1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 7 CENTER FOR FOOD SAFETY, et al., Case No. 20-cv-00256-JSW

8 Plaintiffs,

ORDER RESOLVING CROSS-

9 v. MOTIONS FOR SUMMARY

JUDGMENT

10 SONNY PERDUE, et al., Re: Dkt. Nos. 65, 68 Defendants. 11

12 13 Now before the Court for consideration are the motion for summary judgment filed by 14 Plaintiffs Food & Water Watch (“FWW”), Center for Food Safety (“CFS”), the Humane Farming 15 Association (“HFA”), and Robin Mangini (“Mangini”) (collectively, “Plaintiffs”) and the cross- 16 motion for summary judgment filed by the U.S. Department of Agriculture (“USDA”), Secretary 17 of Agriculture Thomas Vilsack, the Food Safety and Inspection Service (“FSIS”), and Deputy 18 Under Secretary for Food Safety Sandra Eskin (collectively, “Defendants”). The Court has 19 reviewed the parties’ papers, relevant legal authority, and the record in the case, and it finds this 20 matter suitable for disposition without oral argument.1 See N.D. Civ. L.R. 7-1(b). For the 21 following reasons, the Court DENIES Plaintiffs’ motion for summary judgment and GRANTS 22 Defendants’ cross-motion for summary judgment. 23 BACKGROUND 24 Plaintiffs are three non-profit consumer organizations and one of their individual members 25 who bring this challenge to the USDA’s Modernization of Swine Slaughter Inspection Rule. See 26 84 Fed. Reg. 52,300 (Oct. 1, 2019) (“Final Rule”). Plaintiffs allege that the new rule, which 27 1 implements a new inspection system for swine slaughter, is contrary to the Federal Meat 2 Inspection Act (“FMIA”) and violates the Administrative Procedure Act (“APA”). 3 Plaintiffs allege that the FMIA requires federal government inspectors to examine and 4 inspect every animal prior to slaughter and every carcass and body part after slaughter. Plaintiffs 5 contend, however, that the new inspection system established by the Final Rule transfers these 6 federal inspection duties to the slaughter plant employees, which prevents federal inspectors from 7 conducting an adequate appraisal of animals, carcasses, and parts as required by the FMIA. 8 Plaintiffs also allege that the rulemaking process was arbitrary and capricious in violation 9 of the APA because the Final Rule (1) irrationally departs from prior inspection regulations and 10 practices; (2) is based on a flawed pilot project; and (3) relies on a flawed risk assessment for 11 which the public was not provided adequate time to provide notice and comment. 12 A. The Federal Meat Inspection Act. 13 Congress enacted the FMIA in response to unsanitary conditions in meatpacking plants. 14 The FMIA authorizes the FSIS, a component of USDA, with protecting consumer health and 15 welfare by ensuring that meat products are safe and unadulterated. 21 U.S.C. § 602. The FMIA 16 authorizes FSIS to appoint inspectors and public health veterinarians (“PHVs”) to conduct ante- 17 mortem and post-mortem inspections of carcasses intended for use as human food. Id. §§ 603- 18 604. 19 In furtherance of its mission of protecting consumer health and welfare, FSIS has 20 promulgated regulations that govern ante- and post-mortem examinations at swine slaughter 21 establishments. See, e.g., 9 C.F.R. pts 309 & 310; Swine Post-Mortem Inspection Procedures and 22 Staffing Standards, 50 Fed. Reg. 19,900 (May 13, 1985); Cattle & Swine Post-Mortem Inspection 23 Procedures and Staffing Standards, 47 Fed. Reg. 33,673 (Aug. 4, 1982). 24 B. The Traditional Inspection System. 25 The traditional inspection system established by the FSIS requires federal inspectors to 26 inspect all swine before and after slaughter. During ante-mortem inspections, federal inspectors 27 examine all livestock offered for slaughter while at rest and in motion and direct establishment 1 further inspection by a PHV. See 9 C.F.R. § 309.1. During post-mortem inspections, 2 establishments present the head, viscera, and carcass of each animal for inspection by FSIS. 9 3 C.F.R. § 310.1(b)(3). Federal inspectors look for disease through “organoleptic inspections” and 4 by incising and palpating certain lymph nodes. Modernization of Swine Slaughter Inspection, 83 5 Fed. Reg. at 4780, 4783 (Feb. 1, 2018) (“Proposed Rule”). The traditional inspection system 6 requires up to seven FSIS inspectors per evisceration line. Id. Under the traditional inspection 7 system, establishment employees are not required to sort carcasses before the post-mortem 8 inspection to identify or remove contamination or to flag potentially condemnable carcasses or 9 parts for further scrutiny by FSIS. Id. 10 C. The HIMP Pilot Program. 11 In the 1990s, FSIS proposed changes to the inspection process. FSIS acknowledged that 12 the FSIS’s inspection procedure had not kept up with advancements in the control and eradication 13 of animal diseases. See HACCP-Based Meat and Poultry Inspection Concepts, 62 Fed. Reg. 14 31,553 (June 10, 1997). FSIS also found that the traditional inspection method failed to efficiently 15 use its resources, including failing to adequately target foodborne pathogens such as Salmonella 16 and E. coli. Id. 17 As a result, FSIS began to look for ways to modernize and increase the efficiency of the 18 inspection process. As a first step, FSIS required establishments to develop a system of 19 preventative controls designed to ensure their products are safe. See Pathogen Reduction/Hazard 20 Analysis and Critical Control Point (“HACCP”) Systems, 61 Fed. Reg. 38,806 (1996). With these 21 measures, FSIS adopted a new approach that required establishments to play a larger role in 22 ensuring product safety. See 83 Fed. Reg. at 4787; 61 Fed. Reg. at 38,808. 23 To assess the efficacy of the new approach, FSIS developed the HACCP-Based Inspection 24 Models Project (“HIMP”) pilot, which was designed to test the new inspection models in 25 volunteer meat and poultry slaughter establishments. See HACCP-Based Meat and Poultry 26 Inspection Concepts, 62 Fed. Reg. 31,553 (June 10, 1997). Under the HIMP pilot, plant 27 employees performed the task of separating normal from abnormal carcasses and parts; federal 1 performance standards and regulatory requirements. 2 The D.C. Circuit struck down the initial HIMP pilot program as contrary to 21 U.S.C. 3 section 604 (“Section 604”) of the FMIA, finding that the proposed model impermissibly 4 “[d]elegat[ed] the task of inspecting carcasses to plant employees” and relegated federal inspectors 5 to an oversight and verification role in violation of the clear mandate of the FMIA. Am. Fed’n of 6 Gov’t Emps., AFL-CIO v. Glickman, 215 F.3d 7, 11 (D.C. Cir. 2000) (“AFGE I”). 7 Following AFGE I, FSIS modified the HIMP pilot to comply with the D.C. Circuit’s 8 ruling. The modified pilot program placed federal inspectors at up to three fixed locations along 9 the slaughter lines with the responsibility of examining the head, carcass, and viscera of all hogs; 10 other federal inspectors would be assigned to plants to verify the effectiveness of the plant’s 11 process control systems. The D.C. Circuit held that the modified HIMP pilot program satisfied the 12 FMIA because it required “federal inspectors in participating hog plants to inspect all hog 13 carcasses, heads and viscera, as the statute demands.” Am. Fed’n of Gov’t Emps., AFL-CIO v. 14 Veneman, 284 F.3d 125, 130 (D.C. Cir. 2002) (“AFGE II”). However, the court noted that 15 because the program was a test program, were Defendants to undertake rulemaking “to adopt [it] 16 as a permanent change,” the opinion in that case “may not necessarily foreshadow the outcome of 17 judicial review of such future regulations.” Id. at 130-31. 18 In 2000, the revised HIMP pilot program was initiated in five large, high-volume hog 19 slaughter establishments. See 83 Fed. Reg. at 4787-88. Under the HIMP pilot, establishment 20 employees remove and dispose of animals that are dead, moribund, or display condemnable 21 conditions before federal inspectors conduct ante-mortem inspection. Id. at 4788. The 22 establishment employees also sort animals that appear to be healthy into “Normal” pens and sort 23 animals that appear to have condemnable diseases or conditions into “Subject” pens. Id. Federal 24 inspectors then examine, at rest, all animals deemed “Normal”, and they examine five to ten 25 percent of those animals in motion. Id. If any of the animals exhibit signs of condemnable 26 conditions, federal inspectors direct establishment employees to move the animals to “U.S. 27 Suspect” pens for final disposition by a federal PHV. Id. The federal PHVs examine all animals 1 pens for final disposition by a federal PHV. Id. The federal PHV determines if any animals must 2 be identified as “U.S. Condemned” and disposed of in accordance with 9 C.F.R. section 309.13. 3 Id. 4 During post-mortem inspection, establishment personnel sort carcasses and parts and trim 5 dressing defects and contamination (e.g., hair, bruises, feces, ingesta, and milk), mark with ink 6 localized pathology defects intended for removal under FSIS supervision, and identify carcasses 7 and parts intended for disposal under FSIS supervision. Id. Subsequently, FSIS online inspectors 8 visually inspect the head, viscera, and carcass of each hog at fixed locations on the evisceration 9 line. Id. FSIS represents that the plant employees’ pre-sorting permits the federal online 10 inspection to occur much more efficiently under HIMP than under the traditional inspection 11 system. Id. As a result of the increase in online efficiency, FSIS states federal inspectors conduct 12 more offline, food safety related verification inspection activities under HIMP than under the 13 traditional inspection system. Id. 14 In 2013, the U.S. General Accountability Office (“GAO”) and the USDA’s Office of the 15 Inspector General (“OIG”) evaluated the HIMP pilot program. 83 Fed. Reg. at 4781. The GAO 16 report identified certain strengths in the HIMP pilot but also recommended that FSIS collect and 17 analyze additional information to determine the plants’ compliance with the performance 18 standards. AR_101756-806. The OIG report concluded that FSIS did not adequately oversee the 19 HIMP program and made several recommendations related to HIMP procedures. AR_101706- 20 755. 21 D. The Hog HIMP Report. 22 In 2014, FSIS finalized a report detailing and evaluating the results of the HIMP pilot 23 program (“Hog HIMP Report”). 83 Fed. Reg. at 4787-88; see also AR_101990-102037. The Hog 24 HIMP Report assessed the performance of five HIMP market hog slaughter establishments 25 between 2006 through 2010 and 2012 through 2013.2 Id. The five HIMP establishments were 26 2 The results from 2006 to 2010 were based on data from the Performance Based Inspection 27 System (“PBIS”) database. In 2011, FSIS began transitioning establishments to the new Public 1 compared to twenty-one (21) non-HIMP market hog slaughter establishments “of comparable 2 production volume, line speed, and days of operation.” Id. 3 The report evaluated whether the new system performed comparably to traditional 4 inspection by evaluating several proxies including the relative number of carcasses that plant 5 employees and federal inspectors removed from the slaughter lines and the number of federal 6 inspectors’ offline verifications. The report concluded that HIMP establishments “receiv[ed] more 7 off-line food safety related inspection verification checks than the traditional non-HIMP market 8 hog establishments” and that HIMP establishments “have higher compliance with Sanitation SOP 9 and HACCP regulations, lower levels of non-food safety defects, equivalent or better Salmonella 10 verification testing positive rates” than traditional plants. AR_102023. In addition, the report 11 evaluated data from plants for two periods, 2009 through 2010 and 2012 through 2013, to assess 12 HIMP establishments’ compliance with FSIS’s Food Safety and Other Consumer Protection 13 performance standards. Id. The results suggested that the HIMP establishments were performing 14 as well as comparable large non-HIMP market hog establishments. AR_102024. 15 The Hog HIMP Report concluded that market hog slaughter establishments participating in 16 the HIMP pilot were performing as well as comparable large non-HIMP market hog 17 establishments and meeting FSIS requirements for operating under waivers through the HIMP 18 projects. 83 Fed. Reg. at 4788-4789. 19 E. Proposed Rule. 20 In February 2018, FSIS published a notice of proposed rulemaking creating a new, 21 optional inspection system, NSIS, for market hog establishments based on the HIMP pilot 22 program. See 83 Fed. Reg. at 4780. The goals of the Proposed Rule included reducing pathogens 23 in pork products, improving the effectiveness of FSIS inspection, increasing the efficiency of FSIS 24 resources, and allowing establishments flexibility to innovate. Id. FSIS released a draft risk 25 assessment and compliance guidelines regarding plant employee sorting. Id. at 4793-94. FSIS 26 relied on the findings in the Hog HIMP Report and adopted the HIMP pilot’s key features 27 1 regarding plant employee ante- and post-mortem sorting. The Proposed Rule cited the draft risk 2 assessment’s predicted annual reduction of 2,533 Salmonella illnesses as a main public benefit of 3 the new rule. Id. at 4811-12. 4 The proposed rule received 83,000 comments. 84 Fed. Reg. at 52,300. Several comments 5 challenged the rule for preventing federal inspectors from critically appraising animals, failing to 6 provide training to plant sorters. Commenters also questioned the validity of the Hog HIMP 7 Report and the risk assessment. Id. 8 F. The Final Rule. 9 On October 1, 2019, FSIS published the Final Rule, which became effective on December 10 2, 2019. 84 Fed. Reg. at 52,300. The Final Rule adopted the key parts of the Proposed Rule, 11 including that plant employees would engage in pre-inspection sorting.3 Id. The NSIS system 12 established in the Final Rule is voluntary, and FSIS predicted that only high-volume market hog 13 establishments would opt in to NSIS. Id. at 52,322. In 2016, there were 40 high-volume, market- 14 hog-only establishments. Id. FSIS predicted the Final Rule would result in $62 million in savings 15 over the next two years, increase establishments’ compliance with sanitation regulations, and 16 improve welfare by increasing monitoring of humane handling procedures by FSIS inspectors. Id. 17 at 52,335-36. Unlike in the Proposed Rule, which predicted an annual reduction in Salmonella 18 illnesses, the Final Rule stated that the Final Rule may lead to lower prevalence of Salmonella on 19 hog carcasses and thus fewer illnesses. Id. at 52,332. 20 G. The Risk Assessment. 21 In January 2018, FSIS published a risk assessment that assessed the potential change in 22 human risk of Salmonella in HIMP establishments. See AR_104395-104523. The risk assessment 23 found a statistically significant correlation between increased offline inspection procedures and a 24

25 3 The Final Rule also eliminated line speed limits. See 9 C.F.R. § 310.26. A federal district court granted summary judgment in favor of union plaintiffs who challenged the Final Rule’s lifting of 26 line speed limits as arbitrary and capricious. The court found FSIS did not adequately consider the comments it received regarding the effects of eliminating line speeds on worker safety. The court 27 vacated that portion of the Final Rule but did not set aside any other aspect of the Final Rule. See 1 reduction in the prevalence of Salmonella on market hog carcasses at HIMP establishments. 2 Specifically, the risk assessment concluded that establishments operating under HIMP would lead 3 to an expected reduction in market hog attributable Salmonella illnesses to 2,533 cases. The risk 4 assessment concluded that shifting FSIS resources from online inspections to offline inspections 5 would likely result in a decrease in human illnesses from contaminated pork. AR_104410. 6 FSIS published the risk assessment along with the Proposed Rule before it underwent peer 7 review. The risk assessment’s peer review occurred in April 2018 after the public comment period 8 on the Proposed Rule closed. AR_104524. In August 2018, FSIS posted a summary of peer- 9 reviewer feedback and an updated draft risk assessment on its website. See AR_104692-104756; 10 AR_104757-104933. FSIS provided thirty days for comments on the revised risk assessment. 11 AR_103790. FSIS did not reopen comments for the Final Rule because it stated that the revised 12 risk assessment did not change the conclusions in the risk assessment, and FSIS advised that it 13 would respond to the risk assessment comments in the final rule. 14 When FSIS published the Final Rule, it included another updated version of the risk 15 assessment, updated in September 2019. AR_104934-105118. The September 2019 version of 16 the risk assessment projected that the potential benefits reflected increased uncertainty around the 17 human risk from Salmonella, with a potential annual increase of 1,719 cases and a potential 18 decrease of 6,685 cases. See 84 Fed. Reg. at 52,333. However, FSIS maintained that Salmonella 19 illnesses could reduce by an average of 2,533 per year. Id. at 52,334. 20 H. FOIA Requests. 21 Plaintiff FWW sought FSIS’s data for pilot plants from 2014 through 2017 through 22 Freedom of Information Act (“FOIA”) requests. See AR_104331-344. FWW analyzed the data it 23 received and concluded that: (1) offline inspectors at HIMP plants were not able to perform more 24 offline inspections than those at traditional plants from 2014 to 2017; (2) federal PHVs were able 25 to evaluate fewer animals tagged as U.S. Suspect in HIMP plants than in traditional plants from 26 2012 through 2015; and (3) HIMP plants had nearly two times the regulatory violations rate of 27 traditional plants for fecal matter, digestive content, and milk contamination. FWW submitted its

I. Present Litigation.

1 Plaintiffs filed this lawsuit challenging the Final Rule on January 13, 2020 and amended 2 their complaint on April 6, 2020. (Dkt. Nos. 1, 26.) Plaintiffs moved for summary judgment on 3 January 14, 2022, and Defendants filed an opposition and cross-motion for summary judgment on 4 February 24, 2022. On April 20, 2022, the Court granted Plaintiffs’ request to voluntarily dismiss 5 the fourth and fifth claims for relief with prejudice. (Dkt. No. 85.) 6 The Court will discuss additional facts as necessary in the analysis. 7

ANALYSIS

8 A. Applicable Legal Standard. 9 Under the APA, a court must “hold unlawful and set aside agency action ... found to be— 10 arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law,” or “without 11 observance of procedure required by law.” 5 U.S.C. § 706(2). “An agency must ‘examine the 12 relevant data and articulate a satisfactory explanation for its action including a rational connection 13 between the facts found and the choice made.’ ” Gill v. U.S. Dep’t of Justice, 913 F.3d 1179, 14 1187 (9th Cir. 2019) (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. 15 Ins. Co., 463 U.S. 29, 43 (1983)). Agency action is “arbitrary and capricious if the agency has 16 relied on factors which Congress has not intended it to consider, entirely failed to consider an 17 important aspect of the problem, offered an explanation for its decision that runs counter to the 18 evidence before the agency, or is so implausible that it could not be ascribed to a difference in 19 view or the product of agency expertise.” Motor Vehicle Mfrs., 463 U.S. at 43; see also Gill, 913 20 F.3d at 1187. The basis for the agency’s decision must come from the record. Gill, 913 F.3d at 21 1187. 22 A reviewing court’s inquiry must be “thorough,” but “the standard of review is highly 23 deferential; the agency's decision is ‘entitled to a presumption of regularity,’ and [the court] may 24 not substitute [its] judgment for that of the agency.” San Luis & Delta-Mendota Water Auth. v. 25 Jewell, 747 F.3d 581, 601 (9th Cir. 2014) (quoting Citizens to Preserve Overton Park, Inc. v. 26 Volpe, 401 U.S. 402, 416 (1971)). Although a court’s review is deferential, the court “must 27 engage in a careful, searching review to ensure that the agency has made a rational analysis and 1 decision on the record before it.” Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 524 F.3d 2 917, 927 (9th Cir. 2008); see also Brower v. Evans, 257 F.3d 1058, 1067 (9th Cir. 2001) (“The 3 presumption of agency expertise can be rebutted when its decisions, while relying on scientific 4 expertise, are not reasoned.”). A court “must not ‘rubber-stamp’ ... administrative decisions that 5 [it] deem[s] inconsistent with a statutory mandate or that frustrate the congressional policy 6 underlying a statute.” Ocean Advocates v. U.S. Army Corps of Eng’rs, 402 F.3d 846, 859 (9th Cir. 7 2005) (first alteration in original, remaining alterations added). A court, however, may “uphold a 8 decision of less than ideal clarity if the agency’s path may reasonably be discerned.” Gill, 913 9 F.3d at 1187-88 (quoting Motor Vehicle Mfrs., 463 U.S. at 43). 10 The reasoned-decisionmaking requirement the Supreme Court has often observed, includes 11 a duty to explain any “departure from prior norms.” Atchison, Topeka & Santa Fe Ry. Co. v. 12 Wichita Bd. of Trade, 412 U.S. 800, 808 (1973); see also Int’l Union, UAW v. NLRB, 802 F.2d 13 969, 973-74 (7th Cir. 1986) (“[A]n administrative agency is not allowed to change direction 14 without some explanation of what it is doing and why.”). “Unexplained inconsistency” between 15 agency actions is “a reason for holding an interpretation to be an arbitrary and capricious change.” 16 Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981 (2005). 17 A party is entitled to summary judgment if the “movant shows that there is no genuine 18 dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. 19 Civ. P. 56(a). In an action reviewing the merits under the APA, however, the Court does not ask 20 whether there is a genuine dispute as to any material fact. Rather, “the function of the district 21 court is to determine whether or not as a matter of law the evidence in the administrative record 22 permitted the agency to make the decision it did.” Occidental Eng’g Co. v. I.N.S., 753 F.2d 766, 23 769 (9th Cir. 1985). In an APA-review case, “summary judgment is an appropriate mechanism 24 for deciding the legal question of whether the agency could reasonably have found the facts as it 25 did.” Id. at 770. 26 B. Article III Standing. 27 The parties first dispute whether Plaintiffs have Article III standing to pursue their claims. 1 standing to pursue their claims. (See Dkt. No. 39.) Defendants renew their challenge to Plaintiffs’ 2 Article III standing in their cross-motion for summary judgment. 3 “[T]o satisfy Article III’s standing requirements, a plaintiff must show (1) it has suffered 4 an ‘injury in fact’ that is (a) concrete and particularized and (b) actual or imminent, not conjectural 5 or hypothetical; (2) the injury is fairly traceable to the challenged action of the defendant; and (3) 6 it is likely, as opposed to merely speculative, that the injury will be redressed by a favorable 7 decision.” Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 180-81 8 (2000) (citing Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-561 (1992)). 9 The party asserting federal jurisdiction has the burden of proving the three elements of 10 standing, which are “not mere pleading requirements but rather an indispensable part of the 11 plaintiff's case.” Lujan, 504 U.S. at 561. Therefore, “each element must be supported in the same 12 way as any other matter on which the plaintiff bears the burden of proof, i.e., with the manner and 13 degree of evidence required at the successive stages of the litigation.” Id. On a motion for 14 summary judgment, “the plaintiff can no longer rest on such ‘mere allegations,’ but must ‘set 15 forth’ by affidavit or other evidence ‘specific facts,’ ...which for purposes of the summary 16 judgment motion will be taken to be true.” Id. (quoting Fed. R. Civ. P. 56.) 17 Plaintiffs submit numerous declarations in support of standing, including declarations from 18 two former USDA PHVs who attest that the changes to the inspection process in the Final Rule 19 are linked to increased food-safety violations, which will lead to an increased risk that consumers 20 will contract foodborne illness from pork products. (See Basu Decl. at ¶¶ 2, 9-20; Martin Decl. ¶ 21 6-21.) Additionally, Plaintiffs cite evidence in the record linking contamination issues at plants 22 operating under NSIS with foodborne illness. The organizations’ declarant members attest that 23 they have regularly consumed pork and will suffer injury because they will be exposed to pork 24 products from NSIS plants if they continue to consume pork, or will have to spend time and 25 resources to limit their exposure to the higher-risk product. 4 (See Mangini Decl. ¶¶ 9-14, Samuels 26 Decl. ¶¶ 6-19, Sperling Decl. ¶¶ 4-5, 9-17, Rubio Decl. ¶¶ 3-10; Lapp Decl. ¶¶ 5-12, Levenback 27 1 Decl.¶¶ 5-10.) 2 Defendants assert that Plaintiffs’ declarations rest on the erroneous assertion that NSIS- 3 produced pork is “higher risk” than pork produced under the traditional inspection system. 4 According to Defendants, NSIS-produced pork is not “higher risk” than pork produced at non- 5 NSIS establishments, and thus the Final Rule does not injure Plaintiffs or their members. In 6 support of its argument, Defendants point to evidence in the record, including the Hog HIMP 7 Report and risk assessment. 8 Defendants’ argument is effectively an argument on the merits, and “the Court cannot 9 consider the merits when deciding standing.” Grand Canyon Tr. v. Williams, 98 F. Supp. 3d 1044, 10 1057 (D. Ariz. 2015). Defendants contend that the issue of whether the Final Rule exposes 11 “Plaintiffs to an increased risk of contracting foodborne illness is distinct from whether the Final 12 Rule violates the FMIA.” (Defs. Reply at 3:7-8.) However, Defendants’ standing arguments rely 13 on the Hog HIMP Report and risk assessment, the accuracy and validity of which are intertwined 14 with the merits of Plaintiffs’ claims. For purposes of the standing inquiry, the Court must assume 15 the truth of Plaintiff’s claims and that Plaintiffs will succeed on the merits. Grand Canyon Tr., 98 16 F. Supp. 3d at 1057. Thus, the Court concludes that Plaintiffs have put forth sufficient evidence to 17 establish Article III standing. 18 C. Plaintiffs’ Analysis of Data. 19 In support of their motion for summary judgment, Plaintiffs’ counsel submitted analysis of 20 data relating to (1) the evaluations of animals and carcasses by PHVs from 2011 to 2019 for HIMP 21 pilot plants and comparably sized non-HIMP plants and (2) Food Safety Standard 2 (“FS-2”) 22 violation rates. (See Dkt. No. 65-1, Declaration of Zachary B. Corrigan ¶¶ 3, 5.) Plaintiffs assert 23 that the analysis is derived from raw data in the administrative record. Plaintiffs ask the Court to 24 consider their analysis as “calculation[s] to prove the content of voluminous writings, recordings, 25 or photographs that cannot be conveniently examined in court[,]” under Federal Rule of Evidence 26 1006. Alternatively, Plaintiffs contend the analysis of the data is supplementary information that 27 can be considered under the exception in Lands Council v. Powell, 395 F.3d 1019 (9th Cir. 2005) 1 provide[s] it in a more understandable format to the Court.” See Stop B2H Coal v. BLM, 552 F. 2 Supp. 3d 1101, 1121 (D. Or. Aug. 4, 2021). 3 Rule 1006 provides that the proponent of evidence “may use a summary, chart, or 4 calculation to prove the content of voluminous writings, recordings, or photographs that cannot be 5 conveniently examined in court.” The Ninth Circuit has explained: 6 Charts and summaries as evidence are governed by Federal Rule of Evidence 1006. In contrast, charts or summaries of testimony or 7 documents already admitted into evidence are merely pedagogical devices, and are not evidence themselves. However, we have not 8 articulated a bright-line rule against admission of summary charts as evidence. Although we do not approve of receiving summary exhibits 9 of material already in evidence, we have not reversed for that reason. We have also elsewhere recognized a district court’s discretion under 10 Federal Rule of Evidence 611(a) to admit summary exhibits for the purpose of assisting the jury in evaluating voluminous evidence. 11 12 United States v. Anekwu, 695 F.3d 967, 981-82 (9th Cir. 2012). 13 Here, Plaintiffs’ analysis is not a summary of voluminous evidence; it is analysis of raw 14 data that it is already in the record. Although Plaintiffs’ evidence might assist the Court in 15 understanding data, that does not make the evidence admissible as summaries of voluminous 16 evidence under Rule 1006. 17 Plaintiffs alternatively argue the Court may consider its data analysis under the Lands 18 Council exceptions, which permit a court to review extra-record material when: (1) it is necessary 19 to determine whether the agency has considered all relevant factors and explained its decision, (2) 20 the agency has relied on documents not in the record, (3) supplementing the record is necessary to 21 explain technical terms or complex subject matter, or (4) plaintiffs make a showing of bad faith. 22 City of Las Vegas v. F.A.A., 570 F.3d 1109, 1116 (9th Cir. 2009) (discussing what are commonly 23 known as the “Lands Council” exceptions). The exceptions operate to “identify and plug holes in 24 the administrative record.” Lands Council, 395 F.3d at 1030. These exceptions are to be 25 “narrowly construed and applied” to ensure that they do not undermine the general rule limiting 26 review to the administrative record. Id. The party seeking admission of the extra-record material 27 “initially bears the burden of demonstrating that a relevant exception applies.” San Luis & Delta 1 Plaintiffs assert that counsel’s data analysis can be considered under the third Lands 2 Council exception because it takes technical and complex information from the record and 3 provides it in a more understandable format. Plaintiffs’ counsel’s calculations of the HIMP pilot 4 plants violation rates in 2012 and 2013 challenges FSIS’s analysis of that data; it does not provide 5 the data in a more understandable format. (See Corrigan Decl. ¶ 5.) However, the Court 6 concludes it may consider counsels’ analysis of data from after 2013 because FSIS did not analyze 7 that data in the Final Rule; thus, the Court will consider the data for the purpose of determining 8 whether the agency considered all relevant factors and explained its decision. 9 D. Whether the Final Rule Violates the FMIA. 10 1. Section 603(a). 11 Plaintiffs argue that the Final Rule violates 21 U.S.C. section 603(a) (“Section 603(a)”) for 12 two reasons: (1) it prevents federal inspectors from critically appraising hogs before slaughter; and 13 (2) it prevents the inspection and separate slaughter of swine showing symptoms of disease. 14 Section 603(a) of the FMIA provides: 15 [T]he Secretary shall cause to be made, by inspectors appointed for that purpose, an examination and inspection of all amenable species 16 before they shall be allowed to enter into any slaughtering, packing, meat-canning, rendering, or similar establishment, in which they are 17 to be slaughtered and the meat and meat food products thereof are to be used in commerce; and all amenable species found on such 18 inspection to show symptoms of disease shall be set apart and slaughtered separately from all other cattle, sheep, swine, goats, 19 horses, mules, or other equines… 20 21 U.S.C. § 603(a). 21 Plaintiffs assert that “inspection” in the FMIA means that federal inspectors must “pay[] 22 close attention to” and “critical[ly] apprais[e]” each animal prior to slaughter. AFGE I, 215 F.3d 23 at 11. Plaintiffs argue the Final Rule prevents federal inspectors from critically appraising animals 24 prior to slaughter because plant employees—not federal inspectors—conduct a pre-inspection 25 sorting of the animals and remove animals unfit for slaughter. See 9 C.F.R. § 309.19(a) (requiring 26 establishment employees to “conduct market hog sorting activities before the animals are 27 presented for ante-mortem inspection”); see also 83 Fed. Reg. at 4792 (“establishment personnel 1 they are presented to FSIS PHVs for inspection and final disposition.”). Thus, Plaintiffs contend 2 that in NSIS plants, plant employees will take over jobs that are performed by trained federal 3 inspectors in traditional plants. 4 Defendants assert that the pre-inspection sorting conducted by plant employees does not 5 replace federal inspection; rather, it provides an additional step in the process before federal 6 inspection occurs. See 84 Fed. Reg. at 52,311 (“The new inspection system will not eliminate 7 FSIS inspection. NSIS simply requires establishments to take additional steps before FSIS 8 inspection to ensure that their products are safe and wholesome.”). Defendants contend that pre- 9 inspection sorting by plant employees is consistent with the requirements of the FMIA. 10 The Court agrees with Defendants. Under NSIS, federal inspectors still inspect each 11 animal before it is slaughtered for meat. During the pre-inspection sorting process, plant 12 employees sort animals that appear to be healthy into “Normal” pens and animals that appear to 13 have diseases or abnormal conditions into “Subject” pens. 83 Fed. Reg. at 4792. The federal 14 inspectors then inspect all animals found to be normal at rest, and five to ten percent of those 15 animals in motion. Id. If inspectors determine any of the “Normal” animals exhibit condemnable 16 conditions, FSIS inspectors direct establishment employees to move the animals to “U.S. Suspect” 17 pens for final disposition by the FSIS PHV. Id. The PHV inspects all animals in the “Subject” 18 and “U.S. Suspect” pens. Id.; 84 Fed. Reg. 52, 311. Thus, the pre-inspection sorting process does 19 not replace federal inspection; federal inspectors still inspect the animals “before they shall be 20 allowed to enter into any slaughtering…establishment, in which they are to be slaughtered and the 21 meat and meat food products…” 21 U.S.C. § 603(a). 22 Plaintiffs also contend that federal inspectors are unable to critically appraise animals in 23 violation of the FMIA because under NSIS, federal inspectors are required to observe only five to 24 ten percent of the “Normal” animals in motion. Plaintiffs assert that federal inspectors cannot 25 inspect the animals as required by Section 603(a) without observing them in motion. 26 Section 603(a) requires an “examination and inspection of all [swine].” 21 U.S.C. § 27 603(a). It does not require all swine to be examined at rest and in motion. Under NSIS, federal 1 the “[t]he overall condition of each animal, including the head, with attention to the eyes, the legs, 2 and the body of the animal,” “[t]he degree of alertness, mobility, and breathing,” and “[w]hether 3 there are any unusual swellings or any other abnormalities.” AR_100705. Moreover, although it 4 is true that NSIS requires federal inspectors to observe only five to ten percent of the animals 5 sorted as “Normal” in motion, the Final Rule does not preclude federal inspectors from examining 6 a larger percentage of swine in motion if they deem necessary. See 84 Fed. Reg. at 52,312. The 7 Court finds the NSIS approach satisfies the requirements of Section 603(a). 8 Plaintiffs also argue that the Final Rule violates Section 603(a) because it prevents federal 9 inspectors from setting aside and ensuring that animals showing symptoms of disease are 10 slaughtered separately. Plaintiffs assert that under the NSIS system, plant employees place 11 symptomatic animals in “Subject” pens to rest and recover before a PHV evaluation. According to 12 Plaintiffs, the plant employees’ sorting precludes federal PHVs from conducting a critical 13 appraisal of the animals because the animals do not have U.S. Suspect tags and accompanying 14 paperwork detailing the symptoms that caused the animals to be put in the “Subject” pens. As a 15 result, PHVs cannot discern if animals in the “Subject” pen are recovered. Plaintiffs also argue 16 that the animals sorted into the “Subject” pens are not slaughtered separately from other animals 17 and their carcasses are not identified for further post-mortem evaluation as they would be if tagged 18 as “U.S. Suspect.” 19 The Court finds that the pre-sorting of swine by plant employees into “Subject” pens does 20 not run afoul of the FMIA’s ante-mortem inspection requirement. The animals placed in the 21 “Subject” pens by plant employees are subsequently examined by a federal PHV who may 22 condemn the animal as unfit for slaughter, deem the animal healthy and eligible for slaughter, or 23 conclude the animal is diseased and should be marked as “U.S. Suspect” and slaughtered 24 separately. Thus, under NSIS, federal inspectors examine the animals in the “Subject” pens and 25 determine if any need to be slaughtered separately as Section 603(a) requires. 5 26 5 The only animals that do not receive federal inspection under NSIS are the dead, moribund, or 27 otherwise unfit hogs that establishment employees remove and dispose of prior to ante-mortem 1 Accordingly, the Court finds the Final Rule does not violate Section 603(a) of the FMIA. 2 Although the Final Rule permits plant employees to pre-sort animals at NSIS plants, federal 3 inspectors still inspect all hogs prior to slaughter, and the hogs that exhibit signs of disease upon 4 that inspection are set apart and slaughtered separately as required by Section 603(a). 5 2. Section 604. 6 Plaintiffs argue that the Final Rule violates Section 604 of the FMIA because it prevents 7 inspectors from critically appraising carcasses and parts after slaughter. Plaintiffs advance three 8 reasons why the Final Rule violates Section 604: (1) the Final Rule tasks plant employees, not 9 federal inspectors, with palpating and incising lymph nodes; (2) the Final Rule prevents federal 10 inspectors from evaluating carcasses and parts for serious and generalized conditions because the 11 carcasses arrive at the inspection station too late to catch these problems; and (3) the Final Rule 12 reduces the number of federal inspectors on the slaughter lines so that each inspector has half as 13 much time to perform their inspection. 14 Section 604 of the FMIA provides: 15 “[T]he Secretary shall cause to be made by inspectors appointed for that purpose a post mortem inspection and examination of the 16 carcasses and parts thereof of all amenable species to be prepared at any slaughtering…establishment…which are capable of use as 17 human food, and the carcasses and parts thereof of all such animals found to be not adulterated shall be marked…as “Inspected and 18 condemned”; and said inspectors shall label…as “Inspected and condemned” all carcasses and parts thereof of animals found to be 19 adulterated, and all carcasses and parts thereof thus inspected and condemned shall be destroyed for food purposes by the said 20 establishment in the presence of an inspector… 21 21 U.S.C. § 604. 22 Under the traditional inspection process, plant employees do not conduct post-mortem 23 carcass sorting to identify which carcasses and parts contain removable defects correctable 24 (See Opp’n 6-7 n.5.) However, that section provides that “[a]ll livestock offered for slaughter in 25 an official establishment shall be examined and inspected on the day of and before slaughter” and “[s]uch ante-mortem inspection shall be made on the premises of the establishment at which the 26 livestock are offered for slaughter before the livestock shall be allowed to enter into any department of the establishment…” 9 C.F.R. § 309.1(b) (emphasis added). Because the dead, 27 moribund, or otherwise unfit hogs were never to be “offered for slaughter,” the FMIA does not 1 through trimming, which should be condemned because of generalized diseases or conditions, and 2 which are eligible to bear the mark of inspection. 84 Fed. Reg. at 52,311. Rather, those tasks are 3 conducted by federal inspectors. Id. Under NSIS, however, plant employees assume the role of 4 removing contaminated carcasses and identifying defects before federal inspectors conduct a post- 5 mortem inspection. Id. Federal inspectors, who are stationed on the evisceration line, then 6 “perform inspection of the sorted head, viscera, and carcass” including “visually inspect[ing] 7 heads and mandibular lymph nodes incised by establishment employees.” AR_100801; see also 8 84 Fed. Reg. at 52,311. Moreover, federal PHVs conduct a post-mortem examination of U.S. 9 Suspect and U.S. Retained parts and carcasses, which may include testing for chemical residues 10 and specific diseases with implications for food safety. 11 a. Palpation and incision of lymph nodes 12 Plaintiffs first argue that the Final Rule violates Section 604 because it requires plant 13 employees, not federal inspectors, to palpate and incise lymph nodes. Plaintiffs contend that 14 federal inspectors in NSIS plants cannot conduct an adequate post-mortem inspection because they 15 only observe the carcasses and parts and do not palpate lymph nodes, the proper incision of which 16 is important to the detection of disease in the lymphatic system. 17 Defendants argue that the visual inspection conducted by federal inspectors complies with 18 the FMIA’s requirement that federal inspectors conduct a “post mortem examination and 19 inspection of the carcasses and parts” of all swine slaughtered for human food. 21 U.S.C. § 604; 20 see also AFGE II, 284 F.3d at 130. 21 Thus, the issue here is whether federal inspectors are properly performing “inspections” as 22 required by the statute if they visually inspect, but do not incise and palpate, lymph nodes. 23 Section 604 requires a post-mortem inspection of each head, viscera, and carcass; it does not 24 address lymph nodes or palpation, or specifically require federal inspectors to incise and palpate 25 the lymph nodes. Finding that Section 604 includes a requirement that federal inspectors incise 26 and palpate each animal’s lymph nodes would “add[] terms not found in the statute” and 27 “impos[e] limits on [the] agency’s discretion that are not supported by the text” of the statute. 1 (2020) (citations omitted). Under the Final Rule, federal inspectors stationed on the evisceration 2 line visually inspect every head, viscera, and carcass. This satisfies Section 604. 3 Even if inspection is ambiguous, FSIS’s interpretation of inspection as requiring a visual 4 inspection is reasonable. Plaintiffs argue Defendants’ interpretation of inspection as requiring 5 only visual inspection is not entitled to deference because it contradicts Defendants’ prior 6 statements. Plaintiffs are correct that the traditional inspection system requires federal inspectors 7 to conduct the palpation and incision of the lymph nodes while plant employees perform that task 8 under NSIS. However, FSIS explained its reasoning for the change, including that the change 9 frees federal inspectors to perform offline inspections which are more effective in ensuring the 10 current food safety concerns. The agency is permitted to change its position if it provides an 11 explanation for doing so, which it has done here.6 12 The Court also disagrees with Plaintiffs’ argument that the NSIS system is akin to that 13 rejected in AFGE I. In AFGE I, the D.C. Circuit rejected FSIS verification and oversight regime, 14 under which employees would perform all tasks related to slaughter control and the federal 15 inspectors would be limited to observing establishment personnel as they processed carcasses and 16 verifying compliance by conducting random sampling of the carcasses. 215 F.3d at 10. The issue 17 with the regime rejected in AFGE I was that federal inspectors were evaluating people, not the 18 carcasses. See AFGE v. Glickman, 127 F. Supp. 2d 243, 246 (D.D.C. 2001). Here, and like the 19 regime approved of in AFGE II, federal inspectors visually evaluate the head, carcass, and viscera 20 of the animals. 21 b. Evaluation of carcasses and parts for disease 22 Plaintiffs next argue that the Final Rule violates Section 604 because plant employees’ 23 sorting and trimming prevents federal inspectors from evaluating carcasses and parts for serious or 24 generalized conditions necessitating a federal PHV disposition, as well as fecal matter, digestive 25

26 6 Plaintiffs also contend that Chevron deference does not apply because there is no definitive, formal, and judicially reviewable interpretation of the FMIA that warrants deference. The Court 27 disagrees. See United States v. Mead Corp., 533 U.S. 218, 230 (2001) (“[T]he overwhelming 1 contents, and milk contamination. 2 Defendants contend that the sorting and trimming activities performed by plant employees 3 prior to federal post-mortem inspection does not violate Section 604 because the statute does not 4 prohibit establishment employees from removing contamination from carcasses or parts before the 5 post-mortem inspection, or from identifying defects for further FSIS scrutiny. To ensure pre- 6 inspection sorting does not interfere with federal inspection, slaughter establishments must 7 “develop, implement, and maintain written procedures” regarding plant employees’ post-slaughter 8 sorting and trimming activities. See 9 C.F.R. § 310.26(b). Federal PHVs ensure these procedures 9 are adequate, and plant employees are only permitted to trim defects that do not affect the 10 disposition of the carcass. Moreover, the trimming occurs in the presence of federal inspectors 11 who verify that trimming and sorting is carried out correctly. See AR_103835; AR_100800. The 12 Court concludes that pre-inspection sorting and trimming does not prevent federal inspectors from 13 inspecting “each and every carcass and part” as required by the statute. See AFGE II, 284 F.3d at 14 130 (finding the FMIA does not “prohibit[] establishment employees from … removing some 15 adulterated carcasses before they get to FSIS inspectors.”). 16 c. Reduced number of federal inspectors on inspection line 17 Plaintiffs also argue that the Final Rule prevents federal inspectors from critically 18 appraising “each and every carcass and part” because it significantly diminishes the time for 19 federal inspectors to perform their examination. 21 U.S.C. § 604. The Final Rule lifts line speed 20 limits and reduces the number of online inspectors by half compared to the traditional inspection 21 process. As a result, Plaintiffs contend there are fewer total federal inspectors on the lines and the 22 remaining inspectors must inspect carcasses and parts at double their prior pace. Plaintiffs assert 23 that nothing in the record indicates that inspectors can still perform the tasks required for a critical 24 appraisal of carcasses and parts at this pace. 25 However, Section 604 does not require a specific inspection speed limit. FSIS has 26 explained that based on the HIMP pilot program, federal inspectors can conduct effective online 27 inspection of the head, viscera, and carcass of each hog when operating at faster line speeds. 83 1 speeds were appropriate because “fewer adulterated…hog parts and carcasses [would] be 2 presented for federal inspection” because of pre-inspection sorting and trimming”). Additionally, 3 PHVs in NSIS establishments are authorized to regulate line speeds as necessary to ensure 4 adequate carcass-by-carcass inspection. 84 Fed. Reg. 52,312. The Court does not find this 5 approach runs afoul of Section 604. 6 E. Whether the Final Rule Violates the APA. 7 1. Whether the Final Rule Departs from Prior Regulations and Practices. 8 Plaintiffs first argue that the Final Rule departs from the previous regulations governing 9 the traditional inspection process. Plaintiffs contend that FSIS has not reasonably explained why 10 it has abandoned the traditional inspection process in favor of NSIS. 11 The APA requires that when an agency changes its position, “it display an awareness that 12 it is changing position” and “show that there are good reasons for the new policy.” F.C.C. v. Fox 13 Television Stations, Inc., 556 U.S. 502, 515 (2009). The agency may not “depart from a prior 14 policy sub silentio or simply disregard rules that are still on the books.” Id. The agency “need not 15 demonstrate…that the reasons for the new policy are better than the reasons for the old one; it 16 suffices that the new policy is permissible under the statute, that there are good reasons for it, and 17 that the agency believes it to be better.” Id. 18 FSIS has met this burden. First, FSIS has displayed an awareness that it is changing its 19 position. The Final Rule explains that it establishes a new, optional inspection system and further 20 explains the key features of the new inspection process. Second, FSIS has shown that there are 21 good reasons for the policy. FSIS states that the NSIS will “improve the effectiveness of market 22 hog slaughter inspection; make better use of [FSIS’s] resources; and remove unnecessary 23 regulatory obstacles to industry innovation…” 84 Fed. Reg. at 52,300. Third, as discussed above, 24 the new inspection system is permissible under the FMIA. Finally, FSIS has provided good 25 reasons for the change—it believes the new policy will improve the effectiveness of market hog 26 slaughter inspection while “provid[ing] public health protection at least equivalent to the existing 27 inspection system.” See id. The Court concludes that FSIS has adequately explained its change in 2. Whether the Final Rule is Arbitrary and Capricious. 1 2 Plaintiffs also assert that the Final Rule is arbitrary and capricious for several reasons. “A 3 decision is arbitrary and capricious if the agency [1] has relied on factors which Congress has not 4 intended it to consider, [2] entirely failed to consider an important aspect of the problem, [3] 5 offered an explanation for its decision that runs counter to the evidence before the agency, or [4] 6 [has offered an explanation] so implausible that it could not be ascribed to a difference in view or 7 product of agency expertise.” George v. Bay Area Rapid Transit, 577 F.3d 1005, 1010 (9th Cir. 8 2009) (quoting United States v. Snoring Relief Labs, Inc. 210 F.3d 1081, 1085 (9th Cir.2000) 9 (internal quotation marks omitted, bracketed numbers added)). 10 a. Pilot plant sample size evaluated in Hog HIMP Report 11 Plaintiffs first challenge the adequacy of the sample size of the establishments evaluated in 12 the Hog HIMP Report, on which FSIS relied as justification for the Final Rule. Plaintiffs assert 13 the number of plants evaluated in the Hog HIMP Report is not representative of the broader 14 industry. 15 Because of the economic constraints of the program, FSIS anticipated that only large and 16 small high-volume establishments that exclusively slaughter market hogs would choose to 17 participate in NSIS. See 84 Fed. Reg. at 52,322. As of 2016, forty high-volume establishments 18 exclusively slaughtering market hogs existed. Id. These high-volume establishments account for 19 93 percent of total swine slaughter annually. Id. Of the forty high-volume establishments, 20 twenty-seven were large and thirteen were small. Id. The five HIMP pilot plants fell into the 21 large high-volume establishment category. Id. Those five plants produced about fifteen percent of 22 the nation’s pork supply. Id. No small high-volume establishments were included in the HIMP 23 pilot program. In the Final Rule, FSIS conceded that “the live market hog slaughter HIMP 24 establishments represent a small sample size of establishments.” Id. at 52,306. But FSIS 25 defended the adequacy of the sample size because the establishments “collectively represent 26 diversity in geography, corporate structure, management styles, product distributions patterns, and 27 other variables.” Id. 1 small high-volume plants and assert that significant differences exist between the large and small 2 plants. However, as Plaintiffs concede, FSIS was permitted to rely on a study with some 3 limitations to predict the outcome of the rulemaking. Moreover, although Plaintiff contends that 4 significant differences exist between the large and small high-volume establishments, their 5 citations to the record show that the plants slaughter a different number of animals, but this does 6 not establish that the large high-volume plants do not adequately represent outcomes at small 7 high-volume plants. Plaintiffs fail to show that FSIS acted arbitrarily and capriciously in relying 8 on the five HIMP pilot establishments to predict outcomes at the thirty-five other high-volume 9 establishments. 10 b. Hog HIMP Report data 11 Plaintiffs also challenge the adequacy of the data used in the Hog HIMP Report and assert 12 that FSIS’s reliance on this data is arbitrary and capricious. Plaintiffs’ critiques are twofold: (1) 13 the Hog HIMP Report analyzed only seven years of data; and (2) FSIS used performance 14 standards data collected six years after the baseline data. 15 Plaintiffs first argue that the Hog HIMP Report analyzed just seven years of data and did 16 not include data from the early years of the HIMP pilot. The modified HIMP pilot program began 17 in 2000. The Hog HIMP Report, published in 2014, used data from 2006 through 2013.7 18 Plaintiffs argue that FSIS acted arbitrarily and capriciously by failing to include data from the 19 early years of the HIMP pilot. However, Plaintiffs’ arguments fail to show that the FSIS’s 20 reliance on the HIMP Report, which used seven years of data, was arbitrary and capricious. The 21 report evaluated data for a seven-year period, which is not an insignificant period. It was not 22 unreasonable for FSIS to rely on the most recent available data in drafting the Hog HIMP Report, 23 and Plaintiffs have not shown that the absence of the data from the early years renders the Final 24 Rule arbitrary and capricious. 25 Plaintiffs further argue that FSIS’s use of performance standards data collected six years 26 after the baseline data was arbitrary and capricious. At the time the HIMP pilot began in 1999, 27 1 FSIS used an independent contractor to conduct baseline sampling in volunteer plants to create a 2 “performance standard” to compare the progress of these plants operating under HIMP. The 3 performance standards addressed food safety and non-food safety conditions. Id. FSIS calculated 4 the 75th percentile of food-safety results achieved for each category of defects to use as a baseline. 5 The HIMP plants would need to meet those performance standards for food safety and non-food 6 safety concerns. In the Hog HIMP Report, FSIS evaluated performance standards data from April 7 2009 through 2011 and 2012 through 2013 and concluded that the HIMP plants performance 8 compared favorably to the pre-HIMP baseline. See AR_102018. Plaintiffs have not shown that 9 FSIS’s evaluation of four years of performance standard data was arbitrary and capricious. 10 c. FWW’s analysis of HIMP pilot data 11 FWW submitted a FOIA request seeking data from 2014 to 2017. In May 2019, FWW 12 sent a letter to the USDA asking them to delay finalizing the rule based on FWW’s analysis of 13 data it received from FSIS through a FOIA requests. According to FWW’s analysis of the data, 14 offline inspectors in HIMP pilot plants were not performing statistically significantly more offline 15 verifications of compliance with Public Health Regulations (“PHRs”). Violations of PHRs are 16 linked to pathogenic contamination and are an important indicator of food safety issues. 84 Fed. 17 Reg. at 52,308. FSIS declined to consider FWW’s data analysis as part of the rulemaking. 18 Plaintiffs contend that FWW’s data analysis reveals that federal inspectors performed a 19 greater number of offline verifications for PHRs in traditional inspection plants than HIMP pilot 20 plants in 2014 through 2017. Plaintiffs contend that this data contradicts FSIS’s conclusion that a 21 greater number of offline verifications of PHRs would justify reducing the number of online 22 inspectors under NSIS. Plaintiffs contend that FSIS’s failure to consider this data or explain why 23 it was inapposite is arbitrary and capricious. 24 FSIS does not dispute that it did not consider data from 2014 through 2017 in the Final 25 Rule. However, it contends that its failure to do so does not render the Final Rule arbitrary and 26 capricious because it addressed the potential staleness of the Hog HIMP Report’s data in the Final 27 Rule. 84 Fed. Reg. 52,307. In the Final Rule, FSIS defended its use of data ending in 2013 1 Id. FSIS also stated that the findings from 2006 through 2013 were similar, which showed the 2 HIMP model was stable. Thus, although FSIS did not analyze the data from 2014 through 2017, it 3 explained why it did not do so and why its reliance on the data from the Hog HIMP Report was 4 reasonable. 5 Plaintiffs contend, however, that their submission undermined FSIS’s statement that the 6 data was stable, and thus should have been considered by FSIS. Defendants first argue that they 7 did not need to consider the data presented in FWW’s May 2019 submission because it was 8 received after the end of the period for public comment. However, unlike in the cases cited by 9 Defendants, Plaintiffs had raised similar claims during the original notice and comment period 10 based on the data available at that time, so FSIS was on notice of Plaintiffs’ claims challenging the 11 analysis of PHRs. Thus, FSIS was not relieved of its obligation to consider the submission on this 12 basis. 13 However, the Court is not convinced that Plaintiffs’ submission renders the Final Rule is 14 arbitrary and capricious. The May 2019 submission argued that federal inspectors in HIMP pilot 15 plants were unable to perform an equivalent number of PHR verifications as inspectors in non- 16 HIMP plants based on data from 2014 to 2017. A comment in the Final Rule raised a similar 17 substantive concern about the mandatory regulation verifications reported for 2012 and 2013. See 18 84 Fed. Reg. 52,307 (noting that the Hog HIMP Report showed an increase in total offline 19 verification tasks in HIMP establishments during 2012 and 2013, but data in the report showed 20 that the inspectors performed fewer verification tasks in HIMP establishments for more than half 21 of the PHRs). FSIS responded to the commenter’s concern and explained why the 2012 and 2013 22 data did not support the commenter’s argument regarding the number of offline verification tasks. 23 Id. The comment and FSIS’s response addressed the same substantive issue that FWW raised in 24 its May 2019 submission—that FSIS did not properly analyze the compliance verification rates— 25 albeit with regard to mandatory regulations as opposed to PHRs. Thus, the FSIS’s failure to 26 address FWW’s submission was not arbitrary and capricious because FSIS had previously 27 considered and addressed the core issue raised by the submission challenging its data analysis and 1 explained its reliance on the data from 2006 through 2013.8 2 Finally, Plaintiffs fail to establish that any discrepancies in the PHR verification data are so 3 significant as to render arbitrary and capricious FSIS’s conclusion that the NSIS system will 4 provide public health protection at least equivalent to the traditional inspection system or require 5 vacating the Final Rule. 6 d. Food safety regulation compliance data 7 Offline inspectors at all high-volume market hog slaughter establishments periodically 8 verify whether establishments are complying with various food-safety regulations, including FS-2 9 conditions, which include food contamination by fecal material, ingesta, and milk. If such 10 contamination is found, a noncompliance report under 9 C.F.R. section 310.18 (“Section 310.18”) 11 issues. The Hog HIMP Report compared the rate of noncompliance with Section 310.18 between 12 the five HIMP establishments and the twenty-one non-HIMP establishments and concluded that in 13 2012 and 2013, the HIMP establishments had lower noncompliance rates that the non-HIMP 14 establishments in 2012 and 2013. Id. 15 According to Plaintiffs, however, the raw data shows HIMP pilot plants had higher 16 violation rates than comparably sized traditional plants in 2012 and 2013. Plaintiffs contend that 17 the data was in error because FSIS adjusted the rate of noncompliance with FS-2 downwards by 18 more than twofold for HIMP establishments.9 Plaintiffs assert that FSIS was unjustified in 19 applying the downward adjustment to the HIMP pilot plant rates, which made them appear smaller 20 than traditional plants’ rates, without disclosing the assumption for doing so. Thus, FSIS’s 21 reliance on an incorrect interpretation of data was arbitrary and capricious. 22 The downward adjustment FSIS applied to calculate the FS-2 noncompliance rates 23 8 Plaintiffs also contend that FWW’s analysis showed that the Hog HIMP Report’s evaluation of 24 pooled verification data for 2012 and 2013 likely obscured the differences between the pilot and traditional plants. However, this issue was raised in the same comment and similarly addressed by 25 FSIS’s response. See 84 Fed. Reg. 52,307. 9 FS-2 compliance is computed as the number of non-compliances citing Section 310.18 divided 26 by Section 310.18 verifications performed. Section 310.18 addresses the contamination of carcasses, organs, and other parts and requires that those parts be handled in a sanitary manner to 27 prevent contamination, that plants have written procedures to prevent this contamination, and that 1 between the HIMP and non-HIMP establishments was based on that fact HIMP establishments 2 “perform 24 food safety carcass checks versus 11 carcass food safety checks at non-HIMP market 3 hog slaughter establishments.” AR_102000. This fact is disclosed in the Hog HIMP Report. 4 Additionally, FSIS disclosed that it applied the same downward adjustment to other 5 noncompliance rates reported in the Hog HIMP Report based on this difference in the number of 6 food safety carcass checks at HIMP and non-HIMP establishments. See AR_102004-05, Tables 3- 7 2, 3-5. While FSIS should have included the same disclosure in Table 3-13, the Court is able to 8 discern FSIS’s path in reaching its conclusion regarding the comparison of noncompliance rates at 9 HIMP and non-HIMP establishments based on the disclosure of the assumption elsewhere in the 10 Hog HIMP Report. See San Luis & Delta-Mendota Water Auth. v. Locke, 776 F.3d 971, 1001 (9th 11 Cir. 2014). FSIS’s failure to disclose this assumption does not render the Final Rule arbitrary and 12 capricious. 13 e. Evidence of concerns with plant employee performance 14 Plaintiffs next argue that FSIS ignored evidence that plant employees do not adequately 15 perform the inspection tasks required of them under NSIS. In support of this argument, Plaintiffs 16 cite non-compliance reports, two of which occurred at one plant, Clemons Food Group LLC. 17 Plaintiffs argue that FSIS’s failure to address these instances of non-compliance renders the Final 18 Rule arbitrary and capricious. However, the fact that FSIS did not address these specific instances 19 of non-compliance is not a sufficient basis to find the Final Rule arbitrary and capricious. FSIS 20 addressed commenters’ concerns about the ability of plant employees to identify food safety 21 defects and cited data from the Hog HIMP Report supporting the conclusion that the NSIS system 22 would effectively protect food safety. 84 Fed. Reg. at 52,312-13. Thus, the Court cannot 23 conclude that FSIS’s failure to address these specific non-compliance reports related to plant 24 employee performance renders the Final Rule arbitrary and capricious. 25 f. Risk assessment 26 Plaintiffs argue that FSIS’s reliance on the risk assessment in the Final Rule was arbitrary 27 and capricious because FSIS did not adequately address peer reviewers’ comments that questioned 1 reveal problems with the risk assessment’s use of simulated data until after the public comment 2 period closed. 3 Plaintiffs first argue that FSIS’s reliance on the risk assessment was arbitrary and 4 capricious because FSIS failed to address the peer reviewers’ comments that questioned whether 5 the assessment’s projected Salmonella reductions were related to the HIMP pilot as opposed to 6 other variables. The risk assessment was peer reviewed in April 2018. In August 2018, FSIS 7 posted a summary of peer reviewer feedback and an updated risk assessment that included 8 additional analysis to respond to peer reviewer’s feedback regarding the limitations of the earlier 9 risk assessment. Although Plaintiffs disagree with FSIS’s response to the peer reviewer feedback 10 and the conclusions drawn in the risk assessment, the Court cannot say that FSIS failed to consider 11 the issues the peer reviewers raised and does not find FSIS’s reliance on the risk assessment 12 arbitrary and capricious for this reason. 13 Plaintiffs next argue that FSIS failed to afford adequate opportunity for public comment on 14 the revisions to the risk assessment. In the original risk assessment, FSIS relied on certain 15 simulated data as the basis for its projections of avoided Salmonella illnesses. However, in the 16 January 2019 risk assessment that accompanied the Final Rule, FSIS did not rely on this data. 17 Without relying on this data, the Final Rule expressed greater uncertainty in the range of potential 18 benefits related to Salmonella-related illness reduction than was reflected in the Proposed Rule. 19 Plaintiffs assert that FSIS’s failure to permit the public to respond to the revised risk assessment, 20 which reflected greater uncertainty in the benefits, was arbitrary and capricious. 21 It is not disputed that FSIS did not disclose the greater range of uncertainty regarding 22 illness estimates until the Final Rule and did not provide the public an opportunity to comment on 23 the January 2019 risk assessment. However, FSIS’s failure to do so does not render the Final Rule 24 arbitrary and capricious. Although the risk assessment accompanying the Final Rule asserted a 25 larger range of potential benefits than the earlier risk assessments because it no longer relied on 26 the simulated data, FSIS continued to maintain that the estimated decrease in Salmonella illness 27 under the most feasible approach was expected to be 2,533 cases. This was consistent with the ] 104811. While the Final Rule acknowledged greater uncertainty surrounding illness 2 estimates, the conclusion about the likelihood of decrease in illness in the Final Rule remained 3 || unchanged from the Proposed Rule. Compare 83 Fed. Reg. 4780-01 with 84 Fed. Reg. at 52,300- 4 01; see also 83 Fed. Reg. 4791; 84 Fed. Reg. 52,334 n. 84. Thus, FSIS’s failure to allow notice 5 and comment on the revised risk assessment did not alter the conclusion in the Final Rule in a way 6 || that prevented the public from providing comment on proposed illness reductions. The Court 7 || cannot say this renders the Final Rule arbitrary and capricious. 8 F, The Final Rule is Not Otherwise Contrary to Law. 9 Plaintiffs advance two additional arguments why the Final Rule is contrary to law. First, 10 || Plaintiffs contend the Final Rule is arbitrary and capricious because it undermines the FMIA’s 11 purpose to protect the health and welfare of consumers by improperly handing critical inspection 12 || duties to plant employees. However, as discussed above, the pre-inspection sorting tasks E 13 || performed by plant employees under NSIS do not violate the FMIA. Second, Plaintiffs assert the 14 || Final Rule violates the APA’s subdelegation bar because the FMIA requires federal inspectors, not 3 || third-party plant employees, to conduct inspections. See 21 U.S.C. § 621. However, as discussed 16 || above, the Court has concluded that federal inspectors still perform their statutorily required i 17 || inspection duties under the FMIA and have not impermissibly delegated those tasks to plant Z 18 employees. 19 CONCLUSION 20 For the foregoing reasons, the Court DENIES Plaintiffs’ motion for summary judgment 21 and GRANTS Defendants’ cross-motion for summary judgment. 22 A separate judgment shall issue, and the Clerk shall close the file. 23 IT ISSO ORDERED. 24 Dated: September 30, 2022 / “ } / 25 SOF {x / (ike 26 Unted sided bg ict Judge 27 28