Summary
noting that legislative regulations properly issued by agencies "have the force and effect of legislation"
Summary of this case from Sierra Club v. United States Dept. of Transp.Opinion
No. 82-4604.
Argued and Submitted May 9, 1983.
Decided September 15, 1983.
Barry Grossman, Edward T. Hand, William F. Baxter, Nancy C. Garrison, Dept. of Justice, Washington, D.C., for defendants-appellees.
William R. Pendergrast, Hamel, Park, McCabe Saunders, Washington, D.C., J. Bruce McCubbrey, Fitch, Even, Tabin, Flannery Welsh, San Francisco, Cal., for plaintiff-appellant.
Appeal from the United States District Court for the Northern District of California.
This case presents a question of first impression arising under section 201(s) of the Food, Drug, and Cosmetic Act, 21 U.S.C. § 321(s). The issue is whether the Food and Drug Administration (FDA) has properly interpreted the statute, which provides that the safety of substances added to food may be established by experience based on common use in food prior to 1958. We conclude that, even according the deference due an administrative agency's interpretation of a statute that it is responsible for enforcing, the challenged FDA regulation does not fairly reflect either the language or purpose of the "common use in food" portion of section 201(s). We therefore reverse the district court's declaration that the regulation is valid.
I [2] BACKGROUND
Section 201(s) of the Food, Drug, and Cosmetic Act, 21 U.S.C. § 321(s) (1976), defines a "food additive" as a substance, added to food, that is
not generally recognized, among experts qualified by scientific training and experience to evaluate its safety, as having been adequately shown through scientific procedures (or, in the case of a substance used in food prior to January 1, 1958, through either scientific procedures or experience based on common use in food) to be safe under the conditions of its intended use.
The question whether a substance is a "food additive" is significant because all "food additives" must by law undergo costly pretesting procedures before they may be marketed to the public. See 21 U.S.C. § 348 (1976).
In 1974, the FDA promulgated a regulation defining the term "common use in food" for purposes of section 201(s). The regulation states that "[c]ommon use in food' means a substantial history of consumption of a substance by a significant number of consumers in the United States." 21 C.F.R. § 170.3(f) (1982) (emphasis added). Thus, according to the regulation, no substance may be newly introduced into the United States without either pretesting satisfactory to the FDA or general recognition among experts, based solely on scientific evidence, that the substance is safe for the intended use, no matter how widespread or prolonged the history of use of the substance in food in regions outside the United States.
Appellant Fmali Herb, Inc., is an importer of Chinese food products. In its business, Fmali wishes to import foods containing herbs traditional in China that have never been widely used in the United States. One such food is a jelly or honey-like product, known as renshenfengwangjiang, that normally contains schizandra seed. The FDA has ruled that reshenfengwangjiang may be sold in the United States only if it does not contain schizandra seed, because schizandra seed has not been scientifically tested and because it was not commonly used in the United States prior to 1958. The FDA held that evidence of long and widespread use of schizandra seed in China is not admissible in aid of establishing that schizandra seed is safe for human consumption.
Fmali filed an action in district court seeking a declaratory judgment that the FDA regulation codified at 21 C.F.R. § 170.3(f) is invalid as an erroneous interpretation of section 201(s). No fact issues were disputed, and the parties filed cross-motions for summary judgment. The district court, agreeing with the FDA that evidence of use of a substance in food in a foreign country may be unreliable to show safety, dismissed Fmali's action.
II [8] DISCUSSION
A. Ripeness.
Fmali brought the present action in order to obtain pre-enforcement review of the FDA's regulation. Fmali has not attempted to secure FDA clearance for any specific product. The posture of the case thus requires us to decide the validity of the regulation in the abstract, without reference to the regulation's application to a particular product, although prior FDA rulings have been cited. The hypothetical nature of the necessary inquiry suggests that the case might not be ripe for adjudication.
However, the Supreme Court in Abbott Laboratories v. Gardner, 387 U.S. 136, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967), sanctioned pre-enforcement review of an FDA regulation in a similarly-postured case. The Court held that such a challenge is neither statutorily nonreviewable, id. at 141-48, 87 S.Ct. at 1511-15, nor unripe for declaratory judgment adjudication, id. at 148-56, 87 S.Ct. at 1515-20. Following Abbott Laboratories, we conclude that the district court properly assumed jurisdiction of the declaratory judgment action filed by Fmali.
B. Scope of Review.
Administrative regulations fall into two categories: legislative regulations and interpretive regulations. 2 K. Davis, Administrative Law Treatise § 7.8, at 36 (2d ed. 1979). Legislative regulations are "issued by an agency pursuant to statutory authority and . . . implement the statute." Batterton v. Francis, 432 U.S. 416, 425 n. 9, 97 S.Ct. 2399, 2405 n. 9, 53 L.Ed.2d 448 (1977). Such regulations have the force and effect of legislation, and may not be set aside because a court would have read the statutory mandate differently. Id. at 425, 97 S.Ct. at 2405. An interpretive regulation, by contrast, is one issued pursuant to an agency's interpretation of a governing statute, without delegated legislative power. See 2 K. Davis § 7.8, at 36. Since the FDA is not instructed by statute to define the term "common use in food," we assess 21 C.F.R. § 170.3(f) as an interpretive regulation.
Regulations interpreting statutory terms are given important but not controlling significance. Our task is limited to determining whether the FDA's construction of section 201(s) is "sufficiently reasonable" to be adopted by a reviewing court. Zenith Radio Corp. v. United States, 437 U.S. 443, 450, 98 S.Ct. 2441, 2445, 57 L.Ed.2d 337 (1978) (quoting Train v. Natural Resources Defense Council, 421 U.S. 60, 75, 95 S.Ct. 1470, 1479, 43 L.Ed.2d 731 (1975)). To uphold the regulation, we need not find the FDA interpretation to be the only reasonable one possible. Train, 421 U.S. at 75, 95 S.Ct. at 1479. However, the regulation cannot be sustained if it is contrary to the "plain meaning" of the statute and its legislative history. See, e.g., General Electric Co. v. Gilbert, 429 U.S. 125, 145, 97 S.Ct. 401, 412, 50 L.Ed.2d 343 (1976). We must therefore consider whether the challenged regulation represents a reasonable construction of section 201(s).
Among the factors commonly listed as variables in determining the deference due an interpretive regulation promulgated by an agency with subject-matter expertise are (1) whether the regulation was promulgated contemporaneously with the enactment of the statute; (2) the duration and continuity of the administrative interpretation; and (3) whether the statutory provision has been reenacted without change since promulgation of the regulation. 2 K. Davis, Administrative Law Treatise § 7.14, at 64 (2d ed. 1979); see Batterton v. Francis, 432 U.S. 416, 425 n. 9, 97 S.Ct. 2399, 2405 n. 9, 53 L.Ed.2d 448 (1977). These factors do not weigh in favor of deference here. The regulation at issue was promulgated sixteen years following enactment of the governing statute, and the statute has not since been reenacted. The regulation has never been the focus of active debate or judicial decisions, so deference based upon congressional acquiescence is inappropriate. Cf. United States v. Rutherford, 442 U.S. 544, 554, 99 S.Ct. 2470, 2476, 61 L.Ed.2d 68 (1979) (FDA regulation barring Laetrile accorded particular deference where the interpretation had caused "considerable public controversy" and Congress had not acted to change effect of regulation).
C. History, Structure and Language of Section 201(s).
During the early 1950's, a House Select Committee chaired by James J. Delaney, following investigation, registered concern about the many recently developed chemicals being added to improve the texture, color, flavor and shelf life of packaged foods. See H.R.Rep. No. 2284, 85th Cong., 2d Sess. 2 (1958). Legislation was introduced in the 83rd, 84th, and 85th Sessions of Congress for the purpose of subjecting such chemical additives to pretesting for safety. Id. One problem confronted by the draftsmen was how to deal with chemical substances already in general use as food ingredients.
The House subcommittee reached a relatively early consensus that the bill should not contain a blanket "grandfather clause" automatically exempting from coverage all substances already used in food. See, e.g., Federal Food, Drug, and Cosmetic Act (Chemical Additives in Food): Hearings on H.R. 4475 Before a Subcomm. of the House Comm. on Interstate and Foreign Commerce, 84th Cong., 2d Sess. 28 (1956) (statement of Rep. Delaney) [hereinafter cited as Chemical Additives Hearings]. At the same time, members of Congress recognized that many ordinary food ingredients had never been scientifically tested, but had long been used without evidence of health hazards. To require exhaustive laboratory analysis and pretesting of these additives was opposed because it would serve no useful purpose and would unduly burden both the food industry and the Government. Id. at 40 (statement of Rep. Miller).
As the legislation moved toward passage, several alternative approaches were written into the bills submitted to the House subcommittee for consideration. See Food Additives: Hearings on Bills Before a Subcomm. of the House Comm. on Interstate and Foreign Commerce, 85th Cong., 2d Sess. 45-46 (1958) [hereinafter cited as Food Additives Hearings]. One type of bill allowed exceptions to the pretesting requirement only for substances generally recognized as safe by scientific experts, based on testing. E.g., H.R. 7700, H.R. 7798, and H.R. 7938, 85th Cong., 1st Sess. (1957). The second variety allowed exceptions to pretesting for only those substances that had been approved, formally or informally, by the FDA. E.g., H.R. 8390 and H.R. 8629, 85th Cong., 1st Sess. (1957). Finally, a number of bills allowed an exception for substances shown "through prolonged use in food" to be safe for the intended use. E.g., H.R. 6747, H.R. 8112, and S. 1895, 85th Cong., 1st Sess. (1957). It was one of these latter bills, H.R. 6747, that eventually won Administration approval and was enacted, with modifications, by Congress.
Arguably, however, these bills would have exempted substances already in use, because the category of substances defined to require pretesting was entitled "new food additives." See Food Additives Hearings, supra, at 497 (statement of Rep. Delaney).
Assistant Health, Education, and Welfare Secretary Elliott L. Richardson described in House hearings the effect of the "prolonged use" language in the H.R. 6747. He commented that,
with respect to existing additives generally speaking, a large group, such as table salt, would be classified as generally recognized as safe by competent experts. These would be additives as to which there was no substantial question of safety, which have been used over a long period. These . . . would be outside of the definition of chemical additives.
Food Additives Hearings, supra, at 429. FDA Commissioner George P. Larrick, asked to discuss the scope of the exception, stated that it "would include staple foods such as salt, sugar, vinegar, caramel, and many, many others that are common in our food supply." Id. at 460. From the remarks of these and other witnesses, it appears that provision allowing an exception for substances with a history of long use was favored because of a general feeling that such use could provide adequate evidence of the safety of a substance for the historical use. See, e.g., id. at 126 (statement of Lawrence A. Coleman on behalf of Manufacturing Chemists' Association); but cf. id. at 497 (statement of Rep. Delaney) (criticizing "prolonged use" language of exception).
Commissioner Larrick, when referring to "our food supply," probably meant the collective human food supply rather than just that of Americans. The broader focus of the subcommittee discussion is shown by an exchange that occurred shortly after the Commissioner made the quoted statement:
Mr. DINGELL: What would you figure the length of time for this period of continuous use as a means of determining whether the additive is safe?
Mr. LARRICK: I would hate to put a time limit on it. It certainly would be a matter of many years. I would think that the extent to which it is used and the likelihood that the type of harm that it might conceivably cause would be observable; I think you would have to take all of those things into account. But bear in mind that the definition of chemical additive for the purpose of this bill is tremendously broad.
Mr. DINGELL: Mr. Commissioner, would you be able to give us some kind of example of something which has been used for a long period of time other than salt? In other words, something which had not been used in the early unrecorded history of mankind?
Mr. LARRICK: I would say vanilla extract.
Food Additives Hearings, supra, at 464.
The legislative history of section 201(s) does not suggest that only "prolonged use" within the United States can be used to establish the safety of an additive. The sole manifestation of possible intent to impose such a limitation appears in an early bill submitted to the House subcommittee, which would have allowed an exemption from pretesting for only those substances "used in the United States in a particular food or class of foods for the intended purpose." H.R. 7607, 84th Cong., 1st Sess. (1955). The proposed geographic limitation was not retained in any subsequent bill containing a "prolonged use" exemption. We take this sequence to indicate that members of Congress considered incorporating an explicit geographic element in the exception to pretesting under section 321(s), but declined to do so. We are very hesitant to imply such a limitation when Congress failed to adopt a bill that would have included the same restriction.
In the final version of the bill that was enacted, the phrase "prolonged use in food" was changed to read "common use in food." The House hearings indicated that the change may have resulted from one of two criticisms. Representative Delaney objected to the term because he felt that it was not adequately defined and therefore would not establish a proper criterion for establishing safety. Food Additives Hearings, supra, at 497. Delaney, however, opposed any form of "grandfather clause" for existing additives. Id. The more likely source of the language modification was the suggestion of the general counsel for the Grocery Manufacturers of America that the word "prolonged" be deleted because "considerations other than time" can form a proper basis for a finding of safety based on use. Id. at 80 (statement of Charles W. Dunn).
Neither criticism suggests that the "prolonged use" phrasing was viewed as overly broad or that the term should also include a geographical limitation.
The principal limit on the "prolonged use" exemption discussed by the House drafters did not relate to its geographical scope, but rather focused on the concern that use, no matter how widespread or prolonged, must reliably demonstrate safety before a substance may be marketed without pretesting. See 104 Cong.Rec. 17,424 (1958) (remarks of Rep. Sullivan); Food Additives Hearings, supra, at 464 (statement of Commissioner Larrick).
The structure of section 201(s) reflects this overriding concern with the safety of food ingredients. Under the statute, "common use in food" of an ingredient does not automatically exempt the substance from pretesting requirements. Instead, "common use in food" merely describes one form of evidence that may be introduced by a proponent for the purpose of meeting the ultimate standard, which is whether the ingredient is safe for human consumption. As we have observed, neither the language of the statute nor its history indicate that only "common use" within the United States may be cited as evidence of safety.
By promulgating the regulation defining "common use in food" to mean only common use within the United States, the FDA has imposed a restriction not required by the literal terms of the statute. The FDA's reasons for doing so were expressed at the time the regulation was issued:
The Commissioner believes that the type of experience based on common use in food that will support a GRAS determination must involve use in the United States, and not solely in foreign countries. Reported use in foreign countries often cannot be verified, and in any event the experience based on such use cannot be monitored or evaluated. Food consumption patterns and differences between cultures make it impossible to assess whether a history of use abroad would be comparable to a history of use in the United States.
39 Fed.Reg. 34,195 (1974).
We agree for the most part with the concerns expressed by the FDA. Many residents of areas outside the United States have shorter life expectancies than do Americans. Foreign populations also suffer from higher incidences of some diseases, so that generalizations about the safety of their dietary habits may be unreliable. But it is an illogical and, we think, unwarranted constriction of the statute to rule that evidence of long use of a substance in food outside the United States can never provide probative evidence of safety. Counsel for the FDA admitted this point in argument before the district court, when he stated that "there are countries in the world, like West Germany or something, where one might be able to accept" prior use of an ingredient to establish safety, but added that the FDA does not believe that evidence of use in "Guam" or "Southeast Asia" could be accepted. The statute provides no basis for a purely ethnocentric distinction of this kind, divorced from demographic considerations.
Counsel for the FDA made the following argument before the district court:
[T]he loophole the plaintiffs would drive through this statute is a loophole that could bring in an enormous number of things that would not just be used in some esoteric product but could be on every breakfast cereal a year from now. The only way they could get it off is to prove it would be unsafe.
I think the history shows, even with such things as saccharin and the great doubt about safety, it is very difficult to remove those things. What we are trying to do is keep a lid on the bottle and not allow everything to come in. Undoubtedly, there are countries in the world, like West Germany or something, where one might be able to accept their prior use, although West Germany or I might add, was the one that invented thalidomide.
I do not think the United States should have to allow in what has been tolerated or used in Guam or in Southeast Asia on the basis that no one is known to have died from it. That is, if it couldn't be shown that no one [sic] has ever been harmed through eating a particular commodity, we really would have no alternative to let it in if they had the types of letters from professors and so on that said it was safe.
The FDA argues that interpreting section 201(s) to allow evidence of common use in food of an ingredient would open a "loophole" in the statute. We cannot agree. Under section 201(s), the important inquiry is whether a substance added to food is "generally recognized, among experts qualified by scientific training and experience to evaluate its safety, as having been adequately shown . . . to be safe under the conditions of its intended use." Experience based on common use in food serves only as a means of showing safety. If the foreign experience cited by the proponent does not clearly demonstrate safety, because of doubts about cultural comparability or the adequacy of public health data, then the substance must be deemed a food additive. As the overriding purpose of the Food, Drug, and Cosmetic Act is to protect the public health, the burden of proof of safety to be borne by a proponent of an ingredient is heavy. See United States v. Naremco, Inc., 553 F.2d 1138, 1141 (8th Cir. 1977). In practice, evidence of foreign use of an ingredient, standing alone, may rarely or never be enough to establish safety. An FDA regulation establishing a rigorous threshold of evidence needed to constitute sufficient proof of safety based on common use would be entitled to substantial judicial deference. See CIBA Corp. v. Weinberger, 412 U.S. 640, 643-44, 93 S.Ct. 2495, 2497, 37 L.Ed.2d 230 (1973); Naremco, 553 F.2d at 1141.
But 21 C.F.R. § 170.3(f) does not establish an evidentiary standard. Rather, it operates as a blanket exclusion of evidence of safety based on use of food outside the United States. As such, it fails to comport either with the express terms of the statute that contain no such restriction, or with the purpose of the "common use" exception as articulated by legislators, that was to allow use of "any substances which over the years have been clearly demonstrated by long use to be completely safe." 104 Cong.Rec. 17,424 (remarks of Rep. Sullivan). The regulation is not a reasonable interpretation of section 201(s) and cannot be sustained in its present form.
We note that in many cases, substances used in foreign food might be shown safe based on a combination of evidence of traditional dietary use and a chemical analysis that reveals no toxic ingredients. The present regulation would not allow evidence of the foreign use to supplement the scientific data to establish safety. Yet we see nothing in section 201(s) to preclude using a combination of the two different forms of evidence to demonstrate safety.
III [29] CONCLUSION
Because 21 C.F.R. § 170.3(f) is not a permissible interpretation of 21 U.S.C. § 321(s), the judgment sought by Fmali declaring the regulation invalid should have been granted. Such a declaration does not, of course, establish that any particular substance is not a "food additive" as defined by statute.
REVERSED.
The Supreme Court stated in Federal Election Commission v. Democratic Senatorial Campaign Committee, 454 U.S. 27, 31-32, 102 S.Ct. 38, 42, 70 L.Ed.2d 23 (1981) that the "interpretation put on a statute by the agency charged with administering it is entitled to deference." An agency's interpretation will be upheld if its construction of the statute is "sufficiently reasonable." Id. at 39, 102 S.Ct. at 46. "To satisfy this standard it is not necessary for a court to find that the agency's construction was the only reasonable one or even the reading the court would have reached if the question initially had arisen in a judicial proceeding." Id. See also Horizon Mutual Savings Bank v. Federal Savings Loan Insurance Corp., 674 F.2d 1312, 1316 (9th Cir. 1982).
The FDA's regulation that limits the common usage test to use in the United States is a reasonable interpretation of the statute. The purpose of the amendment was to ensure that additives in foods consumed by Americans are safe to eat, taking into consideration such factors as the other items in the consumer's diet and the consumer's life span. S.Rep. No. 2422, 85th Cong., 2nd Sess. 3, reprinted in 1958 U.S. Code Cong. Ad.News 5300, 5305. It is impossible to know if an additive used by a large number of persons in a foreign country is safe unless that country has an accurate reporting system. The problem is further complicated by the fact that the harmful effect of some additives may not come to light until after a lengthy incubation period.
In view of the legislative intent to ensure safe foods for Americans, I think that the FDA's regulation was sufficiently reasonable and, therefore, should be upheld.