The default rule for judicial review of agency action is that review is available for final agency action. Courts typically enforce this requirement by citing an explicit finality requirement in the text of a review statute. When a review statute does not mention finality, courts apply a presumption in favor of requiring final agency action. The Clean Water Act’s (CWA) direct-appellate-review provision does not contain an explicit finality requirement. This Comment analyzes whether the CWA’s direct review provision can overcome the presumption in favor of requiring final agency action. The Comment proposes a test to determine whether a statute can overcome the presumption, applies the test to the CWA's direct-review provision, and concludes that it can overcome the presumption. Finally, the Comment argues that the ripeness doctrine is sufficient to ensure administrative and judicial economy in the absence of the finality rule.


The default rule for judicial review of agency action is that review is available under the Administrative Procedure Act1 (APA) for agency actions that are both ripe and final. Review is categorically unavailable under the APA for a claim that fails to meet both the finality and ripeness requirements.2 Judicial review is also available when Congress has allowed for review of specific actions via direct-review statutes.3 Although courts must still address the ripeness issue, it is not always clear whether Congress intended these statutes to restrict direct review to final agency actions. This is because the language of the statutes varies. Some statutes explicitly require final agency action, while others are silent on a finality requirement.4 When facing this uncertainty, courts typically apply an interpretive presumption in favor of requiring final agency action for judicial review and, almost uniformly, rule that finality is required.5 The Clean Water Act6 (CWA) contains one such direct-appellate-review provision that is silent on the question of finality.

Congress passed the CWA in 1972 as an amendment to the Federal Water Pollution Control Act of 1948.7 Congress enacted the amendment to “restore and maintain the chemical, physical, and biological integrity of the Nation’s waters.”8 The CWA outlaws the “discharge of toxic pollutants”9 and then requires dischargers to secure a permit to pollute.10 Known as the National Permit Discharge Elimination System (NPDES), the system prohibits the discharge of all pollutants from pollution sources subject to the issuance of a permit by the appropriate issuing authority.11 The CWA tasks the Environmental Protection Agency (EPA) with overseeing the CWA’s implementation and allows either individual states or the federal government to manage the permit program within a state’s boundaries.12 Most states have opted to adopt and manage their own NPDES program.13 Pollution control is achieved through the permit system and the enforcement of strict federal technology-based standards, performance-based standards, and water quality standards.14

Section 509(b) of the CWA authorizes direct judicial review of seven categories of EPA actions but does not explicitly require that those actions be final actions to qualify for review.15 The circuits are split as to whether courts should read a finality requirement into the statute. This Comment analyzes whether the CWA’s direct-review provision can overcome the presumption in favor of requiring finality, concludes that it can overcome the presumption, and argues that the ripeness doctrine is sufficient to ensure administrative and judicial economy in the absence of a finality rule. While this Comment is limited to analyzing the CWA, the analysis may also apply to other direct-review statutes that do not explicitly require final agency action.

Part I of this Comment explains the relevant judicial review doctrines and statutes, as well as the presumption in favor of requiring finality. Part II details the case law on whether § 509(b) requires final agency action for direct review and suggests an appropriate framework to use in deciding whether a statute overcomes the presumption in favor of requiring finality. Part III applies the suggested test to § 509(b) and, finding that the presumption is overcome, concludes that the statute should be read to allow for review of nonfinal action. Additionally, Part III argues that the ripeness doctrine is sufficient to address concerns about administrative and judicial economy and is welfare maximizing when not paired with a finality rule. Part III also argues that eliminating the requirement of finality will not lead to an expansion of protective lawsuits as parties seek to avoid the effects of § 509(b)’s preclusion provision.

I.  Setting the Table for Judicial Review

This Part explains the requirements for judicial review of agency action and the CWA’s direct-review provision. Part I.A explains the relevant judicial review doctrines of finality and ripeness. Part I.B describes the presumption in favor of requiring final agency action for judicial review and demonstrates how courts have applied the presumption across a range of statutes. Moving to the statutory vehicles for judicial review, Part I.C details the APA’s and CWA’s judicial review provisions.

A.    The Finality and Ripeness Doctrines: A Rules-versus-Standards Redux

This Section explains the judicial review doctrines of finality and ripeness by framing the doctrines within the classic rules-versus-standards framework.16 At the most abstract level, the two doctrines serve judicial economy and seek to balance this efficiency interest with the concern that an agency action might “unduly burden private parties.”17 While similar enough that courts often blend the doctrines, they are analytically distinct.18 Separating the two doctrines is vital to gaining an understanding of each rule and its potential applications.

1.   Finality.

Finality is best thought of as a hard-and-fast rule of judicial review. When the underlying review statute requires finality, a court must look to whether the challenged action is a final action. Either an action is final and review is appropriate, or it is nonfinal and therefore not subject to review.19 In Bennett v Spear,20 the Supreme Court announced a two-part test for identifying a “final agency action.”21 First, the challenged action must “mark the ‘consummation’ of the agency’s decisionmaking process—it must not be of a merely tentative or interlocutory nature.”22 Nevertheless, the “mere possibility that an agency might reconsider” its action “does not suffice to make an otherwise final agency action nonfinal.”23 Second, “the action must be one by which ‘rights or obligations have been determined,’ or from which ‘legal consequences will flow.’”24 If an action satisfies both prongs, it constitutes final agency action.25 While the finality rule has become more standard-like in recent years, it maintains the primary characteristics of a rule.26 Many (if not most) administrative-review statutes allow for review only of a final agency action.27 As will be discussed in Part I.B, the Supreme Court has recognized the “strong presumption [ ] that judicial review will be available only when agency action becomes final,” even for review statutes that do not explicitly require final agency action.28 Consequently, many administrative-review cases hinge on the question whether an agency action is final.29

2.   Ripeness.

If finality is the epitome of a judicial review rule, ripeness is its standard-like cousin. Ripeness is a constitutional and prudential doctrine. It ensures that courts hear actual cases or controversies as required by Article III and allows courts to refuse to exercise jurisdiction if doing so would be imprudent.30 The doctrine is designed to prevent premature adjudication and court entanglement in abstract disagreements over policy and to ensure that potential plaintiffs have felt the effects of an administrative decision before challenging it.31 The doctrine avoids expending judicial resources on unrealized problems and precludes judicial intervention into the policy domains of the politically accountable branches until the “government’s position has crystallized to the point at which a court can identify a relatively discrete dispute.”32

The Supreme Court established the modern ripeness test in Abbott Laboratories v Gardner.33 The Court framed the ripeness inquiry as determining “the fitness of the issues for judicial decision” and “the hardship to the parties of withholding court consideration.”34 If the issue is fit for review and the parties would suffer hardship were court consideration withheld, the case is ripe for review. In Ohio Forestry Association, Inc v Sierra Club,35 Justice Stephen Breyer further clarified the fitness analysis by framing the question as examining “whether judicial intervention would inappropriately interfere with further administrative action” and “whether the courts would benefit from further factual development of the issues presented.”36 Unlike the finality rule, the ripeness standard requires the judge to weigh different factors to determine whether the claim is ready for adjudication.37 The Court has described the ripeness inquiry as involving “the exercise of judgment, rather than the application of a black-letter rule.”38

B.    Bell v New Jersey: The Presumption in Favor of Requiring Finality

The Supreme Court has recognized a “strong presumption [ ] that judicial review will be available only when agency action becomes final,” even if the direct-review statute does not mention finality.39 The presumption has never been overcome. The origin of the presumption in favor of finality is the 1938 Supreme Court case Federal Power Commission v Metropolitan Edison Co.40 There, the Federal Power Commission (FPC) issued an order launching an investigation into the Metropolitan Edison Company and other power companies.41 The corporations brought suit under the relevant direct-review statute, which provided for review of an order after an application for rehearing occurred.42 The Court rejected the companies’ argument that the direct-review provision allowed for review of a procedural order. In the Court’s view, the language of the direct-review provision should not “be construed as authorizing a review of every order that the Commission may make, albeit of a merely procedural character.”43 The Court opined that such a construction would allow for constant delays and do violence to the purpose of the provision.44 Citing prior precedent in which the Court had rejected jurisdiction of appeals of “mere preliminary or procedural orders” sought on the basis of “a supposed or threatened injury . . . before the prescribed administrative remedy has been exhausted,” the Court held that the appellate court had lacked jurisdiction over the appeal of the FPC’s decision to order a hearing.45

In Bell v New Jersey,46 the Court distilled Metropolitan Edison’s language into a presumption in favor of requiring finality.47 There, the question was whether final agency action was required for appellate jurisdiction when the direct-review statute was silent as to any finality requirement and provided for review of “any action.”48 The Court held that jurisdiction was contingent on the presence of a final agency action.49 Writing for the Court, Justice Sandra Day O’Connor cited Metropolitan Edison and established the “strong presumption [ ] that judicial review will be available only when agency action becomes final.”50 Examining the relevant statutes, the Court held that there was no indication in the provision’s text to “overcome that presumption.”51 Moreover, the relevant statutes all mentioned finality and strongly suggested that “only a ‘decision’ of the [agency] is subject to review.”52 The absence of a reference to final orders was not enough to overcome the presumption in favor of requiring finality as, in O’Connor’s view, the statutory scheme required final agency action.53

Lower courts have taken a similar approach and declined to find that silence is sufficient to overcome the presumption in favor of requiring finality. Most often, the reviewing court will cite Bell’s presumption and then look to some combination of the legislative history, statutory scheme, and purpose of the statute. For example, in Columbia Riverkeeper v United States Coast Guard,54 which involved the Coast Guard’s recommendation to the Federal Energy Regulatory Commission (FERC) regarding a proposed facility, the Ninth Circuit considered a statute that allowed for review of “an order or action . . . to issue, condition, or deny any permit, license, concurrence, or approval.”55 The court held that the statute did not overcome Bell’s presumption for several reasons. First, the “order” or “action” language was similar to the “any action” language in Bell and “order” language in Metropolitan Edison.56 Second, decisions that “issue, condition, or deny any permit, license, concurrence, or approval” are final decisions anyway.57 Third, courts have a long history of interpreting “an order” to mean “final orders” in FERC-related contexts.58 Finally, allowing review of every interim action would be contrary to the statute’s purpose of expediting siting decisions for power projects.59 The statute’s silence on finality was not sufficient to overcome Bell’s presumption.

The DC Circuit tread the same path to find that a review statute required final agency action in Meredith v Federal Mine Safety and Health Review Commission.60 The statute dictated that “any order issued by the Commission . . . shall be subject to judicial review” under 30 USC § 816(a)(1), which itself provides for review of “an order of the Commission.”61 Employees of the Mine Safety and Health Administration (MSHA), who were seeking review of an internal MSHA policy, argued that Bell’s presumption was overcome because the statute referred to “final order[s]” in other sections but used the term “order” in the section relevant to the case.62

The DC Circuit disagreed.63 Admitting that a direct expression from Congress would control, the court found no reason to deviate from Bell’s presumption requiring final agency action.64 The court cited the statute’s legislative history, which described the judicial review section as “providing for the review of final orders; no mention is made of earlier review.”65 Absent clear evidence to the contrary, and with legislative history strongly suggesting that only final orders were subject to review, the court held that Bell’s presumption controlled.66

A third example of a court refusing to find that a statute overcomes Bell’s presumption is Carolina Power and Light Co v United States Department of Labor.67 Following the pattern of Columbia Riverkeeper and Meredith, the Fourth Circuit rejected the argument that, because a statute used the term “order” rather than “final order,” Bell’s presumption was defeated.68 The judicial review statute at issue in Carolina Power provided for review of “an order” by the secretary of labor granting or denying relief to the parties involved in a whistle-blowing suit at the Nuclear Regulatory Commission.69 The court held that despite the statute’s reference to an “‘order’ rather than ‘final order,’ [the] omission alone is insufficient to overcome” Bell’s presumption.70 Looking to the statutory scheme, the court stated that any action taken by the secretary of labor would be inherently final anyway.71 Therefore, the employee’s challenge to the secretary’s decision to remand the case to an administrative-law judge was not reviewable under the statute.72

On the one hand, the cases described above demonstrate that courts applying Bell’s presumption have consistently held that neither the omission of the word “final” nor the inclusion of the word “any” is enough to overcome the presumption. On the other hand, courts are inconsistent in that they will look to different sources of evidence of congressional intent on an ad hoc basis. No court describes a default standard or test to use when deciding whether Bell’s presumption is overcome. Courts will look to the statutory scheme, the legislative history, and the relevant act’s purpose to determine whether the presumption is overcome. If Congress has not indicated an intent to overcome the presumption, or if the evidence supports either outcome, courts have held that the presumption controls.

C.    Judicial Review Provisions: The APA and the CWA

This Section discusses the judicial review provisions of the APA and the CWA. It demonstrates how the APA codifies the finality doctrine for most administrative-review actions but explicitly allows for exceptions to the finality principle. It then discusses the CWA’s direct-review provision, § 509(b)(1), explaining which actions are directly reviewable and, importantly, the provision’s review-preclusion rule.

1.   The APA: 5 USC § 704.

The APA established operating procedures for agencies and the default framework for judicial review.73 Drawing on judicially established rules for administrative review,74 the APA codifies the finality doctrine for certain agency actions. The APA’s judicial review provision authorizes review of “[a]gency action made reviewable by statute and final agency action for which there is no other adequate remedy in a court.”75 It also states that a “preliminary, procedural, or intermediate agency action or ruling not directly reviewable is subject to review on the review of the final agency action.”76

The text of the statute thus contemplates two forms of administrative review. The first form is review of final agency actions. The second form is direct review of nonfinal actions that have been made reviewable by another statute. The “final” modifier in § 704 applies to agency action with no other remedy in court, not “agency action made reviewable by statute.”77 That is, a party may bring a claim under the APA to challenge a nonfinal action directly reviewable under another statute. Recent Supreme Court decisions agree with this reading.78

Review of nonfinal actions thus comports with the framework of the APA. Despite codifying the finality requirement, the APA contemplated that some “preliminary, procedural, or intermediate” actions could be made reviewable by statute. The Supreme Court and the circuit courts have acknowledged this distinction.79 The challenge, then, is that courts have been unwilling to recognize that some statutes authorize review for nonfinal actions. While courts are willing to pay lip service to the idea that such a review statute exists, no court has construed a statute to that effect.

2.   The CWA: § 509(b).

Section 509(b) of the CWA provides a private right of action for direct appellate review of nonfinal EPA actions. Based on similar provisions in the Clean Air Act80 (CAA) amendments of 1970,81 the CWA creates a bifurcated system of direct judicial review of actions taken by the EPA administrator.82 For some actions, the CWA grants original jurisdiction in the district courts through § 505.83 For other actions, the courts of appeals have original jurisdiction under § 509(b)(1).84 Section 509(b)(1) also contains a review-preclusion provision requiring litigants to file suit within 120 days of the administrator’s action.85 If a party fails to file within the time limit, review is barred unless new reasons for review arise after the 120th day.86 The statute also prohibits litigants from challenging the promulgation of a regulation and other listed actions in an enforcement proceeding.87 For example, if the EPA enforced a regulation against a farmer, the preclusion provision prevents the farmer from challenging the promulgation of the underlying regulation.

Deciding which court to file suit in is often a complex and messy affair.88 If a party decides to file suit under § 505, it risks being subject to § 509(b)’s preclusion effect. This would occur when a party files a § 505 suit, but the court decides that the litigant should have brought suit in the court of appeals via § 509(b). If 120 days have passed, the litigant will be unable to bring its challenge in the proper forum.89 This quandary has resulted in CWA litigants filing suit in both the district court and the court of appeals to protect themselves from § 509(b) preclusion. Section 509(b) also contains a claim-preclusion provision requiring litigants to file suit within 120 days of the administrator’s action.90 If a party fails to file within the time limit, review is barred unless reasons for review arise after the 120th day.91 The statute also prohibits a party from seeking judicial review of the administrator’s actions in a later enforcement proceeding if that action was reviewable under § 509(b).92

Section 509(b) allows for review of the “Administrator’s action in” (1) promulgating rules, (2) making determinations, and (3) approving, issuing, or denying rules or permits.93 Unlike the CAA’s almost identical direct-review provision, which explicitly mentions a finality requirement, § 509(b) does not mention final agency action.94 The Supreme Court has never ruled on whether § 509(b) requires final agency action for judicial review, and the question has split the circuits. This Comment seeks to resolve the split by offering a framework to guide courts in deciding whether a finality requirement exists. Employing that framework, this Comment argues that § 509(b) allows for review of nonfinal actions.

II.  Finality in § 509

This Part describes the circuit split over whether § 509(b) contains a finality requirement. Part II.A discusses holdings from six circuits that have directly ruled on whether § 509(b) requires final agency action, as well as the Eighth Circuit’s ruling that no such requirement exists. Part II.B recognizes that the conflicting opinions are operating without a basic test to judge whether § 509(b) can overcome Bell’s presumption. To fill this need, Part II.B proposes a test that courts should use when analyzing whether a statute can rebut the presumption in favor of requiring finality.

A.    The Circuit Split

This Section describes the circuit split over whether § 509(b) contains a finality requirement. First, this Section discusses the holdings of various circuits that have interpreted § 509(b) to require final agency action for review. The Fifth and the Eleventh Circuits required finality based on the text of the statute. The Third and Seventh Circuits based their decision to require final action on judicial-economy concerns. The Fourth Circuit looked to legislative history, and the First Circuit relied solely on the Bell presumption. The First, Third, Fourth, Fifth, Seventh, and Eleventh Circuits have all found that § 509(b) should be read to include a finality requirement.95 This Section also discusses the Eighth Circuit’s recent holding that § 509(b) contains no finality requirement.

1.   Finality is required.

In National Pork Producers Council v United States Environmental Protection Agency,96 the Fifth Circuit held that § 509(b) contains a finality requirement.97 There, groups representing agricultural interests opposed CWA feed lot regulations by challenging guidance letters that the EPA sent to members of Congress and a farm executive.98 The court ultimately held that it did not have jurisdiction to consider the challenge to the EPA letters.99 It based its conclusion on the belief that the text of § 509(b) included an explicit finality requirement.100 This was a mistake; § 509(b) does not contain the words “final action.”101

In Riverkeeper v United States Environmental Protection Agency,102 the Eleventh Circuit stated that § 509(b) “is not the type of provision that overcomes the strong presumption that judicial review is available only when there is final agency action.”103 Though the court admitted that the question “is a close case,” it ultimately held that § 509(b) requires final agency action for judicial review.104 Environmental groups had brought suit seeking to have the EPA withdraw Alabama’s authorization under the CWA to administer its own NPDES permit.105 After considering the petition and the response from the Alabama Department of Environmental Management, the EPA issued an “interim response,”106 stating that twenty-two of the twenty-six alleged deficiencies “did not warrant [the] initiation of program withdrawal proceedings.”107 The EPA expressed “significant concerns” about the remaining four alleged deficiencies, but decided to allow Alabama to address the EPA’s concerns before determining whether to initiate program withdrawal.108 The nonprofits then appealed the EPA’s findings on the twenty-two alleged deficiencies.109 Claiming that § 509(b) allows for judicial review of “any determination as to a State permit program,” the nonprofits argued that they could bring suit without a “final agency action.”110

Citing Bell’s presumption, the Eleventh Circuit held that final agency action was required to establish subject-matter jurisdiction.111 Turning its attention to the statute’s text, the court identified “any determination” as the provision’s “critical words.”112 Citing two dictionaries, the court explained that the word “determination” suggests a “final decision made at the end of a deliberative process.”113 Finding that that “‘any’ [does not] take[ ] away from ‘determination’ the general notion of a conclusive and irreversible decision,” the court concluded that § 509(b) did not overcome the Bell presumption.114

The Third Circuit held that § 509(b) requires final agency action in Pennsylvania Department of Environmental Resources v Environmental Protection Agency.115 In that case, petitioners sought review under § 509(b)(1)(A) to challenge regulations regarding standards of performance for new sources in the coal mining industry.116 Petitioners alleged that the EPA lacked authority to defer promulgating new source performance standards for closed or abandoned mines.117

The court readily admitted that “[u]nlike many other statutes providing for judicial review of agency action in the court of appeals, section 509 is not in terms limited to final agency action.”118 Confessing that “the absence of any finality language in section 509 points to the availability of [nonfinal] review,” the court reasoned that “it does so only slightly.”119 By “stressing the word ‘promulgation’ rather than the word ‘action,’” the court found “the equivalent of a finality requirement.”120 Anticipating Judge Frank Easterbrook’s finality rationale in American Paper Institute, Inc v United States Environmental Protection Agency,121 the court explained that because “unlimited interlocutory review could seriously impede the performance of the EPA’s rulemaking functions, there are strong policy reasons for that reading.”122

A similar defense of the finality requirement comes from a Seventh Circuit opinion authored by Easterbrook. In American Paper, an industry group challenged a policy statement supplement concerning the tolerances for dioxin in new and renewed permits of paper mills using chlorine bleaching.123 The EPA’s regional office in Chicago had published its own policy statement that was, in some respects, more onerous than the statement issued by the EPA’s central office.124 The court rejected the reviewability of the supplemental statement. It found that the policy statement was outside the scope of actions reviewable under § 509(b)(1) because it was created by the regional office rather than the administrator and was not “promulgated” because it lacked legal effect.125

As to the question of finality, the court stated that if the policy statement ever led to an actual denial or modification of a permit, the paper mill would be entitled to judicial review.126 Alternatively, if “the Administrator adopts [the regional office’s] position and a permit is turned down, modified, or rescinded, review will be available in state or federal court. That review, on a full record, will disclose the EPA’s final position, as applied to the plant in question.”127 As there was no “‘final’ agency action” to review, the court dismissed the petition for lack of jurisdiction.128

The court also emphasized the power of § 509(b)’s review-preclusion provision. Fearing that a procedural morass could develop if the court set forth an expansive reading of § 509(b), the court stated that “the more we pull within § 509(b)(1), the more arguments will be knocked out by inadvertence later on—and the more reason firms will have to petition for review of everything in sight.”129 Envisioning a nightmare scenario, the court explained that if it asserted jurisdiction in this case, it “might need to hear challenges to every intra-office memo in which the associate deputy director for 4" pipes told the deputy associate director for nozzles what he planned to do about 4" nozzles, if some should be built.”130

The Fourth Circuit has weighed in on the finality question, as well. In Champion International Corp v United States Environmental Protection Agency,131 the court ruled that § 509(b) incorporates a requirement of final agency action.132 A paper mill had brought an action against the EPA regarding a NPDES permit decision.133 One of the questions in the case was “whether Congress has provided for judicial review of the objections made by the EPA to the . . . permit prior to final action by the EPA.”134 The court answered that question bluntly, stating “[w]e think it has not.”135

To support its holding, the court relied on legislative history from the 1977 CWA amendments. Senator Edmund Muskie had made a floor statement about the EPA’s authority to assume permitting power, stating that:

The Administrator’s action in objecting to a permit would generally not be subject to judicial review since it will always be followed by further administrative action. The final issuance of a permit by EPA would be subject to judicial review pursuant to section 509(b)(1)(F).136

Because Muskie was the manager of the conference bill, the court gave his statement “significant weight” and said that the legislative history was “clear with respect to judicial review of the EPA’s decision to assume issuing jurisdiction.”137 Having decided that § 509(b) requires a final agency action for judicial review, the court concluded that judicial review would be appropriate “after EPA either grants or denies a permit.”138

The First Circuit also addressed the finality question in Rhode Island v United States Environmental Protection Agency139 and found that Bell’s presumption applied.140 In that case, Rhode Island filed a petition for interlocutory review of an order of the US Environmental Appeals Board (EAB).141 Specifically, Rhode Island challenged a denial of its motion to intervene in a pollution-discharge permit proceeding.142 Rhode Island asserted jurisdiction under § 509(b)(1)(F).143 Construing the statute to apply only to an issuance or denial of a permit, the court rejected Rhode Island’s claim of jurisdiction for lack of finality because a permit had neither been issued nor denied.144 The court also supported its reasoning by citing Bell’s presumption.145

In sum, courts have employed a variety of methods to find an implied finality requirement in § 509(b). Many courts cite Bell’s strong presumption of finality.146 Some courts look to the specific action listed in § 509(b) to see if the action inherently requires finality.147 Other courts cite legislative history.148 Lastly, courts will justify reading in a finality requirement based on concerns of judicial and administrative economy and efficiency.149

2.   Finality is not required.

Only the Eighth Circuit has found that § 509(b) does not require final agency action.150 In Iowa League of Cities v Environmental Protection Agency,151 the court was asked to judge whether letters from the EPA to Senator Charles Grassley of Iowa violated the APA’s notice-and-comment requirements and exceeded the limits of the EPA’s authority under the CWA.152 Arguing that the letters “effectively set forth new regulatory requirements with respect to water treatment processes at municipally owned sewer systems,” the Iowa League of Cities sought review under § 509(b)(1)(E).153

The Eighth Circuit focused on whether the letters counted as “promulgations,” primarily looking to whether the letters were binding.154 It did not—as other courts have done when considering jurisdiction under § 509(b)(1)—undertake a finality analysis to determine subject-matter jurisdiction. Nor did it address Bell’s presumption. The court instead held that it had jurisdiction because the letters had a “binding effect on regulated entities” and could therefore be considered promulgations.155 In holding that it had jurisdiction, the court rejected the EPA’s argument that “no federal court has jurisdiction over this claim because these letters are not ‘final agency actions.’”156

In a footnote, the court explained its justification for refusing to consider whether the EPA’s action was final.157 While admitting that its analysis into whether the EPA’s letters were binding “evokes considerations of finality,” the court countered that it declined “to conjure up a finality requirement for ‘[a]gency actions made reviewable by statute.’”158 Noting that other Eighth Circuit panels cautioned that “a legislature says in a statute what it means and means in a statute what it says there,”159 the court concluded its analysis by stating “[t]he CWA expressly makes specified agency actions reviewable, and our task therefore is to determine whether the asserted agency action falls within the statutory terms.”160 Having established that the letters were promulgations for jurisdictional purposes, the court then undertook a ripeness analysis to see if the claim was ripe for review. The court held that it was.161

The Eighth Circuit is the only circuit that does not require finality for jurisdiction under § 509(b). The court took a quintessentially textualist approach. Because the text of the APA does not require finality for “[a]gency action made reviewable by statute” and § 509(b) does not require finality, the court found that finality was not a prerequisite to an assertion of jurisdiction. Instead, it focused on whether the action could be understood as a promulgation. Notably, it addressed neither Bell’s presumption of finality nor Easterbrook’s concerns about judicial economy from American Paper.

B.    Using Block v Community Nutrition Institute as a Test for Bell’s Presumption

Though courts have held that some statutes do not overcome Bell’s presumption, lower courts (including courts examining § 509) have not articulated a test to determine when a statute does overcome the presumption. Despite the absence of a specific test for the Bell presumption, the Supreme Court has defined a general set of criteria to judge whether a statute overcomes an interpretive presumption. This Section describes these criteria and proposes that courts use this multifactor test when considering whether to apply Bell’s presumption.

1.   The Block factors.

In Block v Community Nutrition Institute,162 decided just one year after Bell, the Court confronted the question whether individual consumers of milk products could obtain judicial review of actions taken by the secretary of agriculture. The issue arose because the relevant statute allowed only milk handlers (middlemen between consumers and farmers) to seek judicial review.163 Acknowledging that there was a presumption in favor of allowing judicial review, Justice O’Connor explained that all such presumptions of statutory interpretation may be overcome by (1) “specific language,” (2) “specific legislative history that is a reliable indicator of congressional intent,” (3) an inference from “contemporaneous judicial construction barring review and the congressional acquiescence in it,” or (4) “the collective import of legislative and judicial history behind a particular statute.164 Narrowing her focus to the presumption favoring judicial review of agency action, O’Connor wrote that “inferences of intent drawn from the statutory scheme as a whole” may also overcome the presumption.165 Looking to the legislative scheme of the statute, the Court held that Congress had not intended to grant consumers the ability to seek judicial review of the secretary’s actions.166 Later Court rulings affirmed the idea that inferences of intent from the statutory scheme may overcome “most presumptions” used in interpreting statutes.167

The Block Court also established the standard of evidence needed to overcome presumptions. Rejecting the lower court’s decision to apply the “clear and convincing evidence” standard requiring “unambiguous proof,” O’Connor wrote that the standard, at least in the context of judicial review preclusion, is met whenever congressional intent is “fairly discernible.”168 When there is “substantial doubt” about congressional intent, the presumption controls.169

In other applications of the Block test, the Court has refused to accept statutory silence, by itself, as sufficient evidence to overcome the presumption favoring judicial review.170 Nevertheless, the Court has found that a judicial review provision in one part of a statute combined with omission of such a provision in another part provides “persuasive evidence that Congress deliberately intended to foreclose further review” in the latter part of the statute.171 The Court has also found that, when a statute grants one class of employees the ability to seek judicial review but does not authorize another class of employees, the statute exhibits a “manifestation of a considered congressional judgment that [the latter class] should not have statutory entitlement to review for adverse action.”172 These cases indicate that statutory silence plus another factor can be sufficient to overcome the presumption in favor of judicial review.

2.   Applying the Block factors to Bell’s presumption.

As the Court has not given lower courts a clear standard against which to judge whether § 509(b) overcomes Bell’s presumption in favor of requiring finality, the most logical standard to apply is the Block standard. Like the “strong presumption” in favor of judicial review,173 Bell’s strong presumption is a “presumption[ ] used in interpreting statutes.”174 Further, both presumptions relate to the jurisdictional power of a court to review administrative action.175 Perhaps most indicative of Block’s relevance to Bell is that the Court issued Block’s standard for overcoming all presumptions one year after establishing Bell’s presumption. The test that applies to “all presumptions” must necessarily apply to the Bell presumption.176

Using the Block standard in the Bell context also makes sense because the presumption in favor of requiring finality is not a “clear statement rule[ ].”177 The Supreme Court has held that, unlike clear-statement rules, interpretive presumptions can be overcome without an explicit textual statement.178 In presumption cases, the Court looks to the intent of Congress. The Block factors set out how to conduct that analysis for presumption rebuttal purposes. While courts have not specifically applied the Block test to other presumptions, the test is consistent with how courts approach presumptions generally and is a useful guidepost for framing the rebuttal analysis.

Given the Court’s declaration that “most presumptions” can be “overcome by inferences of [congressional] intent”179 and that “all presumptions” may be “overcome by specific language or specific legislative history,”180 it is surprising that courts have not explicitly used the Block framework to rule on whether a statute can overcome Bell’s presumption. The reason may be that Block dealt with the threshold presumption in favor of judicial review of agency action. In Bell cases, the availability of judicial review is not in question; it is only the timing of that review that is in dispute. Still, that does not explain why criteria used to judge most, if not all, presumptions would not apply to the Bell presumption.181

Another possible explanation is that courts use the Bell presumption as a substitute for ripeness and not as a litmus test for congressional intent. Courts may assume that, if an action is not final, there is not an adequate record on which to base judicial review and that any review may disrupt further administrative processes.182 Framing an issue as one of finality, rather than as a ripeness question, may provide a common basis on which appellate judges can agree. Citing to the Bell presumption and cherry-picking factors that reinforce the presumption is likely an easier path than delving into the messy world of ripeness.183 That is, courts may be employing the Bell presumption as a bright-line rule that will always require finality rather than as a surmountable interpretive presumption.184 If this is the case, the Block factors are irrelevant because a court raising the Bell presumption will likely ignore whatever congressional intent the factors may reveal.

In any event, courts already usually apply at least one of the Block factors in Bell cases (even though they do not explicitly reference Block). As discussed above, courts confronting the Bell presumption have refused to find the presumption overcome by the omission of the word “final.”185 Courts have searched for another indicator of congressional intent. Furthermore, like the Block test, when doubt or weighty evidence exists on both sides, courts tend to hold that Bell’s presumption controls. Because the Court has explicitly held that Block’s presumption test applies to all interpretive presumptions186 and the case law surrounding Bell’s presumption involves similar considerations, it is appropriate to analyze whether § 509(b) overcomes Bell by using the Block test as the primary analytical framework.

III.  Relying on Ripeness

Part III.A applies the Block test to § 509(b), concluding that § 509(b) overcomes Bell’s presumption in favor of requiring final agency action. Part III.B.1 suggests that the ripeness doctrine can effectively filter out frivolous claims and that removing finality will not lead to administrative disruption. Then, using the cases described in Part II, Part III.B.2 explores the instances in which an appeal, had it not been barred by finality concerns, should have been found ripe for review and allowed to proceed. Finally, Part III.C addresses Judge Easterbrook’s concern that allowing review of nonfinal action would lead to chaos as litigants file protective lawsuits to avoid being subject to § 509(b)’s preclusion provision. Part III.C then argues that a flexible reading of the preclusion provision, consistent with Supreme Court interpretations of § 509(b), avoids Easterbrook’s nightmare scenario.

A.    Defeating Bell’s Presumption

Of the courts that have cited Bell to justify requiring final agency action for review under § 509(b), none has articulated a test to determine what would be sufficient to overcome Bell’s presumption. Other courts have neglected to grapple with Bell at all. As discussed in Part II.B, the Block standard is the appropriate test with which to judge whether a statute can overcome Bell’s presumption. This Section applies the Block framework to § 509(b) by examining the statute’s (1) text, (2) statutory scheme and structure, (3) legislative history and purpose, and (4) congressional acquiescence in the judiciary’s expansive reading of § 509(b). The application of the test demonstrates that § 509(b) overcomes Bell’s presumption.

1.   The text does not require final agency action.

Section 509(b) does not mention any final agency action requirement. As established in Part I.B, an omission of the word “final” is necessary but insufficient to overcome Bell’s presumption.187 Section 509(b) further differentiates itself from other direct-review statutes by not authorizing review for an “order” or “action.” The statute specifically allows for review of the administrator’s “action in” promulgating, approving, denying, or making any determination under other CWA statutes.188 Giving meaning to each word of the statute requires ascribing some effect to the word “in.”189 In this context, the most natural meaning to assign to “in” is that it extends the scope of the review authorization to include actions taken in the course of taking the final action. This interpretation authorizes review of the many different actions the administrator may take before making a final determination or other final action.190

Other statutes, such as the provision in Columbia Riverkeeper, authorize review for an “order or action . . . to issue, condition, or deny any permit, license, concurrence, or approval.”191 There, Congress focused on the order or action that, itself, would effectuate a result. In contrast, “action in” suggests Congress contemplated review of the many actions taken toward effectuating a result. This construction gives meaning to the final action192 and the several actions taken in making—or whatever verb the final action requires, like promulgating, issuing, approving, or denying—that determination.193

2.   The statutory scheme does not implicate final action.

The statutory scheme of § 509(b) also supports the inference that Congress intended to allow review of nonfinal actions. Unlike other statutes authorizing review of actions taken under ancillary statutes,194 which themselves mention finality or other inherently final actions, § 509(b) references statutes that contain a bevy of intermediate actions. For example, § 509(b)(1)(A) allows for review of the administrator’s action in promulgating a new standard of performance under § 306 of the CWA. To promulgate a new performance standard, the administrator must take several intermediate actions. First, the administrator must add a new category to its list of source categories.195 Then, the administrator must propose and publish regulations “establishing Federal standards of performance for new sources.”196 The administrator may also make distinctions among “classes, types, and sizes within categories of new sources.”197 These are three actions that the administrator can take “in promulgating any standard of performance” and are therefore subject to review under § 509(b)(1)(A). These actions are unlike final decisions of a board198 or entry into a settlement199 because they are actions taken in anticipation of a final promulgation and are not qualified by “final agency action” language.200

As discussed in Part III.A.4, the Supreme Court has held that the underlying regulations governing actions reviewable through § 509(b) are themselves subject to review under the statute.201 The range of actions that the administrator can take under § 509(b)’s ancillary statutes reinforces the argument that § 509(b)’s authorization of review of “actions in” making determinations or denying permits should be read literally and given meaning.

3.   The legislative purpose and history are open to immediate review.

Under Block and cases applying Bell, courts look to legislative history to see if congressional intent is fairly discernible.202 In the cases using legislative history to rebut the claim that a statute overcomes Bell’s presumption, the legislative history clearly discusses final actions.203 No such discussion exists in the legislative history of the CWA’s enactment. The legislative history for the CWA shows that the drafters did not intend to limit judicial review to final actions by the administrator. In fact, the enacting Congress envisioned a judicial review system that allowed the public to meaningfully participate in government decisionmaking through litigation.

The CWA drafters appreciated that citizens had increasingly “taken the initiative in having the forum for [environmental] decision-making be in the court room.”204 Decrying the fact that agencies had closed themselves off from public input, Congress was interested in “open[ing] wide the opportunities for the public to participate in a meaningful way in the decisions of government.”205 With that intent in mind, the CWA included a provision for “judicial review of the Administrator’s actions . . . in section 509(b).”206

The report from the House Committee on Public Works supports the proposition that § 509(b)(1) is to be construed liberally. The committee made clear its belief that:

[W]ith the number and complexity of administrative determinations that the legislation requires there is a need to establish a clear and orderly process for judicial review. Section 509 will ensure that administrative actions are reviewable, but that the review will not unduly impede enforcement [on account of the review-preclusion provision in § 509(b)(2) that prevents litigants from raising challenges to actions in an enforcement proceeding if they are reviewable under § 509(b)(1)].207

The committee’s reference to the “number” of administrative determinations does not implicate finality in the same way as a reference to a “final order” or “decision.”208 Further, Congress evidently thought that the anticipated increase in litigation was an adequate reason to consider establishing a special environmental court.209

A supplemental view included in the committee report suggests that administrative review of nonfinal actions was a palatable idea to at least one member of the enacting Congress. Supporting the bill, Representative John Terry of New York stated that “an administrative review procedure of Section 304 guidelines, prior to their promulgation, and EPA approval or disapproval of State permit programs is essential.”210 Actions before promulgation would constitute nonfinal actions, as they would not represent the “consummation” of the agency’s decisionmaking process. Though this is the recorded view of only one legislator and should not, absent other factors, control a court’s interpretation of the statute, Terry’s statement demonstrates that Congress was at least aware of the possibility of allowing review of nonfinal actions.

The legislative history of the 1972 amendments creating the modern CWA thus shows that Congress was interested in opening wide the opportunities for public participation and used the phrase “any determination” literally. Further, the history shows that at least one member of Congress actively encouraged review of nonfinal action. Nowhere in the legislative history of the 1972 amendments does Congress mention that review is limited to final actions.

In Champion, the Fourth Circuit erroneously used Senator Muskie’s 1977 CWA amendment statements on the availability of judicial review of EPA approval or disapproval of state permit programs211 as evidence that Congress intended review of only final actions.212 The Fourth Circuit held that statement to be significant evidence that Congress did not intend review of nonfinal actions under § 509(b).

Muskie’s statement does not settle the finality question. Unlike Terry’s statements approving of review of nonfinal actions, Muskie’s floor speech occurred after Congress passed the CWA and cannot illuminate the legislative intent of the CWA’s enacting Congress. This is especially true because Muskie’s statement was primarily about § 402(d), not § 509(b)(1).213 As the Supreme Court has held, “the views of a single legislator, even a bill’s sponsor, are not controlling.”214 Moreover “the views of a subsequent Congress,” much less an individual legislator, “form a hazardous basis for inferring the intent of an earlier one.”215 Even “when it would otherwise be useful, subsequent legislative history will rarely override a reasonable interpretation of a statute that can be gleaned from its language and legislative history prior to its enactment.”216 Muskie’s subsequent floor statement—only tangentially related to § 509(b)—should therefore carry little, if any, weight in interpreting § 509(b). The legislative history from the original enactment should control.

The legislative history of the 1972 CWA shows that Congress intended to allow review of nonfinal actions taken by the administrator. The sole excerpt of subsequent legislative history that contradicts Congress’s original intent is that of a single legislator concerning an amendment to a different statute half a decade later. Given Supreme Court precedent on subsequent legislative history, Muskie’s statement should be accorded no weight in construing § 509(b). The legislative history does not contain the type of language that courts use to support finding an implied finality requirement.

4.   Congress has acquiesced in the judiciary’s expansive reading of § 509(b).

Block instructs interpreters seeking guidance as to whether a statute overcomes a presumption to look to “contemporaneous judicial construction” and “congressional acquiescence in it.”217 History provides the guideposts for this analysis. Informative clues are (1) Congress’s 1977 amendments to the CAA and (2) Supreme Court cases defining the reach of § 509(b). This Section proceeds in chronological order: it first discusses the 1977 case, E.I. du Pont de Nemours & Co v Train,218 then turns to the history of the 1977 CAA and CWA amendments, and finally returns to § 509(b) in the 1980 case, Crown Simpson Pulp Co v Costle.219 As this Section will demonstrate, the Court has taken an expansive view of § 509(b)’s reach. To date, Congress has not intervened and can reasonably be said to have acquiesced in an expansive reading of § 509(b).

In E.I. du Pont, the Court addressed the burgeoning question of whether the EPA was responsible for setting NPDES permit standards or if it could offer only guidelines and allow variances on individual permits.220 The EPA argued that it had the authority to set national standards. The Court agreed.221 In ruling for the EPA, the Court rejected the petitioners’ argument that § 509(b)(1)(E)’s authorization of review of approvals or promulgations of standards under § 301 applied only to § 301(c).222 The Court held that review was available for all actions taken pursuant to § 301 because § 509(b) did not explicitly single out § 301(c).223 Furthermore, the Court stated that it would be perverse to allow the court of appeals to review individual actions issuing or denying permits under § 402 (permits that allow discharges under the NPDES program) but leave it “no power of direct review of the basic regulations governing those individual actions.”224 The Court thus construed the statute to authorize review of all the actions listed in § 509(b)’s ancillary statutes unless otherwise specified. In other words, the Court interpreted the statute to allow review of the administrator’s action in approving or promulgating any effluent limitation under § 301 as long as that action was itself listed in § 301.

Congress amended the CWA in 1977 and did not revise § 509(b).225 Keeping in mind the Court’s longstanding presumption that Congress is “aware of an administrative or judicial interpretation of a statute and [ ] adopt[s] that interpretation when it re-enacts a statute without change,”226 Congress signaled its acquiescence in the Court’s expansive reading of § 509(b)(1)(E) in E.I. du Pont by not amending § 509(b) in the 1977 CWA amendments. Four months earlier, Congress modified the CAA’s similar review provision by adding the phrases “or final action taken” and “any other final action.”227 Despite having inserted “final action” language into a parallel review provision just a few months earlier, Congress did not make the same revision to § 509(b).

Congress’s decision to modify the CAA to explicitly reference final action, but to leave § 509(b) untouched, was a meaningful choice made in light of the Supreme Court’s ruling in E.I. du Pont. In Gwaltney of Smithfield, Ltd v Chesapeake Bay Foundation, Inc,228 the Court emphasized the significance of differences in the language of various environmental review statutes. There, the Court ruled on a jurisdictional question regarding the CWA’s citizen-suit provision229 by looking to the statute’s “most natural reading.”230 The Court ruled that the statute’s failure to provide jurisdiction for past violations was not a “careless” error on the part of Congress because “Congress used identical language in the citizen suit provisions of several other environmental statutes that authorize only prospective relief” and that “Congress has demonstrated in yet other statutory provisions that it knows how to avoid this prospective implication by using language that explicitly targets wholly past violations.”231 The same reasoning applies here. Congress inserted an explicit finality requirement in the CAA and could have done the same with respect to the CWA. The inclusion of finality language in other review statutes demonstrates that Congress knows how to make its intent to restrict review to final agency action clear. Congress did not do so here, and that choice must be given meaning.

Three years after Congress amended the CWA, the Court again interpreted § 509(b) in Crown Simpson. The question before the Court was whether the EPA’s veto of a state-issued permit qualified as a § 402 approval or denial of the permit under § 509(b)(1)(F).232 Section 402 allows the EPA to either veto a proposed NPDES permit or do nothing for ninety days and passively approve it.233 The lower court had determined that a veto of the state-issued permit did not constitute “issuing or denying” a permit, rendering § 509(b)(1)(F) inapplicable.234 The Court disagreed, holding that the veto had the “precise effect” of denying a permit within the meaning of § 509(b), and thus reversed the Ninth Circuit’s decision.235 Agreeing with a concurring opinion in the lower-court decision, the Court highlighted the argument that allowing jurisdiction over this type of action “would best comport with the congressional goal of ensuring prompt resolution of challenges to EPA’s actions.”236 Once again, the Court employed a “practical rather than a cramped construction” of § 509(b) that emphasized function over form.237 Due to these functional concerns, the Court allowed for direct review of an action listed in § 509(b)(1)(F)’s ancillary statute but not enumerated in § 509(b). That is, the Court allowed for review of the EPA veto despite the absence of an actual permit denial.238 Congress has not amended § 509(b)(1) since the Court’s ruling in Crown Simpson.

While E.I. du Pont and Crown Simpson do not squarely address the finality question, these cases both demonstrate that the Court will look beyond the final actions listed in § 509(b)(1) when it has prudential reasons for doing so.239 In E.I. du Pont, the Court declared that § 509(b) allowed for review of any action the administrator could take under § 301.240 Because § 301 contains nonfinal actions, the Court impliedly held that those actions were available for § 509(b) review. There, the Court was concerned with judicial economy and the problem of bifurcated review.241 Three years later, the Crown Simpson Court ruled that the lack of a final issuance or denial of a permit did not deprive the appellate court of jurisdiction under § 509(b)(1).242 In that instance, the Court was guided by the congressional goal of “ensuring prompt resolution of challenges to EPA’s actions” and established that its interpretation of § 509(b) was motivated by practicality rather than formality.243 Congress has not amended § 509(b) since either of these rulings and can be understood to have acquiesced in the Court’s interpretation of § 509(b).

5.   Section 509(b) overcomes Bell’s presumption.

To overcome Bell’s presumption under the Block standard, there must be both an omission of explicit finality language and other factors indicating congressional intent.244 The intent to override the presumption must be “fairly discernible.”245 That standard is met here. The text of § 509(b) contains no finality language, and so the “specific language” requirement is satisfied. Additional Block factors indicate that Congress intended to allow for review of nonfinal action. The CWA’s legislative history supports overcoming Bell’s presumption. Importantly, the legislative history does not mention that review is to be limited to final actions. The statutory scheme presents nonfinal actions in § 509(b)’s ancillary statutes that govern, influence, and have the precise effect of determining the substance of the EPA’s final action. The Court has found these types of actions to be within the meaning of § 509(b)’s enumerated actions. Further, unlike the statutory schemes in other statutes that courts have found to contain a finality requirement, the statutes related to § 509(b) do not discuss a final agency action requirement. Finally, Congress acquiesced in the Court’s practical construction of § 509(b) in E.I. du Pont by not amending the statute in the 1977 revisions. That Congress explicitly required finality in the CAA but not the CWA also serves as an interstatutory argument in favor of overcoming the Bell presumption. Moreover, there has been no revision of § 509(b) in response to the Court’s pragmatic holding in Crown Simpson, which allowed for review of a nonfinal action. Block’s factors246 all weigh in favor of finding that § 509(b) overcomes Bell’s presumption. Section 509(b) should not be read as subject to Bell’s presumption in favor of requiring final agency action for judicial review.

B.    Relying on Ripeness in the Absence of Finality

This Section argues that, even without the finality requirement to filter out frivolous actions, courts can rely on the ripeness doctrine to determine whether judicial review is appropriate in § 509(b) cases. The doctrine’s two-pronged inquiry into whether an issue is “fit” for review and the potential hardship to the parties of withholding review247 should allow courts to take a more flexible approach that is unconstrained by a rigid finality requirement. First, this Section demonstrates that some courts, though using the language of finality, have relied on the ripeness doctrine to determine whether to allow direct review. Then, this Section argues that relying on ripeness in the absence of a finality requirement in § 509(b) provides a cohesive analytical switch that is consistent with Supreme Court jurisprudence and largely consistent with lower-court case law surrounding § 509(b)(1).248 This Section highlights the types of cases in which removing the finality requirement and applying only the ripeness doctrine would have led to a different outcome. Finally, this Section argues that removing the finality requirement from § 509(b) is welfare maximizing.

1.   Courts have been employing ripeness considerations disguised as finality concerns.

Courts considering jurisdiction under § 509 have already used ripeness considerations as a basis for upholding Bell’s presumption in favor of requiring finality. In American Paper, for example, the court acknowledged it could review the case if the policy statement “ever leads to the denial or modification of a permit.”249 The court held that review at that time would be “too soon” in that it would have to review the policy “without knowing how (or even whether) it would affect any plant.”250 The Seventh Circuit, though couching its decision in the language of finality, was speaking to the concerns of the Abbott Laboratories ripeness test.251 There were no facts to review, only a legal question, but there was also no hardship to the parties in withholding review because the policy came from a regional office, not the administrator of the EPA. This reasoning by the court is consistent with the doctrine’s purpose of preventing premature adjudication of abstract disagreements.252

The Fifth Circuit’s holding in National Pork also fits within a ripeness framework. Like the letters in American Paper, the letters in National Pork were exclusively legal documents and, therefore, fit for review.253 However, the court’s conclusion that the letters “merely restate[d]” existing law meant that the petitioners would suffer no specific hardship from the denial of review.254 Instead of conjuring up a textual finality mandate when none exists, the court could have dismissed the case for lack of hardship under the ripeness doctrine.

In Rhode Island, the state appealed a decision while EAB proceedings were ongoing.255 The court read § 509(b) as requiring final action and stated that review would be available when the EPA issued a final permit.256 In discussing whether to apply an exception to the finality rule—the collateral order doctrine—the court reasoned that Rhode Island’s interests would not be irreparably harmed in the absence of immediate review.257 The court also acknowledged that refusing to intercede came with a cost because it introduced the “prospect of duplicative proceedings should the denial of intervention eventually be deemed improvident.”258 Thus, the court calculated that the hardship of denying review to Rhode Island did not outweigh the cost of adhering to the finality rule.259 While the court engaged in a cost-benefit analysis for purposes of the collateral order doctrine analysis, it engaged in the same type of weighing of the hardships required by the ripeness inquiry. Having found that Rhode Island would not be irreparably harmed by postponing review, the court could have dismissed the case solely on ripeness grounds instead of the two-step process of citing to Bell’s presumption and then declining to apply an exception to the finality rule.

2.   Changing the course of Riverkeeper and Champion.

If removing the finality analysis from § 509(b) results in no change from the status quo, this discussion is largely academic. There will be no change in outcome when a party attempts to appeal a nonripe and nonfinal action. Conversely, removing a finality requirement will not affect whether a party can appeal a ripe, final action. Removing the finality requirement will affect judicial review only in cases in which a party seeks to appeal a nonfinal action that is ripe for review. Allowing for review of nonfinal actions would have significantly changed the outcome in the Eleventh Circuit’s Riverkeeper decision and the Fourth Circuit’s Champion ruling. This Section discusses why both cases were ripe for review, despite their nonfinality, and then the next Section argues that removing finality and relying on ripeness to preserve judicial economy is welfare maximizing.

The Eleventh Circuit dismissed the petition for review in Riverkeeper because it failed Bennett’s finality test.260 The EPA’s partial findings did not “mark the ‘consummation’ of the agency’s decision making process” because, as the court noted, the EPA had “the power to revise its findings.”261 While it is impossible to know how the court would have ruled based on ripeness, applying the ripeness test shows that the case was ripe for review.

First, the question was fit for review. The EPA had determined that twenty-two of the alleged deficiencies did not warrant initiating withdrawal of the state’s NPDES program.262 The EPA had already conducted the relevant factual analysis; no further factual development was needed in making determinations as to the twenty-two alleged deficiencies, and only legal questions remained. Second, the petitioners would have suffered a hardship if the court delayed review. The interim response provided that the EPA would defer a final decision to allow more time “for EPA to work with [Alabama] to address the issues and/or to allow additional time to monitor [Alabama’s] program implementation and progress.”263 Not specifying a timeline or date by which the EPA would make a decision constituted a hardship for the petitioners in that the claimed environmental deficiencies would be ongoing during the time the EPA and Alabama worked through the remaining issues. The possibility of a monitoring period meant that the EPA and Alabama could postpone any final decision by claiming that a longer period was needed to monitor the situation.264 It is not the case that judicial intervention would inappropriately interfere with further administrative action.265 Because the question was fit for review and hardship to the parties existed, the case was ripe for review.

Forgoing finality would also have led to a different outcome in Champion. There, the court held that judicial review of the EPA’s assumption of permitting authority would be challengeable once the EPA issued or denied a permit to Champion.266 The court admitted that the EPA’s actions “were actions of the administrator subject to judicial review” under § 509(b) “if those actions were allowed to proceed to their logical completion, i.e., EPA either granting or denying a permit.”267 Removing the finality requirement would have freed Champion from waiting until a permit had been either denied or approved. The question was fit for review because whether the EPA had properly assumed permitting authority was a purely legal question that did not require additional factfinding. Champion also suffered hardship from withholding review because the EPA’s action had deprived it of its renewed state-issued NPDES permit.268

3.   Removing the finality rule would be welfare maximizing.

As demonstrated above, not applying a finality requirement and relying on the ripeness doctrine when deciding whether to allow immediate review can change the outcome of § 509(b) cases. This is a good thing. Ripeness is a flexible doctrine that, compared with the finality rule, results in more equitable and efficient outcomes. Some courts have declined to rule that § 509(b) overcomes Bell’s presumption because of a perceived danger of squandering judicial and administrative resources.269 In fact, a robust application of the ripeness doctrine should allay these concerns and, in some cases, conserve judicial resources. In the Riverkeeper case, by withholding review of twenty-two of the claims until the EPA made determinations on the remaining four claims, the court incentivized petitioners—nonprofit and industry alike—to separate claims. The separation of claims will lead to judicial redundancy in the next lawsuit; instead of presiding over one case, a judge (or several different judges) would have to preside over twenty-six discrete actions. Likewise, in Champion, postponing review created the possibility of the EPA engaging in the laborious process of permit drafting only to later have a court rule that the agency improperly assumed permitting authority in the first place. In these instances, it is better to allow immediate review.

Allowing for § 509(b) review of these types of intermediate decisions realizes Congress’s vision of “open[ing] wide” the possibilities of meaningful input in government decisions270 and ensuring the “prompt resolution of challenges to EPA’s actions.”271 Congress’s concern with opening wide avenues of review was prescient. Scholars have demonstrated that agencies seek to insulate their actions from review.272 The Riverkeeper and Champion cases are good examples. By postponing a final decision or permit issuance, the EPA can postpone review of its determinations and other actions. Professor Nina Mendelson has argued that expanding the availability of suits by interested parties can counter informal (and often unchecked) agency action.273 Though not a citizen-suit provision in the mold of CWA’s § 505 provision, removing the finality rule from § 509(b) would “increase regulatory beneficiaries’ [and regulatory losers’] ability to hold agencies externally accountable for their implementation of a statute.”274 One of the purposes of § 509 was to create judicial review that would have that precise effect.275

Finally, it is important to consider the costs and benefits of using ripeness in the absence of a strict finality rule. At the outset, such a switch would lead to more litigation, not less. This outcome alone should not be cause to worry; our legal system favors the availability of judicial review of agency action.276 The more salient question is whether, because review is already allowed once the EPA has taken final action, the benefits of allowing immediate review of nonfinal action outweigh the costs. If not, perhaps it would be more prudent for courts to continue to enforce a finality requirement in § 509(b) cases.

The additional costs of allowing review are the associated litigation costs, the potential disruptions of administrative proceedings, and the judiciary’s costs in overseeing the litigation.277 For example, allowing immediate review of the twenty-two claims in Riverkeeper would impose additional administrative costs on the courts, but some of those costs would have been imposed anyway when the parties appealed after the EPA’s final determination. It would likely have been the case that review of the EPA’s determination regarding the twenty-two alleged deficiencies would not have affected the EPA’s actions in working with Alabama to correct the other four deficiencies. In the meantime, irreparable environmental harm could have occurred. Further, it is not clear that the administrative cost of reviewing the twenty-two determinations before the EPA’s final decision is greater than the cost of a future case in which a plaintiff, following Riverkeeper’s precedent, files separate suits for each alleged deficiency in a different state to circumvent the finality problem.

Courts should therefore allow review when the harm of withholding review is greater than the cost of additional litigation. If the court anticipates that withholding review would create perverse incentives, resulting in undesirable behavior like the claim splitting discussed above, it should include those considerations in its ripeness calculation. This formulation looks to the extent of hardship on the parties and the cost of litigation on the court and the parties (including the cost on the agency of administrative delay). It is the ripeness doctrine reframed.278 The first prong of the ripeness inquiry is whether the issue is “fit” for review. An issue is fit for review if it is legal in nature and does not require additional factfinding.279 Empirical scholarship has revealed factfinding to be the largest cost of litigation.280 If no additional factfinding is required, then the only cost on the court and parties is briefing and ruling on the legal issue, as well as any cost resulting from delay of the administrative process. The fitness prong of the ripeness analysis can thus be seen as a proxy for allowing review when the cost of doing so is minimal.281 If a court can settle an issue without incurring additional expenses, the issue should be fit for review because deferring review would not add any marginal benefit to the court’s ability to rule on the issue.282 The second prong of the ripeness doctrine is explicitly a cost-measuring inquiry. The court must measure the hardship to the parties of withholding review.283 In other words, the ripeness doctrine allows review when the cost of withholding review is greater than the cost of resolving the issue immediately.

Unlike the ripeness doctrine, the finality rule does not weigh the costs and benefits of allowing review. With a finality requirement, some actions for which review is cost justified are not allowed to proceed in court. Judges have likely realized the welfare-negative effect of the finality doctrine. The judicially created “pragmatic finality” doctrine and the collateral order doctrine allow for judicial review of nonfinal action in circumstances in which the courts are bound by an explicit or Bell-imposed finality requirement.284 These doctrines allow for review of nonfinal actions when those actions are supported by an adequate record and withholding review would cause hardship to the parties.285 The doctrines give courts flexibility to conduct a “ripeness-light” analysis.286 Because § 509(b) does not have an explicit finality requirement and can overcome Bell’s presumption, there is no need to adopt the contorted pragmatic finality doctrine or worry about the collateral order doctrine; ripeness suffices.

It is possible that Congress intended to restrict judicial review for final agency actions. If courts excluded finality considerations and applied only the ripeness analysis in the context of § 509(b) cases, Congress would always be free to amend the statutes and insert explicit finality requirements. But there is scant evidence that Congress intended to disallow review of nonfinal action in the § 509(b) context. If anything, an increase in meritorious litigation would certainly lift the spirits of the CWA’s drafters, who were impressed by the public’s willingness to “turn to legal action as a remedy for what they consider to be errors on the part of the Government.”287 Allowing this type of input into government decisionmaking therefore not only comports with Congress’s desire to strengthen the public’s ability to take part in government action but also leads to societal welfare gains.

C.    The Preclusion Question

Having dispensed with Bell’s presumption and demonstrated that ripeness by itself is preferable to ripeness in addition to the finality rule, one last issue remains—the role of § 509(b)’s review-preclusion provision. Easterbrook’s fear of expanding § 509(b) so much that parties routinely file protective challenges in both district and circuit court has been validated.288 If the lack of a finality requirement increases the amount of protective duplicative filings, the increase in the cost of additional litigation would limit any welfare gains made by allowing for review of ripe nonfinal actions. Ideally, changing how courts interpret § 509(b) would reduce the number of protective filings, but in this case, relying on ripeness does not have to affect the volume of protective filings one way or the other.

The text of § 509(b)’s review-preclusion provision starts the 120-day clock on the date of “such determination, approval, promulgation, issuance or denial.”289 As discussed in Part III.A, the Court has interpreted these words expansively to include actions that govern or affect the listed actions. The Court’s broad interpretation of what actions are subject to review necessarily affects when parties must file for review of those actions. The statute requires that parties file for review of an action within 120 days of the date of the action. This gives litigants two options: either challenge the qualifying nonfinal action within the 120-day timeframe or wait until the administrator takes a final action and then challenge the nonfinal action under the statute’s grant of review of the actions the administrator took in making the final action.

For instance, imagine that the EPA makes a pivotal determination at time T1 that satisfies the Court’s pragmatic approach to § 509(b) review under Crown Simpson and issues a final decision denying a permit at time T2. The party that suffered a hardship at T1 could challenge the nonfinal EPA action within 120 days of T1 or within 120 days of T2 as part of a challenge to the administrator’s action in the final denial of the permit.290 This interpretation of the preclusion provision allows for review but does not force parties to meticulously challenge every action made by the administrator; parties should challenge only actions that meet the ripeness requirement. This framework incentivizes parties to bring a separate lawsuit on a nonfinal action only when the hardship of delaying review is greater than the benefit of waiting until the final action and then consolidating all alleged deficiencies into one claim.

While Easterbrook’s example of a nightmare world in which litigants might challenge interoffice memoranda on four-inch nozzles makes a (rather dramatic) point, it is also unrealistic. Under the proposed framework, litigants will have no additional incentive to challenge “everything in sight” because, in most instances, it would make sense to bring all the claims at once instead of taking a piecemeal approach to litigation. A party should only challenge a nonfinal action if the cost of filing a separate suit is less than the expected hardship of postponing a challenge.291 Moreover, any hardship a party might bear from delaying review of a rote bureaucratic memo would likely not survive even the most lenient ripeness analysis. Easterbrook can rest assured that no procedural morass will result from interpreting § 509(b) to authorize review of nonfinal actions.

* * *

Relying on ripeness in the absence of a finality rule is a promising approach to dealing with § 509(b) review. Like finality, the ripeness doctrine precludes review in cases in which the costs of allowing immediate review of nonfinal action outweigh the gains. Ripeness’s flexibility to allow review when the benefits of review outweigh the costs makes relying on ripeness without a finality requirement a welfare-maximizing approach to § 509(b) review. Relying on ripeness alone also furthers the intent of the Congress that enacted the CWA by allowing for public input in government decisions through judicial review. Employing the ripeness doctrine in the absence of finality thus leads to efficiency gains and desirable normative outcomes. While this Comment has focused on § 509(b), interpreting other statutes to allow for review of nonfinal action and applying this Comment’s suggested ripeness framework could lead to additional social benefits.


This Comment has argued that the CWA’s direct-review provision overcomes Bell’s presumption in favor of requiring finality and that courts can rely on the ripeness doctrine as the primary framework to decide whether to grant direct appellate review. Ripeness furthers the normative interest in allowing for judicial review of agency actions and Congress’s intent to increase the opportunities for the public to have meaningful input in government decisionmaking. Finally, the absence of a finality requirement facilitates social welfare gains currently precluded by the finality rule’s strict requirements. Judges fearing the consequences of declining to find a finality requirement in § 509(b)—and other direct-review statutes—need not worry; relying on ripeness in the absence of a finality requirement will lead to beneficial results.

  • 1. 60 Stat 237 (1946), codified as amended in various sections of Title 5.
  • 2. See, for example, National Park Hospitality Association v Department of the Interior, 538 US 803, 807–12 (2003) (concluding that a final agency action was not ripe for review).
  • 3. The term “review statute” can refer to any statute that provides for judicial review of agency action. A “direct-review statute” refers to a review statute that provides for direct appellate review. This Comment examines direct-review statutes and uses the two terms interchangeably to refer to statutes that provide for direct appellate review.
  • 4. See Sundeep Iyer, Comment, Jurisdictional Rules and Final Agency Action, 125 Yale L J 785, 790–92 (2016) (describing variations in the text of direct-review statutes).
  • 5. See Part I.B.
  • 6. Pub L No 92-500, 86 Stat 816 (1972), codified at 33 USC § 1251 et seq.
  • 7. 62 Stat 1155, codified as amended at 33 USC § 1251 et seq.
  • 8. 33 USC § 1251(a).
  • 9. 33 USC § 1251(a)(3).
  • 10. 33 USC § 1342(a).
  • 11. Richard L. Revesz, Environmental Law and Policy 558 (Foundation 3d ed 2015).
  • 12. See 33 USC § 1251(d); 33 USC § 1342.
  • 13. NPDES State Program Information (EPA, Feb 19, 2016), archived at
    G2JC-BVTM (“Withdrawal Petitions” tab) (“EPA has authorized 46 states and the U.S. Virgin Islands to implement the NPDES program.”).
  • 14. Revesz, Environmental Law and Policy at 558 (cited in note 11).
  • 15. CWA § 509(b), 86 Stat at 892, 33 USC § 1369(b).
  • 16. For a description of the rules-versus-standards framework, see Ward Farnsworth, The Legal Analyst: A Toolkit for Thinking about the Law 163–71 (Chicago 2007).
  • 17. Ticor Title Insurance Co v Federal Trade Commission, 814 F2d 731, 745 (DC Cir 1987) (Williams concurring).
  • 18. See id at 745–46 (Williams concurring).
  • 19. Chicago & S Air Lines, Inc v Waterman Steamship Corp, 333 US 103, 112–13 (1948). See also Farnsworth, The Legal Analyst at 164–65 (cited in note 16).
  • 20. 520 US 154 (1997).
  • 21. Id at 178.
  • 22. Id (citation omitted).
  • 23. Sackett v Environmental Protection Agency, 566 US 120, 127 (2012).
  • 24. Bennett, 520 US at 178, quoting Port of Boston Marine Terminal Association v Rederiaktiebolaget Transatlantic, 400 US 62, 71 (1970).
  • 25. See Farnsworth, The Legal Analyst at 165 (cited in note 16) (explaining that the consequences of a rule are triggered once the facts are settled).
  • 26. See Carol M. Rose, Crystals and Mud in Property Law, 40 Stan L Rev 577, 578–80 (1988) (describing how rules devolve into standards over time and vice versa).
  • 27. See, for example, 5 USC § 704.
  • 28. Bell v New Jersey, 461 US 773, 778 (1982).
  • 29. See Trudeau v Federal Trade Commission, 456 F3d 178, 188–89 (DC Cir 2006); Ticor, 814 F2d at 746 (Williams concurring).
  • 30. Reno v Catholic Social Services, Inc, 509 US 43, 57 n 18 (1993). See also Stolt-Nielsen SA v AnimalFeeds International Corp, 559 US 662, 670 n 2 (2010).
  • 31. See National Park Hospitality Association v Department of the Interior, 538 US 803, 807–08 (2003) (explaining the rationale behind and evaluative method in the ripeness test), quoting Abbott Laboratories v Gardner, 387 US 136, 148–49 (1967).
  • 32. Richard J. Pierce Jr, 2 Administrative Law Treatise § 15.12 at 1334 (Wolters Kluwer 5th ed 2010).
  • 33. 387 US 136 (1967).
  • 34. Id at 149.
  • 35. 523 US 726 (1998).
  • 36. Id at 733.
  • 37. See Farnsworth, The Legal Analyst at 163–64 (cited in note 16).
  • 38. National Park Hospitality Association, 538 US at 814.
  • 39. Bell, 461 US at 778.
  • 40. 304 US 375 (1938).
  • 41. Id at 376–77.
  • 42. The relevant statutory text in the Federal Power Act provided that “[a]ny party to a proceeding under this Act aggrieved by an order issued by the Commission in such proceeding may obtain a review of such order in the Circuit Court of Appeals.” Id at 383, quoting Public Utility Act of 1935 § 313(b), 49 Stat 838, 860, codified as amended at 16 USC § 825l(b).
  • 43. Metropolitan Edison, 304 US at 383.
  • 44. Id at 383–84.
  • 45. Id at 385–87.
  • 46. 461 US 773 (1983).
  • 47. Id at 778.
  • 48. Education Amendments of 1978 § 1232, Pub L No 95-561, 92 Stat 2143, 2350, codified at 20 USC § 1234d. The statute was later amended to specify “final agency action.” Augustus F. Hawkins-Robert T. Stafford Elementary and Secondary School Improvement Amendments of 1988 § 3501(a), Pub L No 100-297, 102 Stat 130, 356, codified at 20 USC § 1234g.
  • 49. Bell, 461 US at 778–79 (holding that 20 USC § 1234d grants federal courts jurisdiction only over a “final order of the Department” before evaluating whether the case at bar met that requirement).
  • 50. Id at 778.
  • 51. Id.
  • 52. Id.
  • 53. Bell, 461 US at 778–79.
  • 54. 761 F3d 1084 (9th Cir 2014).
  • 55. Id at 1085, 1091, quoting 15 USC § 717r(d)(1).
  • 56. Columbia Riverkeeper, 761 F3d at 1091–92.
  • 57. Id at 1092, quoting 15 USC § 717r(d)(1).
  • 58. Columbia Riverkeeper, 761 F3d at 1092.
  • 59. Id.
  • 60. 177 F3d 1042 (DC Cir 1999).
  • 61. Id at 1047.
  • 62. Id.
  • 63. Id at 1047–48.
  • 64. Meredith, 177 F3d at 1048.
  • 65. Id, citing Federal Mine Safety and Health Act of 1977, S Rep No 95-181, 95th Cong, 1st Sess 13 (1977), reprinted in 1977 USCCAN 3401, 3413, and Federal Mine Safety and Health Amendments Act of 1977, HR Rep No 95-655, 95th Cong, 1st Sess 53 (1977), reprinted in 1977 USCCAN 3485, 3501.
  • 66. Meredith, 177 F3d at 1048.
  • 67. 43 F3d 912 (4th Cir 1995).
  • 68. Id at 914.
  • 69. Id at 913–14.
  • 70. Id at 914, quoting Jim Walter Resources, Inc v Federal Mine Safety and Health Review Commission, 920 F2d 738, 743 (11th Cir 1990).
  • 71. Carolina Power, 43 F3d at 914 (observing that the Energy Reorganization Act of 1974 was structured so that the secretary of labor must issue an order that is “inherently ‘final’ in nature”).
  • 72. Id at 915.
  • 73. See Roni A. Elias, The Legislative History of the Administrative Procedure Act, 27 Fordham Envir L Rev 207, 221 (2016).
  • 74. See Office of the Attorney General, Final Report of the Attorney General’s Committee on Administrative Procedure 85 (Government Printing Office 1941) (explaining that the finality requirement “has been formulated by the courts in the absence of legislation”).
  • 75. 5 USC § 704.
  • 76. 5 USC § 704.
  • 77. Iowa League of Cities v Environmental Protection Agency, 711 F3d 844, 863 n 12 (8th Cir 2013) (finding that “final” does not modify “agency action made reviewable by statute”). But see Carter/Mondale Presidential Committee, Inc v Federal Election Commission, 711 F2d 279, 284 n 9 (DC Cir 1983) (concluding that Congress intended that review of “agency action made reviewable by statute” would be restricted to final agency actions).
  • 78. See Norton v Southern Utah Wilderness Alliance, 542 US 55, 61–62 (2004) (“Where no other statute provides a private right of action, the ‘agency action’ complained of must be ‘final agency action.’”); Lujan v National Wildlife Federation, 497 US 871, 882 (1990) (“When, as here, review is sought not pursuant to specific authorization in the substantive statute, but only under the general review provisions of the APA, the ‘agency action’ in question must be ‘final agency action.’”).
  • 79. See, for example, Iowa League, 711 F3d at 863 n 12 (determining whether a nonfinal agency action was reviewable under the terms of the CWA). See also Flue-Cured Tobacco Cooperative Stabilization Corp v United States Environmental Protection Agency, 313 F3d 852, 857 (4th Cir 2002) (specifying that courts, when assessing jurisdiction, must look to whether agency action is final or made specifically reviewable by statute).
  • 80. Pub L No 88-206, 77 Stat 392 (1963), codified as amended at 42 USC § 7401 et seq.
  • 81. Clean Air Amendments of 1970, Pub L No 91-604, 84 Stat 1676, codified at 42 USC § 7401 et seq.
  • 82. Daniel P. Selmi, Jurisdiction to Review Agency Inaction under Federal Environmental Law, 72 Ind L J 65, 72 (1996).
  • 83. 33 USC § 1365 (authorizing citizen suits against any person or government actor violating the CWA and against the administrator of the EPA for failure to perform nondiscretionary actions).
  • 84. 33 USC § 1369(b)(1).
  • 85. 33 USC § 1369(b)(1). The statute’s review-preclusion provision functions as a 120-day statute of limitations. If 120 days have passed since the EPA action, no party can challenge the action’s validity.
  • 86. 33 USC § 1369(b)(1).
  • 87. 33 USC § 1369(b)(2).
  • 88. See David P. Currie, Judicial Review under Federal Pollution Laws, 62 Iowa L Rev 1221, 1225–47 (1977) (criticizing § 509(b) and cataloguing wasteful litigation over whether a case has been brought in the right court under the CWA’s and CAA’s review statutes).
  • 89. See note 85.
  • 90. 33 USC § 1369(b)(1).
  • 91. 33 USC § 1369(b)(1).
  • 92. 33 USC § 1369(b)(2).
  • 93. See 33 USC § 1369(b)(1):

    Review of the Administrator’s action (A) in promulgating any standard of performance under section 1316 of this title, (B) in making any determination pursuant to section 1316(b)(1)(C) of this title, (C) in promulgating any effluent standard, prohibition, or pretreatment standard under section 1317 of this title, (D) in making any determination as to a State permit program submitted under section 1342(b) of this title, (E) in approving or promulgating any effluent limitation or other limitation under section 1311, 1312, 1316, or 1345 of this title, (F) in issuing or denying any permit under section 1342 of this title, and (G) in promulgating any individual control strategy under section 1314(l) of this title, may be had by any interested person in the Circuit Court of Appeals of the United States for the Federal judicial district in which such person resides or transacts business which is directly affected by such action upon application by such person.

  • 94. The CAA’s direct-review provision mentions “final action taken” and “any other final action of the Administrator.” 42 USC § 7607(b)(1).
  • 95. The Second and Ninth Circuits have not explicitly addressed the finality issue but have issued rulings in which they have restricted review to the “iss[uance] or den[ial]” of a permit or when “approval or disapproval [of a state permit program] itself is being challenged.” See Central Hudson Gas & Electric Corp v United States Environmental Protection Agency, 587 F2d 549, 556 (2d Cir 1978); Southern California Alliance of Publicly Owned Treatment Works v U.S. Environmental Protection Agency, 853 F3d 1076, 1085–86 (9th Cir 2017). But see note 150 (discussing conflicting case law in the Ninth Circuit). These rulings suggest that both circuits would take a narrow approach to § 509(b) jurisdiction and would not allow for review of nonfinal action under the other § 509(b) provisions. Neither circuit has dealt directly with the Bell presumption. The Eighth Circuit has taken a similar approach to § 509(b)(1)(F) but has ruled that, as a general matter, § 509(b) does not require final agency action. See note 77.
  • 96. 635 F3d 738 (5th Cir 2011).
  • 97. See id at 756.
  • 98. Id at 741.
  • 99. Id at 748–49.
  • 100. National Pork, 635 F3d at 755, quoting 33 USC § 1369(b)(1).
  • 101. See 33 USC § 1369(b)(1).
  • 102. 806 F3d 1079 (11th Cir 2015).
  • 103. Id at 1082.
  • 104. Id at 1081.
  • 105. Id at 1080.
  • 106. Riverkeeper, 806 F3d at 1080.
  • 107. Id (quotation marks omitted).
  • 108. Id.
  • 109. Id at 1081.
  • 110. Riverkeeper, 806 F3d at 1081.
  • 111. Id at 1081–82, citing Bell, 461 US at 778.
  • 112. Riverkeeper, 806 F3d at 1081. See also 33 USC § 1369(b)(1) (“Review of the Administrator’s action . . . (D) in making any determination as to a State permit program submitted under section 1342(b) of this title . . . may be had by any interested person in the Circuit Court of Appeals.”).
  • 113. Riverkeeper, 806 F3d at 1082.
  • 114. Id.
  • 115. 618 F2d 991 (3d Cir 1980).
  • 116. Id at 993.
  • 117. Id at 993–94.
  • 118. Id at 994.
  • 119. Pennsylvania Department of Environmental Resources, 618 F2d at 997.
  • 120. Id. The court ultimately concluded that the suit sought to compel the EPA to undertake a nondiscretionary action and was therefore appropriately brought under the CWA’s citizen-suit provision. Id at 995.
  • 121. 882 F2d 287 (7th Cir 1989).
  • 122. Pennsylvania Department of Environmental Resources, 618 F2d at 997.
  • 123. American Paper, 882 F2d at 288.
  • 124. Id.
  • 125. Id.
  • 126. Id at 289.
  • 127. American Paper, 882 F2d at 289.
  • 128. Id at 289–90.
  • 129. Id at 289.
  • 130. Id.
  • 131. 850 F2d 182 (4th Cir 1988).
  • 132. Id at 187.
  • 133. Id at 185.
  • 134. Id at 187.
  • 135. Champion, 850 F2d at 187.
  • 136. Id at 188 (emphasis omitted), citing Environmental Policy Division of the Congressional Research Service, 3 A Legislative History of the Clean Water Act of 1977 470 (Government Printing Office 1978).
  • 137. Champion, 850 F2d at 188.
  • 138. Id at 190.
  • 139. 378 F3d 19 (1st Cir 2004).
  • 140. Id at 23.
  • 141. Id at 21. The EAB is an adjudicatory body within the EPA.
  • 142. Id.
  • 143. Rhode Island, 378 F3d at 22.
  • 144. Id at 23.
  • 145. Id.
  • 146. See, for example, id.
  • 147. See, for example, Riverkeeper, 806 F3d at 1081.
  • 148. See, for example, Champion, 850 F2d at 188.
  • 149. See, for example, American Paper, 882 F2d at 289.
  • 150. In Iowa League of Cities v Environmental Protection Agency, 711 F3d 844, 861–62 (8th Cir 2013), and in City of Ames v Reilly, 986 F2d 253, 255–56 (8th Cir 1993), the Eighth Circuit focused on whether the EPA’s action fit into one of the actions listed in § 509(b). While City of Ames used similar language to other courts that have required final agency action under § 509(b)—and even cited American Paper and Champion International—it did not hold that § 509(b) contained a finality requirement. It held only that the EPA’s action indicating disapproval of the locality’s NPDES permit did not constitute an issuance or denial of a permit under § 509(b)(1)(F).

    The Ninth Circuit has also issued conflicting case law. Compare Southern California Alliance, 853 F3d at 1080–86 (declining to find jurisdiction under § 509(b)(1)(E) or (F) for “interim” actions), with Defenders of Wildlife v United States Environmental Protection Agency, 420 F3d 946, 955–56 (9th Cir 2005), revd and remd on other grounds, National Association of Home Builders v Defenders of Wildlife, 551 US 644 (2007) (finding jurisdiction under § 509(b)(1)(D) to review an interim determination that was virtually determinative of the final agency action).

  • 151. 711 F3d 844 (8th Cir 2013).
  • 152. Id at 854.
  • 153. Id at 854, 861. See also 33 USC § 1369(b)(1)(E).
  • 154. Iowa League, 711 F3d at 862.
  • 155. Id at 863.
  • 156. Id at 863 n 12.
  • 157. Id.
  • 158. Iowa League, 711 F3d at 863 n 12, citing 5 USC § 704.
  • 159. Iowa League, 711 F3d at 863 n 12, quoting Yankton Sioux Tribe v Podhradsky, 606 F3d 994, 1012 (8th Cir 2010).
  • 160. Iowa League, 711 F3d at 863 n 12.
  • 161. Id at 868.
  • 162. 467 US 340 (1984).
  • 163. Id at 344–46.
  • 164. Id at 349.
  • 165. Id.
  • 166. Block, 467 US at 351.
  • 167. See Sackett v Environmental Protection Agency, 566 US 120, 128 (2012) (Scalia). Justice Antonin Scalia unsurprisingly took a textually inclined approach to presumptions. See, for example, United States v Fausto, 484 US 439, 444 (1988) (Scalia) (limiting the Block inquiry to “the purpose of the [statute], the entirety of its text, and the structure of review that it establishes”).
  • 168. Block, 467 US at 350–51, quoting Association of Data Processing Service Organizations, Inc v Camp, 397 US 150, 157 (1970). What is required to meet the “fairly discernible” standard is, itself, fairly undiscernible. The Court has not provided any guidance beyond referencing the relevant evidence to which a judge should look. For example, the Court has stated that to determine whether intent is fairly discernible, it looks to the statute’s “text, structure, and purpose.” Elgin v Department of the Treasury, 567 US 1, 10 (2012).
  • 169. Block, 467 US at 351.
  • 170. See Bowen v Michigan Academy of Family Physicians, 476 US 667, 674 (1986), quoting Abbott Laboratories, 387 US at 141.
  • 171. United States v Erika, Inc, 456 US 201, 208 (1982).
  • 172. Fausto, 484 US at 448–49 (citing the structure of the statutory scheme as a reason to overcome the presumption in favor of judicial review).
  • 173. Mach Mining, LLC v Equal Employment Opportunity Commission, 135 S Ct 1645, 1651 (2015), quoting Bowen, 476 US at 670.
  • 174. Block, 467 US at 349.
  • 175. See, for example, Columbia Riverkeeper, 761 F3d at 1097 (dismissing the case for lack of jurisdiction due to the absence of a final agency action). See also Block, 467 US at 353 n 4; Arkansas Dairy Cooperative Association, Inc v United States Department of Agriculture, 573 F3d 815, 829 (DC Cir 2009) (applying Block’s “specific language” factor to the presumption that Congress was aware of background law when legislating).
  • 176. Block, 467 US at 349.
  • 177. See William N. Eskridge Jr and Philip P. Frickey, Quasi-Constitutional Law: Clear Statement Rules as Constitutional Lawmaking, 45 Vand L Rev 593, 595 n 4 (1992) (describing clear-statement rules as requiring a clear statement “to rebut a policy presumption the Court has created”).
  • 178. See Astoria Federal Savings & Loan Association v Solimino, 501 US 104, 108–09 (1991). While the Astoria Court did not cite Block, its test for whether to apply the relevant presumption rested on almost identical factors. The Court looked to whether application of the presumption would be “consistent with Congress’ intent in enacting the statute.” Id at 110 (alteration omitted), quoting University of Tennessee v Elliott, 478 US 788, 796 (1986).
  • 179. Sackett, 566 US at 128.
  • 180. Block, 467 US at 349.
  • 181. See Sackett, 566 US at 128 (“[A]s with most presumptions, this one ‘may be overcome by inferences of intent drawn from the statutory scheme as a whole.’”), quoting Block, 467 US at 349.
  • 182. Indeed, for APA cases the ripeness inquiry includes determining whether the agency action is final. The Supreme Court has included finality as a component of ripeness in the APA context when final agency action is required. See Abbott Laboratories, 387 US at 148–50 (finding that an action was ripe because it qualified as final agency action “within the meaning of [the APA]”). See also National Park Hospitality Association v Department of the Interior, 538 US 803, 812 (2003) (finding that the relevant action constituted “final agency action” within the meaning of the APA for ripeness purposes). The Court has not mentioned finality as an element of ripeness in other administrative contexts. See, for example, Whitman v American Trucking Associations, Inc, 531 US 457, 479 (2001) (adopting the Ohio Forestry ripeness inquiry and differentiating the ripeness requirements for direct-review statutes from the requirements of APA review). See also Shalala v Illinois Council on Long Term Care, Inc, 529 US 1, 13 (2000) (recognizing in a non-APA context that “early review” is available when “the legal question is ‘fit’ for resolution and delay means hardship”), citing Abbott Laboratories, 387 US at 148–49. Lower courts have innovated by including finality as part of the ripeness inquiry in other administrative-review contexts. See, for example, General Electric Co v Environmental Protection Agency, 290 F3d 377, 380 (DC Cir 2002) (including finality as a component of ripeness in a direct-review case). But see Iowa League, 711 F3d at 863 n 12 (finding an action ripe for review without examining whether it qualified as final agency action).
  • 183. See, for example, National Park Hospitality Association, 538 US at 812 (finding that a final action was not ripe for review). But see id at 814–15 (Stevens concurring in the judgment) (finding that the case was ripe for review but voting to order dismissal because petitioner had not alleged sufficient injury); id at 817 (Breyer dissenting) (finding that the case was ripe for review).
  • 184. Courts may be treating the Bell presumption as a clear-statement rule. See note 177 and accompanying text. See also Equal Employment Opportunity Commission v Arabian American Oil Co, 499 US 244, 261 (1991) (Marshall dissenting) (describing the conversion of a presumption to a clear-statement rule as relieving “a court of the duty to give effect to all available indicia of the legislative will”).
  • 185. See Part II.A.
  • 186. Block, 467 US at 349 (explaining that “all presumptions used in interpreting statutes [ ] may be overcome by specific language or specific legislative history that is a reliable indicator of congressional intent”). See also Sackett, 566 US at 128 (“[A]s with most presumptions, this one ‘may be overcome by inferences of intent drawn from the statutory scheme as a whole.’”), quoting Block, 467 US at 349.
  • 187. See Part I.B.
  • 188. 33 USC § 1369(b)(1) (emphasis added). The review statute in the Resource Conservation and Recovery Act of 1976 (RCRA), Pub L 94-580, 90 Stat 2795, codified at 42 USC § 6901 et seq, uses the phrase “review of action of the Administrator in” and gives the DC Circuit exclusive jurisdiction for judicial review. RCRA § 2, 90 Stat at 2827, 42 USC § 6976(a)(1). In American Portland Cement Alliance v Environmental Protection Agency, 101 F3d 772 (DC Cir 1996), the court ruled that it would have jurisdiction “only after the final regulations are promulgated.” Id at 779. It based this decision on the fact that, unlike the CWA, the RCRA statute did not provide for review of determinations. Id at 775. The court held that, by its “plain terms,” the statute provided for review only of the administrator’s action in “the promulgation of final regulations, the promulgation of requirements, and the denial of petitions for the promulgation, amendment, or repeal of RCRA regulations.” Id, citing 42 USC § 6976(a)(1).
  • 189. See Loughrin v United States, 134 S Ct 2384, 2390 (2014) (describing “the ‘cardinal principle’ of interpretation that courts ‘must give effect, if possible, to every clause and word of a statute’”), quoting Williams v Taylor, 529 US 362, 404 (2000).
  • 190. See 33 USC § 1369(b)(1)(B), (D).
  • 191. 15 USC § 717r(d)(1) (emphasis added).
  • 192. In other words, the promulgations, approvals, denials, issuances, and determinations mentioned in 33 USC § 1369(b)(1).
  • 193. See 33 USC § 1369(b)(1).
  • 194. See note 166.
  • 195. 33 USC § 1316(b)(1)(A).
  • 196. 33 USC § 1316(b)(1)(B).
  • 197. 33 USC § 1316(b)(2).
  • 198. See Bell, 461 US at 777–78.
  • 199. See Carolina Power, 43 F3d at 914–15.
  • 200. See 5 USC § 704.
  • 201. See Part III.A.4.
  • 202. See Parts I.B, II.A.1, II.B.
  • 203. See, for example, note 65 and accompanying text.
  • 204. Environmental Policy Division of the Congressional Research Service, 1 A Legislative History of the Water Pollution Control Act Amendments of 1972 819 (Government Printing Office 1973).
  • 205. Id.
  • 206. Id.
  • 207. Id at 823 (emphasis added).
  • 208. See Monterey Coal Co v Federal Mine Safety and Health Review Commission, 635 F2d 291, 292 (4th Cir 1980) (discussing a Senate report stating that “[p]ersons adversely affected by the Commission’s final order” may seek review). See also Bell, 461 US at 778 (highlighting that the statutes in question state that review is available for the “decision” of the board).
  • 209. 1 A Legislative History of the Water Pollution Control Act Amendments of 1972 at 830 (cited in note 204) (“The increase in litigation on environmental matters . . . may well justify a new environmental court.”).
  • 210. Id at 892 (emphasis added).
  • 211. See text accompanying note 136.
  • 212. Champion, 850 F2d at 188. See also 3 A Legislative History of the Clean Water Act of 1977 at 470 (cited in note 136).
  • 213. See 3 Legislative History of the Clean Water Act of 1977 at 470 (cited in note 136).
  • 214. Mims v Arrow Financial Services, LLC, 565 US 368, 385 (2012), citing Consumer Product Safety Commission v GTE Sylvania, Inc, 447 US 102, 118 (1980).
  • 215. GTE Sylvania, 447 US at 117, quoting United States v Price, 361 US 304, 313 (1960).
  • 216. GTE Sylvania, 447 US at 118 n 13.
  • 217. Block, 467 US at 349.
  • 218. 430 US 112 (1977).
  • 219. 445 US 193 (1980) (per curiam).
  • 220. See E.I. du Pont, 430 US at 124.
  • 221. Id at 132–36.
  • 222. Id at 136.
  • 223. Id.
  • 224. E.I. du Pont, 430 US at 136.
  • 225. See Selmi, 72 Ind L J at 78 (cited in note 82).
  • 226. Forest Grove School District v T.A., 557 US 230, 239–40 (2009), quoting Lorillard v Pons, 434 US 575, 580 (1978).
  • 227. Selmi, 72 Ind L J at 74–75 (cited in note 82), quoting 42 USC § 7607(b)(1).
  • 228. 484 US 49 (1987).
  • 229. 33 USC § 1365.
  • 230. Gwaltney, 484 US at 57.
  • 231. Id.
  • 232. Crown Simpson, 445 US at 194.
  • 233. Id. See also 33 USC § 1342(d)(2).
  • 234. Crown Simpson, 445 US at 196.
  • 235. Id.
  • 236. Id.
  • 237. Natural Resources Defense Council, Inc v U.S. Environmental Protection Agency, 673 F2d 400, 405 (DC Cir 1982).
  • 238. The Crown Simpson Court acknowledged that the 1977 amendments were passed after the original controversy in the case occurred. The court did not “consider the impact, if any, of this amendment on the jurisdictional issues presented.” Crown Simpson, 445 US at 194 n 2.
  • 239. The Court also did not raise any jurisdictional issues related to its review of a case in which the Ninth Circuit asserted § 509(b) jurisdiction to review a nonfinal determination. See National Association of Home Builders v Defenders of Wildlife, 551 US 644 (2007). See also note 150 (describing the Ninth Circuit case). But because the Court did not explicitly address the jurisdictional question, its assertion of jurisdiction is not protected by stare decisis. See American Portland, 101 F3d at 776 (“That the court has taken jurisdiction in the past does not affect the analysis because jurisdictional issues that were assumed but never expressly decided in prior opinions do not thereby become precedents.”), citing Brecht v Abrahamson, 507 US 619, 631 (1993).
  • 240. E.I. du Pont, 430 US at 136.
  • 241. Id at 127–28 & n 18.
  • 242. Crown Simpson, 445 US at 196.
  • 243. Id.
  • 244. See note 164 and accompanying text.
  • 245. See note 168 and accompanying text.
  • 246. The factors consist of the statute’s specific language, legislative history, structure, the congressional acquiescence in the Supreme Court’s expansive functional reading of the statute, and the statute’s purpose to provide prompt resolution of challenges to EPA action. See Part II.B.2.
  • 247. See text accompanying notes 33–38.
  • 248. The Third Circuit’s ruling in Pennsylvania Department of Environmental Resources was based on the argument that the district court had proper jurisdiction as the issue was covered under the CWA’s citizen-suit provision. See Pennsylvania Department of Environmental Resources, 618 F2d at 994. As finality was not dispositive, this Section does not try to explain the decision in ripeness terms.
  • 249. American Paper, 882 F2d at 289.
  • 250. Id at 289–90.
  • 251. See Part I.A.2.
  • 252. Abbott Laboratories, 387 US at 148–49.
  • 253. National Pork, 635 F3d at 741.
  • 254. Id at 756.
  • 255. Rhode Island, 378 F3d at 23.
  • 256. Id.
  • 257. Id at 27–28.
  • 258. Id at 28.
  • 259. The court recognized that in some instances, “the costs of finality may outweigh its benefits” in the administrative context. Rhode Island, 378 F3d at 24 (referencing the collateral order doctrine).
  • 260. See Riverkeeper, 806 F3d at 1082.
  • 261. Id.
  • 262. Id at 1080.
  • 263. Letter Brief of the Alabama Department of Environmental Management, Riverkeeper v United States Environmental Protection Agency, Civil Action No 14-13508, *3 (11th Cir filed Aug 7, 2015) (available on Westlaw at 2015 WL 4873278).
  • 264. Courts have recognized the potential for abuse of this situation in the APA context (in which finality is required) and have cautioned that “an administrative agency cannot legitimately evade judicial review forever by continually postponing any consequence-laden action and then challenging federal jurisdiction on ‘final agency action’ grounds.” National Parks Conservation Association v Norton, 324 F3d 1229, 1239 (11th Cir 2003). Construing § 509(b) to allow review of ripe, nonfinal action resolves this same issue in the CWA context.
  • 265. See Riverkeeper, 806 F3d at 1083 (discussing the “‘pragmatic’ interpretation of administrative finality” doctrine).
  • 266. Champion, 850 F2d at 190.
  • 267. Id.
  • 268. Id at 185.
  • 269. See notes 115–30 and accompanying text.
  • 270. 1 A Legislative History of the Water Pollution Control Act Amendments of 1972 at 819 (cited in note 204). See also Crown Simpson, 445 US at 196.
  • 271. Crown Simpson, 445 US at 196.
  • 272. See, for example, Jennifer Nou, Agency Self-Insulation under Presidential Review, 126 Harv L Rev 1755, 1756–64 (2013).
  • 273. Nina A. Mendelson, Regulatory Beneficiaries and Informal Agency Policymaking, 92 Cornell L Rev 397, 451–52 (2007).
  • 274. Id at 451 (discussing expanding citizen suits in the context of enforcement actions).
  • 275. 1 A Legislative History of the Water Pollution Control Act Amendments of 1972 at 819 (cited in note 204).
  • 276. See Abbott Laboratories, 387 US at 140 (finding a preference for judicial review).
  • 277. See American Petroleum Institute v Environmental Protection Agency, 683 F3d 382, 387 (DC Cir 2012) (describing the benefits of declining to review a nonripe action as “potentially eliminating the need for (and costs of) judicial review,” “simplify[ing] the factual context,” avoiding “inefficient and unnecessary piecemeal review,” and ensuring that courts “make decisions only when they have to, and then, only once”) (quotation marks omitted). See also Devia v Nuclear Regulatory Commission, 492 F3d 421, 424 (DC Cir 2007) (explaining that the “unspoken element of the [ripeness] rationale” is “[i]f we do not decide [the claim] now, we may never need to. Not only does this rationale protect the expenditure of judicial resources, but it comports with our theoretical role as the governmental branch of last resort”); McInnis-Misenor v Maine Medical Center, 319 F3d 63, 70 (1st Cir 2003) (noting that “[i]n the fitness inquiry, . . . prudential concerns focus[ ] on the policy of judicial restraint from unnecessary decisions”).
  • 278. The DC Circuit has similarly framed the ripeness inquiry in the APA context. Action Alliance of Senior Citizens of Greater Philadelphia v Heckler, 789 F2d 931, 940 (DC Cir 1986) (“In order to outweigh institutional interests in the deferral of review, the hardship to those affected by the agency’s action must be immediate and significant.”).
  • 279. In APA cases, in which final agency action is required, courts include the finality requirement in its determination of whether an issue is fit for review. See, for example, American Petroleum Institute, 683 F3d at 387 (explaining that the fitness of an issue for review “depends on whether it is purely legal, whether consideration of the issue would benefit from a more concrete setting, and whether the agency’s action is sufficiently final”). The Iowa League court, which declined to conjure up a finality requirement when none existed in § 509(b), framed the fitness criteria as “rest[ing] primarily on whether a case would ‘benefit from further factual development,’ and therefore cases presenting purely legal questions are more likely to be fit for judicial review.” Iowa League, 711 F3d at 867. Because § 509(b) does not contain an explicit finality requirement, this Comment does not adopt the APA-based fitness inquiry. See Thomas v Union Carbide Agricultural Products Co, 473 US 568, 581 (1985) (finding a non-APA claim fit for review because the “issue presented in [the] case is purely legal, and will not be clarified by further factual development”).
  • 280. See Paula Hannaford-Agor and Nicole L. Waters, Estimating the Cost of Civil Litigation, 20 Ct Statistics Project 1, 6–7 (Jan 2013), archived at (finding that discovery and trial costs account for fifty percent of the cost of certain types of civil litigation). See also Emery G. Lee III and Thomas E. Willging, Defining the Problem of Cost in Federal Civil Litigation, 60 Duke L J 765, 783–84 (2010) (observing that a one-unit increase in a subjective one-to-seven scale of factual complexity was associated with an 11 to 13 percent increase in litigation cost); United States v Tohono O’odham Nation, 563 US 307, 315 (2011) (“Developing a factual record is responsible for much of the cost of litigation.”).
  • 281. For whatever reason, courts appear to be more comfortable asking whether an issue is fit for review than whether hearing a case is cost justified. On a related note, one author has proposed that courts impose a finality requirement “to avoid addressing the merits in litigation that would otherwise unnecessarily burden their dockets.” Iyer, Comment, 125 Yale L J at 792 (cited in note 4). This suggestion is consistent with the idea that courts are using Bell to the same effect. See note 184 and accompanying text.
  • 282. Dean Erwin Chemerinsky has questioned whether an issue may be fit for review even if additional factfinding may be needed. He notes that “it is unclear whether a greater hardship might compensate for less in the way of a factual record or vice versa.” Erwin Chemerinsky, Constitutional Law: Principles and Policies § 2.6.3 at 117–18 (Wolters Kluwer 5th ed 2015). Chemerinsky concludes that, unlike the hardship inquiry, the concern about a record is likely “given less weight when there is a compelling need for immediate judicial review.” Id § 2.6.3 at 118. Chemerinsky’s conception of the doctrine suggests that, if the hardship to the parties of withholding review is great, courts should find an action ripe even if it would not otherwise be considered fit for review for lack of an adequate record. Id § 2.6.3 at 117–18. In the same vein, Justice John Paul Stevens left open this possibility by stating that “[i]f there were reason to believe that further development of the facts would clarify the legal question, or that the agency’s view was tentative or apt to be modified, only a strong showing of hardship to the parties would justify a prompt decision.” National Park Hospitality Association v Department of the Interior, 538 US 803, 815 (2003) (Stevens concurring in the judgment). The DC Circuit has also stated that to outweigh any “institutional interests in the deferral of review,” the hardships caused by withholding review must be “immediate and significant.” American Petroleum Institute, 683 F3d at 389, quoting Devia, 492 F3d at 427. See also Iowa League, 711 F3d at 867 (stating that the fitness and hardship prongs are “weighed on a sliding scale”).
  • 283. See Part I.A.2. The presumption that an administrative record is complete further shows that courts would prefer to address the merits rather than open litigation to further factual development. See WildEarth Guardians v Salazar, 670 F Supp 2d 1, 5 (DDC 2009).
  • 284. For a discussion of the pragmatic finality doctrine, see United States Army Corps of Engineers v Hawkes Co, 136 S Ct 1807, 1814–15 (2016). For an explanation of the collateral order doctrine in administrative law, see Federal Trade Commission v Standard Oil Co, 449 US 232, 246 (1980). See also Rhode Island, 378 F3d at 23–25 (recognizing that “preserving crucial collateral claims and avoiding potentially irreparable harm occasionally justifies construing statutorily created finality requirements with a modicum of flexibility seems to apply with equal (or, at least, nearly equal) force to the review of both judicial and administrative orders”). Courts have also created a narrow doctrine allowing for review before exhaustion called the “Leedom exception.” See Scottsdale Capital Advisors Corp v Financial Industry Regulatory Authority, Inc, 844 F3d 414, 420–22 (4th Cir 2016).
  • 285. Courts have held that actions that could lead to future hardship can qualify as final agency action. For example, a jurisdictional determination that “exposes [a party] to double penalties in a future enforcement proceeding” has been found to be a final agency action. Sackett v Environmental Protection Agency, 566 US 120, 126 (2012). See also Rhea Lana, Inc v Department of Labor, 824 F3d 1023, 1032 (DC Cir 2016) (relying on Sackett to reach a similar conclusion regarding a letter classifying a party as subject to civil penalties by the Department of Labor).
  • 286. As described above, courts use these doctrines to deem nonfinal action to be final if they perceive that some legal harm flows from the action itself. When the action is procedural or truly benign in nature, as it was in Metropolitan Edison or in the case of Easterbrook’s nozzle hypothetical, courts tend to hew closely to a formalist interpretation of finality. See National Association of Home Builders v United States Army Corps of Engineers, 417 F3d 1272, 1279 (DC Cir 2005) (noting that “the doctrine of finality would be no more than an empty box if the mere denial of a procedural advantage constituted final agency action subject to judicial review”) (quotation marks omitted).
  • 287. 1 A Legislative History of the Water Pollution Control Act Amendments of 1972 at 819 (cited in note 204).
  • 288. See, for example, Petition for a Writ of Certiorari, National Association of Manufacturers v United States Department of Defense, Docket No 16-299, *3–4 (US filed Sept 2, 2016) (available on Westlaw at 2016 WL 4698748) (explaining how petitioner’s coplaintiffs filed “‘protective’ petitions for review” in a court of appeals).
  • 289. 33 USC § 1369(b)(1).
  • 290. This idea is like the “merger rule” in civil litigation. The merger rule states that “prior interlocutory orders . . . merge with the final judgment . . . [and] may be reviewed on appeal from the final order.” Camesi v University of Pittsburgh Medical Center, 729 F3d 239, 244–45 (3d Cir 2013). Moreover, courts have fashioned an exception to the merger rule to avoid the type of procedural morass described by Easterbrook. The Seventh Circuit has noted that the rule is “inapplicable where adherence would reward a party for dilatory and bad faith tactics.” Sere v Board of Trustees of the University of Illinois, 852 F2d 285, 288 (7th Cir 1988).
  • 291. In other words, the potential litigant also must engage with the second prong of the ripeness analysis—the potential hardship of withholding review.