Center for Environmental Health v. Michael Regan

103 F.4th 1027

Opinion

PUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 23-1476

CENTER FOR ENVIRONMENTAL HEALTH; CAPE FEAR RIVER WATCH; CLEAN CAPE FEAR; TOXIC FREE NC,

Plaintiffs – Appellants,

and

DEMOCRACY GREEN; THE NC BLACK ALLIANCE, Plaintiffs,

v.

MICHAEL S. REGAN, Administrator of the U.S. Environmental Protection Agency; UNITED STATES ENVIRONMENTAL PROTECTION AGENCY,

Defendants – Appellees.

Appeal from the United States District Court for the Eastern District of North Carolina, at Wilmington. Richard E. Myers, II, Chief District Judge. (7-22-cv-00073-M)

Argued: January 23, 2024 Decided: June 10, 2024

Before AGEE and WYNN, Circuit Judges, and John A. GIBNEY, Jr., Senior United States District Judge for the Eastern District of Virginia, sitting by designation.

Affirmed by published opinion. Judge Agee wrote the opinion, which Judge Gibney joined and Judge Wynn joined in part. Judge Wynn wrote a dissenting opinion.

2

ARGUED: Robert Matthew Sussman, I, SUSSMAN & ASSOCIATES, Washington, D.C., for Appellants. Michelle N. Melton, UNITED STATES DEPARTMENT OF JUSTICE, Washington, D.C., for Appellees. ON BRIEF: Thomas J. Lamb, LAW OFFICES OF THOMAS J. LAMB, P.A., Wilmington, North Carolina; Michael Connett, WATERS, KRAUS AND PAUL, El Segundo, California, for Appellants. Todd Kim, Assistant Attorney General, Robert P. Stockman, Environment and Natural Resources Division, UNITED STATES DEPARTMENT OF JUSTICE, Washington, D.C.; Sonja Rodman, Stephanie Schwarz, Margaret Clark, UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, Washington, D.C., for Appellees.

3

AGEE, Circuit Judge:

This litigation concerns the Toxic Substances Control Act (the “TSCA”), which allows citizens to petition the Administrator of the Environmental Protection Agency (the “EPA”) to initiate a proceeding for the issuance of a rule or order requiring the testing of certain chemical substances. 15 U.S.C. § 2620(a). If the EPA denies that petition, the petitioner is entitled to a de novo review by a district court. Id. § 2620(b)(4). If, however, the EPA grants the petition, a district court lacks jurisdiction to review the petition. See id.

Pursuant to the TSCA, four North Carolina-based citizen groups (“Petitioners”)

petitioned the EPA to require the testing of fifty-four Per- and Poly- Fluoroalkyl Substances (“PFAS”) likely prevalent in their community. The EPA granted that petition, agreeing to require testing on PFAS as a class through its own testing protocol. Petitioners sought judicial review of the EPA’s decision, contending it was in effect a denial of their petition. The district court dismissed Petitioners’ complaint for lack of jurisdiction. Petitioners appeal.

We affirm.

I. The TSCA

To understand the claims and arguments on appeal, some familiarity with the TSCA is necessary.

Congress enacted the TSCA after finding that there was a lack of information on a large number of potentially harmful chemicals that humans and the environment are

4

exposed to each year. 1 Believing that this information gap should be filled by those who manufacture and process the potentially harmful chemicals, Congress directed the EPA, through the TSCA, to require those manufacturers and processers to develop information on certain chemicals they produce. See id. § 2601(b)(1).

In that vein, under § 2603 of the TSCA, the EPA shall require the manufacturers and processers of a specific chemical to conduct testing on that chemical if the EPA finds that three conditions are met:

(1) the manufacture, distribution in commerce, processing, use, or disposal of a chemical substance or mixture, or that any combination of such activities, may present an unreasonable risk of injury to health or the environment,

(2) there is insufficient information and experience upon which the effects of such manufacture, distribution in commerce, processing, use, or disposal of such substance or mixture or of any combination of such activities on health or the environment can reasonably be determined or predicted, and

(3) testing of such substance or mixture with respect to such effects is necessary to develop such information[.]

Id. § 2603(a)(1)(A)(i). If these conditions are met, the EPA can mandate testing through a rule, order, or consent agreement. Whatever the format, the resulting EPA mandate must identify the chemical to be tested, the protocol and methodologies necessary for the development of information, and a specific period within which the relevant party must submit that information. Id. § 2603(b)(1).

1

The TSCA refers to “chemical substances and mixtures.” 15 U.S.C. § 2603. For ease of reference, we use “chemicals” to mean the same.

5

As to the particular chemical to be tested, the TSCA directs the EPA to “encourag[e]

and facilitat[e] . . . the grouping of 2 or more chemical substances into scientifically appropriate categories in cases in which testing of a chemical substance would provide scientifically valid and useful information on other chemical substances in the category.” Id. § 2603(h)(1)(B)(ii). It also mandates a “tiered screening and testing process, under which the results of the screening-level tests or assessments of available information inform the decision as to whether 1 or more additional tests are necessary.” Id. § 2603(a)(4).

The TSCA does not provide a specific way in which the EPA must determine the protocols and methodologies to be utilized, but it does provide a significant number of guidelines for the EPA to follow when doing so.

Additionally, § 2620 of the TSCA allows citizens to participate in the chemical-

identifying process. If a citizen believes that there is a lack of information regarding a potentially harmful chemical, he or she can petition the EPA to “initiate a proceeding for the issuance, amendment, or repeal of a rule . . . or an order under [§] 2603.” Id. § 2620(a). The petition must explain why it is “necessary” for the EPA to act, utilizing the same three requirements found in § 2603. Id. § 2620(b)(1). 2 The EPA has ninety days to grant or deny such a petition. Id. § 2620(b)(3). If the EPA grants the petition, it “shall promptly commence an appropriate proceeding” for the issuance of a rule or order. Id. If the EPA

2

The requirement that a petitioner must demonstrate that testing is necessary does not mandate that he or she provide the relevant testing protocols. It simply requires that he or she show that the information cannot be gathered through other avenues, such as modeling.

6

denies the petition, it “shall publish in the Federal Register the [EPA]’s reasons for” that denial. Id.

A petitioner whose petition was either denied or not acted on within the ninety-day period has the right to “commence a civil action in a district court of the United States to compel the [EPA] to initiate a rulemaking proceeding as requested in the petition.” Id. § 2620(b)(4)(A). 3 If the petitioner chooses to bring such an action, the district court conducts a de novo proceeding in which the petitioner must demonstrate by a preponderance of the evidence that:

(I) information available to the [EPA] is insufficient to permit a reasoned evaluation of the health and environmental effects of the chemical substance to be subject to such rule or order; and

(II) in the absence of such information, the substance may present an unreasonable risk to health or the environment, or the substance is or will be produced in substantial quantities and it enters or may reasonably be anticipated to enter the environment in substantial quantities or there is or may be significant or substantial human exposure to it[.]

3

The parties agree that § 2620(b)(4)(A)’s failure to state that the petitioner can commence an action to compel the EPA to initiate a proceeding to issue an order (in addition to a rulemaking proceeding) was a drafting error. This error seems to be an oversight carried over from an earlier version of the TSCA, when the EPA did not have the authority to issue orders. Other provisions in § 2620 make clear that petitioners can request that the district court mandate the initiation of a proceeding for an order as well. See 15 U.S.C. § 2620(b)(4)(B)(i) (explaining what the petitioner must show in an action in federal court “for the issuance of a rule . . . or an order under [§] 2603” (emphasis added)).

7

Id. § 2620(b)(4)(B)(i). 4 If the district court finds that both requirements are met, “the court shall order the [EPA] to initiate the action requested by the petitioner.” Id. § 2620(b)(4)(B). It can also award costs of suit and reasonable attorney’s fees in appropriate circumstances.

II. Background

With that pertinent statutory background in mind, we turn to the facts of this case.

As noted, Petitioners filed a § 2620 petition with the EPA, requesting that the EPA require the testing of fifty-four PFAS they alleged were prevalent in their community.

A. PFAS

For context, there are more than 6,500 substances classified as PFAS, which are a group of long-lasting synthetic chemicals, sometimes known as “forever” chemicals, that do not break down or degrade over time and are therefore highly persistent. J.A. 44. PFAS are released into the environment during the manufacturing of PFAS-containing products and through the use and disposal of those products. Because PFAS are highly mobile, they are easily and widely distributed and have been detected in drinking water, soil, and the blood of humans and animals worldwide.

Although there is limited information available on the health and environmental effects of most PFAS, the available data shows that PFAS may pose a serious risk to human health. Animal studies show that PFAS are linked to cancer, hormone disruption, liver and

4

There are different requirements if the petitioner is seeking a rule or order under a different section of the TSCA. We address only requirements involving a petition for a rule or order under § 2603 because that is all that is at issue in this appeal.

8

kidney damage, and developmental and reproductive harms, among other serious health conditions.

Nonetheless, the use of PFAS is common. For example, they are used in firefighting foams, coatings for clothing and upholstery, manufacturing other chemicals, and fast-food wrappers and boxes.

B. The Petition

Relevant to this appeal, PFAS are also used at the Chemours Company’s (“Chemours”) manufacturing plant south of Fayetteville, North Carolina. Petitioners assert that the PFAS from Chemours’ plant pollute the Cape Fear River, which flows to the City of Wilmington, North Carolina, where over 300,000 residents use it for drinking water. Various PFAS linked to the Chemours plant have been identified in private wells, wastewater, stormwater, sediment, groundwater, soil, and local produce in the Cape Fear watershed. Some PFAS have even been detected in the blood of the general population.

Concerned about the effects that PFAS could be having on their community, Petitioners filed a § 2620 petition, requesting that the EPA issue a § 2603 rule or order requiring Chemours to conduct testing on fifty-four PFAS manufactured at its facility.

Petitioners addressed each of the § 2603 requirements in a forty-nine-page petition.

See J.A. 152–200. They first alleged that the “serious health and environmental concerns presented by PFAS as a class” demonstrate that the fifty-four specific PFAS likely present an unreasonable risk of injury to health or the environment. J.A. 162 (emphasis added). Petitioners then explained that “there is an absence of sufficient data to determine risks to the large exposed population within range of the [Chemours] facility and the surrounding

9

ecosystem.” J.A. 155–56. Petitioners lastly claimed that testing was necessary and proposed a detailed testing program consisting of, among other things, human health effects studies, experimental animal studies, and ecological effects studies. Notably, throughout their petition, Petitioners presented information regarding PFAS as a class to support their position. See, e.g., J.A. 175 (attempting to justify the insufficient information element by noting that “there are many PFAS of potential concern to the public that may be found in the environment. Most of these PFAS lack sufficient toxicity data to inform our understanding of the potential for adverse human or ecological effects” (cleaned up)).

C. The EPA’s Response

Prior to granting the petition, the EPA, on its own initiative, had developed a National PFAS Testing Strategy in order to “deepen understanding of the impacts of PFAS, including potential hazards to human health and the environment.” J.A. 209. As part of its strategy, the EPA grouped the more than 6,500 PFAS by “structural and physical-chemical properties into 70 total terminal categories” and decided to test the PFAS in phases. J.A. 209. The first phase will test PFAS that meet three criteria: (1) there is limited existing toxicity information on the PFAS; (2) there is a known manufacturer of the PFAS who could be ordered to do the testing; and (3) the PFAS are representative of their category as a whole. J.A. 265–66. The EPA identified twenty-four priority candidates from different categories that will be tested in the first phase. It expects that the first phase of testing will provide data that can be extrapolated to the 2,950 PFAS that belong to those categories. As new information is developed, the EPA will review the remaining categories “to determine what testing may be required” of them. J.A. 209.

10

The EPA relied on this developed protocol when it issued a decision granting Petitioners’ petition. 5 Specifically, the EPA decided to “issue a rule or order . . . compelling health and environmental effects testing regarding PFAS” as a class. J.A. 202. The EPA stated that it did not “mak[e] any final determinations . . . regarding whether the TSCA [§ 2603] criteria have been met with respect to any particular chemical substance.” J.A. 208. However, it found that “the petition sets forth facts demonstrating that it is appropriate to issue a [§ 2603] order to address the health and environmental effects of PFAS” generally. J.A. 208. It noted that, as it considers the information developed after each phase of its Testing Strategy, “[s]pecific determinations regarding whether the criteria for issuing TSCA [§ 2603] orders have been met will be made” as to additional PFAS. J.A. 208.

Additionally, the EPA did not adopt Petitioners’ proposed testing strategy because, as noted, it instead decided to use its National PFAS Testing Strategy to address Petitioners’ request. The EPA noted that Petitioners’ intent “is to develop information that would enable the Cape Fear River watershed communities to better understand the potential effects to their health from PFAS exposures.” J.A. 201. It further explained that if the EPA attempted to research each of the PFAS “one at a time, it will be impossible for [the] EPA, states, or communities to expeditiously understand, let alone address, the risks these substances may pose to human health and the environment.” J.A. 209. In contrast, through the National PFAS Testing Strategy’s first phase of testing—which covers twenty-

5

The EPA originally denied the petition. Petitioners then filed an action challenging that decision and submitted a request for reconsideration with the EPA. The EPA then granted reconsideration and Petitioners stayed this suit. The grant of the petition followed.

11

four categories of PFAS—Petitioners (and the public) will learn “data that cover[s] 30 of the 54 petition chemicals.” J.A. 202. It elaborated that seven of Petitioners’ identified PFAS would be tested directly in the first phase and another twenty-three are in the same categories as the ones being tested first, meaning that the first phase will provide data that can be extrapolated to those other, not-individually-tested PFAS. An additional nine PFAS identified by Petitioners are in categories that the EPA may subsequently test. As a consequence, Petitioners should get data on thirty-nine of the fifty-four chemicals. The other fifteen of the chemicals that Petitioners identified, however, “do not fit the definition of PFAS used in developing the Testing Strategy,” because “there is robust data on some of them available to the [EPA].” J.A. 203. Nevertheless, the EPA noted that it “is conducting more in-depth analyses of the existing data, which will inform later phases of testing.” J.A. 203.

III. Procedural History

Unhappy with the EPA’s testing strategy, Petitioners sought review of the EPA’s decision in the United States District Court for the Eastern District of North Carolina. They requested that the district court “direct [the] EPA to initiate a proceeding for the issuance of a rule or order requiring Chemours to carry out the studies on the 54 PFAS specified in [their] petition.” J.A. 64. The EPA filed a motion to dismiss for lack of jurisdiction, asserting that the district court only had the authority to review the denial of a petition and Petitioners’ petition was granted. Petitioners responded that the EPA’s “grant” of their petition was in effect a denial and the court therefore had jurisdiction to review the petition.

12

They reasoned that, despite the demands in their petition, the EPA (1) did not require direct testing on forty-seven of the fifty-four PFAS identified by Petitioners and (2) did not adopt their testing strategy. All in all, Petitioners asserted that the EPA actually denied more of their requests than it granted.

The district court disagreed. Ctr. for Env’t Health v. Regan, 666 F. Supp. 3d 509 (E.D.N.C. 2023). Although the district court recognized that it could not simply accept the EPA’s characterization of its decision as a “grant,” upon review, it determined that the EPA’s decision was in fact a grant. The court reasoned that the EPA reasonably chose to grant Petitioners’ request to test the fifty-four PFAS as a category—PFAS generally— which the TSCA encourages the EPA to do. As to the EPA’s failure to adopt Petitioners’ specific testing program, the district court explained that Petitioners “have a right to petition [the] EPA to initiate proceedings for the issuance of rules and orders, but [they] do not have a right to compel the content of [the] EPA’s proceedings or to compel [the] EPA to issue a specific rule or order.” Id. at 521. Given its determination that the EPA’s decision was a grant, the district court determined it lacked jurisdiction to review the petition and dismissed Petitioners’ complaint accordingly. 6 Petitioners timely appealed, arguing that the district court erroneously concluded that the EPA’s decision was a grant in fact. We have jurisdiction to review the district court’s final order under 28 U.S.C. § 1291.

6

The district court also concluded that “even if [it] could entertain [Petitioners’] suit under Section 21 of the TSCA, the suit is moot” because the only relief the district court could give was to compel the EPA to initiate proceedings, which it had already done. Ctr. for Env’t Health, 666 F. Supp. 3d at 522.

13

IV. Standard of Review

We review a district court’s dismissal for lack of jurisdiction de novo, accepting as true all non-conclusory factual allegations contained in the complaint and drawing all reasonable inferences from those facts in favor of the plaintiff. Pond v. United States, 69 F.4th 155, 160–61 (4th Cir. 2023).

V. The EPA’s Decision was a Grant in Fact As an initial matter, we note that we do not accept as definitive the EPA’s characterization of its decision. To do so would allow the EPA to evade judicial review by simply mislabeling its treatment of a petition as a grant when it is actually a denial. We instead “look[] to the contents of the [EPA]’s action” and decide whether its decision was a grant in fact. Azar v. Allina Health Servs., 139 S. Ct. 1804, 1812 (2019); see also Sorenson Commc’ns, Inc. v. F.C.C., 567 F.3d 1215, 1223 (10th Cir. 2009) (“The agency’s own label for its action is not dispositive.”). 7 Petitioners contend that two aspects of the EPA’s decision render it an effective denial of their petition: the EPA’s choice (1) to test PFAS as a class, rather than individually

7

We also do not accept Petitioners’ characterization of the decision. Petitioners seem to imply that because we review the facts in the light most favorable to them at this stage of the litigation, we are required to accept Petitioners’ contention that the EPA’s treatment of their petition was a denial. Not so. Whether the EPA’s decision was a grant is a question of law reserved for the Court. See Turner v. Thomas, 930 F.3d 640, 644 (4th Cir. 2019) (“[W]e need not accept legal conclusions couched as facts[.]” (cleaned up)).

14

testing all fifty-four PFAS identified by Petitioners, and (2) to utilize its own testing protocols instead of Petitioners’ proposed program. We address each aspect in turn.

A. The EPA’s Treatment of PFAS as a Class Petitioners first contend that the EPA’s decision was actually a denial because by choosing to test PFAS as a class, the EPA refused to directly test forty-seven of the fifty- four substances specified in their petition. We disagree.

There is no doubt that Petitioners requested that the EPA individually test fifty-four specific PFAS that they allege are produced at a particular facility in their community. See J.A. 155 (“[Petitioners] request that . . . the [EPA] require health and environmental effects testing on 54 [PFAS] manufactured by [Chemours] at its chemical production facility in Fayetteville, North Carolina.”). There is also no doubt that the EPA recognized Petitioners’ precise request. See J.A. 201 (“The petitioners requested such testing in connection with 54 chemical substances the petition identifies as PFAS and alleges are released into the environment by [Chemours].”). The question, therefore, is whether the EPA was authorized to grant Petitioners’ specific request through its testing of PFAS as a class. The answer is yes.

Once the EPA determines that testing on particular chemicals is necessary, the TSCA mandates that the EPA “encourage[] and facilitat[e] . . . the grouping of 2 or more chemical substances into scientifically appropriate categories in cases in which testing of a chemical substance would provide scientifically valid and useful information on other chemical substances in the category” in order to reduce testing on vertebrates. 15 U.S.C.

15

§ 2603(h)(1)(B)(ii). 8 Here, the EPA implicitly determined that each of the more than 6,500 chemicals in the PFAS class met the requirements for testing. See J.A. 258 (“To address this data gap and fundamentally advance our understanding of [PFAS], EPA has developed this National PFAS Testing Strategy (Strategy) to deepen understanding of the impacts of PFAS, including potential hazards to human health and the environment.”). Having done so, it created a testing strategy that groups PFAS into categories and then tests specific representative PFAS within each category. And there is no dispute that the testing of the PFAS selected by the EPA will provide valid and useful information on PFAS in general. Even Petitioners recognize as much. See J.A. 162 (Petitioners acknowledging that the EPA’s “authority to address ‘categories’ could be applied to all PFAS or to subgroups, such as the PFAS manufactured at the Chemours Fayetteville facility” in their petition (emphasis added)); J.A. 171 (“For groups of chemicals that qualify as a ‘category’ under [§] 26(c) because of similarities in chemical structure and/or toxicity, these determinations need not be made for every individual substance but can be based on the common characteristics of the class.”). The EPA therefore appropriately followed the TSCA’s requirements when choosing to treat PFAS as a class.

In fact, Petitioners’ framing of their petition supports the EPA’s decision to treat PFAS as a class. Throughout their petition, Petitioners repeatedly relied on information about “PFAS as a class” to demonstrate that they met the § 2603 requirements. For example, to show that the specifically identified PFAS may present an unreasonable risk

8

Petitioners do not dispute the applicability of this provision.

16

of injury to health or the environment, Petitioners relied on the “health effects of PFAS as a class.” J.A. 171–74; see, e.g., J.A. 171 (“These effects have been observed consistently in both short-chain and long-chain substances and should be presumed to be of concern for all PFAS.”). To demonstrate that there is insufficient information about the specific PFAS, Petitioners relied on the EPA’s statement in its PFAS Action Plan that “there are many PFAS of potential concern to the public that may be found in the environment. Most of these PFAS lack sufficient toxicity data to inform our understanding of the potential for adverse human or ecological effects.” J.A. 175 (cleaned up). The same is true for Petitioners’ attempt to illustrate that testing is necessary. See J.A. 176 (explaining that animal testing is necessary, in part, because “other PFAS are known to cause common modes of toxicity in vivo, such as effects on liver biochemistry and cholesterol, pup survival, and immunotoxicity” (endnotes omitted)). Given that Petitioners therefore demonstrated that testing was necessary for PFAS as a class, it was reasonable for the EPA to grant the petition as to PFAS as a class.

Although Petitioners would prefer that the EPA localize their testing mandates to Petitioners’ community, the TSCA does not promise that an agency will limit testing to a particular geographic area or a particular facility. It simply ensures that when a petitioner identifies a data gap regarding a potentially harmful chemical, a manufacturer of that chemical will develop the required information about it.

The EPA is doing that here and is actually gathering information at a much faster rate than would likely be the case if it mandated testing of Petitioners’ specific PFAS individually. By considering PFAS as a class, Petitioners will receive information on thirty

17

of the PFAS they identified in the EPA’s first phase of testing. And some of that information is currently being developed through the testing of the particular PFAS Petitioners identified that the EPA mandated Chemours to administer. See, e.g., United States Environmental Protection Agency, Order Under Section 4(a)(2) of the Toxic Substances Control Act (June 16, 2022), https://perma.cc/YF5H-5Y3N (requiring Chemours to test fluorotelomer sulfonamide betaine); United States Environmental Protection Agency, Order under Section 4 of the Toxic Substances Control Act (TSCA) (August 15, 2023), https://perma.cc/FQ2J-ZX7T (requiring Chemours to test hexafluoropropylene oxide-derived acyl fluoride). Additionally, the EPA’s testing strategy is dynamic and will evolve as it receives information from the initial phases of testing. As a result, it is certainly possible that more of Petitioners’ identified PFAS will be tested than was originally expected and the testing results more focused and informative.

Moreover, the EPA’s tiered strategy is set by the TSCA. “When requiring the development of new information,” the TSCA mandates that the EPA “employ a tiered screening and testing process, under which the results of screening-level tests or assessments of available information inform the decision as to whether 1 or more additional tests are necessary.” 15 U.S.C. § 2603(a)(4). This strategy ensures that the information gaps are filled while also preventing unnecessary costs and delay. To illustrate, the EPA will receive information on 2,950 PFAS after testing only twenty-four of them. See Mobil Oil Expl. & Producing Se. v. United Distribution Cos., 498 U.S. 211, 230 (1991) (“An agency enjoys broad discretion in determining how best to handle related, yet discrete,

18

issues in terms of procedure.”). This TSCA-mandated tiered process is much more efficient than individually testing fifty-four PFAS that may provide identical information. 9 At bottom, Petitioners are getting what they requested—companies that manufacture potentially harmful PFAS, like Chemours, are being required to develop information on the human and environmental effects of those substances. Petitioners will receive information on the health effects of PFAS, and this information will help them, the public, and the EPA protect the health of communities and the environment. 10

9

Petitioners also complain that the EPA hasn’t identified the chemicals that will be tested. But the EPA is only required to specify the chemical in the rule or order requiring testing and the EPA has done that in the rules it is has promulgated thus far. See 15 U.S.C. § 2603(b)(1)(A) (stating only that the EPA’s “rule, order, or consent agreement . . . shall include . . . [an] identification of the chemical substance or mixture for which testing is required”).

10

The dissent agrees that the EPA has the authority to test PFAS as a class and, for the most part, did so properly here. Its disagreement stems from the EPA’s statement that fifteen of Petitioners’ proposed PFAS did “not fit the definition of PFAS used in developing the Testing Strategy” because the EPA “determined that there is robust data on some of them available to [the EPA].” J.A. 203. In the dissent’s view, the fact that these fifteen chemicals were excluded from the EPA’s testing strategy necessarily means that the EPA excluded them from the class of PFAS and accordingly denied the petition as to those substances. We don’t read the EPA’s statement in that way. The EPA explained that it developed the National PFAS Testing Strategy by placing the thousands of PFAS into seventy different categories. J.A. 265. It then “identified a total of 56 terminal categories that lack any data about the toxicity of the PFAS in that category” and chose candidates for initial testing from “each of those 56 terminal categories.” J.A. 265 (emphasis added). In other words, to determine which of the thousands of substances to initially test, the EPA chose to prioritize those lacking any data about their toxicity. The fifteen PFAS identified by the Petitioners did not meet that criterion because the EPA determined that “there is robust data on some of them”—though the EPA may test them as their strategy is updated. J.A. 203. And, as will be discussed, the EPA is free to fashion its testing strategy as it sees fit. We therefore take no issue with the EPA’s decision not to test the fifteen relevant PFAS.

19

B. The EPA’s Utilization of Its Own Testing Strategy Petitioners next contend that the EPA’s decision was actually a denial because the EPA allegedly declined to require 97 percent of the individual studies requested in their petition. In their view, when the EPA grants a petition, it is required to adopt the petitioner’s proposed testing strategy. Petitioners are mistaken.

To understand why Petitioners are incorrect, it’s important to fully understand the EPA’s process when mandating testing under § 2603. As noted, the TSCA instructs the EPA to mandate testing if it makes three findings—a certain chemical may present an unreasonable risk of injury to health or the environment, there is insufficient information on the effects of that chemical, and testing is necessary to develop such information. 15 U.S.C. § 2603(a)(1)(A)(i).

But determining that testing is necessary does not end the EPA’s involvement. Once it is determined that testing is necessary, the TSCA then tasks the EPA with “determining the protocols and methodologies” necessary to conduct those tests so that it can specify how testing must be conducted in its final rule or order. Id. § 2603(b)(1). While the EPA has some discretion to develop the protocols and methodologies, its discretion is not unfettered. The TSCA provides certain guidelines for the EPA to follow.

For example, the TSCA instructs the EPA to consider the “relative costs of the various test protocols and methodologies” and the “reasonably foreseeable availability of the facilities and personnel needed to perform” the required testing. Id. § 2603(b)(1). It also prescribes the health and environmental effects to be tested, id. § 2603(b)(2)(A), and directs the EPA to “reduce and replace, to the extent practicable . . . the use of vertebrate

20

animals in the testing of chemical substances or mixtures,” id. § 2603(h)(1). Additionally, as noted, the TSCA requires the EPA to “employ a tiered screening and testing process.” Id. § 2603(a)(4). These provisions—among others—guide the EPA’s determination of the best testing strategies to require in a particular case.

When a citizen files a § 2620 petition for the EPA to initiate a proceeding for a rule or order under § 2620, the EPA’s role doesn’t materially change. Section 2620 simply allows citizens to identify a potentially harmful chemical and attempt to demonstrate that testing is necessary. Having had a chemical brought to its attention, the EPA must determine whether the § 2603 requirements were met. If they were, the EPA must grant the petition and then “commence an appropriate” proceeding for the issuance of a rule or order. Id. § 2620(b)(3). In that proceeding, the EPA determines the protocols and methodologies to include in its final rule, utilizing the guidelines provided in § 2603.

This is the same process as when the EPA determines on its own that testing is required for a specific chemical under § 2603: the EPA first determines that testing is necessary and then conducts a proceeding to determine what testing protocols and methodologies are appropriate. Nothing in the TSCA suggests that a petitioner is entitled to demand a particular testing program. The petitioner’s only role is to bring to the EPA’s attention a potentially harmful chemical.

It would be unreasonable to require the EPA to adopt a petitioner’s proposed testing strategy—after having only ninety days to review the petition—simply because the EPA agrees that testing on a specific chemical is necessary. And the EPA should not have to deny a petition that would otherwise be successful because the petitioner may not have

21

considered the costs or feasibility of his or her proposed testing program or because he or she unnecessarily proposed the utilization of vertebrate animals. Although § 2320 undoubtedly provides citizens with a powerful tool to be involved in the chemical- identifying process, it does not give them the unrestrained ability to force companies to conduct specific testing when the § 2603 requirements are met.

However, not satisfied with this reading of the TSCA provisions that govern the petition, Petitioners turn to a TSCA provision that governs the district court’s review of a denied petition. Their argument rests on § 2620’s statement that when a district court finds—in its de novo review of a petition that has been denied—that the petitioner has met its burden, “the court shall order the [EPA] to initiate the action requested by the petitioner.” Id. § 2620(b)(4)(B). Petitioners contend that “the action requested by the petitioner” is a specific testing program. So, they argue that, because the district court can mandate that the EPA adopt a proposed testing program, it must be that the EPA is always obligated to adopt their program. Again, we disagree.

Petitioners’ argument relies on a false premise. The action requested by the petitioner referenced in § 2620(b)(4)(B) is not the adoption of a specific testing program. What the petitioner requests is that the EPA initiate a proceeding for the issuance of a rule or an order. See id. § 2620(a) (“Any person may petition the [EPA] to initiate a proceeding for the issuance, amendment, or repeal of a rule . . . or an order.” (emphasis added)). So, all the district court can provide a successful petitioner is an order directing the EPA to

22

initiate a proceeding for a rule or order—depending on what was requested in the petition. 11 In the proceeding, the EPA will determine the best testing protocols and methodologies to utilize. 12 Therefore, the EPA did not effectively deny Petitioners’ petition by declining to adopt their proposed testing program. In fact, by promptly commencing a proceeding for determining how to best test PFAS, the EPA gave Petitioners all that they were entitled to receive. 13 Indeed, as noted, since granting the petition, the EPA has required Chemours to conduct testing on multiple PFAS identified by Petitioners, and there are more to follow.

****

The EPA’s decision was a grant, and the district court therefore did not have jurisdiction over the petition. See id. § 2620(b)(4)(A) (giving the district court jurisdiction over only the EPA’s denial of or failure to act on a petition).

11

This seems especially obvious considering that the petitioner is not required to show that testing is necessary when attempting to meet his or her burden in federal court. See 15 U.S.C. § 2620(b)(4)(B)(i) (stating that the district court need only find by a preponderance of the evidence that there is insufficient information about the relevant chemical and that chemical may present an unreasonable risk to health or the environment). It would be illogical to think that the district court could require the EPA to adopt a petitioner’s testing program when it need not even consider whether testing is necessary.

12

Petitioners also argue that the legislative history of the TSCA supports their argument. But “[l]egislative history is irrelevant to the interpretation of an unambiguous statute.” In re Moore, 907 F.2d 1476, 1478–79 (4th Cir. 1990) (citation omitted). And the TSCA clearly tasks the EPA, not private citizens, with developing the testing protocols and methodologies it requires manufacturers and processers to employ. Moreover, to the extent the Court considers the relevant legislative history, it is conflicting at best.

13

The fact that Petitioners have received all the district court can offer them—the initiation of a proceeding regarding the testing of PFAS—also supports our conclusion that the EPA’s decision was in fact a grant.

23

Although Petitioners are thus not entitled to a de novo review of their petition, that does not mean they are without a future remedy depending on the circumstances of the EPA’s implementation of its testing program. Petitioners are free to seek judicial review of any final order the EPA may issue. See id. § 2618(a)(1)(A) (stating that “any person may file a petition for judicial review” of any of the EPA’s final rules or orders issued under § 2603). They are also free to file a new petition with the EPA. Our decision precludes only one potential avenue for relief.

V. Conclusion

For the foregoing reasons, we affirm the district court’s order.

24

WYNN, Circuit Judge, concurring in part and dissenting in part:

I agree with the majority opinion’s determination that the Environmental Protection Agency (“EPA”) properly granted Petitioners’ petition with respect to the thirty-nine chemicals that fell within the 6,504-member group of Per- and Polyfluoroalkyl Substances (“PFAS”) as defined in the EPA’s National PFAS Testing Strategy (the “Testing Strategy”). But the EPA’s decision not to include within that group fifteen of the chemicals set forth in the petition rendered its decision a partial denial subject to de novo review. Because the EPA effectively denied the petition as to those fifteen chemicals, I must respectfully dissent as to the portion of the majority opinion that reaches the contrary conclusion. I otherwise join the majority’s opinion.

I.

From the outset, it is helpful to understand the framework that the Toxic Substances Control Act sets forth regarding testing of potentially harmful chemicals.

In 1976, Congress enacted the Toxic Substances Control Act to address the risks of injury to human health or the environment posed by the “large number of chemical substances and mixtures” that were being developed and produced each year. Toxic Substances Control Act, Pub. L. No. 94-469, 90 Stat. 2003 (1976) (codified as amended at 15 U.S.C. §§ 2601 et seq.). Specifically, Congress made clear that it is the policy of the United States that “adequate information should be developed with respect to the effect of chemical substances and mixtures on health and the environment” and, as such, provided the EPA with the authority to enable the development of such information. 15 U.S.C. §§ 2601(b)(1), 2603.

25

One way that the EPA can develop such information is by requiring testing of potentially harmful chemicals. The procedure the EPA follows when it decides to require testing and implement a testing regime is found in 15 U.S.C. § 2603. That section sets forth a two-part procedure to which the EPA must adhere when it recognizes a potentially harmful chemical for which it thinks testing may be needed: first, it must make certain findings that testing is necessary for the chemical; second, it must implement a testing regime within certain restrictions set forth by Congress.

The findings the EPA must make before it can conclude that testing is necessary, as set forth in § 2603(a), are: (1) “the manufacture, distribution in commerce, processing, use, or disposal of a chemical substance or mixture, or . . . any combination of such activities, may present an unreasonable risk of injury to health or the environment”; (2) “there is insufficient information and experience upon which the effects of such manufacture, distribution in commerce, processing, use, or disposal of such substance or mixture or of any combination of such activities on health or the environment can reasonably be determined or predicted”; and (3) “testing of such substance or mixture with respect to such effects is necessary to develop such information.” Id. § 2603(a)(1)(A)(i). 1 If the EPA determines that the § 2603(a) requirements are met for a particular chemical substance, it “shall . . . require that testing be conducted on such substance.” Id § 2603(a).

1

The EPA shall also require testing if it makes three alternate findings under 15 U.S.C. § 2603(a)(1)(A)(ii). No party has suggested that those alternative findings are relevant here.

26

Also, Congress recognized that the public might have concerns about certain chemicals that the EPA had not yet taken the initiative to test. Thus, in 15 U.S.C. § 2620, Congress provided a mechanism for individuals and groups to petition the EPA to initiate a proceeding for the issuance of a rule or order requiring the testing of chemicals about which they are concerned. 15 U.S.C. § 2620(a). The § 2620 petition must demonstrate that testing is “necessary” for certain chemicals, which means petitioners must set forth facts establishing that § 2603(a)’s three requirements are met. Id. § 2620(b)(1).

This means that, when the EPA receives a § 2620 petition requesting testing pursuant to § 2603, it considers whether it agrees with the petitioners that § 2603(a)’s requirements have been satisfied. 2 Id. § 2603(a), 2620(b)(1). And if the EPA agrees that § 2603(a)’s requirements are satisfied as to a particular chemical, it must grant the petition and “promptly commence an appropriate proceeding.” Id. § 2620(b)(3). But if it disagrees, it must deny the petition.

If the EPA denies the petition—or if it effectively denies the petition by failing to act on the petition within ninety days—then the petitioners can seek judicial review. Id. § 2620(b)(4)(A). This allows the district court to consider the petition de novo. Id. § 2620(b)(4)(B). In that de novo proceeding, the petitioners must demonstrate by a preponderance of the evidence that (1) “information available to the [EPA] is insufficient to permit a reasoned evaluation of the health and environmental effects of the chemical

2

Section 2620 petitions can also request that the EPA act pursuant to its authority under §§ 2604, 2605, or 2607. See 15 U.S.C. § 2620(a). But because the petition here requested action under § 2603, I discuss only the requirements relevant to that section.

27

substance” identified in the petition; and (2) “in the absence of such information, the substance may present an unreasonable risk to health or the environment, or the substance is or will be produced in substantial quantities and it enters or may reasonably be anticipated to enter the environment in substantial quantities or there is or may be significant or substantial human exposure to it.” Id. § 2620(b)(4)(B). If the district court is satisfied that these requirements are met, it “shall order the [EPA] to initiate the action requested by the petitioner.” Id. § 2620(b)(4)(B).

II.

Accordingly, when the EPA received the petition at issue in this case, it had ninety days to determine whether the fifty-four chemical substances set forth in the petition (“the petition chemicals”) satisfied § 2603(a)’s requirements. The EPA purported to do this—it noted that it had determined that testing was “appropriate” for PFAS as a class, as defined in the Testing Strategy, and, accordingly, it would conduct testing of representative members of that group. J.A. 208. The Toxic Substances Control Act anticipates this type of category-based action—15 U.S.C. § 2625 provides that “[a]ny action” the Toxic Substances Control Act authorizes or requires the EPA to take “with respect to a chemical substance or mixture may be taken . . . with respect to a category of chemical substances or mixtures.” 15 U.S.C. § 2625(c)(1).

Thus, in exercising its authority to take action with respect to a category pursuant to § 2625, the EPA was permitted to conclude that testing was appropriate for PFAS as a class. Majority Op. at 15. Additionally, the EPA necessarily implicitly determined that each chemical included in the 6,504-member class would independently satisfy § 2603(a)’s

28

requirements. 3 Id. And, once the EPA has determined that it is “necessary” to test a particular chemical—or in other words, that the § 2603(a) requirements are satisfied as to a particular chemical—the petitioner’s role in the testing process is complete. Majority Op. at 20. So, if the 6,504-member class defined by the Testing Strategy included all fifty-four petition chemicals, I would agree with the majority opinion that the EPA granted the petition in its entirety.

But the 6,504-member class did not include all fifty-four petition chemicals. It explicitly excluded fifteen of the petition chemicals from the 6,504-member class, stating that those fifteen chemicals “do not fit the definition of PFAS used in developing the Testing Strategy.” 4 J.A. 203. Because the EPA did not otherwise state that it was taking

3

I reach this conclusion despite the EPA’s statement in its response to the petition that it was “not making any final determinations in this letter regarding whether the Toxic Substances Control Act section 4 criteria have been met with respect to any particular chemical substance.” J.A. 208. As the majority correctly determines, in acting with respect to a category pursuant to § 2625, the EPA must necessarily have considered whether the members of that category met § 2603(a)’s requirements.

4

The point that the majority opinion seeks to make with its footnote 10 is not clear to me. But to be sure, when the EPA developed its Testing Strategy, it started with a database of over 900,000 chemical substances. J.A. 260. It then applied various definitional filters to whittle down that large number of substances “to generate the ‘starting list’ of PFAS considered for the [Testing] Strategy.” J.A. 260. After applying these definitional filters, the EPA was left with 6,504 chemical substances. J.A. 261. The EPA’s explanation of how it developed this group of 6,504, and its explanation regarding its action as to fifteen of the petition chemicals, was its attempt to explain to Petitioners that fifteen of the petition chemicals were filtered out during the process of developing the 6,504-member group; in other words, these fifteen chemicals did not make the “starting list” of chemicals that the EPA would later categorize into seventy groups and from which the EPA would pick initial candidates for testing. J.A. 260. Thus, contrary to what appears to be the majority opinion’s belief, the excluded fifteen petition chemicals were not included in the group of 6,504 chemicals. Majority Op. at 18 n.10.

29

action as to those fifteen chemicals, and because more than ninety days have passed since the Petitioners filed their petition, the EPA effectively denied the petition as to the excluded fifteen chemicals. That denial affords the Petitioners the opportunity for judicial review of the partial denial of the petition. 15 U.S.C. § 2620(b)(4)(A).

In explaining why it chose to exclude fifteen petition chemicals from the 6,504-

member class, the EPA stated that there is already “robust data on some of them available.” J.A. 203. Perhaps the EPA was attempting to explain that the excluded fifteen chemicals could not satisfy § 2603(a)(1)(A)(i)(II)’s requirement that “there is insufficient information” from which to determine the risks of a particular chemical. Of course, if the EPA thinks that a chemical in a petition does not satisfy one or more of the § 2603(a) threshold requirements, then the Toxic Substances Control Act mandates that it deny the petition. But, in accordance with the procedures set forth in § 2620, the petitioners must then have the opportunity to demonstrate to a district court that the EPA’s view is incorrect.

Petitioners here maintain that they can make such a demonstration regarding the fifteen excluded chemicals. Opening Br. at 45 n.10 (“[P]laintiffs and their scientific advisors believe that these 15 substances are properly defined as PFAS and in any case independently meet the criteria for testing in sections 4 and 21 of [the Toxic Substances Control Act].”). Regardless of the merits of their argument that the fifteen excluded

30

chemicals meet § 2603(a)’s requirements, the Toxic Substances Control Act requires that Petitioners have the opportunity to make their case to the district court. 5 Accordingly, I respectfully dissent as to the portion of the majority opinion that concludes the EPA granted the petition regarding the fifteen excluded chemicals.

5

The majority correctly recognizes that, in excluding these fifteen chemicals, “the EPA noted that it ‘is conducting more in-depth analyses of the existing data, which will inform later phases of testing.’” Majority Op. at 11 (quoting J.A. 203). But the fact that the EPA may later reconsider its exclusion of these fifteen chemicals should not allow the EPA to evade judicial review now, particularly given the ninety-day time limit that the Toxic Substances Control Act sets for the EPA to rule on a petition. See 15 U.S.C. § 2620(b)(3).