No. 79-1587.United States Court of Appeals, District of Columbia Circuit.Argued February 13, 1980. Judgment Filed February 15, 1980.
Decided March 20, 1980.
Page 605
[EDITORS’ NOTE: THIS PAGE CONTAINS HEADNOTES. HEADNOTES ARE NOT AN OFFICIAL PRODUCT OF THE COURT, THEREFORE THEY ARE NOT DISPLAYED.]Page 606
Diane B. Cohn, Washington, D.C., with whom William B. Schultz and Robert B. Stulberg, were on brief, for plaintiffs-appellants.
Ronald G. Whiting, of the Bar of the Supreme Court of Iowa, Washington, D.C., pro hac vice by special leave of court was allowed to argue for appellee.
Alvin Bramow, Deputy Associate Sol., Washington, D.C., was on brief, for Secretary of Labor.
Appeal from the United States District Court for the District of Columbia. (D.C.Civil Action No. 79-1044).
Before WRIGHT, Chief Judge, BAZELON, Senior Circuit Judge, and WILKEY, Circuit Judge.
Opinion for the Court filed by Senior Circuit Judge BAZELON.
BAZELON, Senior Circuit Judge:
[1] This case presents a question familiar to this court[1] : How can an administrator set safety standards in the absence of adequate scientific evidence? The question here arises in the context of action by the Secretary of Labor (the Secretary) to implement section 13(c)(4)(A) of the Fair Labor Standards Act Amendments of 1977 (the 1977 Amendments). This provision permits the Secretary to waive restrictions on employment of 10 and 11 year olds in short season agricultural harvesting, but only if eight enumerated conditions are fulfilled as demonstrated by “objective data.” 29 U.S.C.A. § 213(c)(4)(A) (1979). [2] One such condition, hereinafter described as “the pesticide condition,” requires objective data that children will not be adversely affected by pesticides and chemicals used on crops.[2] Without notice or comment, the Secretary promulgated a list of pesticides that would satisfy the statutory pesticide condition. The Secretary had no studies or data concerning the effect of any pesticide exposure on children. Because this statutory scheme so clearly requires objective proof of safety, we find the challenged regulations listing approved pesticides inconsistent with governing law. We also find the challenged regulations procedurally flawed as they were promulgated without the notice and comment procedures required under the Administrative Procedure Act (APA), 5 U.S.C. § 553. We therefore reverse the district court’s denial of a motion for preliminary injunction, and remand for proceedings not inconsistent with this opinion.[3][3] I. BACKGROUND[4] A. Statutory Framework
[5] As a general rule, Federal law prohibits employment of children under 12 years of age.[4] The 1977 Amendments permit employers
Page 607
to apply to the Secretary for a waiver of the child labor laws in order to employ 10 and 11 year olds for harvesting short-season crops. Such waivers can be granted under the Amendments only if the following conditions,[5] among others, are met:
[6] 29 U.S.C.A. § 213(c)(4)(A) (1979). These requirements are satisfied only if the Secretary makes findings “based on objective data submitted by the applicant.” Id. [7] B. The Regulations(ii) the employment of the individuals to whom the waiver would apply would not be deleterious to their health or well-being;
(iii) the level and type of pesticides and other chemicals used would not have an adverse effect on the health or well-being of the individuals to whom the waiver would apply.
[8] The history of the regulations at issue reveals an agency struggling to establish uniform standards in the face of little or no evidence. Initially, the Department of Labor attempted to set uniform evidentiary benchmarks to establish the requisite safety assurances for work with pesticide-treated crops. Thus, on April 4, 1978, the Department proposed that waivers permitting employment of 10 and 11 year olds would be granted only if employers produced evidence that their pesticides meet health and safety standards of the Environmental Protection Agency (EPA), Occupational Safety and Health Administration (OSHA), National Institute for Occupational Safety and Health (NIOSH), or other “comparable authority.” 43 Fed.Reg. 14070 (April 4, 1978). This first agency action was the only one announced with advance notice and opportunity for public comment. The agency solicited responses in a two-day public hearing. [9] Comments and testimony on the April 4 proposal pointed to the complete absence of relevant health and safety standards for children exposed to pesticides.[6] The EPA Assistant Administrator for Toxic Substances, a member of EPA’s Scientific Advisory Panel for Pesticides, and one of the instant plaintiffs alerted the Department of Labor to the fact that neither the agencies cited in the regulation nor any comparable authority had set standards that would protect children from the adverse effects of pesticide exposure. The public and governmental response demonstrated that the
Page 608
state of knowledge was simply too inadequate to support safety standards.[7]
[10] The final regulation was published in the Federal Register on June 21, 1978. The preamble acknowledged that the EPA’s pesticide exposure standards[8] “have not been shown to be safe for 10 and 11 year olds.” 43 Fed.Reg. 26563 (June 21, 1978) (codified at 29 C.F.R. § 575.5 (1979)). Section 575.5(d) of the regulation provided that employers seeking a waiver to permit employment of 10 and 11 year olds would have to submit either (1) a statement that they do not use pesticides, or (2) data proving that 10 and 11 year olds can safely be exposed to the particular pesticides used. The regulation also indicated that it would remain subject to future modification in light of studies conducted by the Secretary or his designees. [11] After promulgation of the regulations a Federal district court in the State of Washington temporarily enjoined enforcement of the general statutory minimum age restrictions[9] because the Secretary had failed to act on pending waiver applications before the June 1979 strawberry season. See Washington State Farm Bureau v. Marshall, No. C78-135T (W.D.Wash. 1978). Although that suit ultimately was dismissed, some 3900 children under the age of 11 were employed during that strawberry harvest as a result of that court’s action. Plaintiff’s Br. at 13 n. 2 (citing Department of Labor Memorandum in J.A. Vol. A at A31). [12] At this point the Department apparently felt pressed to facilitate the processing of waiver applications by specifying safety standards. On June 15, 1978, the Secretary asked a private consulting firm, Clement Associates, Inc., to review existing scientific literature in an effort to develop criteria for evaluating waiver applications. Clement was directed to focus specifically on strawberry hand-harvesting in Washington and Oregon, and potato hand-harvesting in Maine. What followed was a series of Clement reports, triggering a series of agency modifications of its regulations, each published in the Federal Register without advance notice or opportunity forPage 609
comment.[10] Each of the Clement studies proposed the use of “preharvest intervals” or “minimum entry times” which provide for a time lag between the spraying of the pesticides and the entrance of harvesters on the fields.[11] Each of the studies qualified its recommendations by observing they could “not assure safety” to the 10 and 11 year olds[12] and consistently acknowledged the need for more exhaustive study — including new studies, not merely reviews of existing literature.[13] Nonetheless, the Secretary
Page 610
relied exclusively on these studies in his regulations implementing the waiver provision.
[13] The first Clement study recommended “tentative” standards that essentially doubled the “preharvest intervals” set by the EP for adults.[14] The Department adopted this recommendation in an August 18, 1978 amendment of section 575.5(d) of its regulation.[15] The Department approved 22 pesticides for use with “preharvest intervals” before children could enter the field. The list included one known carcinogen and other chemicals found dangerous by Clement.[16] Although the Secretary did not provide advance notice or opportunity for comment before putting the regulations into effect, the EPA submitted comments challenging the adequacy of the standards. The EPA noted the special susceptibility of children to toxic effects from pesticides, and concluded “we don’t believe the data available allows us to make a decision on the safety of children either in preharvest or reentry situations.”[17] [14] The next Clement report was requested on March 19, 1979 — some six months after the earlier report. Rather than seeking new field studies to address the special question of pre-adolescent health, the Department asked Clement to review the list of approved pesticides and to investigate additional pesticides. Ten days later, Clement issued a report addressing the carcinogencity of five pesticides. Shortly thereafter, the Department issued another rule removing three known carcinogens from the list of approved pesticides.[18] This action was not based on new data, as the three chemicals had already been identified as carcinogens before the adoption of the August 18 rule.[19] [15] Clement then issued another report discussing 16 other pesticides it researched at the request of the Department.[20]Although noting the dangerous health effects associated
Page 611
with some of these substances, Clement again merely recommended doubling EPA’s preharvest intervals as a standard for children. The agency responded by adding these 16 substances to the approved list with minimum entry times established in the Clement report.[21]
[16] C. Proceedings Below[17] Plaintiffs-appellants, two private non-profit organizations representing farm-worker families,[22] sought: 1) a finding by the district court that the Secretary’s approval of pesticide use with pre-harvest intervals for 10 and 11 year olds violates the statutory waiver provision; 2) a declaratory judgment that the regulations violated the notice and comment provisions of the APA; and 3) injunctive relief enjoining the Secretary from applying the regulations that establish entry times for 10 and 11 year olds. [18] On May 21, 1979, the district court held a hearing on plaintiffs’ motion for temporary relief.[23] The government defendant presented two witnesses. The Assistant Administrator for the Wage and Hour Division described the agency’s decisionmaking process. He said that the agency turned to Clement Associates when it found no federal agency had set satisfactory standards for children exposed to pesticides. He explained that it was on the basis of the Clement Associates reports — and these reports alone[24] — that the agency issued the lists of approved pesticides with minimum entry times.[25] [19] The Department’s second witness was a representative of Clement Associates. She explained that Clement’s research team reviewed all available studies on known health effects of pesticides listed by the Department and also all existing standards for adult exposure. She further testified that the greater susceptibility of children was taken into account in the recommended entry times. She admitted on cross-examination, however, that further studies were needed to advance knowledge of the effects
Page 612
of pesticide exposure for children.[26] But, she said, her group had only been asked by the Secretary to review the existing literature, none of which directly addressed the risk to children.
[20] The district court denied the motion for preliminary relief. The judge first determined that the statute demanded findings by the Secretary, “based on objective data, that the level and type of pesticides or other chemicals will not have an adverse effect on the health or well-being of the 10 and 11 year old hand harvesters employed under a waiver.”[27] He then noted that the “Clement reports found that the recommended standards do not within the limits of current scientific knowledge assure absolute safety.”[28] Nonetheless, the judge concluded that the regulations, based solely on the Clements reports, were not “arbitrary, capricious, or otherwise not in accordance with law.”[29] [21] The core of the court’s reasoning rested on its acceptance of the Department’s view that the 1977 Amendments mandated the issuance of at least some waivers, so safety standards, no matter how arbitrary, had to be set. The court also reasoned that because absolute assurance of safety could never be obtained, the Secretary was justified in acting despite scientific uncertainty about the effect of pesticides on 10- and 11-year-old children.[30] [22] The district court also ruled on the merits of plaintiffs’ procedural claim. The court held that the APA’s “good cause” exception to notice and comment procedural requirements was satisfied by the Secretary’s desire for expeditious issuance, in time for the harvest season.[31] In support of that conclusion, the court also reasoned that the Secretary’s continuing review procedures showed further good cause for suspending notice and comment as ongoing review of submissions from interested parties could be considered.[32] This appeal followed.[23] II. ANALYSIS
[24] The district court declined to grant a preliminary injunction by reasoning that the plaintiffs would lose on the merits.[33]
In so doing the court ignored three of the four factors required to be considered before issuance of a preliminary injunction.[34] The court also wrongly evaluated the one factor it did consider: plaintiffs’ likelihood of success on the merits. We analyze these matters at some length because of the importance
Page 613
of the issues presented and the broad reach of the district court’s conclusions.
[25] A. The Test for Granting Preliminary Injunctions[26] Over twenty years ago, this court articulated these four factors to be weighed by a court before granting a stay or a preliminary injunction:
[27] Virginia Petroleum Jobbers Association v. FPC, 259 F.2d 921, 925 (D.C. Cir. 1958). These factors have assisted analysis ever since, and we see no reason to depart from them now. [28] The district court indicated no consideration of factors (2), (3), and (4) listed above, each of which weighs in plaintiffs’ favor. Indeed each of these factors supports a preliminary injunction to preserve the status quo pending a decision on the merits,[35] based on a balance of the equities between injuries to the parties and the public.[36] Therefore, we address these factors. [29] 1. Irreparable injury to plaintiffs.(1) Has the petitioner made a strong showing that it is likely to prevail on the merits of its appeal? . . . (2) Has the petitioner shown that without such relief, it will be irreparably injured? . . . (3) Would the issuance of a stay substantially harm other parties interested in the proceedings? . . . (4) Where lies the public interest?
[30] Interlocutory injunctions are sometimes necessary to assure that rights sought are not so eviscerated during trial that final relief would be to no avail. The injury to plaintiffs, absent interim relief, must also be evaluated to permit a comparison with harms to other parties and to the public. [31] Here plaintiffs represent children[37] who might work as hand harvesters if the Secretary is not enjoined from administering the waiver provision according to the challenged regulations. As a result, these children would be exposed to the pesticides and chemicals approved by the Secretary for use according to the listed “minimum entry times,” intended to reduce risks of exposure.[38] [32] The risk of harm from such exposure pendente lite would not be eliminated even if plaintiffs ultimately were to win on the merits.[39] Thus, plaintiffs convincingly make out a case of irreparable harm, absent interlocutory relief. In arguing that plaintiffs have not provided evidence of adverse effects to children, the government neglects the evidence in the record. The EPA, the agency most expert in hazards from pesticide exposure, identified some substances approved by the challenged regulations as “high-risk pesticides” and others as “highly toxic.”[40] The EPA further noted that the
Page 614
regulation’s reentry intervals may not be sufficient in light of the complex nature of pesticide degradation.[41] Further, OSHA and Clement Associates emphasized the known heightened susceptibility of pubescent children to harm from pesticide exposure.[42] It is in this context that the EPA, OSHA, and Clement Associates all stressed the insufficiency of current scientific information to assist the setting of safety standards for children exposed to pesticides.[43] These expert statements evidence insufficient knowledge of methods to combat known hazards. The hazards exist, and children’s exposure to them constitutes the kind of irreparable departure from the status quo that necessitates interlocutory relief.[44]
[33] 2. Harm to other parties.[34] The obviously interested parties here are the defendant Department of Labor, and the agricultural employers, or the growers, who are advantaged by the availability of waivers under the Department’s regulation that lists approved pesticides. A preliminary injunction enjoining application of that list would permit waiver only on satisfaction of the statutory requirement of “objective data” — reviewed by the Department and provided by applicant employers — establishing that employment conditions and pesticide exposure would not adversely affect 10 and 11 year olds.[45] A foreseeable result is that fewer waivers would be granted. The proof process would be more difficult, and the case-by-case evaluation more time-consuming, than would simple application
Page 615
of the list of approved pesticides here under challenge.[46]
[35] Growers seeking waivers thus would suffer from a diminished labor pool and would either be short on labor or have to pay more to attract employees. We assume that this result would indeed be burdensome, because under the statutory and regulatory schemes, waiver applications can be approved only based on objective data that the industry would suffer “severe economic disruption” without the child employees available through a waiver.[47] We note also, however, that the industry obviously has survived without any 10- and 11-year-old hand harvesters during the extended period preceding the passage of the statutory waiver provision in 1977.[48] With the restricted availability of waivers even subsequent to 1977,[49] we must conclude that the industry would not be significantly harmed if the Secretary is enjoined from approving applications supported by statements of pesticides used. [36] The other interested party possibly harmed by a preliminary injunction is the defendant Department of Labor. The harm to it would be the suspension of its list of approved pesticides, and the pressure to respond to demands by both growers and child protection groups. These results do not constitute substantial harm for the purpose of delaying injunctive relief. Indeed, these consequences are no different from the Department’s burdens under the statutory scheme. The Department could still seek sufficient information to support reliable safety standards, and the Department could also initiate notice and comment rule-making for that same purpose. Especially when compared with the irreparable harm from children’s exposure to known dangers, the foreseeable consequences to the DepartmentPage 616
of Labor and to growers are insignificant.
[37] 3. The Public Interest[38] The public interest is a uniquely important consideration in evaluating a request for a preliminary injunction. As the Supreme Court has held, “Courts of equity may, and frequently do, go much farther both to give and withhold relief in furtherance of the public interest than they are accustomed to go when only private interests are involved.”[50] [39] Here, two competing problems for the public interest arise. First, how will the price of produce be affected if a preliminary injunction limits the availability of 10-and 11-year-old hand harvesters? Second, what health effects would 10- and 11-year-old children risk if allowed employment? Thus, consideration of the public interest requires us to replay the analysis of the two previous factors. We must ask what is the proper balance between the economic burdens to growers denied waivers, a burden ultimately shifted to consumers, and the irreparable harm to children exposed to pesticides and chemicals through employment allowed waivers? Once again, the balance must be struck in favor of the protection of children.[51] Especially in the context of a motion for preliminary relief, equity requires protection against irreparable harm. Plainly, any possible reduction in the price of produce that might result from denying preliminary relief would be only short-term, and would never approach the value of the children’s health to the nation. [40] In sum, we conclude that the three factors ignored by the district court clearly favor the plaintiffs. It may well be that a preliminary injunction was warranted even if the district court had correctly concluded that plaintiffs would not be likely to prevail on the merits. Even if that were the case, plaintiffs needed only to present a “serious legal question” for preliminary relief to be granted under the other circumstances of their case.[52] Moreover, we conclude that the district court incorrectly assessed plaintiffs’ likelihood of prevailing on the merits. [41] B. The Merits
[42] “The duty to appraise the merits at the stage of preliminary injunction is a duty of appellate as well as trial courts.”[53]
In addition, the mistaken premises of law behind the district court’s denial of preliminary relief require review as they are bound to reappear during further proceedings in the case.[54] [43] Presented only with a motion for preliminary injunction, the district court assessed not only plaintiffs’ likelihood of prevailing on the merits, but also issued conclusions of law.[55] It found that the plaintiffs
Page 617
failed to demonstrate that the challenged regulations are arbitrary, capricious, or otherwise not in accordance of law. The court also concluded that the Secretary had “good cause” to waive the requisite notice and comment procedures. In reaching both conclusions, the district court incorrectly presumed that the existence of the waiver provision in the 1977 Amendment required the issuance of some waivers. This premise neglects the legislation’s clear burden on both waiver applicants, and the Department of Labor to establish, through “objective data,” the safety of employment conditions for 10- and 11-year-old children.[56]
[44] 1. The substance of the regulation.[45] The challenged regulations were issued amidst acknowledged uncertainty. Indeed, assuring safety in exposure to pesticides may be beyond the range of scientific certainty at present.[57]
At the same time, analytic tools are only beginning to take shape for comparing costs and benefits of safety regulations.[58] The task for a court asked to review an administrative safety standard is to determine whether the basis for the standard satisfies the requirements of authorizing statute.[59] [46] Here, the district court concluded that the Secretary “could not delay the issuance of waivers until he received assurances that certain pesticides were absolutely safe or presented zero risk because the state of scientific knowledge could not in the near future, if ever, provide such assurances.”[60] The court’s reasoning is flawed in at least three respects. [47] First, its concern with the problem of proving absolute safety obscures the simpler question posed by this case: did the Department meet the statutory requirement for issuing waivers in announcing a list of approved pesticides solely on the basis of the Clement reports? The statute requires a finding by the Secretary “based on objective data submitted by the applicant, that . . . the level and type of pesticides and other chemicals used would not have an adverse effect on the health or well-being of the individuals to whom the waiver would apply.”[61] The Secretary here chose to simplify the process by announcing a list of approved pesticides. This approach is compatible with the statute, but by law the Secretary still had to find that the pesticides and chemicals used would not adversely affect the child hand harvesters. Yet here the Secretary had no data even tending to point in that direction. The Secretary relied completely on the solicited Clement reports which, as we have discussed, expressly disclaimed any assurances of safety.[62]
None of the reports provides
Page 618
any data to show that even with the recommended entry times, pesticide exposure would not be adverse to children’s health. Indeed, each Clement report recommended supervision and medical surveillance of children exposed.[63]
[48] This is not a case in which there are data supporting the Secretary’s view, but subject to criticism.[64] The Secretary issued lists of approved pesticides solely on the basis of secondary reviews of the research literature which revealed tha “most critical information required for evaluating potential hazards to children is lacking.”[65] The problem of proving absolute safety can in no way excuse the Secretary’s failure to abide by the governing statute. [49] The reasoning of the district court is further flawed because of a factual error. The court concluded that “the state of scientific knowledge could not in the near future, if ever” provide requisite assurances of safety.[66] In fact, the Clement reports acknowledged that their findings were limited in no small part by the time constraints imposed by its contracts with the Department.[67] At the hearing on the motion for preliminary injunction, the witness from Clement testified that because of the time restrictions imposed by the contracts, Clement had never evaluated the basis for EPA pesticide standards,[68] even though EPA notified the Secretary that its data did not cover thePage 619
special case of pubescent children.[69] The Clement witness also testified at the hearing that existing methodologies, applied to this special case, would be helpful, and feasible, “time permitting.”[70] Moreover, the EPA said the Department could and should “examine retrospectively the effects of past harvest practices [to determine] whether any adverse effects are observed.”[71] Nonetheless, the Department did not seek any new studies before issuing its lists of approved pesticides; it merely asked Clement to review existing literature, none of which dealt with children’s exposure to pesticides.
[50] Bolstered by the unsupported belief that no better evidence could be obtained, the district court concluded that the Secretary was justified in relying on “the best available evidence.”[72] This standard, while approved in other Department of Labor statutes, does not appear anywhere in the waiver provision.[73] This provision instead specifies the need for “objective data” of no adverse health effects.[74] [51] Finally, the district court wrongly reasoned that the regulations had sufficient basis because absolute safety would be impossible to assure. Proof of an absolute — the absence of danger — is indeed beyond reach.[75] Yet such a proposition must not permit a court to substitute its view for the statute’s explicit requirements. Apparently, the district court believed the requirements set by statute are impossible to fulfill. But if there ever were an instance in which Congress did not delegate discretion to an agency to establish less than absolute safety standards, this might be that case. Congress permitted a limited exception to the prohibition against child labor by imposing most stringent conditions to assure safety. The proposed bill passed by the House required merely objective proof that the employment permitted through waivers “would not be deleterious to [the children’s] health or well-being.”[76] The Senate, however, refused to pass the bill without adding the further condition to guard specifically against adverse health effects from pesticide exposure.[77] The Senate sponsor of the initial language said new language was added “just to make sure that there is no danger whatsoever from any toxic material” encountered by children employedPage 620
through waivers.[78] Thus, neither the legislative history nor the language of the waiver provision provides any leeway. The directive is clear. Exceptions to the prohibition against child labor can be made only upon findings, based on objective data, that the employment will not have adverse effects on the children’s health or well-being.
[52] The district court similarly erred in concluding that a requirement of absolute safety “would in effect nullify the congressionally authorized waiver provision.”[79] In fact, a situation of “zero risk”[80] from pesticides exists and is provided for in the regulations: fields where no pesticides are used.[81] They explicitly provide that employers may support waiver applications by submitting “a statement that no pesticides or other chemicals were used on the crop to be harvested.”[82] [53] Apparently unaware of this possibility, the district court reasoned that stringent safety requirements would result in no waivers, and thus violate Congress’ intention to permit waivers.[83] This reasoning turns the waiver provision on its head. Congress provided that the “Secretary may not grant such a waiver unless” the conditions of the statute are met.[84]Congress authorized the waivers only after underscoring its commitment to protect children, especially from risks due to pesticide exposure. By conditioning waivers on the showing of proof that the work would not be deleterious or adverse to the children’s health and well-being, Congress imposed tough barriers. Absent the requisite proof, Congress’ intent would be violated if waivers are granted. Rather than turning the waiver provision into a nullity, plaintiffs’ claim of statutory violation reflects exactly what Congress intended. Thus, plaintiffs’ likelihood of prevailing on the merits is assured.[85] [54] 2. Procedural violation
[55] Issued pursuant to statutory authority,[86] the regulations here at issue are standards governing conduct and rights. They are clearly rules subject to the notice and comment procedures required by the APA, 5 U.S.C. § 553.[87] Yet only the initial regulation,
Page 621
which proposed reliance on EPA, OSHA and other existing pesticide standards, was subject to notice and comment.[88] Because the comments revealed that existing pesticide standards did not cover children, the Department changed its regulation to provide that the applicant “will either have to submit a statement that no pesticides or other chemicals were used on the crops to be harvested or submit data which upon study by the Secretary or the Secretary’s designees establishes safe reentry times for 10 and 11 year olds.”[89]
[56] Without advance notice or comment, the subsequent regulations issued by the Department first established and then modified the approved list of pesticides with minimum entry times. The district court found no procedural defect because “the public interest in the expeditious issuance of safety standards for the hand harvesting of crops with a short harvest season” falls within the “good cause” exception to the notice and comment requirement.[90] The government here similarly defends the suspension of notice and comment to make the regulations effective during impending harvest seasons.[91] The government argues that once in effect, the regulations (1) protect children and (2) assist waiver applicants with advance notice of acceptable pesticides. [57] These two purposes certainly support the promulgation of rules rather than merely a case-by-case administrative review of pesticides. But good cause to suspend notice and comment must be supported by more than the bare need to have regulations.[92]Especially in the context of health risks, notice and comment procedures assure the dialogue necessary to the creation of reasonable rules.[93] The government concedes that the challenged regulations are its first attempt to set protective standards for children employed under the agriculture waiver provision.[94] This is exactly the kind of standard which especially needs the utmost care in its development and exposure to public and expert criticism. [58] The government also here advances the district court’s reasoning that there was no procedural defect because the Secretary has agreed to accept additional data from interested parties even now that the regulations are in effect.[95] This ongoing sensitivity to developing knowledge is to be encouraged; it is a normal requirement of competent administration.[96] It does not, however, justify suspension of requirements otherwise
Page 622
mandated for the initial promulgation of regulations.
[59] Finally, on a practical note, plaintiffs correctly observe that the time pressure posed by the impending harvest seasons was due in large part to the Secretary’s own delays. The Department waited nearly seven months between the initial regulation — promulgated through notice and comment — and the first modification of it — promulgated without the requisite procedures.[97] During that time, the Department held meetings with state farm bureaus, growers, and congressional staff members.[98] The Department apparently found it quite possible to consult with the interested parties it selected. In this light, we cannot sustain the suspension of notice and comment to the general public which includes parties, such as plaintiffs who are primarily concerned with the health of their children. The Department has failed to demonstrate that the requisite procedures were “impracticable, unnecessary, or contrary to the public interest.”[99][60] III. CONCLUSION
[61] Because of the urgency of this expedited appeal, we issued a judgment immediately after oral argument.[100] As explained more fully in this opinion, we reversed the judgment below because the district court misapplied the standard for granting a preliminary injunction. The district court was directed to enter a preliminary injunction in light of our evaluation of the merits. The irreparable harm to the children affected by the regulations and the fact that the regulations are marred by both procedural and substantive violations lead us to conclude that plaintiffs must prevail on the merits.
Page 623
waivers.[103] The Department has committed itself to ongoing study to develop data specifically addressing the effect of pesticide exposure on pubescent children.[104]
Therefore, we have determined that in the interests of justice[105] — if not plain decency — the preliminary injunction enjoining application of the challenged regulations[106] must remain in effect long enough to permit the completion of appropriate notice and comment rulemaking by the Department. As a result, no waivers may be granted in reliance on the list of approved pesticides until the Department has had an opportunity to apply its developing knowledge, the entire interested public has had an opportunity to comment on new proposed regulations, and the Department has responded to those comments in publishing final rules.
(D.C. Cir. 1979); Ethyl Corp. v. EPA, 541 F.2d 1 (D.C. Cir. 1976).
Although we respect the judge’s reasoning, we agree with the parties that his ruling on preliminary relief in effect disposed of the merits. As both parties agreed during oral argument that the full case is ready for review by this court, and as the scope of the district court’s preliminary ruling reaches all the merits, this court too reviews the merits of the issues presented. See infra p. 622.
Once granted, waivers must require that employment only be during non-school hours; that the children commute daily between work and permanent residence; that the employment extend only during 8 weeks per calendar year; and that any other conditions set by the Secretary will be followed. 29 U.S.C. § 213(c)(4)(B) (1976).
Epidemiologic information for pesticide effects on children of the peripubertal ages is nonexistent. . ..
* * * * * *
[It is] impossible with reasonable certainty to determine that protection of adult workers exposed to pesticides is adequate, let alone children of peripubertal ages.
Memorandum to Grover Wrenn, Director, OSHA Health Standards Program, from Peter Infante, (June 16, 1978), reprinted in
Joint Appendix (J.A.) Vol. A at A25-A28. The memorandum also noted that children of this age are “more vulnerable to environmental insult” than others, especially in light of their developing reproductive organs. Id.
The Environmental Protection Agency, which is the Federal agency designated for determining safety standards for the use of pesticides and chemicals for agricultural workers, advised that the standards it has established for the reentry of agricultural workers after the application of pesticides were established for adult workers and not for the pubescent child. The agency has further advised that it had not and could not on the basis of its present knowledge establish any safe reentry times for 10 and 11 year olds. Several witnesses at the hearing also testified as to their personal experience as to the adverse effects of pesticides on children. Others argued that the regulations should accept the manufacturer’s reentry times (based on EPA’s standards) specified as being safe for 10 and 11 year olds. This suggestion was not adopted since it was established at the hearing that EPA safe reentry times were based on adult tolerance. Moreover, it was apparent from the testimony at the hearings that the currently established EPA and other federal standards have not been shown to be safe for 10 and 11 year olds. Accordingly till such standards are developed or until the Secretary obtains information establishing by objective data that specified reentry times are safe for 10 and 11 year olds, no waivers will be granted to an employer or group of employers who have used pesticides or other chemicals on the crops to be harvested. Therefore, the final applicant, in order to satisfy this condition, will either have to submit a statement that no pesticides or other chemicals were used on the crop to be harvested or submit data which upon study by the Secretary or the Secretary’s designee establishes safe reentry times for 10 and 11 year olds.
43 Fed.Reg. 26564 (June 21, 1978). As the government points out in its brief, the agency also heard two congressmen testify that the proposed regulations would prove too restrictive and would thwart the congressional decision to make waivers available. Gov’t Br. at 12 n. 6.
2. Preharvest intervals (PHIs) are the minimum periods of time required after the crop is treated before it may be harvested. They are designed to ensure that crop residues do not exceed residue tolerances after the pesticide is applied under the prescribed conditions. Preharvest intervals are usually based on the results of residue dissipation studies under filed application conditions.
3. Reentry intervals (REIs) are the minimum periods required after the crop is treated before field-workers are permitted to enter the field to harvest the crop or to perform other work. They are designed to protect the worker from injury resulting from exposure, usually via dermal absorption or inhalation to pesticide residues on the crop surfaces or in the soil. In principle they should be established by comparing the quantities of pesticide residue likely to be absorbed (derived from residue dissipation studies and dermal absorption measurements) with dose levels likely to be toxic (usually derived from acute or subacute toxicity studies). In practice, however, because of the complexity and uncertainty associated with such comparisons, reentry intervals are often based substantially upon accident histories.
Clement Final Report at J.A. Vol. B. at 3-4. Clement called the interim standards it recommended for children “minimum entry times (METs).” Id. at 11.
The Department initially referred to “safe reentry times” and “preharvest intervals” in its regulations. In its April 24, 1979 announcement, the agency explained that these terms
do not express the standard here, i. e., the amount of elapsed time required from the last application of the pesticide or chemical to the entry to the 10- or 11-year-old hand harvester into the field. Therefore, this document deletes those terms wherever they appear and inserts the term, “minimum entry time.”
44 Fed.Reg. 24059 (April 24, 1979).
Clement also concluded:
much of the critical information required for evaluating potential hazards to children is lacking. Specifically, there is essentially no information about rates of exposure or absorption by children, nor is there a body of experimental data on the toxic effects of pesticides in preadolescent animals. According, although we have reviewed data on metabolism, residue dynamics, and toxicity for each pesticide, we have not attempted to use these data in a formal way to set standards. For the reasons given above, we conclude that basing standards on these factors systematically would not be possible without the generation of extensive new data.
Clement Final Report, J.A. Vol. B at 5-6.
Establishment of Preharvest Intervals that would assuredly protect the health of 10- and 11-year-olds would require a thorough scientific review of existing data on residue dynamics, metabolism, and toxicity of each of the pesticides to preadolescents and adolescents.
J.A. Vol. A at A49, A81. The research group’s final report included the following statement:
We wish to make clear that the suggested METs cannot assure safety to 10- and 11-year-old harvesters. Because of major uncertainties in the data, we have used scientific judgment in deriving the proposed METs. The METs have been established on the basis of currently available data and may be subject to change as more data become available. We therefore strongly urge that a program of additional research be instituted.
Clement Final Report, J.A. Vol. B at 13 (emphasis in original).
In each report, the research group also recommended medical surveillance of all children exposed to pesticides or chemicals because of waivers. J.A. Vol. A at A50, A83; J.A. Vol. B at 13.
In its final report nearly a year later, Clement Associates found “convincing data” that Captan is carcinogenic and that Benomyl is “moderately or highly persistent and [has] serious toxic effects.” Clement Report, J.A. Vol. B at 16. Therefore, Clement in its final report did not recommend minimum entry times for these and seven other substances with similar effects. At the same time, it increased the recommended delay period to 16 days for Endosulfan and 40 days for Carbaryl.
(May 16, 1979). Also in these amendments, the Department created a new category for substances lacking sufficient data to set minimum entry times.
The Washington and Oregon State Farm Bureaus filed suit to enjoin implementation of the regulation, this time attacking the removal of Captan and Benomyl from the approved list (Captan removed altogether; Benomyl lacking sufficient data to set minimum entry times). A preliminary injunction was granted but then vacated in that case. Washington State Farm Bureau v. Marshall, Civil Action No. C79-197T; appealed, No. 79-4342. Plaintiffs here participated in that suit as amici curiae.
We’re dealing in a scientific area where, I think, everyone and certainly the scientists, first of all, will admit that there are no absolutes to be dealt with here and all we’re suggesting is given the fact that Congress changed this law and seemed to indicate that there had to be some kind of standards applied, that we had to look at what data was available. . . . [Clement Associates] look[ed] at the data which was, in fact, available and then applying scientific judgment, came up with recommended entry times and since the Secretary is charged basically that, [sic] reasonably it is not arbitrary and capricious, again I submit that acting on the basis of that, fully recognizing that the hedges that must be contained for scientific purposes. His acting in just accepting that data as scientific data submitted to him is, in fact, adequate.
Tr. at 17, J.A. Vol. A at A145.
The Secretary’s decision to issue waivers and set preharvest intervals in the face of some scientific uncertainty over the precise effects of pesticide exposure upon the health and well-being of 10 and 11 year old children was consistent with the congressional intent to provide for the employment of 10 and 11 year olds in the hand harvesting of short season crops. The Secretary, in issuing regulations, could not delay the issuance of waivers until he received assurances that certain pesticides were absolutely safe or presented zero risk because the state of scientific knowledge could not in the near future, if ever, provide such assurances. A requirement of absolute safety or zero would in effect nullify the congressionally authorized waiver provision.
Id. at J.A. Vol. A at A126-27 (¶ 8, Conclusions of Law).
(D.C. Cir. 1958).
Clement Associates similarly identified factors likely to make children “more susceptible than adults to the effects of toxic substances.” Clement Associates, Inc., Safety Factors for Children Employed as Harvesters of Strawberries in Washington and Oregon (Aug. 7, 1978), J.A. Vol. A at A42; Clement Associates, Safety Factors for Children Employed as Harvesters of Potatoes in Maine (Aug. 7, 1978), J.A. Vol. A at A64.
Moreover, both claims are beside the point. The issue is not how many children will be irreparably harmed but whether any children represented by the plaintiffs would be. On this record, the plaintiffs have established that the regulations would cause exposure of at least some children to known health risks.
The provisions of [the child labor prohibition] shall apply to an employee below the age of sixteen employed in agriculture in an occupation that the Secretary of Labor finds and declares to be particularly hazardous for the employment of children below the age of sixteen, except where such employee is employed by his parent or by a person standing in the place of his parent on a farm owned or operated by such parent or person.
Since 1974, children under 12 could be employed in agriculture only if 1) employed by their parents, or persons standing in the place of their parents or 2) employed with the consent of their parents, or persons standing in the place of their parents, on small farms exempt from the Fair Labor Standards Act, 29 U.S.C. § 213(c)(1)(A) (1976) (codifying Fair Labor Standards Act Amendments of 1974, P.L. 93-259, § 25).
The current provision, which requires objective data that pesticide exposure will not adversely affect 10- and 11-year-old child hand harvesters, was enacted in 1977. 29 U.S.C.A. § 213(c)(4) (1979) (codifying Fair Labor Standards Amendments of 1977, P.L. 95-151, § 8).
Memorandum from Donald Elisburg, Assistant Secretary for Employment Standards, Department of Labor, to Ray Marshall, Secretary of Labor (June 14, 1978), Plaintiffs’ Exhibit 13 in Support of Plaintiffs’ Motion for Preliminary Injunction reprinted in J.A. Vol. A at A96 (several hundred initial waiver applications denied, a few asked to provide more information). Due to a temporary restraining order issued by a district court in the State of Washington, some 3900 children aged 11 and under were employed during the June 1979 strawberry harvest. Plaintiffs’ Br. at 13 n. 2 (citing Washington State Farm Bureau v. Marshall, No. C78-135T (W.D.Wash. 1978), and Memorandum from Charles Pugh, Director, Office of Program Development and Accountability, Department of Labor, to Donald Elisburg, Assistant Secretary for Employment Standards, Department of Labor (Oct. 13, 1978), Plaintiffs’ Exhibit 4 in Support of Plaintiffs’ Motion for Preliminary Injunction, reprinted in J.A. Vol. A at A31). Employers’ reliance on this employment pool would surely be mistaken as it includes children beneath the statutory minimum age even under the waiver provision.
An order maintaining the status quo is appropriate when a serious legal question is presented, when little if any harm will befall other interested persons or the public and when denial of the order would inflict irreparable injury on the movant. There is substantial equity, and need for judicial protection, whether or not movant has shown a mathematical probability of success.
559 F.2d 841, 844 (D.C. Cir. 1977). Similarly, the Supreme Court concluded that:
Where the questions presented by an application for an interlocutory injunction are grace, and the injury to the moving party will be certain and irreparable, if the application be denied and the final decree be in his favor, while if the injunction be granted the injury to opposing party, even if the final decree be in his favor, will be inconsiderable, or may be adequately indemnified by a bond, the injunction usually will be granted.
Ohio Oil Co. v. Conway, 279 U.S. 813, 49 S.Ct. 256, 73 L.Ed. 972 (1929).
W. Lowrance, Of Acceptable Risk 99 (1976).
Clement indicated that the Department merely asked it “to propose reentry standards that will provide reasonable assurance of safety for 10- and 11-year-old fieldworkers engaged in the hand harvesting of potatoes and strawberries.” Clement Final Report, J.A. Vol. B at 5 (emphasis added). This assignment itself departed from the statutory requirement of “objective data . . . that . . . [t]he level and type of pesticides and other chemicals used would not have an adverse effect on the health or well-being” of the 10-and 11-year-old field-workers. 29 U.S.C.A. § 213(c)(4)(A), (c)(4)(A)(iii) (1979) (emphasis added).
In addition, EPA staff apparently found fault with the Clement reports’ methodology. The Department of Labor received a letter including EPA staff criticisms that 1) both the EPA adult standards and Clement standards for children confused the risk of dermal and inhalation with the risk of chronic ingestion; and 2) the Clement studies did not address the pesticides generally applied just before potato harvests. Letter from Edwin L. Johnson, Assistant Deputy Administrator for Pesticide Programs, EPA, to Donald E. Elisburg, Assistant Secretary for Employment Standards, Department of Labor (Sept. 26, 1978). Plaintiffs’ Exhibit 9 in Support of Plaintiffs’ Motion for a Preliminary Injunction, reprinted in J.A. Vol. A at A92.
The time constraints were severe. The Department first contracted with Clement on June 1978. Clement submitted final reports less than two months later, on August 7, 1978, in time for the September potato harvest. The Department adopted the Clement recommendations on August 18, 1978. The Department next contracted with Clement on March 19, 1979, just before the spring strawberry harvest season. Clement issued one report on March 30, 1979, while recommendations were adopted by the agency on April 10, 1979. Then on April 20, 1979, Clement issued another report, followed by two more Department modifications of its regulations on April 24, and May 16, 1979. Clement issued its final, summary report two days later, on May 18, 1979.
The government nevertheless argues that Congress intended to invest discretion in the Secretary because the Senator proposing the added language explained it directed the Secretary to “give appropriate consideration to the problem of chemicals.” Gov’t Br. at 27 n. 16 (quoting 123 Cong.Rec. S16635 (daily ed. Oct. 7, 1977) (remarks of Sen. Williams)). The plaintiffs correct this misconception in noting that the same Senator supported placing stringent conditions on the waivers “to insure that young children will not be exposed to the hazardous effects of pesticides and herbicides.” Plaintiffs’ Br. at 26 (quoting 123 Cong.Rec. S16626 (daily ed. Oct. 7, 1977) (remarks of Sen. Williams)).
(D.C. Cir. 1976); Pacific Gas Electric Co. v. FPC, 506 F.2d 33 (D.C. Cir. 1974).
[W]hen a health-related standard such as this is involved, the good cause exemption may not be used to circumvent the legal requirements designed to protect the public by ensuring that interested persons will have the opportunity to bring to the agency’s attention all relevant aspects of the proposed action and thereby enhance the quality of agency decisions.
Community Nutrition Institute v. Butz, 420 F.Supp. 751
(D.D.C. 1976). Where the knowledge base for rules is uncertain, public confidence is possible only if the procedures followed promote full and open debate.
[A]s previously stated by the Secretary, the Department of Labor has undertaken a continuing study of the use and effect of pesticides and other chemicals used in short season crops in order to establish minimum entry times for specified pesticides and chemicals for use on specified crops.
At oral argument, the Department’s counsel indicated that the EPA was entering an agreement to assist in developing pesticide exposure standards for 10 and 11 year olds.
Page 1360