comment rulemaking in promulgating the Final Rule. 14 (Compl. at 37–39.) Plaintiffs filed the pending motion for a preliminary injunction on November 15 30, 2020. (Doc. No. 5.) On December 7, 2020, defendants filed their opposition to plaintiffs’ 16 motion, and plaintiffs replied thereto on December 11, 2020. (Doc. Nos. 31, 34.) The California 17 Attorney General’s Office filed an amicus curiae brief on behalf of the State of California in 18 support of plaintiffs’ pending motion for a preliminary injunction on December 9, 2020.1 (Doc. 19 No. 32-1.) 20 LEGAL STANDARD 21 “The proper legal standard for preliminary injunctive relief requires a party to demonstrate 22 ‘that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the 23 absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction 24 is in the public interest.’” Stormans, Inc. v. Selecky, 586 F.3d 1109, 1127 (9th Cir. 2009) (quoting 25 Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008)); see also Ctr. for Food Safety v. 26 Vilsack, 636 F.3d 1166, 1172 (9th Cir. 2011) (“After Winter, ‘plaintiffs must establish that

27 1 The court granted the State of California’s motion for leave to file its amicus curiae brief on 28 December 10, 2020. (Doc. No. 33.)

7

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 7 of 39

1 irreparable harm is likely, not just possible, in order to obtain a preliminary injunction.’”) 2 (quoting All. for Wild Rockies v. Cottrell, 632 F.3d 1127, 1131 (9th Cir. 2011)). The Ninth 3 Circuit has also held that an “injunction is appropriate when a plaintiff demonstrates . . . that 4 serious questions going to the merits were raised and the balance of hardships tips sharply in the 5 plaintiff’s favor.” All. for Wild Rockies, 632 F.3d at 1134–35 (quoting Lands Council v. McNair, 6 537 F.3d 981, 987 (9th Cir. 2008) (en banc), overruled on other grounds by Winter, 555 U.S. 7.2 7 The party seeking the injunction bears the burden of proving these elements. See Klein v. City of 8 San Clemente, 584 F.3d 1196, 1201 (9th Cir. 2009); Caribbean Marine Servs. Co. v. Baldrige, 9 844 F.2d 668, 674 (9th Cir. 1988) (“A plaintiff must do more than merely allege imminent harm 10 sufficient to establish standing; a plaintiff must demonstrate immediate threatened injury as a 11 prerequisite to preliminary injunctive relief.”). Finally, an injunction is “an extraordinary remedy 12 that may only be awarded upon a clear showing that the plaintiff is entitled to such relief.” 13 Winter, 555 U.S. at 22. 14 ANALYSIS 15 Here, plaintiffs allege various violations of the APA and seek a preliminary injunction (1) 16 preventing the DOL from implementing the regulatory changes announced in its Final Rule 17 published on November 5, 2020, and (2) ordering the DOL to issue AEWRs. (Doc. No. 5 at 33.) 18 “The APA sets forth the procedures by which federal agencies are accountable to the 19 public and their actions subject to review by the courts.” Dep’t of Homeland Sec. v. Regents of 20 the Univ. of California, __U.S.__, 140 S. Ct. 1891, 1905 (2020) (internal quotation marks and 21 citation omitted). Only “final agency actions” are reviewable under the APA. 5 U.S.C. § 704; 22 see also 5 U.S.C. § 701 (for purposes of the APA’s judicial review provisions, “agency action” 23 has “the meaning[] given” by § 551). An “‘agency action’ includes the whole or a part of an 24 agency rule, order, license, sanction, relief, or the equivalent or denial thereof, or failure to act.”

25 2 The Ninth Circuit has found that this “serious question” version of the circuit’s sliding scale 26 approach survives “when applied as part of the four-element Winter test.” All. for the Wild Rockies, 632 F.3d at 1134. “That is, ‘serious questions going to the merits’ and a balance of 27 hardships that tips sharply towards the plaintiff can support issuance of a preliminary injunction, so long as the plaintiff also shows that there is a likelihood of irreparable injury and that the 28 injunction is in the public interest.” Id. at 1135.

8

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 8 of 39

1 5 U.S.C. § 551(13). Under § 706 of the APA, the court is “to assess only whether the decision 2 was based on a consideration of the relevant factors and whether there has been a clear error of 3 judgment.” Regents, 140 S. Ct. at 1905 (internal quotation marks and citation omitted). 4 The APA “requires agencies to engage in reasoned decisionmaking, and directs that 5 agency actions be set aside if they are arbitrary or capricious.” Regents, 140 S. Ct. at 1905 6 (internal citations and quotation marks omitted). An agency’s “determination in an area 7 involving a ‘high level of technical expertise’” is to be afforded deference. McNair, 537 F.3d at 8 993 (citing 5 U.S.C. § 706(2)(A)). The district court’s role “is simply to ensure that the [agency] 9 made no ‘clear error of judgment’ that would render its action ‘arbitrary and capricious.’” Id. 10 (citing Marsh v. Or. Nat. Res. Council, 490 U.S. 360, 378 (1989)). “Factual determinations must 11 be supported by substantial evidence,” and “[t]he arbitrary and capricious standard requires ‘a 12 rational connection between facts found and conclusions made.’” League of Wilderness 13 Defs./Blue Mountains Biodiversity Project v. Connaughton, 752 F.3d 755, 759–60 (9th Cir. 2014) 14 (internal citations omitted).

15 This requires the court to ensure that the agency has not, for instance, “relied on factors which Congress has not intended it to consider, 16 entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence 17 before the agency, or [an explanation that] is so implausible that it could not be ascribed to a difference in view or the product of agency 18 expertise.” 19 McNair, 537 F.3d at 987 (quoting Motor Vehicle Mfrs. Assn., Inc. v. State Farm Mut. Auto. Ins.

20 Co., 463 U.S. 29, 43 (1983)). 21 As noted, plaintiffs must make a sufficient showing as to all four prongs of the Winter test 22 in order to be entitled to the requested preliminary relief. All. for the Wild Rockies, 632 F.3d at 23 1135. The court will begin its analysis by considering plaintiffs’ likelihood of success on their 24 claims. 25 A. Likelihood of Success on the Merits 26 Plaintiffs bear the burden of demonstrating that they are likely to succeed on the merits of 27 this action or, at the very least, that “serious questions going to the merits were raised.” All. for 28 the Wild Rockies, 632 F.3d at 1131.

9

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 9 of 39

1 1. Whether the Final Rule is Arbitrary and Capricious Because It Contravenes Federal Law by Failing to Protect United States Workers Against Adverse Effects 2

3 In their first claim, plaintiffs allege that the Final Rule is arbitrary and capricious because 4 it contravenes the INA’s mandate that the DOL ensure that the hiring of temporary foreign 5 guestworkers “will not adversely affect the wages and working conditions of workers in the 6 United States similarly employed.” (Compl. at ¶¶ 99–103.) 7 “Agencies cannot exceed the scope of their authority as circumscribed by Congress.” 8 Planned Parenthood of Greater Washington & N. Idaho v. U.S. Dep’t of Health & Human Servs., 9 946 F.3d 1100, 1112 (9th Cir. 2020). Under § 706(2)(A) of the APA, regulations that contravene 10 federal law or that are “‘contrary to clear congressional intent’” must be declared invalid and set 11 aside. Id. “When reviewing an agency’s statutory interpretation under the APA’s ‘not in 12 accordance with law’ standard,” the court must “adhere to the familiar two-step test of Chevron.” 13 Nw. Envtl. Advocates v. U.S. E.P.A., 537 F.3d 1006, 1014 (9th Cir. 2008). At Chevron step one, 14 “if Congress ‘has directly spoken to the precise question at issue . . . the court, as well as the 15 agency, must give effect to the unambiguously expressed intent of Congress.’” Harkonen v. U.S. 16 Dep’t of Justice, 800 F.3d 1143, 1149 (9th Cir. 2015) (quoting Chevron, U.S.A., Inc. v. NRDC, 17 467 U.S. 837, 842 (1984)). If the court determines that the statute is ambiguous with respect to 18 the precise question at issue, and Congress therefore left a gap for the agency to fill, the court 19 must proceed to Chevron step two and ask “whether the agency’s answer is based on a 20 permissible construction of the statute.” Id. (quoting Chevron, 467 U.S. at 843). While the 21 standard of review at Chevron step two is highly deferential to the agency determination, id., the 22 analysis is the same test applied to agency changes in policy: that is, “[a] permissible 23 construction is one that is not ‘arbitrary, capricious, or manifestly contrary to the statute.’” Altera 24 Corp. & Subsidiaries v. Comm’r of Internal Revenue, 926 F.3d 1061, 1075 (9th Cir. 2019), cert. 25 denied, No. 19-1009, 2020 WL 3405861 (U.S. June 22, 2020). 26 Here, the court finds that plaintiffs have demonstrated a likelihood of success on the 27 merits of this claim. The undersigned finds the decision in American Federation of Labor & 28 Congress of Industrial Organizations (AFL-CIO) v. Brock, 835 F.2d 912 (D.C. Cir. 1987) to be

10

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 10 of 39

1 instructive under the circumstances presented here. There, the plaintiffs challenged the DOL’s 2 new methodology for computing the AEWRs as both contrary to congressional intent and 3 arbitrary and capricious. Id. at 914. Over the preceding twenty years, the DOL had periodically 4 increased the AEWRs to compensate for past adverse wage effects. Id. Based on the assumption 5 that farm wages had stagnated due to the influx of foreign guestworkers, the DOL had linked 6 AEWRs to manufacturing wages and enhanced those wages by a USDA data-based index, 7 producing AEWRs that exceeded farm wages by approximately 20 percent. Id. In 1987, the 8 DOL issued a new methodology that set AEWRs for all states equal to the average hourly 9 agricultural wages paid the prior year. Id. Because farmworkers faced possible wage cuts under 10 the new rule, plaintiffs challenged it. Id. 11 After the district court invalidated the new rule, the D.C. Circuit reversed. The court first 12 determined that

13 Congress, indeed, has never paid any attention to the method or policy of calculating AEWRs. . . . [C]alculating AEWRs has been 14 left entirely to the Department’s discretion. The committee and floor discussion on the IRCA, cited by both parties, confirms only 15 Congress’ general intent to protect United States workers against adverse effects from imported labor.

16

17 Id. at 915. At Chevron step two, however, the court found that the new AEWR rule was not a 18 permissible construction of the statute. Id. at 917–19. “Because the Department ma[de] no 19 explanatory attempt to ‘forthrightly distinguish or outrightly reject’ [its] contradictory precedent, 20 [the court] and the public [we]re ‘left with no guideposts for determining the consistency of 21 administrative action . . . or for accurately predicting future action’ by the agency.” Id. at 918. In 22 discounting the DOL’s arguments that its previous AEWR methodology caused wage rate 23 anomalies and was difficult to calculate, the court stated that

24 [i]f [the DOL] is saying that there is no wage depression from past foreign workers, it must make that case forthrightly. Inability to 25 secure persuasive data as to any effects of past wage depression might indeed justify ending the enhancement or contribute to such a 26 decision. But the example given by the Department could just as logically suggest that adjustments were needed, but in an upward 27 direction rather than a total elimination. The Department simply does not explain why such variances justify the Department’s total 28 abandonment of its policy of enhancing AEWRs to compensate for

11

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 11 of 39

1 past wage depressions rather than changes in methodology to assure more accurate estimates. Thus, even if the Department does have a 2 case for changing its premises about adverse wage effects on American workers, that case has yet to be made.

3

4 Id. at 919. 5 As an initial matter, the court agrees with the D.C. Circuit’s reasoning expressed in the 6 Brock decision as it relates to Chevron step one. Defendants assert in this case, and the court 7 agrees, that the INA does not prescribe the methodology for calculating the AEWR—or even 8 require that the DOL set an AEWR—but instead broadly delegates to the DOL the responsibility 9 to craft a mechanism to certify that hiring H-2A workers “will not adversely affect the wages of 10 workers in the United States similarly employed.” (See Doc. No. 31 at 19) (quoting 8 U.S.C. § 11 1188(a)(1)(B)); United Farm Workers v. Solis, 697 F. Supp. 2d 5, 9 (D.D.C. 2010) (rejecting a 12 challenge to the DOL’s 2008 AEWR rule and stating “Congress did not . . . define adverse effect 13 and left it in the [DOL’s] discretion how to ensure that the importation of farmworkers met the 14 statutory requirements”). 15 Accordingly, the court must move to the second question under Chevron: whether the 16 DOL’s Final Rule is a permissible construction of the INA. The gravamen of plaintiffs’ claim in 17 this case is that the Final Rule’s determination to permit employers to pay guestworkers below 18 market wage rates is contrary to Congress’s clear intent because the wages of U.S. farmworkers 19 will decline or stagnate in response. (Doc. No. 5 at 15–17.) Plaintiffs note that the DOL has 20 historically recognized that while admitting H-2A workers can address a labor shortage, “[a]bsent 21 an increase of workers under the H-2A program, wages would rise above the currently observed 22 wage in order to dispel the labor shortage.” (Doc. No. 5 at 15) (quoting 75 Fed. Reg. at 6,891). 23 Indeed, the DOL’s 2010 Rule explained that AEWRs guard against wage stagnation “[b]y 24 computing an AEWR to approximate the equilibrium wage that would result absent an influx of 25 temporary foreign workers, . . . put[ting] incumbent farm workers in the position they would have 26 been in but for the H-2A program.” (Doc. No. 5 at 15) (quoting 75 Fed. Reg at 6,891–92). That 27 approach, the DOL explained, “avoids adverse effects on currently employed workers by 28 /////

12

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 12 of 39

1 preventing wages from stagnating at the local prevailing wage rate when they would have 2 otherwise risen to a higher equilibrium level over time.” (Id.) (quoting 75 Fed. Reg at 6,891–92). 3 According to plaintiffs, the Final Rule reaffirms this reasoning while nonetheless adopting 4 a completely contrary policy. As plaintiffs characterize it, the Final Rule severs the relationship 5 between the AEWR and current farm labor market conditions; freezes the AEWRs for two years 6 at 2020 levels, even though recent trends establish that agricultural wages have been rising and 7 will likely continue to do so; then lifts the wage freeze in 2023 based on the ECI, a generic index 8 of wages from across the economy, without ever compensating for the wage growth lost during 9 the two-year wage freeze. (Doc. No. 5 at 16.) At the hearing on the pending motion, plaintiffs 10 explained that the historical trend of the FLS data conclusively shows that the Final Rule will 11 result in reduced wages. The May 2020 FLS data also indicates that the AEWRs are increasing 12 as they have been over the last number of years and doing so at a faster rate than the general wage 13 index. (See Doc. No. 5-3.) Plaintiffs also note that employers’ complaints about increasing H-2A 14 wages reflect that AEWRs were expected to be higher in 2021 than in 2020. (Doc. No. 5 at 21.) 15 Thus, plaintiffs aver that even AEWRs based upon the May 2020 FLS data would be higher than 16 AEWRs based on 2019 data. Moreover, plaintiffs argue that the ECI’s lack of farm labor data is 17 particularly significant because agricultural wages have been rising faster than average wages 18 across the economy. (Doc. No. 5 at 16) (quoting 85 Fed. Reg. at 70,455). Further, because the 19 ECI reflects national trends, not state-wide markets, the new methodology under the Final Rule 20 does not track local labor market conditions. (Id. at 17.) 21 Defendants counter that the Final Rule “strikes a reasonable balance” between “providing 22 employers with an adequate legal supply of agricultural labor while protecting the wages and 23 working conditions of” U.S. workers. (Doc. No. 31 at 20) (quoting 85 Fed. Reg. at 70,453). At 24 the hearing on the pending motion, plaintiffs characterized that balance as a clear attempt to 25 protect employers, an objective not contemplated by the INA. The court notes, however, that 26 defendants are obligated to “serve the interests of both farmworkers and growers—which are 27 often in tension,” and “[t]hat is why Congress left it to the DOL’s judgment and expertise to 28 strike the balance.” See Am. Fed’n of Labor & Cong. of Indus. Organizations (AFL-CIO) v.

13

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 13 of 39

1 Dole, 923 F.2d 182, 187 (D.C. Cir. 1991). Defendants contend that they strike that balance by 2 disaggregating the wage rate paid to farmworkers of different occupational categories. (Doc. No. 3 31 at 20.) Defendants posit that the new methodology provides significant wage increases for 4 highly skilled workers, pointing to a chart attached to their opposition brief demonstrating that 5 wages for construction laborers, first-line supervisors of farm workers, and heavy trucking/truck 6 drivers will increase significantly under the Final Rule in comparison to the methodology to be 7 employed under the 2010 Rule.3 (Doc. No. 31-1 at ¶12) (citing 85 Fed. Reg. at 70,458–59). As 8 for field and livestock workers, defendants discount plaintiffs’ assertion that farmworkers’ wages 9 will decline or stagnate, noting that the ECI has never gone down year-over-year, whereas the 10 FLS has. (Doc. No. 31 at 21) (citing Doc. No. 31-3). Defendants emphasize that wages will be 11 frozen and not decline for the next two years and are expected to only increase after that. 12 In the end, the court finds defendants’ arguments to be unpersuasive. “[T]he requirement 13 that an agency provide reasoned explanation for its action would ordinarily demand that it display 14 awareness that it is changing position.” F.C.C. v. Fox Television Stations, Inc., 556 U.S. 502, 15 515–16 (2009). The agency “must show that there are good reasons for the new policy,” but it 16 need not always show “to a court’s satisfaction that the reasons for the new policy are better than 17 the reasons for the old one.” Id. It does have to make such a showing, however, “[w]hen the new 18 policy rests upon factual findings that contradict those which underlay its prior policy; or when its 19 prior policy has engendered serious reliance interests that must be taken into account.” Id. 20 Here, defendants have failed to “‘forthrightly distinguish[] or outrightly reject[]’ [the] 21 contradictory precedent” as required since the Final Rule departs from accurate livestock and 22 fieldworker market wage data and intentionally depresses and stagnates the wages of those 23 workers. Brock, 835 F.2d at 918. While the DOL “is obliged to balance the competing goals of 24 the statute—providing an adequate labor supply and protecting the jobs of domestic workers”— 25 that balance must be supported by at least adequate reasoning. Dole, 923 F.2d at 187. In the

26 3 Attached to defendants’ opposition brief is the declaration of Brian Pasternak, the current 27 Administrator of the Office of Foreign Labor Certification, Employment and Training Administration (“OFLC”) at the DOL. (Doc. No. 31-1 at ¶ 1.) The Pasternak Declaration 28 appends the referenced chart. (Id. at ¶ 12; see also id. at 5.)

14

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 14 of 39

1 2010 Rule, the DOL recognized that lower-skill workers are most likely to suffer adverse wage 2 effects. 75 Fed. Reg. at 6,894 (“Additional research not previously considered suggests that any 3 adverse wage effects would be more likely to affect lower-skill workers.”). The DOL explicitly 4 “consider[ed] the forward-looking need to protect U.S. workers whose low skills make them 5 particularly vulnerable to even relatively mild—and thus very difficult to capture empirically– 6 wage stagnation or deflation that has the potential to result from the hiring of immigrant labor.” 7 Id. at 6,893. The current Final Rule also explicitly recognizes that the lower-skill work, which is 8 field and livestock work, make up most H-2A job opportunities. 85 Fed. Reg. at 70,461 (“[T]he 9 overwhelming majority of H-2A job opportunities . . . fall within the FLS field and livestock 10 workers (combined) category.”). 11 Yet the Final Rule skirts analyzing its impact on those U.S. workers who make up most of 12 the target demographic that the INA mandates the DOL to protect. The Final Rule both 13 “acknowledges the concerns of some commenters that fluctuating wages can be harmful to 14 workers, and their concerns that changes to the methodology could result in stagnating or 15 decreasing wages for farmworkers,” and it “recognizes the possibility that the revised 16 methodology in this final rule may result in the AEWRs for field workers and livestock workers 17 being set at slightly lower levels in future years than would be the case under the [2010 Rule’s] 18 methodology.” 85 Fed. Reg. 70,453. The Final Rule frames its new methodology as “more 19 accurately calculat[ing] than before the wages of certain more highly skilled farmworkers, for 20 which the Department has reason to believe the AEWRs have artificially depressed wages.” Id. at 21 70,454. Indeed, the DOL justifies the Final Rule’s methodology by expressing the concern that 22 the 2010 Rule’s methodology “may have an adverse effect on the wages of workers in higher paid 23 agricultural occupations . . . whose wages may be inappropriately lowered by an AEWR 24 established from the wages of field and livestock workers (combined), an occupational category 25 from the FLS that does not include those supervisory workers.” Id. at 70,470. While defendants 26 choose to characterize this as “striking a reasonable balance,” the court finds that this analysis 27 most certainly tips heavily to one side. It does so despite the fact that defendants are required to 28 protect against adverse effects on workers generally. Where, as here, the DOL has historically

15

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 15 of 39

1 acknowledged that most adverse wage effects impact field and livestock workers, the agency is 2 required to justify a Final Rule that predominantly benefits higher-skilled workers’ wages while 3 “slightly lower[ing] wages” for field and livestock workers. However, the Final Rule fails even 4 to “display awareness that it is changing position” by intentionally deviating from the most 5 accurate data on field and livestock worker market wages. See Fox Television Stations, Inc., 556 6 U.S. at 515–16. 7 For these reasons, and for reasons explained in further detail below, the court is persuaded 8 that plaintiffs have established that they are likely to prevail on their claim that the Final Rule is 9 arbitrary and capricious because it is unsupported by and lacks adequate reasoning. At the very 10 least, there can be no doubt that plaintiffs have raised serious questions about the Chevron step 11 two analysis: whether the Final Rule is a permissible construction of the INA.

12 2. Whether the Final Rule is Arbitrary and Capricious Because the DOL Failed to Offer a Reasoned Explanation in the Final Rule 13

14 In their second claim, plaintiffs allege that the DOL has failed to offer a reasoned 15 explanation in the Final Rule, and thus the Final Rule is arbitrary and capricious. (Doc. No. 5 at 16 17–25.) It is “a fundamental requirement of administrative law . . . that an agency set forth its 17 reasons for decision; an agency’s failure to do so constitutes arbitrary and capricious agency 18 action.” Amerijet Int’l, Inc. v. Pistole, 753 F.3d 1343, 1350 (D.C. Cir. 2014). “[C]onclusory 19 statements will not do; an agency’s statement must be one of reasoning.” Id. (internal quotation 20 marks omitted); see also Dep’t of Commerce, 139 S. Ct. at 2569 (an agency must “articulate[] a 21 satisfactory explanation for [its] decision”). 22 Here, plaintiffs assert that the DOL’s Final Rule failed to offer a reasoned explanation 23 with respect to three obvious issues: (1) justification for its imposition of a two-year wage freeze; 24 (2) the adoption of a generic wage index to adjust AEWRs starting in 2023, contradicting the 25 Final Rule’s emphasis elsewhere regarding the need to rely upon accurate data on agricultural 26 labor markets; and (3) its failure to analyze the economic effects of the new AEWR methodology 27 on U.S. workers. (Compl. at ¶¶ 104–10.) The court will consider each argument in turn. 28 /////

16

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 16 of 39

1 a. Whether the DOL Explained Its Decision to Freeze H-2A Wages for Two Years

2

3 Plaintiffs argue that the Final Rule arbitrarily freezes H-2A wages for two years, 4 stagnating wages in a manner wholly at odds with the AEWR’s purpose and the DOL’s statutory 5 mandate. (Doc. No. 5 at 18.) They contend that the evidence, as described above, demonstrates 6 that farmworkers’ market wages are already above the level of the wage freeze. (Doc. No. 5 at 7 21) (citing 85 Fed. Reg. at 70,456). Plaintiffs also object to the Final Rule’s characterization that 8 the USDA’s FLS Suspension Notice has created uncertainty about the FLS, noting that the Final 9 Rule does not acknowledge that FLS data will be available because this court enjoined the 10 application of the FLS Suspension Notice.4 (Doc. No. 5 at 18–19) (citing 85 Fed. Reg. at 70,446, 11 70,456); see also Perdue, 2020 WL 6318432. 12 In opposition, defendants assert that the predictability offered by the wage freeze followed 13 by reliance on the ECI is beneficial for farmers and farmworkers alike. (Doc. No. 31 at 23.) In 14 this regard, defendants reason that “[f]or farmers, large and unpredictable wage fluctuations make 15 it harder to plan and ‘ultimately discourage domestic agriculture production,’ which harms U.S. 16 farmworkers who depend on robust agricultural output in the United States.” (Id.) (citing 85 Fed. 17 Reg. at 70,452). Additionally, defendants explain “farmworkers can be assured that wages will 18 not decline from one year to the next and that, beginning in 2023, they will steadily increase— 19 from a starting point well above the minimum wage.” (Id. at 23–24.) Defendants also contend 20 that “[there is nothing contradictory about defending the rigor of the FLS and deciding its use is, 21 on balance, not appropriate in this context.” (Id. at 25.) 22 4 23 In their opposition, defendants argue that the DOL did not act unreasonably in predicting that it was at least possible that new FLS data might not be available by the time the OFLC 24 Administrator needed to publish new rates on December 31, 2020. (Doc. No. 31 at 25.)

Defendants state that the Final Rule “acknowledged this Court’s injunction requiring USDA to 25 resume data collection and publication,” but it remained concerned with the timing of those 26 resumed efforts. (Id.) (citing 85 Fed. Reg. at 70,446). Defendants note that following this court’s October 28, 2020 order granting a temporary restraining order and preliminary injunction, the 27 USDA provided a declaration on November 16, 2020, stating that the FLR would not be prepared by the end of the year. (Id.) The court addresses these arguments further below in its 28 consideration of irreparable harm.

17

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 17 of 39

1 “[T]he question before this court is not whether Labor adopted the best wage possible. It 2 is only whether Labor’s selection . . . falls within the broad realm of reason, and whether Labor 3 sufficiently explained the basis for its judgment.” Hispanic Affairs Project v. Acosta, 901 F.3d 4 378, 395 (D.C. Cir. 2018). Thus, defendants are correct that “the FLS data may be the best data 5 available concerning wages of livestock workers, but if a wage based on that data would 6 adversely affect U.S. workers, then the agency was within its broad authority not to use the FLS 7 data.”5 (Doc. No. 31 at 25.) 8 Nonetheless, the court is persuaded at this stage of the litigation that the Final Rule fails to 9 justify freezing wages below market rate. In the 2010 Rule, the DOL stated that it was required 10 to ensure that U.S. “workers receive the greatest potential protection from adverse effects on their 11 wages and working conditions, including the adverse effect of being denied access to the 12 opportunity to earn a higher equilibrium wage that would have resulted as the market (perhaps 13 slowly) adjusted in the absence of the guest workers.” 75 Fed. Reg. 6,893. Throughout the Final 14 Rule at issue here, the DOL notes the importance of the AEWR reflecting the market rate. See 85 15 Fed. Reg. at 70,461–62. For example, in rejecting certain proposals, the Final Rule states that 16 “[t]he AEWR is meant to approximate the wage paid to workers in the United States similarly 17 employed” and “a single national AEWR . . . that covers all occupations would not meet that 18 purpose” because it “would immediately and dramatically reduce the wages of both H-2A and 19 similarly employed workers, particularly those performing work in dozens of states currently 20 being paid a wage above the FY 2020 national AEWR based on the FLS”; and that, “‘a single 21 national AEWR applicable to all agricultural jobs in all geographic locations would prove to be 22 below market rates in some areas and above market rates in other areas, resulting in all of the 23

24 5 Similarly, the court discounts plaintiffs’ contention that that the DOL fails to explain why the two-year wage freeze serves as a “transition period” needed to “provide[] employers with a 25 reasonable amount of time to plan their labor needs and agricultural operations under the new 26 wage requirements.” (Doc. No. 5 at 22) (quoting 85 Fed. Reg. at 70,467). Plaintiffs assert that the DOL could have best ensured continuity for employers and farmworkers alike by leaving the 27 existing methodology in place and using the FLS data that the USDA is enjoined to produce.

(Id.) That argument is misplaced, since that decision falls within the DOL’s broad discretion so 28 long as the decision made is sufficiently explained.

18

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 18 of 39

1 associated adverse effects that have been previously discussed.’” Id. at 70,461. Elsewhere, the 2 Final Rule discounts a proposal that would “produce[] ‘artificially lower [wages] to a point that 3 [they] no longer represent[ed] a market-based wage.’” Id. at 70,462. 4 Still, the Final Rule implements a methodology freezing wage rates that are already below 5 the current market rate. Defendants skirt this issue by contending that the wage freeze ensures 6 that farmworkers’ wages will not decrease and “will steadily rise ‘with the changes in wages and 7 salaries in the broader economy,’—two beneficial features absent from the current AEWR 8 methodology.” (Doc. No. 31 at 23) (citing 85 Fed. Reg. at 70,454). However, the closest that the 9 Final Rule gets to addressing the intentional departure from accurate market wages is its 10 statement that “even if more recent, 2020 FLS wage data were available, relying on it to set 2021 11 AEWRS would only serve to perpetuate the very wage volatility that the Department seeks to 12 ameliorate through this rule.” 85 Fed. Reg. at 70,453. First, the court agrees with plaintiffs that 13 the USDA’s FLS Suspension Notice should not factor into this equation, at least with regard to 14 setting the 2021 AEWRs, because the undersigned enjoined that decision and 2020 FLS data 15 should therefore be available in a timely fashion. Second, although it is a factor to be considered, 16 employers’ concerns about an FLS-based AEWR methodology being “unpredictable” and 17 “volatile” is alone an insufficient justification for departing from the methodology. Despite 18 claiming that it concluded “on balance” that use of the FLS was “not appropriate in this context” 19 (Doc. No. 31 at 25), the DOL has not in fact addressed the impact that freezing below-market 20 rates will have on field and livestock workers. As explained above, the 2010 Rule focused on 21 protecting these lower-skilled workers—the vast majority of the very farmworkers the DOL is 22 charged with protecting—from adverse wage effects. Here, the DOL has failed to justify the 23 Final Rule’s shift to primarily benefiting a smaller group of higher-skilled workers. See Fox 24 Television Stations, Inc., 556 U.S. at 515–16. 25 Accordingly, plaintiffs have demonstrated that they are likely to prevail on their claim that 26 the Final Rule is arbitrary and capricious because defendants have failed to explain the decision to 27 freeze H-2A wages below market rate for two years. 28 /////

19

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 19 of 39

1 b. Whether the DOL Adequately Explained Its Decision to Use the ECI to Adjust AEWRs Starting In 2023 2

3 Next, plaintiffs argue that the Final Rule is arbitrary and capricious because it fails to 4 justify its decision to replace reliance on the FLS with a system of annual adjustments to the 2020 5 AEWRs using the ECI. (Doc. No. 5 at 22–23.) Plaintiffs assert that the Final Rule does not 6 explain how a metric with no connection whatsoever to the farm labor market can adequately 7 protects U.S. farmworkers from adverse effects. (Id. at 23; see also Doc. No. 5-4, table 1 n.2). 8 According to plaintiffs, the DOL’s decision to rely on the ECI is even more significant given that 9 farmworker wages have been increasing faster than wages in the general economy. (Doc. No. 5 10 at 23.) Though the Final Rule states that reliance on the ECI would lead to “greater stability” for 11 employers, nowhere does the Final Rule explain why ECI data is an economically valid proxy for 12 agricultural labor market outcomes. (Id.) Pointing to the decision in AFL-CIO v. Brock, plaintiffs 13 note that when the DOL previously used manufacturing wage rates to calculate the AEWR for 14 agricultural guestworkers, the DOL recognized the mismatch and justified its decision as 15 necessary to address adverse effects from temporary foreign labor. (Doc. No. 5 at 23.) Here, by 16 contrast, the DOL celebrates its switch to reliance upon ECI data because it yields a “reduction in 17 wage growth relative to the previous methodology,” a goal at odds with the very purpose of the 18 rule itself which is to prevent adverse effects on U.S. farmworkers’ wages. (Id.) (citing 85 Fed. 19 Reg. at 70,456). 20 In opposition, defendants argue that the ECI “provides an accurate measure of annual 21 increases in wages across the private sector and ‘is particularly well suited as a vehicle to adjust 22 wage rates to keep apace with what is paid by other employers.’” (Doc. No. 31 at 27) (citing 85 23 Fed. Reg. at 70,455). Similarly, the Final Rule states that the ECI-based adjustments to the 24 AEWRs “will ensure field and livestock worker wages continue to rise apace with wages in the 25 broader U.S. economy in a consistent and predictable manner.” (Id.) (citing 85 Fed. Reg. at 26 70,445). According to defendants, “indexing the AEWRs to the ECI will produce steadily 27 increasing AEWRs for field and livestock workers that fulfill the statutory requirement to prevent 28 adverse effect on the wages of workers in the United States similarly employed, while providing

20

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 20 of 39

1 consistency and predictability to the agricultural economy.” (Id.) (citing 85 Fed. Reg. 70,445). In 2 this same vein, the Final Rule notes that using the ECI provides programmatic consistency as it is 3 “the current means by which the monthly AEWR is adjusted for range occupations.” (Id.) (citing 4 85 Fed. Reg. 70,445). 5 The court is not persuaded that plaintiffs have met their burden with regards to this issue 6 in seeking preliminary injunctive relief. The relevant section of the Final Rule thoroughly 7 explains why the DOL chose to rely on the ECI. See 85 Fed. Reg. at 70,455. The DOL’s 8 decision, even if one “of less than ideal clarity,” must be upheld because in making that choice 9 “the agency’s path may reasonably be discerned.” State Farm, 463 U.S. at 43. Unlike the 10 manufacturing wages-based methodology in Brock, here the Final Rule still relies on field and 11 livestock worker wage data—albeit, not current data—and then adjusts using the ECI because it 12 measures changes in wages. The Final Rule explains that “[t]he FLS-based, ECI-adjusted AEWR 13 methodology in this final rule is, in the Department’s judgment, the most effective available 14 methodology that addresses the oft-cited concern among many commenters that under the 15 proposed approach, AEWRs would be too unpredictable and based on a methodology that would 16 be too complex.” 85 Fed. Reg at 70,456. The Final Rule also addresses the concerns expressed 17 by commenters with the old methodology by emphasizing that “ECI-based adjustments are 18 straightforward to calculate and, based on the substantial historical data available, predictable.” 19 Id. Moreover, the Final Rule notes that the DOL already uses the ECI “to make AEWR 20 determinations for H-2A herding and livestock jobs on the range.” Id. at 70,455. 21 As previously stated, the court is not tasked with assessing whether the DOL “adopted the 22 best wage possible,” but is only to determine whether the DOL “sufficiently explained the basis 23 for its judgment.” Hispanic Affairs Project, 901 F.3d at 395. Here, the court finds the Final 24 Rule’s use of the ECI to be sufficiently explained. Thus, the court concludes that plaintiffs have 25 not demonstrated likelihood of success, or even presented serious questions, with respect to the 26 merits of this aspect of their claim. 27 ///// 28 /////

21

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 21 of 39

1 c. Whether the DOL Analyzed Harm to U.S. Farmworkers 2 Next, plaintiffs argue that the DOL failed to analyze the economic harm to U.S.

3 farmworkers of its Final Rule. (Doc. No. 5 at 24.) Plaintiffs take issue with the Final Rule 4 simply labeling those harms as “Unquantifiable Transfer Payments,” accompanied by a bare 5 assertion that the DOL lacks the data “about the number of [U.S.] workers in corresponding 6 employment and their wage structure.” (Id.) (citing 85 Fed. Reg. at 70,472). Plaintiffs contend 7 that consideration of the central issue underlying DOL’s rule—whether an AEWR methodology 8 prevents adverse effects on U.S. farmworkers—requires more than merely conceding harm but 9 asserting at the same time that it is “unquantifiable.” (Id.) (citing NRDC v. Rauch, 244 F. Supp. 10 3d 66, 71 (D.D.C. 2017)). 11 Notably, defendants have little to say in opposition to this contention. Rather, defendants 12 merely reiterate that while the DOL has access to the number of H-2A worker positions sought, 13 the nature of the occupations, the applicable AEWR, and the location of the employers, the 14 agency has no access to similar information for the number of farmworkers in “corresponding 15 employment.” (Doc. No. 31 at 28) (citing 85 Fed. Reg. at 70,472). Defendants also contend that 16 plaintiffs have offered no suggestion as to how the DOL might go about performing this 17 calculating. (Id.) 18 In reply, plaintiffs note that “DOL’s regulations require that H-2A program employers 19 provide similar information for U.S. workers.” (Doc. No. 34 at 14.) Plaintiffs point to 20 C.F.R. 20 § 655.156(a)(2), which requires H-2A employers to prepare, sign, and date written recruitment 21 reports that include “the name and contact information of each U.S. worker who applied or was 22 referred to the job opportunity up to the date of the preparation of the recruitment report, and the 23 disposition of each worker.” These reports “must be made available in the event of a post- 24 certification audit or upon request by authorized representatives of the Secretary.” Id. 25 § 655.156(b). 26 The court finds plaintiffs’ arguments in this regard to be persuasive based on the present 27 record. The D.C. Circuit’s opinion in AFL-CIO v. Dole is instructive. After that court reversed 28 and remanded the action considering the 1987 AEWR rule in Brock, the DOL published a new

22

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 22 of 39

1 final rule providing additional reasoning for its new methodology. Dole, 932 F.2d at 185. 2 Plaintiffs challenged the new rule, arguing that the DOL “impermissibly abandoned its past 3 policy of adding an upward adjustment for past wage depression, merely because such adverse 4 effect cannot be precisely measured, rather than attempting to find another proxy for supposed 5 wage depression.” Id. at 187. The court found the arguments meritless, explaining why as 6 follows:

7 Were we to require that DOL hew to its old methodology until it had conclusive data on which to base a change, we would lock the 8 Department into its previous policy. Instead, we are entitled to ask only that DOL demonstrate that the data is inconclusive and “identify 9 the considerations it found persuasive in making its decision.”

Because the record clearly shows that the data on adverse effect is 10 inconclusive, and the Department provided an ample explanation of the considerations it found persuasive, we reverse the district court 11 and uphold the final rule. 12 Id. at 187–88. Thus, despite a lack of precise measurements, the court in Dole upheld the rule

13 specifically because the DOL made a case for why that data was inconclusive. 14 Here, the DOL has not pointed anywhere in the administrative record where it has made a 15 similar showing in this instance. Rather than providing data points or citing attempts to identify 16 relevant data, the DOL simply concludes that it “does not have sufficient information about the 17 number of workers in corresponding employment affected and their wage structure to reasonably 18 measure the wage transfer to or from these workers.” 85 Fed. Reg. at 70,472. This conclusion is 19 expressed despite the Final Rule’s acknowledgement that “the overwhelming majority of H-2A 20 job opportunities . . . fall within the FLS field and livestock workers (combined) category.” 85 21 Fed. Reg. at 70,461. At the hearing on the pending motion, defense counsel pointed to the 2010 22 Rule, which stated that the DOL could not quantify with precision the “transfer of costs from 23 government entities to employers as a result of lower expenditures on unemployment insurance 24 benefit claims.” 75 Fed. Reg. at 6,947. This difficulty arose in part because of “uncertainty about 25 . . . the quantity of corresponding U.S. workers, . . . [and] the ranges of wages in the areas of 26 actual employment.” Id. Defendants assert that this lack of data has been a long-standing 27 problem that sufficiently explains why the DOL did not quantify the transfer payment from non- 28 H-2A employees to employers under the Final Rule. However, as defendants have conceded, the

23

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 23 of 39

1 2010 Rule’s failure to attempt to provide data with respect to U.S. workers does not justify the 2 current Final Rule’s failure to do so. Additionally, the court is less concerned with the 2010 3 Rule’s failure to quantify transfer payments from government entities to employers than it is with 4 the current Final Rule’s failure to quantify transfer payments from U.S. workers to H-2A 5 employers. This is because the latter shift is a significant aspect to consider in light of the INA’s 6 mandate that the DOL protect U.S. workers from adverse effects on wage. This is especially 7 important considering that the Final Rule concludes that “[t]he new AEWR methodology may 8 further encourage U.S. employers to use more H-2A workers for field and livestock work in the 9 absence of available U.S. workers.” 85 Fed. Reg. at 70,472. 10 Although the court cannot require the DOL to suspend issuing a rule until the data is 11 conclusive, it is “entitled to ask [] that DOL demonstrate that the data is inconclusive and 12 ‘identify the considerations it found persuasive in making its decision.’” Dole, 932 F.2d at 187. 13 The DOL has failed to do so in the Final Rule at issue here. For these reasons, plaintiffs have 14 demonstrated a likelihood of success on the merits of this aspect of their claim. 15 3. Whether the Final Rule Violates the APA’s Notice-and-Comment Requirement 16 In their third claim, plaintiffs allege that the DOL’s Final Rule should be vacated because 17 defendants did not abide by the requirements of notice-and-comment rulemaking before issuing 18 the Final Rule. (Compl. at ¶¶ 111–18.) Specifically, plaintiffs allege that the Final Rule diverges 19 materially from the NPRM published in the Federal Register on July 26, 2019, and thus the public 20 was denied sufficient notice regarding the Final Rule’s contents as well as an adequate 21 opportunity to provide comments. (Id. at ¶ 116.) 22 Under the APA, agency actions taken “without observance of procedure required by law” 23 must be set aside. 5 U.S.C. § 706(2)(D). “The APA requires public notice and comment and a 24 thirty-day grace period before a proposed rule takes effect.” E. Bay Sanctuary Covenant v. 25 Trump, 950 F.3d 1242, 1277 (9th Cir. 2020) (citing 5 U.S.C. §§ 553(b)–(d)). The agency is 26 required to publish a notice of proposed rulemaking in the Federal Register, and it must include 27 “either the terms or substance of the proposed rule or a description of the subjects and issues 28 involved.” 5 U.S.C. § 553(b)(3). This means that “the final rule the agency adopts must be a

24

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 24 of 39

1 ‘logical outgrowth’ of the rule proposed. The object, in short, is one of fair notice.” Long Island 2 Care at Home, Ltd. v. Coke, 551 U.S. 158, 174 (2007) (internal citations and quotation marks 3 omitted). As the Ninth Circuit has explained:

4 [A] final regulation that varies from the proposal, even substantially, will be valid as long as it is in character with the original proposal 5 and a logical outgrowth of the notice and comments. In determining whether notice was adequate, we consider whether the complaining 6 party should have anticipated that a particular requirement might be imposed. The test is whether a new round of notice and comment 7 would provide the first opportunity for interested parties to offer comments that could persuade the agency to modify its rule.

8

9 Envtl. Def. Ctr., Inc. v. U.S. E.P.A., 344 F.3d 832, 851 (9th Cir. 2003) (internal citation and 10 quotation marks omitted). Moreover, “[a]gencies, are free—indeed, they are encouraged—to 11 modify proposed rules as a result of the comments they receive.” Ne. Md. Waste Disposal Auth. 12 v. E.P.A., 358 F.3d 936, 951 (D.C. Cir. 2004). 13 In this case, plaintiffs argue that the NPRM failed to provide the public with adequate 14 notice that the DOL would untether AEWRs from market wages by freezing AEWRs for 15 two years and then relying on the ECI to make future adjustments. (Doc. No. 5 at 26.) Plaintiffs 16 note that the DOL would not have had any reason to address a potential freeze of AEWRs in its 17 NPRM because the USDA did not issue its FLS Suspension Notice until over a year after the 18 DOL issued its NPRM. (Id.) Additionally, plaintiffs argue that the NPRM never suggested that 19 FLS data was in anyway problematic, and indeed stated that “the FLS [w]ould continue to be the 20 basis for the AEWRs covering the vast majority of H-2A workers.” (Id.) (citing 84 Fed. Reg. at 21 36,182). Plaintiffs also argue that the Final Rule’s departure from agricultural labor market data 22 is not a logical outgrowth of the NPRM because “[t]he NPRM stressed the importance of 23 establishing AEWRs based on market wages, explaining that the AEWR protects U.S. 24 farmworkers from adverse effects because it is the wage rate that is determined from a survey of 25 actual wages paid by employers.’” (Id. at 27) (citing 84 Fed. Reg. 36,179). 26 In opposition, defendants argue that the NPRM specifically referred to the ECI as an 27 alternative in the absence of FLS data and specifically asked the public to comment on its 28 possible future use. (Doc. No. 31 at 17) (citing 84 Fed. Reg. at 36,182) Defendants contend that

25

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 25 of 39

1 the NPRM asked whether the DOL “should consider any other methodology that would promote 2 consistency and reliability in wage rates from year to year,” if FLS data could not be used to 3 produce a wage. (Doc. No. 31 at 17) (citing 84 Fed. Reg at 36,182). According to defendants, 4 the NPRM invited comment on “all aspects” of the AEWR methodology, including the “use of 5 the FLS and OES survey, the conditions under which each survey should be used to establish the 6 AEWR,” and “any alternate wage sources the [DOL] might use to establish the AEWRs in the H- 7 2A program.” (Id.) (quoting 84 Fed. Reg. at 36,184). Additionally, defendants argue that the 8 NPRM announced the DOL’s intention to utilize more occupation-specific data and end its 9 exclusive reliance on the FLS. (Doc. No. 31 at 16) (citing 84 Fed. Reg. at 36,179). Defendants 10 also note that several worker advocacy organizations, including plaintiffs, cautioned against the 11 use of OES data mentioned in this aspect of the proposal, thus indicating that those organizations 12 were aware that the new methodology was a possibility. (Id.) (citing 85 Fed. Reg. at 70,452). 13 In analyzing this issue, the undersigned finds the decision in CSX Transportation, Inc. v. 14 Surface Transportation Board, 584 F.3d 1076 (D.C. Cir. 2009) to be instructive. Plaintiffs rely 15 on this case to support the assertion that this court should follow the lead of other “courts [that] 16 have previously invalidated rate-setting rules that made methodological changes never suggested 17 by the relevant NPRM.” (Doc. No. 5 at 27.) In CSX Transportation, Inc., railroad companies 18 asserted that the Surface Transportation Board violated notice-and-comment requirements 19 because it failed to give notice of a significant change to the methodology for resolving rail rate 20 disputes, which surfaced for the first time in the final rule. CSX Transp., Inc., 584 F.3d at 1078. 21 The railroads challenged the decision to depart from calculating rates based on four years’ worth 22 of data to instead relying on only one year’s–worth of data. Id. The D.C. Circuit found that while 23 “the final rule did not amount to a complete turnaround from the NPRM,” the case was “far more 24 like those in which [the court] found that agencies had failed to give adequate notice.” Id. at 25 1081–82. The court was not persuaded by the Board’s contention “that the mere mention of the 26 release of one-year data for comparison groups gave notice that the amount of data available for 27 that purpose might change.” Id. at 1082. The court also saw “no way that commenters here could 28 have anticipated which particular aspects of [the Board’s] proposal [were] open for

26

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 26 of 39

1 consideration,” and if the court “conclude[d] that commenters had notice merely because the 2 NPRM mentioned one year’s worth of data, the Board could issue broad NPRMs only to justify 3 any final rule it might be able to devise by whimsically picking and choosing within the four 4 corners of a lengthy notice.” Id. 5 Similarly, the undersigned concludes that the plaintiffs in this case could not have 6 anticipated that a complete departure from the FLS was “open for consideration,” despite “the 7 mere mention” that the USDA may at some point terminate the FLS. See id. The NPRM did note 8 that the DOL “does not have direct control over the FLS, and that USDA could elect to terminate 9 the survey at some point.” 84 Fed. Reg. at 36,183. The NPRM “addressed such a possibility in 10 this proposal by providing that the OES statewide average hourly wage for the Standard 11 Occupational Classification (“SOC”) will be used if the FLS does not produce an annual gross 12 hourly wage for the occupational classification for a State or region.” Id. The following 13 paragraph also acknowledged “that USDA may make future adjustments to the FLS 14 methodology,” and the NPRM stated that “[i]f the Department decides to later adjust the AEWR 15 calculation based on methodological changes by USDA, the [DOL] will provide the public with 16 notice and the opportunity to provide comment before adopting any changes.” Id (emphasis 17 added). Thus, even where the DOL recognized that the FLS could at some point be terminated or 18 changed, the NPRM explicitly stated that it had already provided an alternative solution and 19 would offer an additional opportunity for the public to comment if the agency had to adjust the 20 AEWR based on methodological changes to the FLS by the USDA. 84 Fed. Reg. at 36,183. This 21 supports the conclusion that a fair reading of the notice conveyed that the DOL did not intend to 22 invite comment on removing the FLS from the methodology, at least for the final rule arising 23 from this NPRM. 24 Additionally, the court does not read the NPRM’s reference to using the ECI in lieu of 25 unavailable FLS data as suggesting any intention to completely supplant the FLS data with ECI 26 and OES data. See 84 Fed. Reg. at 36,182 (“The Department requests comments on whether 27 there are alternate methods or sources that it should use to set the AEWR in the event that the 28 FLS does not produce a wage in an SOC and State or region . . . .”). The court concludes that this

27

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 27 of 39

1 aspect of the NPRM, fairly read, merely solicited suggestions on gap filling were the FLS to lack 2 relevant information. This is especially true considering the NPRM’s repeated references to the 3 FLS as the preferred source of data on livestock and fieldworker wage rates and its statement that 4 the DOL proposed to continue using the FLS despite lacking complete control over its 5 publication. See id. at 36,183. Given the overall context of this broad request for suggestions, the 6 court is not persuaded that the catchall request for comment on “any other methodology” 7 indicated that commenters should have known that a complete departure from the FLS “was ‘on 8 the table.’” Nat. Res. Def. Council v. E.P.A., 279 F.3d 1180, 1188 (9th Cir. 2002). 9 At the hearing on the pending motion, defense counsel contended that the NPRM invited 10 comment on two scenarios in which FLS data might be unavailable: (1) as quoted above, “in the 11 event that the FLS does not produce a wage in an SOC and State or region,” 84 Fed. Reg. at 12 36,182; and (2) in the NPRM’s subsequent paragraph, which states that “[a]s an alternative, the 13 [DOL] invites comments on whether to set AEWRs based on the current FLS occupational 14 classifications of field workers and livestock workers for each State or region.” Id. at 36,182–83. 15 But that paragraph also states that “the [DOL] generally prefers to establish AEWRs based on the 16 FLS rather than the OES survey because the FLS surveys farmers and ranchers, whereas the OES 17 surveys establishments that support farm production, as discussed below.” Id. Thus, contrary to 18 defendants’ position taken in opposition to the pending motion, this paragraph also indicates an 19 intention to continue using the FLS. 20 Plaintiffs have therefore shown that they are likely to prevail on their claim that the DOL 21 failed to comply with the notice-and-comment rulemaking requirements in issuing the Final Rule. 22 Accordingly, the court finds that plaintiffs have demonstrated a likelihood of success on the 23 merits of this claim. 24 3. Irreparable Harm 25 Having found that plaintiffs have shown a likelihood of success on the merits of several 26 aspects of their claims, the court now turns to whether plaintiffs have also shown a likelihood that 27 they will suffer irreparable harm in the absence of the court granting preliminary injunctive relief. 28 The risk of irreparable harm must be “likely, not just possible.” All. for the Wild Rockies, 632

28

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 28 of 39

1 F.3d at 1131. “Speculative injury does not constitute irreparable injury sufficient to warrant 2 granting a preliminary injunction.” Caribbean Marine Servs. Co. v. Baldrige, 844 F.2d 668, 674 3 (9th Cir. 1988). “[E]conomic hardship constitutes irreparable harm.” Kildare v. Saenz, 325 F.3d 4 1078, 1083 (9th Cir. 2003). For low-income individuals, economic loss can mean inadequate 5 access to “food, shelter [and] other necessities.” Id.; cf. Paxton v. Sec’y of Health & Human 6 Servs., 856 F.2d 1352, 1354 (9th Cir. 1988) (“When a family is living at subsistence level, the 7 subtraction of any benefit can make a significant difference to its budget and to its ability to 8 survive.”). 9 Here, plaintiffs assert that their members, including both U.S. and H-2A farmworkers, will 10 suffer material wage reductions under the new AEWR methodology adopted in the Final Rule. 11 (Doc. No. 5 at 28–29.) Plaintiffs point to the most recent FLS data as strongly suggesting that the 12 California AEWR would have increased to approximately $15.58 in 2021. (Id. at 29) (citing Doc. 13 No. 5-3 at 6, 8). Under the Final Rule, plaintiffs assert that those wages would be frozen at 14 $14.77—the wage rate that farmworkers were paid in 2019. (Id.) (citing Doc. No. 5-6 at 26). 15 Most of plaintiffs’ members working for H-2A employers are paid the AEWR, and thus those 16 members working in California will be paid approximately $0.81 less per hour—totaling $139.32 17 per month and $1,393.20 over a ten-month farming season—absent a court order enjoining the 18 DOL from implementing the Final Rule. (Id.) Plaintiffs contend that farmworkers in Oregon and 19 Washington will experience similar losses, as most farmworkers working for H-2A employers in 20 those states will be paid $77.40 less per month, totaling $774 in lost wages over a ten-month 21 farming season, under the Final Rule. (Doc. No. 5 at 29; see also Doc. No. 5-3 at 6, 8, 14, 16.) 22 Plaintiffs aver that these depressed wages will cause substantial economic hardship for 23 many of their members. (Doc. No. 5 at 29.) According to the declarations of Teresa Romero, 24 president of UFW, and Diana Tellefson Torres, executive director of UFW Foundation, 25 farmworkers are among the lowest-paid workers in the United States, with many earning a 26 subsistence income. (Doc. Nos. 5-10 at ¶ 17; 5-11 at ¶ 6.) Reducing farmworkers wages by 27 approximately four or five percent would therefore clearly cause substantial harm to plaintiffs’ 28 members and their families. (Doc. No. 5-10 at ¶¶ 11–16.) Farmworkers who already struggle to

29

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 29 of 39

1 provide for the necessities of life will find that this wage depression will only place further 2 financial strain on their ability to obtain food, shelter, and other necessities. (Doc. No. 5 at 29.) 3 Many of plaintiffs’ members, and other farmworkers across the United States, already struggle to 4 pay for necessities such as shelter and medical care (see Doc. Nos. 5-10 at ¶ 18; 5-11 at ¶¶ 6–7), 5 and many farmworkers suffer with food insecurity and must rely on emergency food programs 6 (see Doc. Nos. 5-7 at 6 (finding that 49 percent of farmworkers fell into a state of food insecurity 7 over a 2-year period); Doc. No. 5-11 at ¶ 6). Plaintiffs contend that the DOL increases the risk 8 that farmworkers will struggle to afford the cost of such necessities which, unlike their wages 9 under the Final Rule, will not be frozen for the next two years. (Doc. No. 5 at 30.) According to 10 plaintiffs, the economic impact of depressed wages on farmworkers will be exacerbated by the 11 substantial increases in consumer prices over the last year. (See Doc. Nos. 5-10 at ¶ 19; 5-8 12 (showing the cost of food increasing by 3.9 percent over the last twelve months.)) The cost of 13 medical care has likewise increased by 4.1 percent between 2019 and 2020. (See Doc. No. 5-9.) 14 Such harms threatened if the Final Rule is implemented are further exacerbated by reduced hours 15 caused by the ongoing COVID-19 pandemic, leaving an already impoverished population even 16 more vulnerable. (See Doc. No. 5-11 at ¶ 8.) 17 Plaintiffs note that the Final Rule itself acknowledges that its methodology could result in 18 employers hiring H-2A field and livestock workers at the expense of U.S. farmworkers, see 85 19 Fed. Reg. at 70,472, and that those lost job opportunities will cause substantial, incurable 20 hardship for plaintiffs’ members (see Doc. Nos. 5-10 at ¶ 26; 5-11 at ¶ 9). Thus, according to 21 plaintiffs, their U.S. farmworker members would also be irreparably harmed through the loss of 22 their jobs. (Doc. No. 5 at 29.) As plaintiffs note, the “‘loss of opportunity to pursue . . . chosen 23 professions’ constitutes irreparable harm,” and “[t]he irreparable nature of [that] injury is 24 heightened by [farmworkers’] fragile socioeconomic position.” (Doc. No. 5 at 30) (quoting Ariz. 25 Dream Act Coal. v. Brewer, 757 F.3d 1053, 1068 (9th Cir. 2014)). Plaintiffs argue that even if 26 retroactive back pay could address any harm caused by the Final Rule, farmworkers would likely 27 be unsuccessful in obtaining those back wages because (1) the H-2A certification process makes 28 /////

30

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 30 of 39

1 it difficult if not impossible to obtain retroactive pay, and (2) farmworkers face barriers to 2 accessing legal resources. (Id. at 30–31) (citing Doc. No. 5-10 at ¶¶ 22–24). 3 In support of plaintiffs’ motion, the State of California noted in its amicus curiae brief that 4 the state’s farmworkers are essential and skilled labor that feeds the nation and boosts the 5 economy. (Doc. No. 32-1 at 10.) As the leading state for cash farm receipts and the nation’s 6 largest agricultural global exporter, California emphasizes the magnitude of the harm that will 7 occur without the granting of the requested injunctive relief. (Id. at 10–11.) California notes that 8 given the lower wages in Mexico, where the vast majority of H-2A workers originate, the Final 9 Rule’s two-year freeze on wages and the subsequent limit on the AEWR’s rate of increase will 10 not discourage H-2A workers from taking U.S. agricultural jobs, which will further depress 11 domestic farmworker wages. (Id. at 17.) California also highlights the issues that are intertwined 12 with farmworker poverty, notably that: the Final Rule will intensify the challenges farmworkers 13 already face in obtaining affordable housing and increase demand on state housing programs (id. 14 at 18–19); farmworkers’ children will be more educationally disadvantaged, experience food 15 insecurity, and suffer poorer health, placing additional demands on state programs (id. at 19–22); 16 and farmworkers, who already suffer from inadequate health care, will suffer poorer health (id. at 17 22–24). Finally, California avers that the increase in the vulnerable H-2A workforce—many of 18 whom fear speaking out against poor working conditions, face language barriers, and fear 19 retaliation—will seriously undermine the state’s ability to enforce its labor protection laws. (Id. 20 at 25–27.) 21 In their opposition to the pending motion, defendants argue that the granting of the 22 requested injunction is actually likely to cause irreparable harm to farmworkers, at least if it is 23 issued in the next few weeks. (Doc. No. 31 at 28.) In this regard, according to defendants, 24 plaintiffs have not demonstrated that “the requested relief is necessary to avoid irreparable harm 25 during the interim period that the relief is to be provided.” (Id. at 29) (quoting S. Yuba River 26 Citizens League v. Nat’l Marine Fisheries Serv., No. 2:13-cv-00042-MCE, 2013 WL 4094777, at 27 *7 (E.D. Cal. Aug. 13, 2013) (emphasis in original)). Defendants contend that under the Final 28 Rule, the AEWRs will be set at the same rate as the 2020 AEWRs for most farmworkers (e.g.,

31

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 31 of 39

1 field and livestock workers), but the AEWRs will increase for higher-skilled workers. (Id.) 2 According to the Pasternak Declaration, under the Final Rule truck drivers will actually see wage 3 gains of approximately 75 to 120 percent in each of the top 25 states using the H-2A Program (or 4 a $8.82 to $13.87 increase in their hourly rates); supervisors in these states will see increases of 5 85 to 150 percent or more (a $10.02 to $18.34 increase in their hourly rates); and construction 6 laborers will similarly see significant wage gains. (Id.) (citing Doc. No. 31-1 at ¶ 12). 7 Defendants also assert that if implementation of the Final Rule is enjoined, the AEWRs 8 for all workers will be the same as the 2020 AEWRs, or worse, the DOL may have to operate the 9 H-2A program without an AEWR at all. (Id. at 29–30.) Defendants note that the DOL’s 10 regulations require the agency to publish “at least once in each calendar year . . . the AEWRs for 11 each State as a notice in the Federal Register,” and AEWRs have not yet been published this year 12 and must be published by December 31, 2020. (Id. at 30.) (citing 20 C.F.R. § 655.120(c)). 13 Current regulations also require the DOL to set the AEWRs using the “annual weighted average 14 hourly wage for field and livestock workers (combined) in the States or regions as published 15 annually by the U.S. Department of Agriculture based on its quarterly wage survey.” (Id.) (citing 16 20 C.F.R. § 655.103(b)). They also observe that because no annual FLS data will be published 17 until the next FLR is published pursuant to the court’s October 28, 2020 order in United Farm 18 Workers v. Perdue, No. 1:20-cv-1452-DAD-JLT, the most recent annual data is the 2019 FLR, 19 which is the data source used to set the 2020 AEWRs. (Id.) Defendants contend that the DOL 20 will have to take one of two actions if the Final Rule is enjoined: (1) the DOL will either set the 21 2021 AEWRs using the 2020 rates, or, if that is not possible given DOL’s current regulations, (2) 22 the DOL will be forced to assess operating the H-2A program without an AEWR, lowering the 23 wage floor established in the 2020 AEWRs to the federal or state minimum wages in the absence 24 of a CBA wage or a prevailing wage survey for all workers. (Id.) According to defendants, the 25 former scenario would keep most farmworkers in the same position as if the injunction did not 26 issue and would place some workers in a worse position; the latter scenario would bring harm to 27 nearly all farmworkers. (Id.) In this way, defendants assert that the granting of the requested 28 injunction would cause harm to at least some farmworkers. (Id.)

32

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 32 of 39

1 Plaintiffs note that defendants do not dispute that thousands of plaintiffs’ members and 2 farmworkers across the country would be paid substantially less under the Final Rule compared to 3 the 2010 Rule’s methodology and instead are merely arguing that the court cannot grant effective 4 relief because FLS data is not yet available. (Doc. No. 34 at 16–17.) Plaintiffs urge this court not 5 to permit the DOL to use the USDA’s delay in complying with the court’s order to justify 6 allowing the DOL’s Final Rule to become effective, since this was the very scenario that the 7 court’s October 28, 2020 order granting preliminary relief in Perdue sought to avoid. (Id. at 17.) 8 In light of the FLR publication timeline (see Doc. No. 34-7), plaintiffs propose that the DOL 9 should issue interim AEWRs that protect farmworkers’ wages while simultaneously notifying 10 employers that updated AEWRs are expected in February of 2021 and that employers are 11 responsible for providing backpay to farmworkers for work those workers perform during the few 12 weeks the interim AEWRs are effective. (Id. at 18.) Plaintiffs also suggest that the DOL issue 13 interim AEWRs based on FLS data published in May 2020, which details the wages paid to 14 farmworkers in January and April 2020. (Id.) 15 As an initial matter, an unlawful agency action cannot be upheld merely because of a 16 predicament that the government itself has created. Defendants argue that “[p]laintiffs created 17 what they now assert is an urgent need for the Court’s intervention” because they waited nearly 18 four weeks to file suit and now assert that their members will suffer irreparable harm if the court 19 does not provide immediate preliminary relief. (Doc. No. 31 at 31.) Defendants assert that this 20 rushed timeline is largely of plaintiffs’ making and they should not be rewarded for their own 21 inaction, particularly since they had advance notice that publication of the Final Rule was 22 forthcoming. (Id.) (citing Perdue, 1:20-cv- 01452-DAD-JLT (Doc. No. 30)). Defendants point to 23 Local Rule 231(b), contending that “[b]ecause Plaintiffs ‘could have sought relief by motion for 24 preliminary injunction at an earlier date without the necessity for seeking last-minute relief,’ . . . 25 the Court should deny Plaintiffs’ Motion here and defer ruling on the merits until the full 26 administrative record is before the Court.” (Doc. No. 31 at 31.) 27 The court finds this argument to be unpersuasive because it fails to acknowledge that, 28 after this court enjoined the USDA’s FLS suspension Notice on October 28, 2020, the USDA

33

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 33 of 39

1 filed a motion seeking to dissolve that order on November 5, 2020—the Final Rule’s publication 2 date. See United Farm Workers v. Perdue, No. 1:20-cv-01452-DAD-JLT, 2020 WL 6939021 3 (E.D. Cal. Nov. 25, 2020). Plaintiffs filed their opposition to that motion to dissolve on 4 November 11, 2020. Id. at *1. On November 25, 2020, the court denied the USDA’s motion to 5 dissolve because the Final Rule had not yet gone into effect, and thus no change in law had 6 occurred that warranted dissolving the temporary restraining order. Id. at *4. The court also 7 noted in the order that the purpose of APA § 553(d)’s time lag of at least 30 days between a 8 substantive rule’s publication and its effective date “is to ‘afford persons affected a reasonable 9 time to prepare for the effective date of a rule or rules or to take any other action which the 10 issuance of the rules may prompt.’” Id. The court underscored this purpose in acknowledging 11 that plaintiffs intended to challenge the Final Rule. Id. at *4 n.4. A mere five days later, on 12 November 30, 2020, plaintiffs submitted their complaint and pending motion in this action. (See 13 Doc. Nos. 1, 5.) Thus, in this court’s view the current situation is in fact one of the government’s 14 own making—not plaintiffs. In any event, the court does not find that the short time between 15 publication of the Final Rule and the filing of the complaint in this action constitutes a delay that 16 weighs in favor of denying injunctive relief, and this conclusion is supported by § 553(d)’s time 17 lag. Moreover, defendants’ reliance on the decision in Lydo Enters. v. City of Las Vegas, 745 18 F.2d 1211 (9th Cir. 1984) is misplaced, because the appellees in that action waited five years 19 before acting, not five days. 20 Furthermore, the court is persuaded by both plaintiffs’ and the State of California’s 21 arguments and finds that plaintiffs—and farmworkers beyond plaintiffs’ members—will suffer 22 irreparable harm absent the granting of the requested injunctive relief. As plaintiffs have noted, 23 defendants do little to dispute plaintiffs’ arguments regarding irreparable harm. As expressed at 24 the hearing on the motion, the court is also persuaded by plaintiffs’ argument that any harm 25 suffered from issuing interim AEWRs governing the period from January 1, 2021 to the FLR’s 26 publication in February 2021 pales in comparison to harm posed if the requested injunctive relief 27 is not granted. Although defense counsel at the hearing noted that mandatory injunctions are 28 disfavored, no such mandatory injunction is being sought here. Just as the court explained to the

34

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 34 of 39

1 USDA when it enjoined the FLS Suspension Notice, the court is not in this case directing the 2 DOL “to take any additional action that it otherwise would not have undertaken.” Perdue, 2020 3 WL 6318432, at *16. Rather, in granting injunctive relief in this case the court is merely 4 preserving the status quo by prohibiting the DOL from implementing the Final Rule and instead 5 ordering defendants to operate in accordance with the 2010 Rule, which was the last uncontested 6 status of the AEWR calculation methodology. See GoTo.com, Inc. v. Walt Disney Co., 202 F.3d 7 1199, 1210 (9th Cir. 2000) (“The status quo ante litem refers not simply to any situation before 8 the filing of a lawsuit, but instead to ‘the last uncontested status which preceded the pending 9 controversy.’”). 10 Contrary to defendants’ assertion, the court sees only one clear path in this case: the DOL 11 is required by its current regulations to publish AEWRs for 2021. See 20 C.F.R. § 655.120. The 12 court acknowledges that the current situation presents some difficulties for the DOL to overcome 13 in satisfying that requirement. As defense counsel noted at the hearing, there is no indication that 14 the May 2020 FLS data could be used to publish AEWRs for 2021 under the current governing 15 regulations. But the defendants are the experts in this area, and the court will not substitute its 16 judgment for theirs by crafting a solution to the predicament that the government and its agencies 17 have created. See Perdue, 2020 WL 6939021, at *2 n.1 (acknowledging plaintiffs’ suggestion 18 that the court may hold the government in contempt for “intentionally, and without adequate 19 [excuse], def[ying] a court order by” delaying compliance). 20 Moreover, there appear to be a number of appropriate solutions from which defendants 21 may choose. Plaintiffs persuasively suggest that interim AEWRs could satisfy the DOL’s 22 requirements because the regulations require that the agency “publish, at least once in each 23 calendar year, . . . the AEWRs for each State as a notice in the Federal Register.” (See Doc. No. 24 34 at 18 n.17) (citing 20 C.F.R. § 655.120(c)). Moreover, the court can extend the DOL’s 25 deadline to equitably facilitate defendants’ compliance with this order and the requirements of 26 law. See Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 524 F.3d 917, 936 (9th Cir. 2008) 27 (“The district court has broad latitude in fashioning equitable relief when necessary to remedy an 28 established wrong, and we review the district court’s choice of remedies within that scope for

35

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 35 of 39

1 abuse of discretion.”) (internal quotation marks and citation omitted); San Luis & Delta-Mendota 2 Water Auth. v. Salazar, 686 F. Supp. 2d 1026, 1050–51 (E.D. Cal. 2009) (extending deadlines for 3 the United States Fish and Wildlife Service to cure regulatory timeline issues created by 4 remanding agency action); accord Indep. Min. Co. v. Babbitt, 105 F.3d 502, 507 (9th Cir. 1997) 5 (“‘The APA provides that a court may compel ‘agency action unlawfully withheld or 6 unreasonably delayed.’ 5 U.S.C. § 706(1).”). Here, on November 16, 2020, the USDA indicated 7 the FLR would be published approximately nine weeks from the date that the USDA’s data 8 collection reinstatement notice was published in the Federal Register. Perdue, 1:20-cv- 01452- 9 DAD-JLT (Doc. No. 40-1 at ¶¶ 8–10). The USDA’s reinstatement notice was published in the 10 Federal Register on December 10, 2020. See Notice of Reinstatement of the Agricultural Labor 11 Survey Previously Scheduled for October 2020, 85 Fed. Reg. 79,463 (Dec. 10, 2020). Thus, the 12 FLR is expected to be published on or about February 11, 2021. The court finds it reasonable to 13 expect the DOL to set the 2021 AEWRs in compliance with this order and its regulations within 14 thirty (30) days of the FLR’s publication. This roughly aligns with the original timeline for 15 setting the AEWRs—“[t]he Farm Labor report was originally scheduled for publication on 16 November 25, 2020,” Perdue, 1:20-cv- 01452-DAD-JLT (Doc. No. 40-1 at ¶ 11), and the DOL 17 would have set the AEWRs no later than December 31, 2020. 18 The status quo to be preserved by this preliminary injunction order is one in which the 19 2010 Rule governs. Defendants will be required to operate under the requirements of the 2010 20 Rule pending further order of this court, except for the deadlines which have or are about to pass 21 as a result of the government’s actions. The parties will be directed to meet and confer as to the 22 setting of new deadlines in light of this order as well as the framework to apply during the brief 23 period of early 2021 before the new 2021 AEWRs can be announced. 24 Accordingly, the court concludes that consideration of the Winter factor of irreparable 25 harm weighs in favor of the granting of the requested injunctive relief. 26 4. Balance of the Hardships 27 Courts “must balance the competing claims of injury and must consider the effect on each 28 party of the granting or withholding of the requested relief,” and “should pay particular regard for

36

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 36 of 39

1 the public consequences in employing the extraordinary remedy of injunction.” Winter, 555 U.S. 2 at 24. “In assessing whether the plaintiffs have met this burden, the district court has a duty to 3 balance the interests of all parties and weigh the damage to each.” Stormans, Inc., 586 F.3d at 4 1138 (internal quotation marks and alteration omitted). “Where the government is a party to a 5 case in which a preliminary injunction is sought, the balance of the equities and public interest 6 factors merge.” Padilla v. Immigration & Customs Enf’t, 953 F.3d 1134, 1141 (9th Cir. 2020) 7 (citing Drakes Bay Oyster Co. v. Jewell, 747 F.3d 1073, 1092 (9th Cir. 2014)). As one district 8 court has stated:

9 There is generally no public interest in the perpetuation of unlawful agency action. To the contrary, there is a substantial public interest 10 in having governmental agencies abide by the federal laws that govern their existence and operations.

11

12 Washington v. DeVos, No. 2:20-cv-1119-BJR, 2020 WL 5079038, at *10 (W.D. Wash. Aug. 21, 13 2020) (citing League of Women Voters v. Newby, 838 F.3d 1, 12 (D.C. Cir. 2016)). 14 Here, plaintiffs argue that the balance of the hardships weighs in favor of the granting of 15 the requested injunctive relief because the public interest is served by preventing the DOL from 16 contravening the INA, preventing the wages of U.S. farmworkers from being depressed, and 17 facilitating the effective administration of the H-2A foreign guestworker visa program. (Doc. No. 18 5 at 32) (citing Gerstein v. CIA, No. 06-cv-4643, 2006 WL 3462659, at *5 (N.D. Cal. Nov. 29, 19 2006) (recognizing public interest is served by promoting Congress’s “core purpose” in enacting 20 regulatory program)). Plaintiffs also note that that the public interest is served by ensuring that 21 the DOL complies with the APA. (Doc. No. 5 at 31–32.) Plaintiffs add that the public interest 22 “in the ‘efficient administration of the immigration laws,’” which includes the H-2A program, “is 23 ‘weighty.’” (Doc. No. 5 at 32) (citing E. Bay, 950 F.3d at 1281). Moreover, injunctive relief 24 would serve the public interest by preserving the status quo. (Id.) (citing Doe #1 v. Trump, 957 25 F.3d 1050, 1069 (9th Cir. 2020) (holding “the public interest favors preserving the status quo”)). 26 Finally, those interests are particularly strong where an agency’s practice has “for countless 27 decades” allowed the government to administer “a stable immigration system.” (Id.) (citing 28 Trump, 957 F.3d at 1068).

37

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 37 of 39

1 In opposition, defendants reiterate that issuing a preliminary injunction will cause 2 irreparable harm to some farmworkers without preventing harm to others. (Doc. No. 31 at 30.) 3 Defendants note that during the comment process, other worker advocacy organizations 4 “commended” the DOL “for realizing the H-2A employers have increasingly utilized the H-2A 5 program for occupations that should be paid at higher wage rates than the historical AEWR 6 approach.” (Id. at 30–31) (citing Doc. No. 31-4 at 5).6 Defendants also assert that higher-skilled 7 farmworkers will be irreparably harmed if plaintiffs’ motion is granted because they will not 8 benefit from higher OES wages, and the DOL may be forced to assess operating the H-2A 9 program without any AEWR, which would harm nearly all farmworkers. (Id.) Additionally, 10 defendants state that preventing the Final Rule from taking effect will create unnecessary 11 confusion for employers and farmworkers alike if multiple rates will be set for 2021. (Id.) 12 Having considered the arguments of the parties, the court concludes that the balance of 13 hardships in this case tips in favor of the granting of injunctive relief. Indeed, considered alone 14 “[t]he government’s failure to comply with the APA—particularly given the strength of the 15 [plaintiffs’] procedural attack on the Rule—weighs in favor of granting injunctive relief.” E. Bay, 16 950 F.3d at 1281. Even assuming that higher-skilled farmworkers will suffer a harm if the Final 17 Rule is enjoined and ultimately reinstated, and that setting multiple rates will be confusing for 18 employers and farmworkers, those hardships are marginal compared to the hardship that field and 19 livestock workers—who make up “the overwhelming majority of H-2A job opportunities,” 85 20 Fed. Reg. at 70,461—will suffer if a likely unlawful rule is implemented. This is particularly true 21 in light of the fact that these farmworkers will face barriers to obtaining any applicable back pay.

22 6 This comment letter—submitted by the general counsel of Justice at Work in Pennsylvania 23 (formerly Friends of Farmworkers, Inc.) and their client Comite de Trabajadores Agricolcas—

states in the subsequent paragraph that the “DOL has failed to recognize that its methodology 24 should require the payment of the highest wage rate that can be determined based on available data as being paid to U.S. workers in the area of employment.” (Doc. No. 31-4 at 5.) The 25 commenter then states that “[w]here the OES occupational survey established an average wage 26 rate for an occupational code that wage rate should be the basis for the Adverse Effect Wage Rate unless farm labor survey data or another valid data source establishes a higher average wage rate 27 in the area of employment.” (Id. at 6.) In sum, while some worker advocacy organizations may have supported the Final Rule, the position taken by the commenter cited by defendants is not 28 itself completely aligned with the Final Rule.

38

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 38 of 39

1 Because of the hardships that the public will face if defendants are not prevented from 2 implementing the November 5, 2020 Final Rule amending the DOL’s regulations governing the 3 AEWR calculation methodology, the court finds that the balance of the hardships in this case 4 weighs in favor of the granting of the injunctive relief requested by plaintiffs. 5 5. Federal Rule of Civil Procedure 65(c) Security Bond 6 Finally, the court will waive Federal Rule of Civil Procedure 65(c)’s security bond 7 requirement. That rule states that the court may issue a preliminary injunction “only if the 8 movant gives security in an amount that the court considers proper to pay the costs and damages 9 sustained by any party found to have been wrongfully enjoined or restrained.” The district court 10 has discretion “as to the amount of security required, if any” and “may dispense with the filing of 11 a bond when it concludes there is no realistic likelihood of harm to the defendant from enjoining 12 his or her conduct.” Johnson v. Couturier, 572 F.3d 1067, 1086 (9th Cir. 2009) (internal 13 quotation marks and citation omitted). 14 Here, the court finds requiring a security bond to be posted in this case not to be 15 warranted, particularly in light of the fact that plaintiffs have already posted a $1,000.00 bond in 16 Perdue. See Perdue, 1:20-cv- 01452-DAD-JLT (Doc. Nos. 33, 35)). Notably, defendants have 17 not requested a bond, the balance of the hardships weighs in favor of plaintiffs, and the granting 18 of injunctive relief is in the public interest. The court therefore finds that requiring a security 19 bond is unwarranted under these circumstances. See E. Bay Sanctuary Covenant v. Trump, 349 F. 20 Supp. 3d 838, 868–69 (N.D. Cal. 2018) (waiving a bond where defendants had not requested a 21 bond, the balance of the hardships weighed strongly in favor of plaintiffs, and there was a 22 significant public interest underling the action), aff’d, 950 F.3d 1242 (9th Cir. 2020), and aff’d, 23 950 F.3d 1242 (9th Cir. 2020). 24 ///// 25 ///// 26 ///// 27 //// 28 ////

39

Case 1:20-cv-01690-DAD-JLT Document 37 Filed 12/23/20 Page 39 of 39

1 CONCLUSION 2 For the reasons set forth above, 3 1. Plaintiffs’ motion for a preliminary injunction (Doc. No. 5) is granted; 4 2. The court orders that defendants shall be prevented from implementing the 5 November 5, 2020 Final Rule amending the DOL’s regulations governing the 6 AEWR calculation methodology and are ordered to operate under the 2010 Rule as 7 it pertains to calculating the AEWRs; 8 3. The parties are ordered to meet and confer within fourteen (14) days to submit a 9 proposed order that includes deadlines by which defendants will set the 2021 10 AEWRs in accordance with this order and other legal requirements; and 11 4. No bond will be required to be posted by plaintiffs pursuant to Rule 65(c) of the 12 Federal Rules of Civil Procedure.

13 IT IS SO ORDERED. 14 Dated: December 23, 2020 15 UNITED STATES DISTRICT JUDGE

16 17 18 19

20 21

22 23 24 25 26