The opinion of the court was delivered by: EDWARD R. BECKER
The plaintiff is Rohm and Haas Company ("Rohm & Haas"), a major producer and supplier of chemical products to world-wide industrial and agricultural markets. Rohm & Haas brought this action on May 4, 1981, a scant 14 days ago, seeking to prevent EPA from issuing to defendant-intervenor, The Mobil Oil Corporation ("Mobil") an Experimental Use Permit ("EUP") pursuant to Section 5 of FIFRA, 7 U.S.C. § 136c, and the regulations which EPA has promulgated to implement § 5, 40 C.F.R. § 172. That permit would allow Mobil, which is a competing producer and supplier of chemical products (through its Mobil Chemical Division), to conduct certain tests on a new pesticide product called "Tackle" which it is developing. Tackle, like "Blazer", a product which Rohm & Haas has commercially marketed since April, 1980, is a postemergent herbicide used principally for broad-leaf weed control on soybeans.
The soybean crop in the United States is planted between May 15 and July 15 of each year, and, in order to be effective, a herbicide must be applied within 14 to 23 days after planting. Because it is undisputed that, unless this matter is resolved in time for Mobil to finalize its arrangement with the soybean farmers who are to test the pesticide and deliver it to the test locations, another year's planting season will go by, we have given this case expedited consideration. On the afternoon of May 4 we heard oral argument on plaintiff's request for a restraining order, but declined to grant relief, concluding that it would be more appropriate to hold a full hearing on plaintiff's motion for a preliminary injunction. As we learned more about the case and determined that what was at issue was essentially a question of law, we exercised our powers under Rule 65(a)(2) of the Federal Rules of Civil Procedure and advanced the trial on the merits, consolidating it with a hearing on the preliminary injunction. The trial took place on the afternoon and evening of May 7th and the morning of May 8, 1981, and was followed by post-trial argument on May 11, 1981. In addition, the parties have filed extensive briefs. This opinion, which constitutes our Findings of Fact and Conclusions of Law pursuant to FRCP 52(a), finally disposes of the matter and places it, as has been our intention, in a posture for a prompt ruling by the Court of Appeals. Thus, the rights of the parties can be adjudicated without the effectuation of relief by a mere default through the passage of time.
Although the problem of statutory construction involved in this case is convoluted, the factual background is rather simple. Both Blazer and Tackle contain the same active ingredient - the sodium salt of a synthetic chemical compound whose common name is acifluorfen. Acifluorfen was apparently developed independently by both Rohm & Haas and Mobil, who have been engaged for several years in litigation over the patent rights to the chemical.
The record does not reflect when Mobil identified the sodium salt of acifluorfen for potential use as a pesticide. It does reflect, however, that Rohm & Haas identified it in 1973 and that, in order to commercialize its new discovery, Rohm & Haas conducted various toxicity, residue, and environmental tests from 1975 through 1976 to demonstrate that the use of a compound containing acifluorfen could be used on soybeans on an experimental basis without risk to health or environment. It is also undisputed that, based upon these tests and upon extensive data submitted by Rohm & Haas to the EPA, on November 1, 1977 the EPA granted Rohm & Haas' application for an EUP. It is also undisputed that, under the authority of the EUP, Rohm & Haas conducted additional tests and submitted voluminous additional data to the EPA, resulting in the full registration of Blazer pursuant to § 3 of FIFRA on April 10, 1980.
This case arises because of the following state of affairs (described more precisely in our formal Findings of Fact). First, when EPA considered Mobil's application for an EUP for Tackle, it compared Mobil's product chemistry data with product chemistry data earlier submitted by Rohm & Haas for Blazer. Second, the EPA's conclusion that granting Mobil a permit for experimental use of Tackle on soybeans and allowing the sale and distribution of Tackle-treated soybeans in commerce would not subject the public to any risks over and above those risks, if any, posed by the use of Rohm & Haas' Blazer on soybeans, depended upon a comparison of the product chemistry data submitted by Mobil for Tackle with that submitted by Rohm & Haas for Blazer and upon a comparison of the EUP label terms and conditions proposed by Mobil for Tackle with those approved for Rohm & Haas' Blazer registration. EPA's conclusion that issuance of the EUP to Mobil was warranted was thus based on Rohm & Haas' registration of Blazer, which, in turn, was issued on the basis of Rohm & Haas' data. EPA would thus not have granted Mobil's EUP without Rohm & Haas' data on file.
Rohm & Haas has advanced three principal arguments in support of its action to enjoin the issuance of the EUP to Mobil. First, Rohm & Haas contends that EPA cannot base the issuance of Mobil's EUP on data submitted by Rohm & Haas because Section 5 of FIFRA does not allow an EUP to be issued to an applicant based upon data submitted by another company. Rohm & Haas insists that an EUP must be based solely on data submitted by the applicant. Rohm & Haas argues alternatively that, even if Section 5 does not prevent the issuance of an EUP based on data other than data submitted by the applicant, certain data consideration restrictions contained in Section 3(c)(1)(D) of FIFRA are applicable to the issuance of EUPs and preclude EPA from basing Mobil's EUP on data previously submitted by Rohm & Haas. These data consideration restrictions, which are described in detail infra and which are at the heart of plaintiff's case, confer upon those who have expended effort and funds in research and development of scientific data certain long-term rights of exclusive use to that data, or at least the right of compensation for its use. They reflect one of the statutory policies at issue in this case - that of encouraging research and innovation by protecting the fruits or the innovator's work. The bulk of the legal argumentation and briefing - and of the discussion herein - will concern the § 3(c)(1)(D) point and will, of course, include consideration of the countervailing policies referenced by defendants - the alleged dominant congressional purpose of achieving safe and effective registration of new pesticides, and administrative convenience, insofar as these policies, too, may inform our construction of the statute.
The third and final point pressed by Rohm & Haas is related to the publication of a Federal Register notice by EPA on May 5, 1981, 46 Fed.Reg. 25138, describing Mobil's application for an EUP and inviting public comments thereon. The EPA has maintained that this publication was unnecessary because it was not required by the EPA's standards for Federal Register publication and that the publication was a reaction (or overreaction) to the persistence of Rohm & Haas' counsel in his efforts to determine with precision EPA's intentions in the matter. EPA has announced its intention to issue the EUP without reference to such comment as may be received, and Rohm & Haas, predictably, has asserted that the issuance of a permit prior to fair opportunity for comment and adequate consideration thereof is unlawful.
In the third segment of this opinion we will describe the contentions of the parties in greater detail, after first having explained the relevant regulatory scheme. At this juncture we note only that defendants counter each of the points raised by the plaintiff.
The trial of the case was preceded by a one and one-half day discovery period which resulted in an extensive stipulation of facts by Rohm & Haas and EPA. During the interim between the temporary restraining order conference and the trial, we granted, over opposition of the plaintiff, the motion of Mobil to intervene as a party defendant. Mobil adopted virtually all the stipulated facts. The trial consumed the equivalent of one and one-half trial days.
In addition to the stipulated facts, various documents and several affidavits, the plaintiff offered two witnesses, Dr. Stephen F. Krzeminski, the Rohm & Haas Manager of Regulatory Affairs for North America and of registration activities throughout the world, and Dr. Jerry Smith, its Chief Toxicologist. The EPA called James Wilson Akerman, Branch Chief for Fungicides and Herbicides in the Registration Division. Mobil called no witnesses, but like the plaintiff relied on certain affidavits of its employees as well as on the stipulation of facts and a few documents.
Notwithstanding the pendency of the motion for preliminary injunction, we devoted the trial to the merits and did not dwell upon the issues of irreparable harm or the balance of harm to the parties and the public, which would have been necessary were we to adjudicate the issues involved in the preliminary injunction motion. That approach was a function of our intention to render a prompt decision on the merits (delivered herein); it enabled us to move more rapidly, unburdened by the difficult and time-consuming balance of harm questions. So as to obviate the necessity of ruling upon the temporary restraining order and preliminary injunction applications in the interim, EPA agreed not to issue the EUP permit to Mobil pending our final decision.
In the ordinary course, findings of fact would be our first order of business after the preliminary statement. However, because the findings of fact in this case are inextricably woven to the fabric of the regulatory scheme, we must first discuss that scheme, starting with the statute and the EPA regulations and then turning to the EPA's modus procedendi in dealing with EUP applications, particularly the genre which the EPA dubs "me-too" applications. The description of the EPA practices should be considered a part of our findings of fact. The balance of the findings will follow our description of the regulatory scheme; they will, in the main, deal with the manner in which the EPA treated Mobil's application for an EUP for "Tackle." Factual matters are not, for the most part, in dispute.
Finally, we shall, in our discussion of the law, explain the reasons for our decision - that the interpretation of the statute advanced by the EPA and Mobil appears to be the correct one, resulting in our decision to find in favor of the defendants. Because we have had so little time to develop an opinion in the matter (we have been on a protracted trial in another case since this case was filed), we have drawn heavily upon the parties' requested findings and summaries of the statutory scheme, and are not entirely happy with the organization of the discussion or the opinion as a whole. This may be the classical example of writing a long opinion because there was not time to write a short one.
EPA is also authorized, in Section 5 of FIFRA, 7 U.S.C. § 136c, to issue an "Experimental Use Permit" ("EUP") to allow a potential registration applicant to test the pesticide prior to seeking registration. EUP's authorize limited use of a product on a limited number of acres under tightly controlled conditions for the purpose of helping to determine whether the product can ultimately be registered for sale. An EUP, once granted, does not represent a decision by EPA to grant registration and does not entitle the holder of the EUP to launch the pesticide on the market. Only a registration granted under Section 3 in compliance with all the terms of that section constitutes authorization to market.
In order for EPA to make the determination that a pesticide will not cause unreasonable adverse effects, an applicant seeking EPA registration of a pesticide must submit or cite data documenting the safety and efficacy of the pesticide.
FIFRA Section 3(c)(1)(D), 7 U.S.C. § 136a(c)(1)(D). For purposes of this litigation, the most important information requirement imposed on applicants for registration is that imposed in Section 3(c)(1)(D), 7 U.S.C. § 136a(c)(1)(D), which requires a variety of test data defining the risks and benefits of the product for which registration is sought. This information, as set forth in the Regulations, includes test data reflecting the effectiveness of the pesticide in performing its past suppression function ("efficacy data") and test results concerning the dangers of the product ("hazard data").
When an applicant seeks to register a product which is the same or substantially the same as a previously registered product, EPA practice has been to not require the subsequent, or "me-too," applicant to submit its own data. Rather, EPA permits the "me-too" applicant to rely on the efficacy and hazard data already submitted. However, in Section 3(c)(1)(D), Congress established certain limitations on EPA's ability to consider data submitted by one company in support of applications filed by another company.
For purposes of this case, the most significant data-consideration limitation is that found in Section 3(c)(1)(D)(i), which prohibits EPA from considering, without the permission of the data submitter, data submitted in support of the registration of a pesticide containing an active ingredient
registered for the first time after September 30, 1978 in support of another company's application. This "exclusive use" period lasts for 10 years after the product is first registered.
In addition, in the case of a pesticide which will be applied to food or feed crops,
an applicant must first obtain a tolerance from EPA before a registration can be issued for that use. The tolerance requirement is imposed by Section 408 of the Federal Food, Drug, and Cosmetic Act ("FFDCA"), 21 U.S.C. § 346a. In essence, a tolerance defines the level of the pesticide product and its degradation products which is not to be exceeded in a food or feed crop to which the pesticide is applied or in animal products (meat, milk, and eggs) derived from animals which have eaten feed crops to which the pesticide has been applied. The level at which a tolerance is set is the level required to protect the public from adverse health effects caused by chronic dietary exposure to pesticides. Although a separate tolerance must be established for each crop to which a particular active ingredient will be applied, once a tolerance has been set for a particular active ingredient it is applicable to all products containing that active ingredient. Thus, EPA does not require "me-too" applicants to ask EPA to reestablish the tolerance when they submit their applications for registration.
A separate section of FIFRA - Section 5, 7 U.S.C. § 136c - governs the issuance of an EUP. An EUP allows the permit holder to apply a limited amount of his unregistered product to a limited number of acres for the purpose of performing tests necessary to obtain registration of that product. Section 5 is silent on the question whether EPA may consider data submitted by one company in support of an EUP application filed by another company, containing no provision analogous to section 3(c)(1)(D).
The regulations that EPA has promulgated to implement the requirements of Section 3 are contained in 40 C.F.R. Part 172. There are two basic types of EUP's: one for a new use for a previously registered product, and the second for a new, as yet unregistered, product. The information which EPA may require to be submitted in support of an application is described in 40 C.F.R. § 172.4. Subsection 172.4(b)(1) describes the general information which all applicants must submit, consisting principally of details of the testing program which the applicant plans to undertake under his permit. Subsection 172.4(b)(2) also describes the tolerance requirements that must be met if the product is to be tested on food or feed crops.
Finally, 40 C.F.R. § 172.4(b)(3) sets forth certain additional information requirements for unregistered pesticides. The additional information required includes information regarding the composition of the pesticide product, some data regarding the existence of residues after application in the environment (if available), and certain toxicity tests which would indicate the potential for causing injury to the users or other persons exposed to the pesticide. The short of it is that considerably less data is required to support the issuance of an EUP than is required to support the issuance of a Section 3 registration. See footnote 5, supra. It should also be noted, however, that the grant of an EUP does not in any way assure the eventual issuance of a registration, but simply provides a mechanism to allow a potential registration applicant to generate the data necessary to support its registration. EUPs are generally routinely granted, provided that EPA does not conclude that the experimental use will cause unreasonable adverse effects on man or the environment.
The EPA uses a similar "me-too" approach when evaluating EUPs to the approach it uses when evaluating registration applications. Under that approach, the EPA simply analyzes a proposed pesticide's product chemistry data and its label, compares its analysis with a registered pesticide's product chemistry data and label, and, if there is substantial similarity, grants an EUP to the applicant. The theory behind the "me-too" approach is that once the EPA is assured that an applicant's product has the same chemistry and will be used in the same manner as an already-registered pesticide, the EPA can rely upon the already-registered pesticide's supportive research and test data as evidence that the "substantially similar" product will not harm health or the environment. When using the "me-too" approach and relying upon the original registrant's research and test data, the EPA does not actually re-review the data submitted by the original applicant. Rather, the agency assumes that because the data were sufficient to obtain an initial registration, the data should also sustain a substantially similar product.
It is EPA's use of this approach in approving a "me-too" EUP application for Mobil's Tackle based upon Rohm & Haas' previously submitted Blazer data which gave rise to this litigation. We turn to the contentions of the parties.
III. The Contentions of the Parties
Rohm & Haas, plaintiff in this litigation, markets a registered pesticide, Blazer. The data submitted by Rohm & Haas to EPA in support of its Blazer registration application is concededly protected under the exclusive use provision of FIFRA. That data was referenced by EPA in evaluating Mobil's application for an EUP for Tackle, a product the EPA found to be "substantially similar."
As we have noted above, Rohm and Haas contends: (1) that EPA cannot base the issuance of Mobil's EUP on data submitted by Rohm & Haas because Section 5 requires that an EUP be based solely on data submitted by the applicant; (2) that the data consideration restrictions contained in Section 3(c)(1)(D) of FIFRA at all events preclude EPA from basing Mobil's EUP on data submitted by Rohm and Haas; and (3) that EPA cannot issue the EUP to Mobil until it has evaluated public comments on the proposed EUP emanating from the Federal Register publication.
The arguments raised by Rohm and Haas essentially present three distinct legal issues: 1) whether Section 5 allows an EUP to be based on data other than data submitted by the applicant; 2) whether, if Section 5 allows EUP's to be based on data other than data submitted by the applicant, the data consideration restrictions established by Section 3(c)(1)(D) are applicable to EUP's; and 3) whether the EUP which EPA proposed to issue Mobil is of national or regional significance so as to require publication of a Federal Register notice pursuant to 40 C.F.R. § 172.11.
On the first question, Rohm & Haas relies upon what it characterizes as the "clear language" of FIFRA's EUP provision,
with reference to the provision that calls for the Administrator to act "after receipt of the application and all required supporting data." In Rohm and Haas' submission, that provision requires that all needed research and test data be "received" by EPA from the applicant, and that nothing already in EPA's files may be used. Secondly, Rohm & Haas cites 40 C.F.R. § ...