New Jersey v. Environmental Protection Agency 663 F.3d 1279

TATEL, Circuit Judge:

This is a motion for fees and costs under section 307(f) of the Clean Air Act, which authorizes courts to “award costs of litigation (including reasonable attorney and expert witness fees) whenever [they] determine[] that such award is appropriate.” 42 U.S.C. § 7607(f). In the underlying litigation, movants, a group of Native American tribes and tribal associations, intervened on behalf of petitioners who were challenging EPA rules regulating mercury emissions from power plants. See New Jersey v. EPA, 517 F.3d 574 (D.C.Cir.2008). The Tribes also filed a petition for review challenging an ancillary regulation not at issue here. We vacated the mercury rules because we agreed with petitioners that the rules violated the Clean Air Act. Id. Petitioners sought fees, and EPA agreed to pay. EPA’s Fees Br. 1.

Tribal Intervenors, who also pressed for vacatur, albeit on the basis of different arguments that we never reached, now ask us to order EPA to pay their fees and costs. EPA objects, claiming that Tribal Intervenors are ineligible for fee shifting. EPA also argues that even were Tribal Intervenors eligible, the size of their fee request is unreasonable and should be reduced by more than two-thirds. For the reasons explained below, we conclude that Tribal Intervenors merit a fee award. We decline, however, to weigh in now on the appropriate amount; instead, we direct the parties to our Appellate Mediation Program.

I.

In support of its argument that Tribal Intervenors are ineligible for fees, EPA relies on two cases, Donnell v. United States and Alabama Power Co. v. Gorsuch, in which we limited intervenor fee recovery to circumstances where intervenors had influenced the outcome of the *1281litigation. But these two cases have no applicability here because intervenors in those cases had entered the litigation on behalf of the government — a fact essential to the disposition in both cases. See Donnell v. United States, 682 F.2d 240, 248 (D.C.Cir.1982) (holding that fees should not be awarded under the Voting Rights Act fee-shifting provision where an intervenor entering the litigation on behalf of the Justice Department contributes “nothing of substance in producing th[e] outcome” of the litigation); Alabama Power Co. v. Gorsuch, 672 F.2d 1, 4 (D.C.Cir.1982) (per curiam) (rejecting fee request under Clean Air Act section 307(f) because “[i]f ever an intervenor can recover attorneys’ fees from a party on whose side it participated,” it must at least make a “unique contribution ... to the strength of that party’s legal position”). In Donnell, we explained that when a party has intervened on behalf of the government, the fee-shifting provision’s “objective is far less compelling”: “when the Justice Department defends a suit ... it is acting on behalf of those whose rights are affected.” 682 F.2d at 246. Because “[w]e will not lightly infer that the Justice Department has violated this statutory obligation,” id. at 247, the protection of intervenors’ interests in such cases normally requires no intervention — and thus no fee shifting to incentivize intervention.

This case is very different. Here, the Tribes intervened on the side of petitioners, not the government, and they offered two substantial arguments based on EPA’s alleged failure to consider intervenors’ treaty rights, arguments that petitioners lacked standing to make. True, we never reached the Tribes’ arguments, but that is immaterial. By giving us alternative bases for resolving the case — bases petitioners were unable to offer — Tribal Intervenors contributed to the “ ‘proper implementation and administration of the act or otherwise serve[d] the public interest.’ ” Sierra Club v. EPA 322 F.3d 718, 722 (D.C.Cir.2003) (quoting Ruckelshaus v. Sierra Club, 463 U.S. 680, 687, 103 S.Ct. 3274, 77 L.Ed.2d 938 (1983)). Indeed, if petitioners had been able to make the Tribes’ arguments, the fact that we never reached them would provide “no basis for reducing the[ir] fee.” Kennecott Corp. v. EPA 804 F.2d 763, 766 (D.C.Cir.1986) (per curiam); see also Hensley v. Eckerhart, 461 U.S. 424, 435, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983) (explaining that “[ljitigants in good faith may raise alternative legal grounds for a desired outcome, and the court’s ... failure to reach certain grounds is not a sufficient reason for reducing a fee”); Am. Petroleum Inst. v. EPA 72 F.3d 907, 911 (D.C.Cir.1996) (awarding fees to petitioners for all arguments supporting “the invalidity of the regulation at issue” where the court reached only one of the five arguments).

EPA’s view, embraced by the dissent— that we should allow fee shifting only where an intervenor affected the outcome of a case as determined after the fact— would discourage interventions that play a useful role. “It is usually impossible to determine in advance of trial which issues will be reached or which parties will play pivotal roles in the course of the litigation. To retrospectively deny attorney’s fees because an issue is not considered or because a party’s participation proves unnecessary would have the effect of discouraging the intervention of what in future cases may be essential parties.” Seattle Sch. Dist. No. 1 v. Washington, 633 F.2d 1338, 1349 (9th Cir.1980), aff'd on other grounds, 458 U.S. 457, 102 S.Ct. 3187, 73 L.Ed.2d 896 (1982); see also Am. Petroleum Inst., 72 F.3d at 912 (“It is not necessary that a fee-petitioning client and its attorney have acted with the 20/20 acuity of hindsight in *1282developing their arguments in order to collect attorneys’ fees.”).

Of course, while ineentivizing interventions that contribute to the proper administration of the Act, we want to be sure that we are not simultaneously encouraging fee-seeking interventions. But we believe that courts have all the tools they need to prevent that from happening. First, to intervene a party must have standing in its own right, see Military Toxics Project v. EPA, 146 F.3d 948, 953 (D.C.Cir.1998), and thus be able to file directly as petitioner. Fee seekers who have standing can already file their own suits, so our decision, at most, encourages them to participate as intervenors rather than as petitioners. See Massachusetts v. Microsoft Corp., 373 F.3d 1199, 1235 (D.C.Cir.2004) (“[E]fficiency gains ... ordinarily make intervention worthwhile when there are common issues[.]”); see also King v. Ill. State Bd. of Elections, 410 F.3d 404, 421 (7th Cir.2005) (“[Awarding attorneys’ fees to the intervenors promotes judicial efficiency. Parties ... should be encouraged to intervene in suits such as this one, rather than bringing their own claims in subsequent suits.”). Second, once in the litigation, intervenors may recover fees only insofar as they avoid wasteful duplication of effort. As we have said about awards of costs .pursuant to Federal Rule of Appellate Procedure 39, “insofar as [intervenors’] briefs duplicate what is presented by” the party on whose behalf they have intervened, their “costs are essentially for their own account, a kind of extra insurance for which they pay the premium.” Am. Pub. Gas Ass’n v. FERC, 587 F.2d 1089, 1099 (D.C.Cir.1978). The same applies in the fee-shifting context, as other circuits have held. See, e.g., Shaw v. Hunt, 154 F.3d 161, 168 (4th Cir.1998) (“Courts should certainly deny fees to meddlesome or officious intervenors whose services have been counterproductive or have, at most, been duplicative of work better left to plaintiffs counsel.”). Thus, intervenors who offer duplicative arguments in violation of our circuit rules, Handbook of Practice and Internal Procedures of the United States Court of Appeals for the District of Columbia Circuit 37 (2011), may not recover for those efforts. Here, intervenors wasted no time duplicating petitioners’ arguments. EPA’s Fees Br. 6. Instead, they focused on arguments petitioners did not and could not make. In our view, their contribution is precisely the type section 307(f) seeks to incentivize. Third, our case law protects against wasteful litigation by barring recovery of fees for arguments not reached where those arguments are frivolous, and so, from the outset, added nothing. Am. Petroleum Inst., 72 F.3d at 912 (“[S]ome issue might be so frivolous that all time spent on it was unreasonable[.]”). In this case, EPA never claims that Tribal Intervenors’ arguments were frivolous. Nor could it, for Tribal Intervenors raised weighty issues with which EPA vigorously engaged. See EPA’s Merits Br. 68-80, 86-98. Finally, under Clean Air Act section 307(f), courts retain broad authority to deny fees when they deem them not “appropriate” — authority courts can exercise in a way that ensures that intervenors get fees only where, aside from its impact on the litigation’s outcome, intervention assists the judicial process.

II.

The dissent argues that Tribal Intervenors’ challenge was “based on a different claim for relief.” Dissenting Op. at 1287. The dissent is mistaken. Petitioners challenged two regulations, the Delisting Rule, 70 Fed.Reg. 15,994 (Mar. 29, 2005), and the Clean Air Mercury Rule, 70 Fed.Reg. 28,606 (May 18, 2005), claiming that both violated Clean Air Act section 112. In *1283support, they argued that the regulations ran afoul of the statute’s plain language and were arbitrary and capricious. Tribal Intervenors challenged the exact same regulations, arguing that they violated the exact same section of the exact same statute. Embracing (but not repeating) petitioners’ arguments, they added their • own — that the regulations violated the statute’s plain language and were arbitrary and capricious because of their effect on tribal fishing rights. If accepted, their arguments would have produced the exact same relief that petitioners sought, i.e., vacatur of the regulations. See Tribal Merits Br. 44 (“Tribal [Movants] ask that this Court: (1) vacate the section 112(n) Revision Rule; (2) vacate the Clean Air Mercury Rule[.]”).

Under American Petroleum, these arguments are all in support of a single claim. There, petitioners challenged EPA regulations on five separate grounds. Am. Petroleum Inst., 72 F.3d at 911. Much like the parties here, they argued that EPA violated the Clean Air Act, failed to consider certain environmental impacts, and acted arbitrarily and capriciously. See Am. Petroleum Inst., 52 F.3d 1113, 1116 (D.C.Cir.1995). In the end, we reached only one of their arguments, invalidating the regulations under the statute’s plain meaning. Am. Petroleum Inst., 72 F.3d at 911. Rejecting EPA’s contention that the non-dispositive arguments were in fact distinct claims, we explained: “Petitioners did not raise any claims distinct and separate from the one on which they prevailed. They pursued only one claim for relief— the invalidity of the regulation at issue. They argued five defensible bases for that invalidity.” Id. That is precisely the case here. Petitioners and Tribal Intervenors “pursued only one claim for relief — the invalidity of the regulation at issue,” id. And, as in American Petroleum, they offered several “defensible bases for that invalidity,” id.

According to the dissent, “[i]t is difficult to see how an argument is especially helpful ... when it in no way contributes to the resolution of a case.” Dissenting Op. at 1287 (internal quotation marks omitted). But our case law is to the contrary. In American Petroleum, we awarded fees for each argument not reached, including the narrowest, finding that none was “so frivolous that all time spent on it was unreasonable.” 72 F.3d at 912. In doing so, we recognized that alternative arguments, though ultimately not dispositive, may nonetheless be helpful to the judicial process and thus to “the proper implementation and interpretation of the Act.” Id. at 911. In Alabama Power, we even awarded fees for an argument petitioners withdrew because “had it remained in the case, [the argument’s] resolution would have contributed importantly to the administration of the Act.” 672 F.2d at 4-5 & n. 18. “[W]e will not use hindsight to deny an otherwise appropriate recovery.” Id at 5 n. 18. Indeed, “narrow” arguments of the kind belittled by the dissent, Dissenting Op. at 1287, can be especially helpful, offering the court a basis for a disposition that makes as little law as possible, best preserves agency discretion, or otherwise promotes the purposes of the Clean Air Act. We cannot imagine why we would want to discourage that kind of assistance by drawing an arbitrary line between petitioners and intervenors.

Here, had we not disposed of the case on the basis of petitioners’ contentions, the Tribes’ arguments, which sought to force EPA to comply with its Clean Air Act obligations, would, like petitioners’ withdrawn arguments in Alabama Power, have “contributed importantly to the administration of the Act,” 672 F.2d at 5 n. 18. And rather than summarily dismissing *1284these supposedly “anemic” arguments, Dissenting Op. at 1285, EPA devoted twenty-four pages of its brief responding to them. EPA’s Merits Br. 86-98 (Part IV, arguing entirely that “EPA Properly Considered Tribal Treaties in the Section 112(n) Rule”); id. at 68-80 (defending itself against “Tribal Petitioners^] contention] that EPA’s freshwater health hazard assessment” is arbitrary and capricious). Only by measuring Tribal Intervenors’ contribution against our ultimate decision — a measure we have rejected for evaluating a petitioner’s eligibility for fees — could one discount the significant role they played in this litigation.

Further attempting to marginalize Tribal Intervenors’ contribution, the dissent criticizes them for “br[inging] a narrow, fact-based challenge to an ancillary regulation,” unlikely to. “play a critical role in this case.” Dissenting Op. at 1287. But this conflates the Tribes’ role as intervenors, in which they challenged the Delisting and Mercury Rules, and their role as petitioners, in which they challenged a thud regulation, the ancillary Reconsideration Rule. Tribal Intervenors seek no fees for the latter. They seek fees only with respect to the vacatur of the Delisting and Mercury Rules — the regulations lying at the very heart of this case. See Tribal Fees Br. 3-4.

The dissent claims that today’s decision takes an “extraordinary step.” Dissenting Op. at 1284. Again, the dissent is mistaken. Other courts have awarded fees in similar circumstances. For instance, the Fourth Circuit rejected an outcome-based rule (like the one urged by the dissent), awarding fees to intervenors who, though ultimately found to lack standing, had nonetheless “play[ed] an active role” in the litigation. Shaw, 154 F.3d at 163. The Ninth Circuit likewise awarded fees to intervenors whose arguments it never reached. Seattle Seh. Dist. No. 1, 633 F.2d at 1350. In that case, in which plaintiffs challenged an initiative banning race-conscious school desegregation, the district court bifurcated the litigation: Phase I would consider plaintiffs’ argument that the initiative was unconstitutional, and Phase II would consider a separate argument — made by intervenors — that the school districts were operating racially dual school systems. Id. at 1341. Because plaintiffs prevailed at trial, the court never even had to hear intervenors’ Phase II arguments. The court nonetheless awarded fees for the “substantial time and effort” intervenors spent “preparing] for trial on the Phase II issues,” declining to “retrospectively deny attorney’s fees” solely because intervenors’ participation ultimately “prove[d] unnecessary.” Id. at 1349-50.

III.

Tribal Intervenors seek a total of $305,389 in fees and costs. EPA insists that the request is excessive and should be reduced to $64,793. Rather than sorting through the parties’ competing claims, we direct the question to our Appellate Mediation Program. See, e.g., Clifton Power Corp. v. FERC, No. 94-1775 (D.C.Cir. Feb. 12, 1997) (per curiam order referring consideration of motion for attorney fees to the Appellate Mediation Division).

So ordered.

BROWN, Circuit Judge,

dissenting:

In today’s decision, the court takes the extraordinary step of awarding fees to Tribal Intervenors, who intervened on the side of a petitioner in a Clean Air Act challenge and offered an argument that other petitioners lacked standing to make. As the court acknowledges, we never reached the Tribal Intervenors’ claim and their argument had no effect on the out*1285come of the litigation. Nevertheless, the court finds the Petitioners are entitled to a fee award for contributing to the “proper implementation and administration” of the Act or otherwise “serv[ing] the public interest.” (Op. at 1281).

Section 807(f) of the Clean Air Act, 42 U.S.C. § 7607(f), permits the award of attorneys’ fees in certain proceedings “whenever [the court] determines that such award is appropriate.” In an early case dealing with this fee provision, Judge Wilkey warned that such broad and ill-defined authority was likely to lead to an “all-but-the-frivolous” standard — a result he considered deeply at odds with Congressional intent. Alabama Power v. Gorsuch, 672 F.2d 1, 16-17 (D.C.Cir.1982) (Wilkey, J., dissenting). “If Congress meant for ‘appropriate’ to mean that all non-frivolous [applicants] would recover, it surely would have said so.” Id. Despite Judge Wilkey’s pessimism, this Court initially resisted the temptation to use the “vagueness of the statute ... to validate the whim of any judge.” Id. at 16. And other courts, interpreting similarly worded provisions, have generally avoided establishing such a dramatic alteration in the fee award structure, concluding that intervenors are eligible for fees only when they play a “significant role in the litigation.” See Shaw v. Hunt, 154 F.3d 161, 168 (4th Cir.1998); Wilder v. Bernstein, 965 F.2d 1196, 1204 (2d Cir.1992); Grove v. Mead Sch. Dist. No. 351, 753 F.2d 1528, 1535 (9th Cir.1985); see also Donnell v. United States, 682 F.2d 240, 248 n. 16 (D.C.Cir.1982) (noting even a prevailing party would not be entitled to fees where it merely caught “a train on its way out of the station,” played “no part in firing the boiler, getting up a head of steam, or opening the throttle,” but “just went along for the ride”).

In Shaw, for example, the court concluded that a fee award was appropriate in the exceptional circumstances where intervenors, deprived of standing during the course of the litigation, had taken an active, unique role in the litigation and contributed to its success. Shaw, 154 F.3d at 166. However, the court emphasized the narrowness of its holding: “Moreover— and this is a point we stress — it is not every permissive intervenor who will be entitled to fees. Courts should deny fees to intervenors who have failed to play a ‘significant role in the litigation.’ ” Id. at 168. Similarly, in Seattle School Dist. No. 1 v. Washington, 633 F.2d 1338, 1341 (9th Cir.1980), plaintiff school districts and intervenors challenged Initiative 305, an anti-busing initiative. Although the court concluded that a fee award was permissible for an issue that had not been fully litigated, it pointedly avoided making any broad generalization about the propriety of such an award under different circumstances. In fact, the court declined to award fees for intervenors’ participation in Phase I of the litigation — during which the case was resolved — because the school districts were fully capable of litigating the unconstitutionality of Initiative 305 and the intervenors’ role was de minimis. Id. at 1349. By contrast, the court found intervenors’ substantial time and effort spent preparing for Phase II could be compensated because the intervenors’ essential role in that phase of the litigation was “apparent from the onset of [the] case.” Id. If Initiative 350 was found to be constitutional in Phase I, which was a “substantial likelihood,” the school districts would be unwilling to argue they were operating discriminatory school districts. Id. at 1349-50. In that case, the entire burden of litigating Phase II would have fallen on intervenors. Id. at 1349. Tellingly, no other court has ever determined an intervenor merited a fee award for the sort of anemic effort the court rewards today. After more than two decades, the “all-but-*1286the-frivolous” standard Judge Wilkey feared has become a reality in this circuit.

Apparently, the court opts for such a radical departure simply because it can. The court says our precedent — which has consistently imposed the more exacting “significant role” standard- — -is distinguishable. That much is true. In Alabama Power v. Gorsuch, 672 F.2d at 4, we rejected a fee request under section 307(f) of the Clean Air Act — the same section at issue here — because an intervenor seeking to recover fees from the party on whose side it participated had failed to make any “unique contribution ... to the strength of that party’s legal position.” In Donnell v. United States, 682 F.2d at 248-49, we rejected a request for fees under the Voting Rights Act where an intervenor on behalf of the Justice Department had “contributed nothing the Government did not also contribute.” That case focused heavily on the fact that the party had intervened on behalf of the government, a circumstance that made the fee-shifting provisions’ objectives “far less compelling.” Id. at 246. But it also endorsed the broader principle that “[i]f a lawsuit is successful, but the intervenor contributed little or nothing of substance in producing that outcome, then fees should not be awarded.” Id. at 248. I find the analytical force of those earlier precedents — -particularly the principles articulated in Donnell — undiminished by the factual differences in this case.

The broad grant of statutory discretion in Section 307(f) has usually been interpreted as a narrow exception to the traditional rule that only prevailing parties are entitled to fees. “Section 307(f) was meant to expand the class of parties eligible for fee awards from prevailing parties to partially prevailing parties — parties achieving some success ...” Ruckelshaus v. Sierra Club, 463 U.S. 680, 682-88, 103 S.Ct. 3274, 77 L.Ed.2d 938 (1983). But the Court was careful to note that neither “trivial success on the merits” nor “purely procedural victories” would automatically justify fee awards. Id. at 688 n. 9, 103 S.Ct. 3274. Tribal Intervenors can claim only a procedural victory. As the government points out, they challenged a regulation that was eventually overturned solely on grounds raised by other parties. This makes their claim much more akin to the wholly unsuccessful plaintiffs whose claim was rejected in Ruckelshaus. More significantly, nothing about Tribal Intervenors’ fact-based challenge to the EPA’s compliance with treaty fishing rights had any impact on EPA’s authority to administer and implement the delisting rule under Section 7412(c)(9) — the basis on which the case was resolved.

We have held that the court need not parse the success of each separate argument in support of a single claim when determining eligibility for fees. See Am. Petroleum Inst. v. EPA 72 F.3d 907, 912 (D.C.Cir.1996). But that does not mean every separate claim — even one completely unrelated to the successful strategy — must be deemed a success because it sought the same remedy. New Jersey and fourteen additional states and environmental organizations challenged the “Delisting Rule,” which had removed power plants from the list of sources whose emissions are regulated under Section 7412. See 70 Fed.Reg. 15,994 (Mar. 29, 2005). Tribal Intervenors elected not to file a petition for review of the Delisting Rule, but rather challenged the ancillary “Reconsideration Rule,” a proceeding in which the EPA made two substantive changes to the Clean Air Mercury Rule but no substantive change to the Delisting Rule. See 71 Fed.Reg. 33,388 (June 9, 2006). They participated in the challenge to the Delisting Rule only as intervenors, and made no arguments with respect to the determinative issue in the case. Tribal Intervenors’ claim, a fact-*1287based challenge to the Reconsideration Rule based on treaty rights, was not resolved by the Court, even though the Court ultimately granted the remedy they sought. See City of Sherrill v. Oneida Indian Nation of N.Y., 544 U.S. 197, 213, 125 S.Ct. 1478, 161 L.Ed.2d 386 (2005) (explaining difference between a claim and a remedy). Intervenors should not be rewarded simply for challenging the same regulation based on a different claim for relief arising out of a different legal theory. See Hensley v. Eckerhart, 461 U.S. 424, 434-35, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983); Sierra Club v. EPA, 769 F.2d 796, 801-02 (D.C.Cir.1985).

The court attempts to justify its decision by claiming that denying attorneys’ fees to Tribal Intervenors would discourage future intervenors from bringing “useful” claims (Op. at 1281). Its concern is overblown. While it is admittedly impossible for parties to determine in advance what issues will be reached by the court, it is entirely possible for a litigant to make a reasonable estimation of how large an impact his issue is likely to have on the litigation, in light of the other challenges being raised. See Seattle Sch. Dist. No. 1, 633 F.2d at 1349-50 (noting there was a “substantial likelihood” that the initiative at issue would be found constitutional and that intervenors’ role would therefore be critical). In this case, the petitioners raised a sweeping challenge to the EPA’s authority to administer and implement the Delisting Rule — a claim that, if successful, would resolve all of Tribal Intervenors’ issues. The likelihood that Tribal Intervenors, who brought a narrow, fact-based challenge to an ancillary regulation, would play a critical role in this case was infinitesimally small from the outset. Refusing to reward them for their decision to pile onto the petitioners’ different- — -and much more substantial — claim would cause little danger of discouraging “useful” interventions in the future. It would merely force potential intervenors to conduct a basic cost-benefit analysis to determine whether their claim is sufficiently likely to make an actual impact to justify the risk they "will bear their own costs. While “20/20 acuity of hindsight” isn’t required, Am. Petroleum Inst., 72 F.3d at 912, willful blindness should not be permitted.

The court also claims that Tribal Intervenors assisted the process by providing an alternative basis for its disposition (Op. at 1283). It is difficult to see how an argument is “especially helpful,” however, when it in no way contributes to the resolution of a case. In fact, tangential arguments piled on by self-interested intervenors force the agency — and the court — to waste valuable resources evaluating and addressing arguments that will have no impact on the court’s ultimate decision. The Court seems to forget that the EPA must respond to all arguments lest it be deemed to have conceded them. Here, the EPA spent twenty-four pages of its reply brief addressing arguments made by Tribal Intervenors, wasting valuable resources that might otherwise have been devoted to rebuttal of the petitioners’ much more substantial- — -and ultimately victorious— claims. Were Tribal Intervenors’ arguments costless, it might be tempting to grant their request. However, “we live in a world of scarce resources and the question inevitably becomes how best to allocate them.” Alabama Power, 672 F.2d at 29 (Wilkey, J., dissenting). In this case, it was clear from the outset that Tribal Intervenors’ fact-based objection to the ancillary Reconsideration Rule would not play a substantial role in petitioners’ sweeping and substantial challenge to the overarching Delisting Rule. Moreover, allowing all but the most frivolous or duplicative intervenors to recover attorneys’ fees encourages “additional filings of dubious value in *1288suits already of notorious complexity.” Alabama Power, 672 F.2d at 30 (Wilkey, J., dissenting).

The court seeks to reassure us that its holding will not encourage fee-seeking interventions, pointing out “all the tools” courts have to prevent that from happening (Op. at 1282). The tools, though, turn out to be few and frail. The majority overstates the value of the standing requirement because it fails to recognize the substantial practical difference between participating in litigation as a petitioner or as an intervenor. Our litigation system forces the petitioner to literally “put his money where his mouth is” by bearing all costs associated with his suit. Even where Congress departs from the “American rule” by allowing recovery of attorneys’ fees, a petitioner must attain some success on the merits of its claim to warrant reimbursement. Ruckelshaus, 463 U.S. at 682-84, 103 S.Ct. 3274. Thus, a rational plaintiff will bring suit if and only if the expected judgment would be at least as large as his expected legal costs, i.e. the total legal costs discounted by his probability of losing at trial. See Steven Shavell, Suit, Settlement, and Trial: A Theoretical Analysis Under Alternative Methods for the Allocation of Legal Costs, 11 J. Legal Stud. 55, 58 (1982). Because intervenors bear far fewer costs — and thus shoulder far less risk — than petitioners, a party with a marginal claim would be substantially more likely to intervene than it would be to file suit in its own right. The majority skews the calculus even further by allowing an intervenor to hitch its completely unrelated claim to a promising challenge to the same regulation. Its holding encourages parties to pile on claims that are not sufficiently meritorious to justify filing in their own right.

The majority conflates, and ultimately eviscerates, the bar against “frivolous” litigation and the court’s discretion to determine what is “appropriate.” Under the court’s newly minted standard, it is no impediment to a fee award that an intervenor’s argument is irrelevant to the outcome, nonsubstantive, and does nothing to strengthen the primary legal position. The court defines “appropriate” so broadly that an intervenor is now entitled to fees unless the challenge is patently frivolous.

It is hard to see what is inappropriate about requiring intervenors to carefully consider what is likely to enhance and promote the purposes of any given litigative effort. Creating a risk — even a small one — that intervenors will be required to bear their own costs will at least force them to undertake a basic cost-benefit analysis to determine when to avoid spending money. See Richard L. Revesz and Michael A. Livermore, Retaking Rationality 12 (2008) (“In the absence of an obvious endpoint [for spending], we need a mechanism that tells us when to stop spending money. Cost-benefit analysis is that mechanism[.]”). And if parties want to pile on just for the sake of piling on — to pursue some frolic of their own — why would it be inappropriate to require them to pay the costs of that indulgence?

Since preserving the public fisc from unreasonable depredations also serves the public interest, I would not be so eager to find new ways to waste Other People’s Money.

I dissent.