View PDF Version

Nos. 07-588, 07-589 and 07-597

 

In the Supreme Court of the United States

ENTERGY CORPORATION, PETITIONER

v.

RIVERKEEPER, INC., ET AL.

PSEG FOSSIL LLC, ET AL., PETITIONERS

v.

RIVERKEEPER, INC., ET AL.

UTILITY WATER ACT GROUP, PETITIONER

v.

RIVERKEEPER, INC., ET AL.

ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE SECOND CIRCUIT

REPLY BRIEF FOR THE FEDERAL PARTIES
AS RESPONDENTS SUPPORTING PETITIONERS

 

GREGORY G. GARRE
Solicitor General
Counsel of Record
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217

In the Supreme Court of the United States

No. 07-588

ENTERGY CORPORATION, PETITIONER

v.

RIVERKEEPER, INC., ET AL.

No. 07-589

PSEG FOSSIL LLC, ET AL., PETITIONERS

v.

RIVERKEEPER, INC., ET AL.

No. 07-597

UTILITY WATER ACT GROUP, PETITIONER

v.

RIVERKEEPER, INC., ET AL.

ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE SECOND CIRCUIT

REPLY BRIEF FOR THE FEDERAL PARTIES
AS RESPONDENTS SUPPORTING PETITIONERS

The court of appeals held that the Environmental Pro tection Agency (EPA) may take cost-effectiveness but not cost-benefit considerations into account under Section 316(b) of the Clean Water Act (CWA or Act), 33 U.S.C.1326(b). The States assert that costs may be a secondary but not a primary consideration under Section 316(b). And Riverkeeper contends that costs may be taken into account only if the environmental consequences of adopting the less expensive technology are de minimis. Because nothing in the statute draws any of those distinctions or otherwise unambiguously precludes consideration of the relationship between costs and benefits in establishing appropriate water-intake standards under Section 316(b), EPA's deter mination of how best to consider costs and benefits is enti tled to deference under the principles set forth in Chevron U.S.A. Inc. v. NRDC, 467 U.S. 837, 842 (1984).

Respondents lean heavily on the text, structure, and legislative history of different best-technology provisions that govern the discharge of pollutants. Some of those pro visions, however, expressly require cost-benefit analysis. Those provisions confirm that Congress regarded a com parison of costs and benefits as potentially relevant to the determination of which technology is "best." Moreover, those discharge provisions are far more specific than Sec tion 316(b)'s single terse sentence, which confers broad gap-filling authority on the agency to determine whether and how to consider the relationship between costs and benefits in addressing the distinct and unique issue of water intake. The court of appeals erred in depriving EPA of that statutory authority.

A. The Text Of Section 316(b) Does Not Unambiguously Pre clude Consideration Of The Relationship Between Costs And Benefits

Section 316(b) requires that "the location, design, con struction, and capacity of cooling water intake structures reflect the best technology available [(BTA)] for minimiz ing adverse environmental impact." 33 U.S.C. 1326(b) (em phases added). That standard does not unambiguously preclude EPA from considering the relationship between costs and benefits-especially considering that Congress did not define any of the key statutory terms or otherwise specify the factors the agency may or must consider. See Pet. App. 20a1; Gov't Br. 15-18.

1. Respondents observe that "best" often means "sur passing all others in excellence," and that Section 316(b) refers to the best technology available for the goal of "mini mizing" adverse environmental impact. Riverkeeper Br. 24, 25 (quoting American Heritage Dictionary of the Eng lish Language 173 (4th ed. 2006)); see States Br. 19. Those observations do not respond to the government's point that, as a matter of common usage, the "best" way of pursuing a goal is not necessarily the one that most single-mindedly achieves that goal at all costs. Gov't Br. 15-16. For exam ple, the "best" way to commute to and from work might not be the quickest and most direct route on a map. That route might be more dangerous than others, more prone to traffic jams, or more expensive (for example, it might require pay ment of a toll). Similarly, the "best" way to win a game does not typically entail violating the rules, even if cheating would improve one's odds of winning, because the rules and other values matter as well. And the "best" way to catch fish is not necessarily the one that nets the most fish in the shortest period of time; to many, fly fishing has offsetting advantages. Ibid.

The States (at 26) object to the first of those examples on the ground that, if one's objective is to commute as quickly or safely as possible, the quickest or safest route is necessarily the "best" method of achieving that goal. But even if speed is the stated objective, a commuter would not ordinarily consider hiring a helicopter or professional race car driver or leaving home at 4:00 a.m. to be the "best" way to travel to work. And if safety is the ultimate goal, one would not ordinarily describe acquiring a tank or walking instead of driving ten miles as the "best" approach. In stead, even when a goal is defined as narrowly as commut ing as quickly or safely as possible, additional practical con siderations such as costs and other countervailing factors are often relevant, in daily life and ordinary English usage, to the selection of the "best" approach.

Moreover, the statute refers to the best technology "for," rather than the technology that is best "at," the achievement of the stated objective. 33 U.S.C. 1326(b). That choice of words reinforces the conclusion that EPA is authorized to take account of competing considerations in determining which technology is "best." For example, an individual regarded as the best person at his trade might not be the best person for a particular job, depending on a range of considerations. Gov't Br. 16. The States respond (at 28) that "if a particular job calls only for a person with superior skills for repairing shoes, then the best cobbler at his trade would also be the best person for the job." But that blinks the reality that other considerations, such as honesty and salary demands, normally matter as well.

2. In any event, Section 316(b) does not refer to the best technology for reducing adverse environmental impact "to the lowest level possible," as the States claim (at 24). Instead, the statute refers to "the best technology available for minimizing adverse environmental impact." 33 U.S.C. 1326(b) (emphases added). Even the court of appeals rec ognized that the availability of a particular technology de pends in part on its cost. Pet. App. 24a. The court erred, however, in holding that the statute unambiguously re stricts EPA to considering whether a technology's cost could be "reasonably borne by the industry." Ibid. Even considering the term "available" in isolation, many people would not think of a luxury item as being "available" simply because its purchase would not bankrupt them. Gov't Br. 17; see Riverkeeper Br. 26 ("There is, to be sure, some po tential ambiguity in terms of what the word 'available' means."). The States argue (at 28) that an item's cost is relevant to its availability only "if the person [is] not re quired to obtain the item." But that merely begs the ques tion of what is required here.

The States also suggest (at 29) that, in determining availability, EPA might consider cost "at some secondary level," but not at "the forefront of its technology evalua tion." The States' apparent recognition that costs may be considered to some degree, however, cannot be reconciled with their contention that Section 316(b) unambiguously precludes the approach that EPA has chosen. Section 316(b) does not distinguish, much less unambiguously dis tinguish, between consideration of cost as a "secondary" factor and consideration of cost as a "primary" decision- making criterion. And, absent contrary specification in the statutory text, the choice of the appropriate degree or man ner of considering a permissible factor involves a classic exercise of agency discretion. See Gov't Br. 39.

Moreover, technology must be available for "minimiz ing" adverse environmental impact. 33 U.S.C. 1326(b). Respondents argue that "'minimize' means 'to reduce to the smallest possible number, degree, or extent.'" States Br. 19 (quoting Webster's Third New Int'l Dictionary of the English Language 1438 (1981)). As the government's open ing brief explained (at 17), however, the terms "minimal" and "minimize" often refer to a lesser degree of reduction. Thus, if a person said that he was trying to minimize the risk of being hit by a car while crossing a street, he presum ably would not mean that he was staying inside his house at all times. Instead, the person presumably would mean that he was trying to reduce that risk consistent with other practical considerations-including economic ones such as the need to travel to work-and thus was taking lesser pre cautions such as looking both ways before crossing streets. Ibid.

While the States complain (at 29) that the government's opening brief cited only "a single source" for the proposi tion that "minimize" commonly refers to reductions that are not to the lowest possible level, that brief cited two per fectly authoritative sources: American Heritage Dictio nary 123 (4th ed. 2006), and Black's Law Dictionary 1016 (8th ed. 2004). Gov't Br. 17. Another example is River keeper's brief, which ultimately concedes (at 29) that "the Agency has some discretion (albeit not boundless) to deter mine that further differences in reduction would be so mi nor as to be unnecessary for compliance with the minimiz ing requirement." Riverkeeper thereby acknowledges that a particular technology can "minimiz[e]" environmental impacts, within the meaning of Section 316(b), even if it does not reduce those impacts to the smallest possible level.

The States also claim (at 19-20) that EPA's Phase I reg ulations adopt their definition of "minimize." In fact, those regulations define "minimize" to mean "to reduce to the smallest amount, extent, or degree reasonably possi ble." 40 C.F.R. 125.83 (emphasis added). Reasonableness, of course, depends on a range of considerations. See, e.g., United States v. Banks, 540 U.S. 31, 35-36 (2003). Thus, EPA determined in the rulemaking below that the appro priate "degree" of reduction may depend in part on "the relationship between costs and benefits." Pet. App. 356a.2

3. The States suggest (at 22, 25) that the standard ar ticulated in Section 316(b) is an "indivisible term of art" and that any ambiguities in the individual words are therefore irrelevant. The BTA standard, however, was by no means a term of art when Congress enacted it; instead, Congress appears to have articulated that standard for the first time in Section 316(b). And while Riverkeeper colorfully argues (at 23) that the government does not rely on "primary, sec ondary, or even tertiary definitions of the relevant terms," that is manifestly untrue, as discussed above. The fact that the critical statutory terms can and do take on different meanings in different contexts only underscores the stat ute's ambiguity. What matters, for Chevron deference pur poses, is that nothing unambiguously compels the alternate usages proffered by respondents. See, e.g., EEOC v. Com mercial Office Prods. Co., 486 U.S. 107, 115 (1988).

B. The Statutory Structure, Context, And History Confirm That EPA May Consider The Relationship Between Costs And Benefits In Establishing Water-Intake Standards Un der Section 316(b)

1. Section 316(b) cross-references Sections 301 and 306 of the CWA by specifying that standards established pursu ant to those sections, which govern the discharge of pollut ants, must require that intake structures reflect BTA. 33 U.S.C. 1326(b). Sections 301 and 306 include numerous other best-technology standards. For all of those stan dards, Congress expressly required EPA to consider costs in determining what technologies are "best." And for two of those standards, Congress specifically required EPA to consider the relationship between costs and benefits in identifying the "best" technologies. Gov't Br. 19-21.

Those express statutory mandates strongly support EPA's determination that consideration of the relationship between costs and benefits is permissible under Section 316(b). In particular, those provisions refute Riverkeeper's notion (at 21) that a comparison between costs and benefits is generally inconsistent with the application of a best-tech nology standard. And Congress's decision to specify the factors that EPA must consider under the various best- technology standards that govern the discharge of pollut ants under Sections 301 and 306, but not under the differ ent best-technology standard that governs the intake of water under Section 316(b), confirms that Congress in tended to grant broad gap-filling authority to the agency to interpret and implement Section 316(b)'s terse and unique provision governing water intake. Gov't Br. 18-26.

Respondents fail to articulate a coherent and consistent theory as to the relevance of the cross-referenced sections to the proper interpretation of Section 316(b). They insist that, "[b]ecause section 316(b) addresses intake structures, not effluent [discharges], it stands apart from the statutory provisions elsewhere in the Act that govern effluent limita tions." States Br. 5; see id. at 32; Riverkeeper Br. 37 n.19; id. at 40 (suggesting that the Section 301 standards are "unrelated" to Section 316(b)). But they nonetheless argue at length (e.g., States Br. 34-36) that, by expressly requir ing consideration of the relationship between costs and benefits under two of the cross-referenced standards, Con gress unambiguously prohibited such consideration under Section 316(b).

Neither of those conclusions logically follows from a comparison between Section 316(b) and the cross-refer enced provisions of the Act. Each of the cross-referenced best-technology standards in Sections 301 and 306 identi fies specific lists of factors that EPA must consider. And as noted, those lists expressly require cost-benefit analysis as one of multiple mandated considerations for some but not all of those standards. See 33 U.S.C. 1314(b)(1)(B), (2)(B) and (4)(B); see also Gov't Br. 19-21.

For intake restrictions, however, Congress enacted only the one-sentence Section 316(b), and it conspicuously de clined to provide an additional list of factors that EPA is required, permitted, or forbidden to consider. Respondents are therefore incorrect in stating that "[t]hroughout the CWA, Congress told EPA when and how it could consider costs." States Br. 16. For water-intake limitations-unlike pollutant-discharge restrictions-Congress provided no specific direction either way beyond the ambiguous BTA standard. The only reasonable conclusion from that statu tory scheme is that Congress thereby delegated broader gap-filling and interpretive authority to EPA for water- intake limitations under Section 316(b) than for pollutant- discharge restrictions under Sections 301 and 306. Gov't Br. 22-23.

The States nonetheless argue (at 35) that "the reference in section 316(b) to section 306 (the section applicable to new rather than existing facilities) confirms Congress' in tent to eliminate EPA's authority to make cost-benefit com parisons when it sets new source performance standards." There is no textual basis whatsoever for treating Section 316(b)'s BTA standard for intake limits as being coexten sive with Section 306's best available demonstrated control technology (BADT) standard for the discharge of pollutants from new sources. See 33 U.S.C. 1316(a)(1). Indeed, the States elsewhere acknowledge (at 33) that, in selecting dif ferent standards for the various limitations under Sections 301, 306, and 316, Congress obviously "intended to give each standard different meaning in its application." Espe cially considering that Section 316(b) addresses a different subject matter than Sections 301 and 306 (intake of water as opposed to discharge of pollutants) and establishes a different standard, there is no reason to believe that Con gress intended to equate BTA with any one of the Section 301 or 306 standards.

There is even less basis for construing Section 316(b) as unambiguously mirroring BADT as opposed to one of the other cross-referenced standards, such as the best prac ticable control technology (BPT) or best conventional pollu tion control technology (BCT) standards-standards under which Congress expressly required consideration of the relationship between costs and benefits. See 33 U.S.C. 1311(b)(1)(A) and (2)(E); 33 U.S.C. 1314(b)(1)(B) and (4)(B); see also Gov't Br. 19, 20. The States contend (at 36) that Congress did not evince an "unambiguous intent" to pat tern BTA on BCT. But that is the government's point-the statute is ambiguous on this issue.3

The States also contend (at 36-37) that Section 316(a) does not permit consideration of the relationship between costs and benefits. That provision states that, if a Section 301 or 306 discharge limitation on heat would otherwise be "more stringent than necessary to assure the pro[t]ection and propagation of a balanced, indigenous population of shellfish, fish, and wildlife," a permitting authority "may impose" a Section 301 or 306 limitation "that will assure the protection and propagation of a balanced, indigenous popu lation of shellfish, fish, and wildlife." 33 U.S.C. 1326(a). Section 316(a) thereby authorizes EPA and state permit ting authorities to grant variances from Section 301 and 306 limitations on discharges of heat. While it is true that Sec tion 316(a) variances are not based on cost considerations, that hardly means that Section 316(b)-which sets forth a different standard for regulating a different matter and does not cross-reference Section 316(a)-unambiguously precludes cost-benefit analysis.

2. Nor (even assuming that legislative history could ever resolve statutory ambiguity for purposes of Chevron) does the legislative history unambiguously demonstrate a congressional intent to preclude consideration of the rela tionship between costs and benefits under Section 316(b). The States argue (at 38) that "[t]he legislative history of section 316(b) is sparse and undeserving of any significant weight." The government generally agrees. Section 316(b) was "added by the conference committee without substan tive comment." Riverkeeper, Inc. v. United States EPA, 358 F.3d 174, 186 n.12 (2d Cir. 2004) (Riverkeeper I). And while the only floor statement concerning Section 316(b) supports EPA's interpretation, that statement reflects the views of a single legislator, Representative Clausen. See Gov't Br. 23; 118 Cong. Rec. 33,762 (1972).

Like the States, Riverkeeper (at 47) also disclaims reli ance on legislative history, but then goes on to discuss at length the legislative history of other provisions of the CWA. Riverkeeper's basic argument (at 6-12, 47-49) is that, by choosing to rely primarily on technology-based discharge standards rather than water-quality standards, Congress rejected cost-benefit analysis in the CWA. But Congress's reasons for generally preferring technology- based discharge standards (which directly address dis charges as opposed to water bodies) to water-quality stan dards (which directly address water bodies as opposed to individual discharges), see generally EPA v. California, 426 U.S. 200, 202-203 (1976), do not imply any particular view as to whether EPA should consider the relationship be tween costs and benefits in formulating appropriate stan dards. That is evident on the face of the BCT and BPT standards, which are technology-based discharge standards that expressly require consideration of the relationship between costs and benefits. See p. 10, supra.

Moreover, the legislative history proffered by respon dents relates almost entirely to discharges of pollutants governed by Sections 301, 306, and 316(a), not to water in takes governed by Section 316(b). The States are clear on that point: after discounting the legislative history of Sec tion 316(b) on the ground that it consists of only a floor statement, the States go on to rely on legislative history- mostly floor statements, ironically-concerning other provi sions of the Act. States Br. 38-41. Going even farther afield, Riverkeeper searched the National Archives for var ious notes, memoranda, and correspondence that, it claims, show that the conferees decided to address thermal dis charges with a water-quality based variance (in Section 316(a)) and water intake through a best-technology stan dard (in Section 316(b)). Riverkeeper Br. 10 n.6, 13 n.8. That much is obvious on the face of the statute. But none of the materials on which respondents rely tackles the question presented here, which is whether Section 316(b)'s best-technology standard precludes, permits, or (like BPT and BCT) requires consideration of the relationship be tween costs and benefits. And even if the assorted materi als relied on by Riverkeeper were on point, and even if they were considered reliable and relevant sources of legislative history, they could hardly establish an unambiguous con gressional intent-especially considering that Representa tive Clausen's floor statement, which is at least a more con ventional form of legislative history, supports EPA's inter pretation.

Respondents' discussion shows only that Section 316(b)'s intake provision received very little consideration in the legislative debates, especially compared to the Sec tion 301, 306, and 316(a) discharge restrictions. But, if any thing, that undercuts respondents' position because the gravamen of their argument is that Congress in enacting Section 316(b) specifically "determined the relationship between the costs and benefits." States Br. 44. If that were true, one would expect at least some indication that Congress had actually considered the question and had determined that the benefits of applying respondents' pro posed water-intake standard justified the costs. As the Second Circuit explained in Riverkeeper I, however, the "brevity" of Section 316(b), combined with the "paucity of legislative history, when measured against the volumes of drafts and speeches devoted to other aspects of the 1972 amendments," suggests that Congress "desire[d] to dele gate significant rulemaking authority to the Agency." 358 F.3d at 186 n.12.

C. The Court Of Appeals Erroneously Usurped EPA's Discre tion

Because Section 316(b) does not "directly sp[eak] to the precise question at issue," EPA's reasonable interpretation of the ambiguous statutory text is entitled to deference. Chevron, 467 U.S. at 842; see Gov't Br. 26-30.

1. The court of appeals turned normal principles of Chevron deference on their head by relying on an artificial presumption that cost-benefit analysis is forbidden unless Congress clearly authorizes it. See Pet. App. 22a-23a. Re spondents do not appear to defend that presumption, see Riverkeeper Br. 41-42; States Br. 43, and for good reason: Congress's silence on whether an agency may consider the relationship between costs and benefits provides no ba sis for inferring an unambiguous legislative prohibition against such consideration. Gov't Br. 30-34. "[S]ilence, after all, normally creates ambiguity. It does not resolve it." Barnhart v. Walton, 535 U.S. 212, 218 (2002). The court of appeals' presumption is especially unwarranted because, far from being aberrational, cost-benefit analysis instead is a routine feature of private and governmental decision-making. See Gov't Br. 13-14.

2. The extent of the court of appeals' departure from normal deference principles is further underscored by its attempt to micro-manage EPA's decisionmaking through rules that cannot be found in the Act. The court concluded, for example, that EPA may consider costs as part of "cost- effectiveness" but not "cost-benefit" analysis-terms that appear nowhere in Section 316(b). See Pet. App. 24a, 26a. Based on that atextual distinction, the court would have allowed EPA to adopt a significantly cheaper technology that would save 99-101 fish instead of 100-103 fish. Id. at 22a-23a & n.10, 27a. By permitting the agency to weigh costs against at least one or two fish (and perhaps more, though the number is unclear), the court of appeals essen tially permitted EPA to consider the relationship between costs and benefits, but only in the most extreme cases. Even on its own terms, therefore, the court of appeals' deci sion lacks a principled basis in the statutory text. More over, the court of appeals agreed that EPA could consider other practical factors such as energy efficiency that, while important, are no more grounded in explicit statutory text than the relationship between costs and benefits. See id. at 26a n.12.

While respondents do not directly address those points, their own positions rest on similar errors. The States (at 16, 21, 29) repeatedly assert that costs can be a "secondary" but not a "primary" consideration under Section 316(b). That approach has no more grounding in the statute than the court of appeals' cost-benefit/cost-effectiveness distinc tion. And as noted above, the States' embrace of cost as a secondary factor only underscores that EPA is entitled to deference, because the precise manner or extent of consid ering a permissible factor is a textbook matter for agency discretion. See p. 5, supra.

Similarly, Riverkeeper concedes (at 29), in the course of assuring this Court that its position would not produce ab surd results, that "the Agency has some discretion (albeit not boundless) to determine that further differences in re duction would be so minor as to be unnecessary for compliance with the minimizing requirement." But there is no more principled basis for cabining EPA's conceded discre tion to "minor" (ibid.) matters than for limiting it to cost- effectiveness or "secondary" considerations. Congress ex pressed no unambiguous intent to draw any of those prof fered distinctions.

3. Riverkeeper argues (at 43-44) that the agency's in terpretation is not entitled to Chevron deference because, in the preamble to a 1976 rulemaking that was vacated on judicial review, EPA initially construed Section 316(b) to preclude cost-benefit analysis. That characterization is nei ther correct nor relevant (especially considering that the rule was vacated, see Appalachian Power Co. v. Train, 566 F.2d 451 (4th Cir. 1977)). EPA determined in the 1976 pre amble that cost-benefit analysis is not "required," but it did not squarely address whether consideration of the relation ship between costs and benefits is permitted. Instead, in a portion of the preamble that Riverkeeper does not discuss, the agency emphasized that "economic practicability" is an important consideration that should be analyzed on a case- by-case basis. J.A. 42.

The following year, in a permitting decision and a Gen eral Counsel opinion, EPA explained that, while Section 316(b) does not require a formal cost-benefit analysis, it would not be "reasonable to interpret Section 316(b) as requiring use of technology whose cost is wholly dispropor tionate to the environmental benefit to be gained." In re Pub. Serv. Co. of N.H. (Seabrook Station, Units 1 & 2), No. 76-7, 1977 WL 22370 (EPA June 10, 1977) (Pub. Serv. Co. of N.H.), remanded on other grounds sub nom. Seacoast Anti-Pollution League v. Costle, 572 F.2d 872 (1st Cir. 1978); accord In re Cent. Hudson Gas & Elec. Corp., Op. EPA Gen. Counsel, NPDES No. 63, 1977 WL 28250, at *8 (July 29, 1977). Ever since, permitting authorities in imple menting Section 316(b) have considered the relationship between costs and benefits to at least that extent. Gov't Br. 27-28.

Riverkeeper contends (at 45) that EPA, in applying the "wholly disproportionate" test, has considered only whether the benefits of additional protective measures would be de minimis. That gloss, however, is flatly inconsistent with the test as EPA has articulated it: EPA directed permit ting authorities to determine whether "cost is wholly dis proportionate to the environmental benefit to be gained," not whether the benefit is de minimis. Public Serv. Co. of N.H., 1977 WL 22370 (emphases added). The court of ap peals' decision is thus incompatible with more than 30 years of settled agency practice. See also, e.g., In re Florida Power Corp. Crystal River Power Plant Units 1, 2, & 3, NPDES Permit No. FL0000159, at 7-8 (EPA Region IV Sept. 1, 1988) (determining that closed-cycle cooling towers were not BTA because their extremely high cost was "wholly disproportionate" to benefits at a particular plant, even though that technology would reduce entrainment by 85% and there were no other technologically practical alter natives to reduce entrainment "to an acceptable level" at the plant).

Riverkeeper is likewise incorrect in contending (at 45- 46) that EPA recognized during the Phase I rulemaking for new sources that "it lacked authority to compare costs and benefits under Section 316(b)." As the government's open ing brief explained (at 38-39), the Phase I rule rejected a technology known as dry cooling because, among other things, dry cooling "costs more than ten times as much per year as closed-cycle wet cooling, but it is estimated to re duce water intake by only an additional 5 percent relative to once-through cooling." Riverkeeper I, 358 F.3d at 194 (footnotes omitted). The Second Circuit upheld EPA's con sideration of the relationship between costs and benefits, along with other factors, as falling within the agency's "considerable discretion." Id. at 195. Riverkeeper makes no effort to reconcile its characterization of the Phase I rulemaking with those facts.

D. EPA's Consideration Of Costs And Benefits In The Rulemaking At Issue Here Fell Well Within Its Discretion

Because EPA acted within the scope of its statutory authority in promulgating the performance standards and site-specific cost-benefit variance provision at issue here, this Court should reverse the judgment of the court of ap peals and reinstate those portions of the rule. Although respondents dispute EPA's authority to give any weight to the relationship between costs and benefits, they raise no substantial challenge to the specific manner in which EPA considered that relationship here.

Nor could they. EPA did not select the performance standards based only on a cost-benefit analysis. Instead, the agency rejected closed-cycle cooling technology as BTA "based on its generally high costs (due to conversions), the fact that other technologies approach the performance of this option, concerns for energy impacts due to retrofitting existing facilities, and other considerations." Pet. App. 255a; see Gov't Br. 35-37. The cost-benefit variance provi sion likewise does not turn on a strict cost-benefit compari son. Instead, it permits the operator of an individual facil ity to apply for a site-specific determination of BTA-which itself would be based on a permitting authority's discretion ary consideration and balancing of all relevant fac tors-only if the facility's cost of complying with the na tional performance standards "would be significantly greater than the benefits." 40 C.F.R. 125.94(a)(5)(ii) (em phasis added); see Gov't Br. 39-41. As the government's opening brief explained (at 14), agencies' consideration of the relationship between costs and benefits may take a number of different forms. EPA's consideration of various relevant factors in the rulemaking below was by no means among the more robust types of cost-benefit analysis.4

Riverkeeper observes (at 52) that EPA "assigned no dollar value" to many of the relevant benefits "because it was too difficult to determine a meaningful [monetary] value for them." The agency did, however, undertake a "qualitative[]" analysis of the benefits that were not quanti fied. Pet. App. 485a; see id. at 482a-515a. EPA ultimately concluded that, "[a]lthough not identical, the ranges of im pingement and entrainment reduction are similar under both" the cooling-towers option and the selected option, while the costs of the former far exceed those of the latter. Id. at 260a. Far from reflecting a defect in EPA's analysis, the agency's conclusion that it would make little sense to try to compare all relevant considerations in purely mone tary terms demonstrates the flexibility and common-sense nature of cost-benefit analysis.

Respondents do not appear to dispute that the validity of the cost-benefit variance depends entirely on the resolu tion of the question presented. See Gov't Br. 40-41. River keeper contends (at 53 n.24), however, that the court of ap peals' invalidation of the performance standards rests on the alternative ground that EPA did not adequately explain the basis for its decisionmaking. But what the court consid ered unclear is whether the agency "based its decision on permissible cost-effectiveness analysis or exceeded its au thority by relying impermissibly upon a cost-benefit analy sis." Pet. App. 36a. Because that conclusion rests entirely on the court of appeals' erroneous holding that EPA may not consider the relationship between costs and benefits, it does not provide an independent basis for a remand to the agency.

Riverkeeper further contends (at 31 n.16) that the court of appeals invalidated EPA's performance standards on the (assertedly) separate ground that they allowed facilities that could comply with the upper end of a range to comply with the lower end instead, and that the validity of the ranges is not fairly included in the question presented. As the government's opening brief explained (at 39 n.3), how ever, the court remanded EPA's chosen ranges based on its view that Section 316(b) requires "as much reduction of adverse environmental impacts as is technologically possi ble." Pet. App. 43a. Because that holding rests on the court's erroneous resolution of the question presented, and Riverkeeper acknowledges (at 31 n.16) that ranges are oth erwise permissible, the court's invalidation of the ranges is fairly encompassed in the question presented and should be reversed.

* * * * *

For the foregoing reasons and those stated in the government's opening brief, the judgment of the court of appeals should be reversed with respect to the perfor mance standards and the site-specific cost-benefit provi sion, those provisions should be reinstated, and the case should be remanded.

Respectfully submitted.

GREGORY G. GARRE
Solicitor General

OCTOBER 2008

1 Citations to the Pet. App. refer to the appendix filed in No. 07-588.

2 The States note (at 19-20) that EPA's preamble to a 1976 rule treated the term "minimizing" as referring to "reducing to the smallest possible amount or degree." J.A. 41. That rule was vacated, however, on judicial review. See Appalachian Power Co. v. Train, 566 F.2d 451 (4th Cir. 1977). The following year, EPA issued a permitting decision and a General Counsel opinion making clear the agency's view that it would not be "reasonable to interpret Section 316(b) as requiring use of technology whose cost is wholly disproportionate to the environmen tal benefit to be gained." In re Pub. Serv. Co. of N.H. (Seabrook Sta tion, Units 1 & 2), No. 76-7, 1977 WL 22370 (EPA June 10, 1977), re manded on other grounds sub nom. Seacoast Anti-Pollution League v. Costle, 572 F.2d 872 (1st Cir. 1978); accord In re Cent. Hudson Gas & Elec. Corp., Op. EPA Gen. Counsel, NPDES No. 63, 1977 WL 28250, at *8 (July 29, 1977); see p. 16, infra. And EPA's current regulatory defi nition of "minimize," which was established in the Phase I rule, is the one quoted in the text.

3 Because the statute does not equate BTA with BAT or BADT, there is no need to decide in this case whether EPA may consider the relationship between costs and benefits in determining BAT or BADT. Riverkeeper correctly argues (at 35-37) that EPA is not required to consider that relationship in making those determinations, and that BAT standards are intended to be stricter than BPT standards. But it does not necessarily follow that no cost-benefit consideration is per mitted for either BAT or BADT. See Gov't Br. 23-24.

4 The States assert (at 8) that EPA has found closed-cycle cooling to be BTA "at more than a dozen existing power plants." That is incor rect. In both of the examples cited by the States, power plants were retrofitted with closed-cycle cooling technology to reduce thermal dis charges pursuant to Section 316(a). See Consolidated Edison Co. v. New York State Dep't of Envt'l Conservation, 726 F. Supp. 1404, 1406 (S.D.N.Y. 1989); California's Coastal Power Plants: Alternative Cool ing System Analysis 6-2 (visited Oct. 27, 2008) <http://www. resources. ca.gov/copc/OTC/Chapter_6_Retrofit_and_Repower_Examples_ 28121840.pdf>. Some other facilities have been retrofitted for opera tional, not pollution-related, reasons. See id. at 6-2, 6-3, 6-4. The States' reliance (at 7) on the preamble to the vacated 1976 rule is like wise misplaced, not only because the rule was vacated, but also because that preamble explained that closed-cycle cooling is not "universally and necessarily the best technology available," and that BTA should instead be determined on a facility-specific basis. J.A. 43.