Opinion
No. 19-15535
08-05-2021
SOUTHERN CALIFORNIA ALLIANCE OF PUBLICLY OWNED TREATMENT WORKS; Central Valley Clean Water Association ; Bay Area Clean Water Agencies, Plaintiffs-Appellants, v. U.S. ENVIRONMENTAL PROTECTION AGENCY ; Deborah Jordan, Acting Regional Administrator, United States Environmental Protection Agency, Region IX, Defendants-Appellees.
Melissa A. Thorme (argued) and Patrick F. Veasy, Downey Brand LLP, Sacramento, California, for Plaintiffs-Appellants. John D. Gunter II (argued), Michael C. Gray, and Leslie M. Hill, Attorneys; Eric Grant, Deputy Assistant Attorney General; Environment and Natural Resources Division, United States Department of Justice, Washington, D.C.; for Defendants-Appellees.
Melissa A. Thorme (argued) and Patrick F. Veasy, Downey Brand LLP, Sacramento, California, for Plaintiffs-Appellants.
John D. Gunter II (argued), Michael C. Gray, and Leslie M. Hill, Attorneys; Eric Grant, Deputy Assistant Attorney General; Environment and Natural Resources Division, United States Department of Justice, Washington, D.C.; for Defendants-Appellees.
Before: Eric D. Miller and Danielle J. Forrest, Circuit Judges, and Patrick J. Schiltz, District Judge.
Formerly known as Danielle J. Hunsaker.
The Honorable Patrick J. Schiltz, United States District Judge for the District of Minnesota, sitting by designation.
MILLER, Circuit Judge: This case involves a challenge to nonbinding guidance that the Environmental Protection Agency issued to recommend a statistical method for assessing water toxicity. The Administrative Procedure Act allows a plaintiff to challenge only final agency action, and an agency's action is final only if it imposes legal consequences. Because the guidance at issue imposes no such consequences, we conclude that the APA does not permit this challenge, and we affirm the district court's judgment in favor of the agency.
The Clean Water Act prohibits "the discharge of any pollutant by any person" into the waters of the United States without a permit. 33 U.S.C. § 1311(a). Although the EPA may issue discharge permits, the Act also allows it to delegate permitting responsibility to the States. Id . § 1342(b). "If [permitting] authority is transferred, then state officials—not the federal EPA—have the primary responsibility for reviewing and approving ... discharge permits, albeit with continuing EPA oversight." National Ass'n of Home Builders v. Defenders of Wildlife , 551 U.S. 644, 650, 127 S.Ct. 2518, 168 L.Ed.2d 467 (2007). State permitting programs must meet minimum requirements set by EPA regulations. 40 C.F.R. §§ 122.44, 123.25(a)(15). The EPA has transferred permitting authority to 47 States, including California. NPDES State Program Authority , EPA, https://www.epa.gov/npdes/npdes-state-program-authority (last visited July 28, 2021).
The EPA takes several measures to ensure that any discharge into public waters is safe and nontoxic. Its regulations entirely ban permitholders from discharging certain pollutants and severely limit discharging others. See, e.g. , 40 C.F.R. §§ 129.100 – 129.105. And the regulations require States to establish similar limitations on the amounts of specific pollutants that permitholders can discharge. Id. § 131.11. But even if a discharge complies with the limits on individual pollutants, it might still be toxic because it contains a combination of pollutants, or because it contains substances that federal or state regulators have not yet found to be toxic. To address those possibilities, the EPA also requires certain permitholders to pass a test called a "whole effluent toxicity" (WET) test. Id. § 122.44(d)(1)(iv). A WET test measures the aggregate effect of a discharge on aquatic organisms such as minnows by exposing a test population of organisms to a discharge and counting how many die or become immobilized. See 60 Fed. Reg. 53,529, 53,532 (Oct. 16, 1995).
Because a WET test does not measure specific levels of pollutants but instead measures toxicity based on the response of aquatic organisms, the regulations must define what is considered toxic in a way that accounts for variations in how different populations of organisms may respond to identical samples. The 1995 regulations incorporated three manuals on WET testing—which in turn included several recommended statistical methods—and noted that any "changes" to the manuals "will be published in the Federal Register prior to their effective date for regulatory purposes." 60 Fed. Reg. at 53,532, 53,540. In 2002, the EPA updated the manuals but declined to "include[ ] ... alternative statistical methods"; it noted, however, that the recommended statistical methods "are not the only possible methods of statistical analysis." 67 Fed. Reg. 69,952, 69,964 (Nov. 19, 2002).
The initial WET test regulations aimed to limit false positive results—results that incorrectly state that a sample is toxic—to no more than 5 percent. See 67 Fed. Reg. at 69,968. In June 2010, the EPA issued the guidance at issue here, explaining how to use a new statistical method called the Test of Significant Toxicity (TST). Among other things, the TST aims to limit false negative results—results that incorrectly state that a sample is nontoxic—by adopting a null hypothesis that a sample is toxic. In other words, the TST presumes that a sample is toxic absent statistically significant evidence to the contrary. The EPA explained that it believed adopting that null hypothesis increases the statistical power of the TST—the likelihood that it will correctly classify samples as toxic or nontoxic—compared to the methods authorized by the 1995 and 2002 regulations, which did not control for false negatives. The EPA has amended the relevant regulations governing WET tests several times since issuing the 2010 guidance, but it has never promulgated the TST as a formal rule. See 77 Fed. Reg. 29,758 (May 18, 2012) ; 80 Fed. Reg. 8,956 (Feb. 19, 2015) ; 82 Fed. Reg. 40,836 (Aug. 28, 2017).
Plaintiffs are trade associations whose members are California municipal agencies that operate wastewater treatment plants. In 2014, plaintiffs brought an action in the Eastern District of California to challenge the EPA's use of the TST. Plaintiffs alleged that the agency violated the APA and the Clean Water Act when it approved California's application to use the TST as an "alternative test procedure" for permits under 33 U.S.C. § 1314(h) and 40 C.F.R. §§ 136.3(a), 136.5. After the complaint was filed, the EPA withdrew its approval of California's alternative test procedure, and the district court dismissed the case as moot.
After unsuccessfully seeking reconsideration of the dismissal, plaintiffs sought to reopen the case to amend their complaint. Although most of the original complaint had focused on the alternative test procedure, some allegations related directly to the EPA's use of the TST and its issuance of the 2010 guidance, and plaintiffs sought to expand on those allegations in the amended complaint. In October 2016, the district court denied the motion, concluding that "[i]t makes no sense ... to clumsily tack such a new claim to [plaintiffs’] original [alternative-test-procedure] challenge via a motion for reconsideration of a prior motion for reconsideration."
In December 2016, plaintiffs brought the action that is now before us. Plaintiffs alleged that the EPA had violated the APA by issuing the TST guidance without following notice-and-comment rulemaking procedures, and that the EPA had violated its own regulations by requiring and using the TST in discharge permits. The district court dismissed the complaint, in relevant part, on the ground that it was barred by the APA's six-year statute of limitations. See 28 U.S.C. § 2401(a). The court reasoned that plaintiffs "fundamentally take procedural issue with the EPA's failure to formally promulgate the 2010 TST Guidance pursuant to notice-and-comment requirements," so the limitations period expired in June 2016, six years after the guidance was adopted. The court stated that because it had determined that plaintiffs’ challenge was untimely, it did not need to "address whether the 2010 TST Guidance ... constitutes a final agency action."
Plaintiffs amended their complaint to allege that the EPA's actions were ultra vires and in violation of the Clean Water Act. The district court determined that "[a]dding this label ... does nothing to change the substance of [p]laintiffs’ allegations," so it again dismissed the complaint, this time with prejudice, in a three-page order that "incorporated" its prior order "in its entirety." Plaintiffs timely appealed from that order. As a threshold matter, the EPA suggests that we should ignore plaintiffs’ challenges to the district court's first dismissal order because plaintiffs named only the second dismissal order in their notice of appeal. The first dismissal order was not an appealable final judgment because the district court had allowed leave to amend; only the second order was a final judgment. See Disabled Rts. Action Comm. v. Las Vegas Events, Inc. , 375 F.3d 861, 870 (9th Cir. 2004). But we have held that "[a]n appeal from a final judgment draws in question all earlier, non-final orders and rulings which produced the judgment." Litchfield v. Spielberg , 736 F.2d 1352, 1355 (9th Cir. 1984). And here, the district court expressly stated that its second dismissal order "incorporated" the first order "in its entirety." We may therefore consider plaintiffs’ arguments relating to both orders.
Although plaintiffs advance a variety of different legal theories, all of them challenge what plaintiffs describe as the EPA's "requirement, use, allowance, and promotion" of the 2010 guidance, which "created and recommended use of statistical and other toxicity testing procedures." That guidance, plaintiffs assert, "is ultra vires and exceeds [the EPA's] statutory authority because the guidance document was not promulgated ... as a formal rule under the APA." As we have explained, the district court determined that plaintiffs’ challenge was untimely. We review the district court's dismissal de novo and "may affirm on any ground supported by the record." Karasek v. Regents of Univ. of Cal. , 956 F.3d 1093, 1104 (9th Cir. 2020). We find it unnecessary to consider the timeliness of the complaint because we affirm the dismissal on the alternative ground that the 2010 guidance was not final agency action.
The APA authorizes district courts to review only "final agency action." 5 U.S.C. § 704. Here, the EPA acknowledges that its guidance was "agency action," a concept that "cover[s] comprehensively every manner in which an agency may exercise its power." San Francisco Herring Ass'n v. Department of the Interior , 946 F.3d 564, 575–76 (9th Cir. 2019) (quoting Whitman v. American Trucking Ass'ns , 531 U.S. 457, 478, 121 S.Ct. 903, 149 L.Ed.2d 1 (2001) ). This case therefore turns on whether the guidance was "final."
In Bennett v. Spear , 520 U.S. 154, 117 S.Ct. 1154, 137 L.Ed.2d 281 (1997), the Supreme Court set out two requirements that must be satisfied for agency action to be deemed final: "First, the action must mark the ‘consummation’ of the agency's decisionmaking process—it must not be of a merely tentative or interlocutory nature. And second, the action must be one by which ‘rights or obligations have been determined,’ or from which ‘legal consequences will flow.’ " Id. at 177–78, 117 S.Ct. 1154 (first quoting Chicago & S. Air Lines, Inc. v. Waterman S.S. Corp. , 333 U.S. 103, 113, 68 S.Ct. 431, 92 L.Ed. 568 (1948) ; and then quoting Port of Bos. Marine Terminal Ass'n v. Rederiaktiebolaget Transatlantic , 400 U.S. 62, 71, 91 S.Ct. 203, 27 L.Ed.2d 203 (1970) ). The EPA concedes that the 2010 guidance meets the first requirement. But the EPA argues that the guidance does not meet the second requirement because the guidance "imposed no rights, obligations, or legal consequences."
As the District of Columbia Circuit has explained, courts must "make Bennett prong-two determinations based on the concrete consequences an agency action has or does not have." California Cmtys. Against Toxics v. EPA , 934 F.3d 627, 637 (D.C. Cir. 2019) ; accord Whitewater Draw Nat. Res. Conservation Dist. v. Mayorkas , No. 20-55777, 5 F.4th 997, 1009–10 (9th Cir. July 19, 2021) ; Gill v. United States Dep't of Justice , 913 F.3d 1179, 1185 (9th Cir. 2019). For example, in United States Army Corps of Engineers v. Hawkes Co. , 578 U.S. 590, 136 S. Ct. 1807, 195 L.Ed.2d 77 (2016), the Supreme Court determined that the Corps of Engineers’ decision about whether a property contained "waters of the United States" was final agency action because it fixed "the Government's position" in subsequent litigation and could "limit[ ] the potential liability a landowner faces for discharging pollutants." Id . at 1814. By contrast, in California Communities Against Toxics , the court determined that an EPA memo was not final agency action because it merely "advise[d] EPA employees of the agency's position" and did not "bind state permitting authorities or assure regulated entities" of any rights. 934 F.3d at 639.
Plaintiffs attempt to demonstrate the requisite concrete consequences through several steps. The Clean Water Act requires the EPA to "promulgate guidelines establishing test procedures for the analysis of pollutants," 33 U.S.C. § 1314(h), and also to "publish" rules relating to water quality, id . § 1314(a)(2)(C), (a)(8). Consistent with that mandate, the EPA's 2002 rule incorporated into published regulations three WET test manuals, which had "selected" and "recommended" certain statistical methods. 67 Fed. Reg. at 69,964 ; see also 40 C.F.R. § 136.3(a) (Table IA). But the EPA's 2010 guidance allowed permitting authorities to use the TST as "another statistical option to analyze valid WET test data for ... permit compliance determinations," even though the agency did not publish the TST as a rule or add it to the methods listed in 40 C.F.R. § 136.3. As a result, plaintiffs contend, the 2010 guidance changed the legal regime by allowing permitting authorities to use the TST.
The EPA disagrees with plaintiffs’ construction of the relevant regulations, relying on language in the 2002 rule that describes the selected methods for interpreting WET test data as "not the only appropriate techniques." 67 Fed. Reg. at 69,964. We find it unnecessary to resolve that dispute because even under plaintiffs’ interpretation, the 2010 guidance is not final agency action.
Even if the 2010 guidance represents a departure from the view reflected in the earlier regulations, it creates no concrete consequences on its own. To be sure, under plaintiffs’ theory, the 2010 guidance suggests that permitting authorities have a new testing option. But it is permits, not guidance documents, that create consequences for regulated entities like plaintiffs. Plaintiffs point out that permit holders may be subject to criminal penalties or civil enforcement actions for failing the TST if a state or federal permit requires it. See 33 U.S.C. § 1319. But the "if" is key. The statute authorizes civil enforcement actions and criminal penalties for violations of "permit conditions." Id. § 1319(a) – (c). In other words, permit holders are subject to concrete consequences only if a state or federal permit incorporates the TST. We have previously recognized that an agency action is not final when subsequent agency decision making is necessary to create any practical consequences. See City of San Diego v. Whitman , 242 F.3d 1097, 1102 (9th Cir. 2001). That principle is controlling here.
Significantly, the guidance document itself disclaims "any legally binding requirements on EPA, states, ... permittees, or laboratories conducting or using WET testing for permittees." Plaintiffs correctly point out that such boilerplate disclaimers are not necessarily controlling. See Appalachian Power Co. v. EPA , 208 F.3d 1015, 1022–23 (D.C. Cir. 2000) ; Regents of the Univ. of Cal. v. Department of Homeland Sec. , 908 F.3d 476, 516 (9th Cir. 2018), rev'd in part on other grounds , ––– U.S. ––––, 140 S. Ct. 1891, 207 L.Ed.2d 353 (2020). But the rest of the guidance confirms that it does not bind anyone to anything. To the contrary, it explains that the "EPA developed the TST approach as another statistical option" to use for evaluating WET test data, and it does not "preclude the use of" the EPA's existing approved methodologies. It advises that "[p]ermitting authorities should consider the practical programmatic shift from the traditional hypothesis testing approach to the TST approach by opening a dialogue with their regulated community," adding that "they might want to begin to identify what changes might be needed to assimilate the TST approach." That is the language of suggestion—and mild suggestion at that—not command. Cf. Appalachian Power Co. , 208 F.3d at 1023.
In urging a contrary conclusion, plaintiffs rely primarily on Barrick Goldstrike Mines Inc. v. Browner , 215 F.3d 45 (D.C. Cir. 2000), but that case does not help them. In Barrick , the District of Columbia Circuit considered EPA guidance clarifying that waste rock was not subject to a de minimis exception applicable to other regulated activities, which meant that Barrick, whose business involved moving waste rock containing trace amounts of toxic substances, had to report those toxins to the agency. Id . at 47. The court held that the guidance was final agency action because if Barrick did not comply, it faced "enforcement action and fines." Id . at 47–48. Here, by contrast, plaintiffs do not explain how the 2010 guidance, standing alone, will cause them to face anything. Instead, "as a bare statement of the agency's opinion," the 2010 guidance "can be neither the subject of ‘immediate compliance’ nor of defiance." Fairbanks North Star Borough v. United States Army Corps of Eng'rs , 543 F.3d 586, 593–94 (9th Cir. 2008) (quoting FTC v. Standard Oil Co. , 449 U.S. 232, 239–40, 101 S.Ct. 488, 66 L.Ed.2d 416 (1980) ). Neither plaintiffs nor anyone else "can rely on it as independently authoritative in any proceeding," and there is "no penalty or liability of any sort in ignoring it." California Cmtys. Against Toxics , 934 F.3d at 638.
Plaintiffs argue that even if the 2010 guidance itself was not final, the EPA's later actions "crystallized" it into final agency action. Plaintiffs first point to a spreadsheet that the EPA circulated to state water regulators in May 2012. The agency described the spreadsheet as an "easy to use, inexpensive way" for States and EPA regional offices to "analyze and evaluate valid WET data." But the spreadsheet has the same finality problems as the 2010 guidance itself: It makes clear that the TST is an option, not a requirement. See Sierra Club v. EPA , 955 F.3d 56, 64 (D.C. Cir. 2020) (explaining that guidance is not final agency action when States "retain discretion to utilize the [guidance] or maintain the status quo in their individual permitting programs").
The same is true of two 2015 emails that the EPA sent to state permitting authorities. In one, the agency assured California regulators that the State was "still able to use" the TST despite the EPA's withdrawal of the alternative test procedure that plaintiffs’ first lawsuit had challenged. In the other, the EPA "strongly recommend[ed]" that California regulators add a detailed description of the TST to a state-issued permit. Setting aside any argument that the EPA's recommendation effectively required state permitting authorities to use the TST—a theory plaintiffs expressly disclaimed at oral argument—the emails reflect the same thing as the 2010 guidance: The EPA considers the TST one option for interpreting the WET test data necessary to obtain a discharge permit.
Of course, as the EPA acknowledges, permits themselves are final agency actions. But plaintiffs have disclaimed any challenge to specific permits in this litigation, and rightly so. Federally issued permits may not be challenged in an APA action in district court because they are subject to exclusive review in the court of appeals. See 33 U.S.C. § 1369(b)(1)(F). As for State-issued permits, we have held—in unrelated litigation brought by these same plaintiffs—that the statute "does not contemplate federal court review of state-issued permits" and that such permits are subject to review only in state court. Southern Cal. All. of Publicly Owned Treatment Works v. EPA , 853 F.3d 1076, 1086 (9th Cir. 2017) (quoting American Paper Inst. v. EPA , 890 F.2d 869, 875 (7th Cir. 1989) ).
Plaintiffs object that if they are unable to challenge the TST in district court, then their challenge cannot be heard in any other forum. That is incorrect. We have previously observed that "state courts can interpret federal law, and thus can review and enjoin state authorities from issuing permits that violate the requirements of the Clean Water Act." Boise Cascade Corp. v. EPA , 942 F.2d 1427, 1434 (9th Cir. 1991). Indeed, California courts have often interpreted the Act. See, e.g. , City of Burbank v. State Water Res. Control Bd. , 35 Cal.4th 613, 26 Cal.Rptr.3d 304, 108 P.3d 862, 869–70 (2005). Plaintiffs’ challenge to the EPA's decision to allow use of the TST in individual permits is appropriately adjudicated in the context of individual permit decisions. Cf. Lujan v. National Wildlife Fed'n , 497 U.S. 871, 894, 110 S.Ct. 3177, 111 L.Ed.2d 695 (1990).
AFFIRMED .