Argued January 11, 2002.
Decided March 29, 2002.
Appeal from the United States District Court for the District of Columbia (98cv00893).
Anne M. Wagner argued the cause and filed the briefs for appellants. With her on the briefs was Mark Roth.
Charles W. Scarborough, Attorney, U.S. Department of Justice, argued the cause for appellees. With him on the briefs were Roscoe C. Howard, Jr., U.S. Attorney, Mark B. Stern and Colette G. Matzzie, Attorneys, U.S. Department of Justice. Barbara C. Biddle, Assistant Director, U.S. Department of Justice, entered an appearance.
Before: GINSBURG, Chief Judge, RANDOLPH and TATEL, Circuit Judges.
Opinion for the Court filed by Circuit Judge RANDOLPH.
This case returns to us after proceedings on remand from Am. Fed'n of Gov't Employees v. Glickman, 215 F.3d 7 (D.C. Cir. 2000) ( "AFGE I"). The United States Department of Agriculture ("USDA") had attempted to test a new inspection model at several hog and poultry processing plants. AFGE I held that, as implemented, the model program violated the Federal Meat Inspection Act ("FMIA"), 21 U.S.C. § 604, and the Poultry Products Inspection Act ("PPIA"), 21 U.S.C. § 455. While the case was on remand in the district court, the USDA modified its model program. The question now is whether the revised model program complies with these statutes.
Both statutes seek to ensure that all meat and poultry products processed for human consumption are wholesome and unadulterated. See 21 U.S.C. § 602; 21 U.S.C. § 451. The FMIA requires USDA inspectors to conduct a postmortem "examination and inspection of the carcasses and parts thereof of all cattle, sheep, swine, goats, horses, mules and other equines." 21 U.S.C. § 604. The PPIA requires USDA inspectors to conduct a "post mortem inspection of the carcass of each bird processed." 21 U.S.C. § 455(b).
The USDA's Food Safety and Inspection Service ("FSIS") traditionally assigned federal inspectors to processing lines where they conducted "organoleptic" inspections of carcasses, relying on sight, touch and smell in examining the head, viscera, and exterior of the carcasses to detect signs of adulteration, such as tumors, parasites and other diseases. See AFGE I, 215 F.3d at 8, 10. In "the mid-1990s, FSIS embarked on a comprehensive food safety initiative targeting the agency's resources at what it perceived as a serious health risk — foodborne pathogens, such as salmonella and E. coli, which cannot be detected by organoleptic inspection. At the same time it determined to make changes in the current inspection process to combat these microbial causes of foodborne illness, FSIS addressed what it considered to be another failure of the present regulatory system — that it provides processing plants with little incentive to detect and eliminate unacceptable carcasses before presenting them for inspection." AFGE I, 215 F.3d at 9.
"In July 1996, FSIS took the first step in implementing its new initiative by promulgating the Pathogen Reduction/Hazard Analysis and Critical Control Points (`HACCP') final rule. See 61 Fed.Reg. 38,806 (1996)." 215 F.3d at 9. The rule required meat and poultry plants to develop and install a system of preventive controls to ensure the safety of their products. See 61 Fed.Reg. 38,814; see also 9 C.F.R. § 417.2(a). Because the HACCP rule gave processing plants greater control over production decisions, FSIS determined that changes in its traditional organoleptic inspection methods would be necessary to realign the roles of federal inspectors with the new HACCP philosophy. See 61 Fed. Reg. 38,818. Accordingly, in 1997 FSIS published a notice in the Federal Register explaining that as a result of the HACCP final rule, "[e]very aspect of traditional FSIS methods of inspection for slaughter and processing needs to be reconsidered." 62 Fed.Reg. 31,553. The notice said the agency was launching the Inspection Models Development Project (the "Models Project"), through which it would design and test various new inspection models in a series of trials in volunteer meat and poultry slaughter establishments. See 62 Fed. Reg. 31,558. If — after gathering data on the various test models and preparing a final report analyzing the different models — the FSIS decided that its traditional inspection methods needed to be changed, then it pledged to initiate rulemaking to alter existing inspection procedures. 62 Fed.Reg. 31,558.
The first new inspection approach, tested at several volunteer processing plants under the Models Project, gave industry personnel the task of separating normal from abnormal carcasses and parts. Federal inspectors had two limited roles: oversight and verification. Oversight inspectors observed establishment personnel as they processed carcasses and removed unacceptable products from the food supply; verification inspectors randomly sampled carcasses to determine if the plant was complying with relevant performance standards. See AFGE I, 215 F.3d at 10.
In 1998, the plaintiffs in this action sought to enjoin implementation of the new inspection model, arguing that it violated the FMIA and PPIA because federal inspectors were not personally inspecting each carcass. The district court, finding the program in compliance with the statutes, granted summary judgment for the government. We reversed and remanded. See AFGE I, 215 F.3d at 7. Because the model program had federal employees "inspecting people not carcasses," the USDA was not fulfilling its statutory duty to conduct post-mortem inspections of carcasses. See id. at 11.
On remand, plaintiffs moved for entry of an order consistent with AFGE I declaring that the USDA's inspection model violated the FMIA and PPIA. The government opposed the motion and moved for a declaratory judgment, arguing that it had modified its Models Project in response to the AFGE I decision. Only the declaration of Michael Grasso, the project manager of the FSIS's inspection program, offered a comprehensive written description of the modification, which was accomplished without a new rulemaking. Grasso's declaration explained that in the eleven poultry establishments voluntarily participating in the Models Project, the new program called for federal inspectors to play two roles: (1) "carcass inspectors" stationed at the end of slaughter lines would examine each poultry carcass for adulteration after the carcasses were eviscerated, sorted, washed and trimmed by establishment employees but before the carcasses were put into the chiller; and (2) "verification inspectors" would oversee the poultry establishment's inspection efforts. See Grasso Declaration at 5. In the three participating hog plants, the modified program called for federal carcass inspectors located at up to three fixed locations along the slaughter lines: in the area where the carcass and head were separated; where the carcass and viscera were separated; and at the pre-wash verification location. See id. at 8. These inspectors would be responsible for examining the carcass, head, and viscera of all hogs. See id. Other inspectors would also be assigned to hog plants to verify the effectiveness of the plant's process control systems.
Due to the physical configuration of the slaughter line in some hog establishments, fewer than three inspectors may be used because "it may be possible for one inspector to inspect both the head and viscera or the viscera and carcass." Grasso Declaration at 8 (explaining that, for example, "the chain that carries the head may be adjacent to the viscera pans, which may enable one inspector to examine both the head and viscera").
The USDA told the district court that it would begin implementing the modified inspection system in two of the eleven poultry plants on September 18, 2000, and that it would complete implementation in all fourteen participating plants no later than November 6, 2000. Plaintiffs objected on the grounds that the modified program violated the FMIA, PPIA, and conflicted with AFGE I because it did not require federal inspectors to subject each carcass to a close and critical appraisal. The government defended on the basis that the main error in the prior program had been remedied because federal inspectors would be directly observing each carcass to determine whether or not it was adulterated.
The district court granted plaintiffs' motion for a remedial order, declaring that the first model program violated the FMIA and PPIA. See Am. Fed'n of Gov't Employees v. Glickman, 127 F.Supp.2d 243 (D.D.C. 2001). As to the modified program, the court determined that Congress had not unambiguously expressed its intent about how much "inspection" is required by the statutes. Relying on Chevron, U.S.A., Inc. v. Natural Res. Def. Council, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), the court concluded that the modified program reflected USDA's reasonable interpretation of the statutes.
In the first appeal in this case, we were faced with the issue "whether the statutes permit federal inspectors to step back from the processing lines and perform their inspection duties by overseeing inspections conducted by plant employees." 215 F.3d at 8. Both the FMIA and PPIA yielded a certain solution: the statutes require federal inspectors to inspect animal carcasses not people. The issue in this second appeal is of a different sort. Federal inspectors are now examining carcasses, but are they properly performing "inspections," as the statutes require?
Courts pronouncing on the meaning of a word or phrase in a statute often declare generally that the meaning is "plain" or that it is "ambiguous." Such declarations may mask a critical analytical step. A statute can be clear or unclear depending on what question we want it to answer. We may borrow H.L.A. Hart's famous hypothetical to illustrate the point. Imagine the following law: "No vehicle may be taken into the park." H.L.A. HART, THE CONCEPT OF LAW 125 (1961). If one asks whether this prohibits taking baseballs into the park, the answer is clearly no. If one asks whether this forbids driving automobiles through the park, the answer is clearly yes. But if the question is whether the law prohibits riding a bicycle or a skateboard in the park or pushing a baby carriage there, uncertainties arise.
So it is with the FMIA and PPIA. Both statutes delineate what must be inspected and by whom, see AFGE I, 215 F.3d at 10-11, but neither statute tells the reader exactly what an "inspection" entails. The district court, citing Chevron, therefore deferred to the USDA's view of "inspection" as reflected in the modified program. The parties, following the district court's lead, assume Chevron applies. But it would be plain error for us to rely on the case. United States v. Mead Corp., 533 U.S. 218, 226-27, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001), decided after the district court ruled, held that Chevron applies only when "Congress delegated authority to the agency generally to make rules carrying the force of law, and . . . the agency interpretation claiming deference was promulgated in the exercise of that authority." The modified program described in the Grasso declaration was not the product of a statutorily-created decision-making process, such as formal adjudication or notice-and-comment rulemaking. Nor does the modified program itself contain any indication that the USDA "ever set out with a lawmaking pretense in mind," Mead, 533 U.S. at 233, 121 S.Ct. 2164. The USDA announced that the Models Project would serve as a temporary test project rather than as its final say on the matter. See 62 Fed.Reg. 31,558 (explaining that after testing various inspection models, the agency will write a final report and "initiate rulemaking, as appropriate, to change existing inspection procedures"). Because the USDA did not intend to act with the force of law when it promulgated the modified program, the program has no more status than opinion letters, policy statements, agency manuals, and enforcement guidelines, all of which are undeserving of Chevron deference. See Christensen v. Harris County, 529 U.S. 576, 587, 120 S.Ct. 1655, 146 L.Ed.2d 621 (2000); Mead, 533 U.S. at 234, 121 S.Ct. 2164. Still, as the reasoned judgment of the federal agency charged with administering our federal meat and poultry inspection laws, the USDA's view about the scope of inspection required by the FMIA and PPIA constitutes "a body of experience and informed judgment" to which we may properly resort for guidance. Skidmore v. Swift Co., 323 U.S. 134, 140, 65 S.Ct. 161, 89 L.Ed. 124 (1944).
The USDA determined that its modified inspection program, which calls for one carcass and one verification inspector per slaughter line at poultry plants, and one verification inspector and up to three carcass inspectors at hog plants, would be sufficient to allow its employees to detect adulterated meat and poultry. With admirable candor, plaintiffs concede that the statutes do not require organoleptic inspections. They believe the modified program is nevertheless flawed because federal inspectors examine poultry carcasses only after industry employees have eviscerated, sorted, trimmed and rinsed the carcasses. As to the statute, the PPIA states that federal inspectors must conduct a "post mortem inspection of the carcass of each bird processed." 21 U.S.C. § 455(b). Another subsection states that "[a]ll poultry carcasses and parts thereof . . . found to be adulterated shall be condemned," 21 U.S.C. § 455(c). Plaintiffs wonder how federal inspectors can fulfill their duty to condemn adulterated "parts" if all they are checking for adulteration are poultry carcasses. The USDA has a ready response, one we find convincing. Inspection of the viscera "is not necessary to determine whether a poultry carcass is adulterated." Brief for Appellees at 23. With only one exception, all generalized poultry diseases and conditions can be readily discerned by inspecting the carcasses alone and therefore are adequately addressed by the practice, incorporated in the modified program, of condemning all viscera that correspond to carcasses found to be adulterated. The one disease that cannot be detected by inspecting the carcasses alone — avian visceral leukosis — is adequately dealt with under the modified program by having an FSIS inspector examine the viscera from the initial 300 birds slaughtered from each flock. If avian visceral leukosis is present, it will be present throughout the flock and therefore will be detected.
Plaintiffs also assert that the line speeds used in the modified program at both hog and poultry plants are too fast to allow federal inspectors to make a critical appraisal of each carcass. But as the USDA points out, higher line speeds are appropriate. Fewer adulterated poultry carcasses and hog parts and carcasses will be presented for federal inspection under the modified program because employees of the processing plants will sort out carcasses before they reach federal inspectors. Neither the PPIA nor the FMIA prohibits establishment employees from paring down the overall number of carcasses by removing some adulterated carcasses before they get to FSIS inspectors. Furthermore, as the USDA tests its models project, it plans to monitor whether a second inspector is necessary in establishments operating at high line speeds. See Grasso Declaration at 6.
We therefore hold that the USDA's current modified inspection model, as described in the Grasso declaration, does not violate the FMIA or the PPIA. Because the modified program calls for federal inspectors in participating poultry plants to personally examine each poultry carcass leaving the slaughter line, the USDA is complying with the PPIA's requirement that "the carcass of each bird processed" be inspected for adulteration. 21 U.S.C. § 455. The modified program also satisfies the FMIA because the program calls for federal inspectors in participating hog plants to inspect all hog carcasses, heads and viscera, as the statute demands. See 21 U.S.C. § 604.
We have reviewed only the USDA's implementation of its current, modified inspection model. This is a test program, a temporary measure intended as an experiment. If the USDA undertakes a rulemaking to adopt as a permanent change something along the lines of the modified program, experience with the program's operation and its effectiveness will doubtless play a significant role. For this and other reasons, our opinion today may not necessarily foreshadow the outcome of judicial review of such future regulations.