|
Supreme Court Briefs | |||||||||||
IN THE
SUPREME COURT OF THE UNITED STATES
CAROL M. BROWNER, ADMINSTRATOR OF
ENVIRONMENTAL PROTECTION AGENCY, et al.
Petitioners,
v.
AMERICAN TRUCKING ASSOCIATION INC., et al.
Respondents.
BRIEF OF AMICI CURIAE SENATOR ORRIN HATCH
AND REPRESENTATIVE TOM BLILEY
IN SUPPORT OP REPONDENTS
QUESTION PRESENTED
Whether the Environmental Protection Agency's interpretation of Section 109 of the Clean Air Act, in conformity with Lead Industries Ass'n v. EPA, 647 F.2d 1130, 1148 (D.C. Cir. 1980), effects an unconstitutional delegation of legislative power?
INTEREST OF AMICI CURIAE1
Amicus Orrin Hatch is a Member of the United States Senate from Utah, and amicus Tom Bliley is a Member of the United States House of Representatives from Virginia. Amici are both concerned that the Environmental Protection Agency's ("EPA") implementation of the Act is contrary to congressional will, undermines the separation of powers, and frustrates effective congressional and judicial oversight. Because EPA has been hamstrung by an erroneous judicial precedent that irrationally prevents the agency from considering the cost effectiveness of its regulations, and that provides no other constitutionally required intelligible principles to constrain the agency's discretion, EPA engages in a science charade wherein policy and economic judgments are masked as scientific determinations.
INTRODUCTION AND SUMMARY OF THE ARGUMENT
This case involves a statutory provision that confers unusually broad administrative power. Under this Court's teaching, the statute¢and the Agency's interpretation of its authority thereunder¢must receive "separation of powers scrutiny commensurate with the potential impacts on society.
On petition for review of the revised national ambient-air quality standards ("NAAQS") for particulate matter ("PM") and ozone set by the EPA pursuant to section 109 of the Clean Air Act ("Act"), the United States Court of Appeals for the District of Columbia held that "the construction of the Clean Air Act on which EPA relied in promulgating the
The court found no enlightenment in the agency preambles for the final rules, which amounted to little more than a showing "that EPA is applying the stated factors and that larger public health harms (including increased probability of such harms) are, as expected, associated with higher pollutant concentrations." id. at lOa. The Agency's identification of relevant factors did not provide any intelligible principle limiting agency discretion; EPA's rationale "could as easily, for any non-threshold pollutant, justify a standard of zero." Id. at 1 Ia. Cost-benefit analysis, the Court observed, would avoid such indeterminacy, but D.C. Circuit precedents "read § I 09(b)( I) as barring EPA from considering any factor other than 'health effects relating to pollutants in the air.' " Pet. App. ISa (citing, inter (ilia, Lead lndus. Ass'n v. EPA, 647 F.2d 1130, 1148 (D.C. Cir. 1980)).2
Lead Industries's exclusion of cost-benefit balancing empowers the Administrator to regulate without limits. Unwilling to admit or surrender the unfettered discretion it exercises, EPA engages in "science charades" that mask inevitable cost-based decisions as science-based, and it effectively abandons¢on covert grounds¢the core Clean Air Act goal of NAAQS attainment. Hatch/Bliley Br. 25-30 (discussing negative implications for attainment of the President's $10,000 per ton cap on "implementation" costs).
As this Court's precedents make clear, whether a given statutory delegation is an unconstitutional grant of arbitrary power to an agency depends critically on context. The Govemment and its supporters miss this point by mechanically citing the language of statutory delegations upheld by this Court in the past, without regard to the statutory and factual contexts defining limits for those delegations. The context for setting NAAQS under the Clean Air Act is fundamentally different because of the conceded indeterminacy of the relevant science and the unparalleled power the Administrator wields under section 109 over states, local govemments, and the American public. The non-delegation decisions of this Court require a narrowing construction of administrative authority to ensure that such power is properly circumscribed.
Lead Industries extinguished intelligible principles that could constrain the Administrator's judgment in this context. The principle suggested by the Government and its supporters¢that the Administrator has a mandate to protect the public health by setting NAAQS based only on scientific or medical evidence of health risks suffered by sensitive populations¢does not in fact provide any decisional principle to guide the Administrator in deciding which health risks she will abate and which she will tolerate. The Administrator has acknowledged that Congress did not intend NAAQS to be set at a level to eliminate all health risks, for that would be harmful to the country. Thus, EPA is essentially rationing public health protections: it has set the ozone and PM NAAQS at levels that will protect some populations from health impairment, but not others.
The Administrator purports to rely exclusively on science-based criteria in making those choices. But risk management, which is what the NAAQS provisions of the Act require, is not that simple. Under the Lead Industries regime, EPA is denied the measuring stick necessary to make comparative judgments about public-health impacts. As a result, EPA cannot forthrightly justify where it has drawn the lines of protection. Lead Industries thus leads to incoherence and science charades, as these rulemakings demonstrate. See infra at 26-29. Congress cannot be presumed to have delegated to the Administrator such massive power over American life without a rational principle to direct her judgments. See FDA v. Brown & Williamson Tobacco Corp., 120 5. Ct. 1291, 1301 (2000). Under such a reading, the Administrator exercises an impermissible power, legislative in nature, to allocate health protections based on arbitrary (and not fully disclosed) factors.3
Section 109 is not unconstitutional simply because Lead Industries does not supply an intelligible limiting principle; rather, this Court's nondelegation jurisprudence requires that this Court, if at all possible, adopt an alternative construction
I. NONDELEGATION IS A VITAL PRINCIPLE OF SEPARATION OF POWERS THAT GUARDS AGAINST GRANTS OF ARBITRARY POWER TO THE EXECUTIVE.
A. Congress May Not Grant The Executive Branch (And The Executive May Not Usurp) Arbitrary Power Unbounded By Law.
Article I, section 1 of the Constitution grants "[a]ll legislative Powers" exclusively to Congress. See U.S. Const. art. I, § 1. "That Congress cannot delegate legislative power to the President is a principle universally recognized as vital to the integrity and maintenance of the system of government ordained by the Constitution." Field v. Clark, 143 U.S. 649, 692 (1892). This principle "ensures to the extent consistent with orderly governmental administration that important choices of social policy are made by Congress, the branch of our Government most responsive to the popular will." Industrial Union Dep't, AFL-CIO v. American Petroleum lnst., 448 U.S. 607, 685 (1980) (Rehnquist, J., concurring in the judgment); Loving v. United States, 517 U.S. 748, 757-58 (1996). The nondelegation doctrine thus "preventts] Congress from forsaking its duties," even when it does not encroach upon the prerogatives or impair the functioning of a coordinate branch. Id. at 758. "Abdication of responsibility is not part of the constitutional design." Clinton v. City of New York, 524 U.S. 417, 452 (1998) (Kennedy, J., concurring).
Nondelegation is not, however, strictly a matter of preventing one branch's exercise of functions properly belonging to another. This doctrine applies not only to circumstances where the executive is given the power to issue prospective rules with the force and effect of law (although such power is granted here), but also to the quintessentially executive function of administering a statute in a specific case.4 All federal executive authority not independently derived from the Constitution must be granted by Congress, Louisiana Public Service Commission v. FCC, 476 U.S. 355, 374 (1986), and no officer may be given power unbounded by law. The fundamental concern in any kind of statutory delegation to the executive branch is whether it grants "arbitrary discretion," Buttfield v. Stranahan, 192 U.S. 470, 496 (1904); Union Bridge, 204 U.S. at 387; Federal Radio Commission v. Nelson Brothers Bond & Mortgage Co., 289 U.S. 266, 282 (1933); Schechter Poultry, 295 U.S. at 537, whereby the executive acts as a law unto itself. See Clinton, 524 U.S. at 452 (Kennedy, J., concurring) (nondelegation is a vertical protection of the citizenry as well as a horizontal protection of the division of power among the
B. The Nondelegation Doctrine Has Emerged As An Important Canon Of Statutory Construction To Constrain Executive Power.
The essential inquiry in this Court's nondelegation jurisprudence is to ascertain when the executive discretion granted by Congress is arbitrary. Delegation of discretionary authority has long been recognized as a legitimate attribute of legislative power, see Wayman v. Southard, 23 U.S. (10 Wheat.) 1, 42 (1825); Field v. Clark, 143 U.S. 649, 693-94 (1892), and this Court has not hesitated to approve even 'broad' standards for administrative action" that are "a reflection of the necessities of modern legislation dealing with complex economic and social problems." American Power & Light, 329 U.S. at 105 ("Necessity ... fixes a point beyond which it is unreasonable and impracticable to compel Congress to prescribe detailed rules . . . ."); Sunshine Anthracite Coal Co. v. Adkins, 310 U.S. 381, 398 (1940). But even so, discretion, in Justice Cardozo's famous words, must be "canalized within banks that keep it from overflowing" and not "unconfined and vagrant." Schechter Poultry, 295 U.S. at 551 (Cardozo, J., concurring); Touby v. United States, 500 U.S. 160, 166 (1991) (the statute must "meaningfully constrain [3 the [officer's] discretion"). Congress must fix the "primary standard," Buttfield, 192 U.S. at 496, against which executive action may be measured so that a court may "'"ascertain whether the will of Congress has been obeyed." ' " Skinner v. Mid-America Pipeline Co., 490 U.S. 212, 218 (1989) (quoting Mistretta v. United States, 488 U.S. 361, 379 (1989) (quoting Yakus v. United States, 321 U.S. 414, 426 (1944))); see also Clinton, 524 U.S. at 484 (Breyer, J., dissenting). In this Court's classic formulation, "IIs]o long as Congress 'lays down by legislative act an intelligible principle to which the person or body authorized to act is directed to conform, such legislative action is not a forbidden delegation of legislative power."' Touby, 500 U.S. at 165 (brackets omitted) (quoting J.W. Hampton, Jr. & Co. v. United States, 276 U.S. 394, 409 (1928)). Cf. AT&T Corp. v. Iowa Utils. Bd., 525 U.S. 366, 388 (1999) (interpreting statute to require agency "to apply some limiting standard, rationally related to the goals of the Act").
Although this Court has struck down statutes as violating the nondelegation doctrine only in two 1935 decisions, Panama Refining and Schechter Poultry, the doctrine remains an important canon of interpretation employed to give "narrow constructions to statutory delegations that might otherwise be thought to be unconstitutional." Mistretta v. United States, 488 U.S. 361, 373 n.7 (1989). Thus, in Kent v. Dulles, 357 U.S. 116, 128-29 (1958), this Court invoked nondelegation principles to narrow a statute to deny the Secretary of State "unbridled discretion" to restrict the right of citizens to travel. See also National Cable, 415 U.S. at 342 (avoiding nondelegation problem by narrowing power of agency to assess fees). And, in Industrial Union Department, AFL-CiO v. American Petroleum Institute, 448 U.S. 607 (1980), a case strikingly parallel to this one, the Occupational Safety and Health Administration (OSHA) had interpreted its organic statute to permit it to promulgate workplace standards to regulate any health risks from dangerous substances. Noting that OSHA's construction suggested a grant of "unprecedented power over American industry" to an administrative agency to engage in "pervasive regulation limited only by the constraint of feasibility," id. at 645, a plurality of this Court narrowed the statute to empower OSHA to regulate only "significant" risks to health and safety. Id. at 642. Otherwise, this Court held, "the statute would make such a 'sweeping delegation of legislative power' that it might be unconstitutional under the [nondelegation doctrine]. A construction of the statute that avoids this kind of open-ended grant should certainly be favored." Id. at 646; see also id. at 672-76 (Rehnquist, J., concurring in the judgment) (voting to invalidate the statute as an unconstitutional delegation of legislative power).
C. The Arbitrariness Of A Grant Of Power Depends On Context.
As this Court's nondelegation precedents make clear, whether a statute delegates arbitrary power depends critically on context. Lichter, 334 U.S. at 785 ("Standards prescribed by Congress are to be read in light of the conditions to which they are to be applied."); American Power & Light, 329 U.S. at 104 (statutory standards are not to "be tested in isolation," but "derive much meaningful content from the purpose of the Act, its factual background and the statutory context in which they appear"); Nelson Bros., 289 U.S. at 285; New York Cent. Sec. Corp. v. United States, 287 U.S. 12, 24 (1932); National Broad. Co. v. United States, 319 U.S. 190. 215-16, 219 (1943) ("NBC"); see also Clinton, 524 U.S. at 488 (Breyer, J., dissenting). The intelligible-principle rule is always the same, but the sufficiency of a given standard under that rule turns not just on the bare language of the statute, but on numerous factors specific to the statutory scheme. Thus, the arbitrariness of a delegation may depend upon the clarity of statutory purposes, American Power & Light, 329 U.S. at 105; New York Central, 287 U.S. at 24-25, the complexity of the subject matter of regulation, NBC, 319 U.S. at 219, the variability of factual conditions, J.W. Hampton, Jr. & Co. v. United States, 276 U.S. 394, 404-05 (1928); United States v. Grimaud, 220 U.S. 506, 516 (1911), the exigency of circumstances, Touby, 500 U.S. at 168; Lichter, 334 U.S. at 779-80, the nature of power being exercised and of the officer exercising it, Loving, 517 U.S. at 772; United States v. Mazurie, 419 U.S. 544, 556-67 (1975),5 past administrative practice, Kent, 357 U.S. at 127-28; Lichter, 334 U.S. at 777-78, 783, the presence of procedural safeguards, Lichter, 334 U.S. at 786-87; Union Bridge, 204 U.S. at 387, whether the regulation concerns government property or the public domain, Grimaud, 220 U.S. at 521; Nelson I3ros., 289 U.S. at 282, whether constitutionally protected rights are at stake, Kent, 357 U.S. at 128-29, and whether unusually broad administrative power is conferred, Industrial Union, 448 U.S. at 645-46; Schechter Poultry, 295 U.S. at 542; Michigan v. EPA, 213 F.3d. 663,680-81 (D.C. Cir. 2000).
The Government and its supporters disregard the crucial element of context by mechanically citing broad language of delegation in statutes that have been upheld by this Court; they argue that, even under Lead Industries, the detailed prescriptions of section 109 cannot be unconstitutional in light of past decisions of this Court upholding standards such as the "public interest," "public convenience," justice and equity, and reasonableness. Cf. U.S. Br. at 22, 25; Mass Br. at 29, 35; see also Pet. App. 59a-60a (Tatel, J., dissenting in part). The cited cases, however, involved circumstances where there were ascertainable historical facts that could be found by the administrative or executive officer; the officer's judgment (whether embodied in an order or a generally applicable regulation) could be applied to those facts and tested by a court against reason and statutory purpose. The broad standards at issue in those cases related to administrative determinations based on the actual costs or conditions of a particular company or industry,6 on historical
6 See Skinner, 490 U.S. at 2 19-20 (imposition of fees on pipeline companies based on "reasonable relationship" to their specific volume-miles, miles, and revenues); Lichter, 334 U.S. at 785-86 (determinatiOn of "excessive profits" earned on a specific government contract based on cost investigation); Yakus v. United States, 321 U.S. 414, 420-21 (1944)
7 See NBC, 319 U.S. at 217 (regulation of radio licensees); Nelson Bros., 289 U.S. at 282 (same); Interstate Commerce Conim'n v. Goodrich Transit Co., 224 U.S. 194. 214-15 (1912) (formulation of uniform systems of accounts to monitor common carriers by water); Grimaud, 220 U.S. at 5 15-16, 521 (authorization of regulations of "occupancy and use" for preservation of national forests, which persons use under an "implied license" from the Government).
8 See Union Bridge. 204 U.S. at 386-87 (determining after hearing whether a particular bridge examined by the Army Corps of Engineers was an unreasonable obstruction to navigation); Buttfleld, 192 U.S. at 494-95 (issuance of standards of "purity, quality, and fitness for consumption" based on expert rankings of imported teas); Field, 143 U.S. at 692-93 (suspension of free importation of articles after investigation as to whether exporting country imposed duties that were "reciprocally unequal and unreasonable").
D. Whether Lead Industries Is Inconsistent With The Nondelegation Doctrine Must Be Determined In Light Of The Statutory Context Of Scientific Indeterminacy And Vast Administrative Power.
The context of the Clean Air Act is fundamentally different from that of statutes reviewed in this Court's prior nondelegation decisions (except for Industrial Union) in two important respects. The first is the inherently subjective nature of regulatory science. As Sheila Jasanoff has documented, there has been a profound shift in administrative regulation from traditional factfinding and industry expertise to the "new" administrative agencies such as EPA engaged in "regulatory science." Sheila Jasanoff, The Fifth Branch: Science Advisers as Policy Makers 40-45 (1990). Regulatory science is different from rigorous research science; it is policymaking requiring "'predictions ... at the frontiers of science."' NRDC v. EPA, 902 F.2d 962, 968 (D.C. Cir. 1990), vacated in part on other grounds, 921 F.2d
10 Cf Daubert v. Merrell Dow Pharms, Inc., 509 U.S. 579, 593 (1993) (citing Jasanoff regarding issues of reliability of scientific evidence).
This case amply demonstrates that proposition. The massive scientific uncertainties about the existence of any health effects from fine PM are uniformly acknowledged (by the President, Congress, CASAC, the National Academy of Sciences, and EPA itself), and Congress had appropriated large sums for further research with the express intent that EPA delay further rulemaking. Hatch/Bliley Br. 2-3; see also Pub. L. No. 105-178, §§ 6101-6104, 1998 U.S.C.C.A.N. (112 Stat.) 107, 463-65 (1998) (codified at 42 U.S.C. § 7407 note)
The second critical element of context that distinguishes other nondelegation cases is the sheer scope of the coercive power in the Administrator's hands here. The authority to set NAAQS is the power to dictate massive regulation of multitudinous sources by States and their political subdivisions in devising implementation plans under section 110. It is also undisputed, even by EPA's conservative estimate, that many billions of dollars in annual compliance costs, above and beyond what is required to meet current NAAQS, turn on every fractional change in the NAAQS. The plurality decision in Industrial Union recognized that the grant of vast power to an agency increases the risk of arbitrary action. 448 U.S. at 645-46. The court below likewise properly recognized that such vast power demands a "'more precise'" standard to ensure against arbitrary administrative action. Pet. App. 12a.
Thus, Judge Silberman, in his dissent from denial of rehearing en banc, was wrong to dismiss the nondelegation reasoning in Industrial Union as "a makeweight, tossed into the analysis, ... to help justify the result." Pet. App. 93a. First, nondelegation was central to the outcome of that decision, and justified the narrow construction adopted by the plurality. Second, because it interpreted the statute to ensure that the grant of power is not arbitrary in its context, Industrial Union is squarely within the tradition of this Court's nondelegation jurisprudence. Although not so much as mentioned in the Government's brief, that decision strongly supports rejecting Lead Industries in favor of a narrower construction. Indeed, the reasoning of Industrial Union applies afortiori here, for EPA's power over the States and industrial sources under section 109 dwarfs that of OSHA to regulate hazardous workplaces.12
II. THE LEAD JNDUS TRIES INTERPRETATION OF SECTION 109 DOES NOT PROVIDE AN INTELLIGIBLE PRINCIPLE CONSTRAINING THE ADMINISTRATOR'S POWER.
A. Under A Proper Interpretation Of Section 109, The Principle Of Cost-Effectiveness Constrains The Administrator's Discretion When The Scientific Evidence Is Indeterminate.
Properly interpreted, section 109 of the Clean Air Act is clearly constitutional. As described more fully in our prior brief, Congress legislated "as far as was reasonably practicable," Buttfield, 192 U.S. at 496, in a scientifically complex area of environmental regulation, and channeled the Administrator's discretion by providing principled constraints that permit the courts to "ascertain whether the will of Congress has been obeyed," Yakus, 321 U.S. at 426. For pollutants emitted from numerous or diverse sources that "cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare," 42 U.S.C. § 7408(a)(l)(A), the Administrator, among other things, sets primary NAAQS "the attainment and maintenance of which in the judgment of the Administrator, based on [air quality] criteria and allowing an adequate margin of safety, are requisite to protect the public health." Id. § 7409(b)(l). Congress intended the Administrator's judgment to be based on science to the extent possible: not only does section 109 require the Administrator to base the standard on air-quality criteria that "shall accurately reflect the latest scientific knowledge" regarding pollution effects, id. § 7408(a)(2), but it also authorizes "an independent scientific review committee," CASAC, to advise the Administrator on the establishment and revision of NAAQS, id. § 7409(d)(2)(A)-(B).
As noted above, science cannot alone guide the Administrator in her judgment. Recognizing the inherent indeterminacy of science about public health risks from pollution, and intending standards that would be "preventative or precautionary," H.R. Rep. No. 95-294, at 49 (1977), Congress directed the Administrator to "allow[] an adequate margin of safety" to account for scientific uncertainty in making a "judgment" about what is "requisite to protect the public health." 42 U.S.C. § 7409(b)(l). Congress did not forbid the Administrator to consider any factor relevant to inform her judgment as to what was "adequate" or "requisite"¢and certainly did not preclude consideration of cost/benefit balancing, which is fundamental to traditional concepts of environmental risk management, Hatch/Bliley Br. 20-23, and of "protect[ing] the public health," see Br. for Cross-Pet'rs 37-43. To the contrary, Congress directed the Administrator to gather "information on air pollution techniques" that "shall include data relating to the cost of installation and operation, energy requirements, emission reduction benefits, and environmental impact of the emission control technology," 42 U.S.C. § 7408(b)(l), and to publish this information along with the air-quality criteria prior to the proposal of a new or revised NAAQS, id. § 7409(a)(2). Congress further required CASAC to "advise the Administrator of any adverse public health, welfare, social, economic, or energy effects which may result from various strategies for attainment and maintenance of such national ambient air quality standards," id. § 7409(d)(2)(C)(iv),13 and required the Administrator to explain any significant departure from "pertinent findings, recommendations, and comments" of CASAC and the National Academy of Sciences, id. § 7607(d)(3). The Administrator also must conduct cost-benefit analysis of the NAAQS. Id. § 7612(a)(1).14
The regulatory scheme essentially replicates that of the Air Quality Act of 1967, except that the authority to set air-quality standards is now vested in EPA rather than the States: standards must be set at or below the level at which science reflected in the air-quality criteria shows dangers to public health, but, in the face of scientific uncertainty, the ultimate standard would reflect a socioeconomic judgment as to what risk levels are tolerable. See Hatch/Bliley Br. 15-18. Thus, scientific evidence of a significant risk of danger to the public health, and, to the extent science is uncertain, the cost-effectiveness of alternative standards, are the intelligible principles that constrain the Administrator's discretion.
Bound by Lead Industries, the court of appeals could not adopt this interpretation, Pet. App. ISa, and thus was compelled to declare unconstitutional the Administrator's construction (which itself was tethered to Lead Industries).
14 EPA also considers costs in apportioning burdens among states pursuant to its authority to order upwind states to curtail emissions that contribute signiticantly to nonattainment by downwind states¢a riskmanagement task that cannot be divorced from the setting of NAAQS. See Hatch/B liley Br. 24-25.
B. The Lead Industries Rule That EPA May Only Rely Upon Scientific Factors Even When Science Is Indeterminate Violates Nondelegation Principles.
Drawing on some of the same statutory provisions discussed above, the Government and its supporters attempt to defend Lead Industries by arguing that there are intelligible principles that bind the Administrator, even if cost-effectiveness may not be considered. Their view, summarized, is that EPA has a statutory mandate to "protect the public health" by setting standards, after consideration of air-quality criteria and CASAC advice and adding a margin of safety, that protect sensitive populations from adverse health effects linked to the presence of a criteria pollutant in the outside air. U.S. Br. 22-25; Mass. Br. 28-34. This, they claim, is specific enough to pass constitutional muster.
Erroneous as a matter of statutory interpretation, this construction cannot provide the "primary standard," Buttfield, 192 U.S. at 496, against which EPA's actions can be measured. The reason is that EPA has acknowledged that section 109 does not require the NAAQS to eliminate all adverse health effects (i.e., achieve zero-risk). See, e.g., PM Rule, 62 Fed. Reg. at 38,653; Ozone Rule, 62 Fed. Reg. at 38,857. In other words, the purported "primary standard" of protecting against adverse health effects upon sensitive populations does not suggest any intelligible principle to guide the Administrator in deciding which adverse health effects to prevent, or which populations to protect. Moreover, all the statutory procedures on which their briefs rely only relate to that part of the NAAQS process that is not in controversy: namely, the congressional directive that NAAQS be based on the best science possible. The procedures do not address the issue of how the Administrator makes the highly consequential choice among alternative risk levels when the science is indeterminate, and when adverse health effects would still occur even under the alternatives selected. The lengthy recounting of statutory procedures in the opposing briefs thus should not deflect this Court's attention from the issue at hand: namely, as the court of appeals rightly identified it, what "intelligible principle" controls the discretion of the Administrator in choosing any risk level above zero when science does not provide the answer. Pet. App. 6a.
On this score, the Administrator offers only a series of so-called decisional factors, and it is on these factors that the validity of Lead Industries turns. The Administrator identifies those factors as "the nature and severity of the health effects involved, the size of the sensitive population(s) at risk, the types of health information available, and the kind and degree of uncertainties that must be addressed." Ozone Rule, 62 Fed. Reg. at 38,883. But, those factors (though certainly relevant) are not "determinate criteria for drawing lines" and "do not themselves speak to the issue of degree." Pet. App. 6a, 7a. More fundamentally, in simply identifying decisional factors, the Administrator has failed in her obligation "to apply some limiting standard, rationally related to the goals of the Act." Iov.'a Utils. Bd., 525 U.S. at 388.
The nub of the issue is this: Because Congress did not expect NAAQS to eliminate all health risks, as EPA acknowledges, the setting of NAAQS is essentially an exercise in the allocation of environmental health protections and the management of risk. See Pet. App. 16a-17a (suggesting that EPA could derive intelligible principle from Oregon plan for rationing health care); Breyer, supra, at 18-19, 23 (discussing impact of finite resources on risk regulation). Put another way, EPA has set standards at levels that will protect some populations from health impairment, but not others. The only reason for drawing such lines is economic, in the broad sense of that word: namely, tighter standards are progressively expensive, and approaching zero risk is counter-productive because it diverts society's resources away from other needs. See, e.g., H.R. Rep. No. 95-294, at 127 (1977) (rejecting argument that NAAQS could be set at a zero-risk level because such an approach "ignores all economic and social consequences and is impractical"). Cost-effectiveness is an intelligible principle by which to allocate health protection: if the scientific basis for choosing among various standards is indeterminate, the Administrator may select a standard because its benefits for the public are in her judgment reasonably commensurate with the costs to soctety.15
The medical-evidence decisional factors forwarded by the Administrator, by contrast, provide no principled basis for balancing and line-drawing. They are not rationally related to any coherent standard for allocating public health protections, nor do they fully "tak[e] into account the objectives of the
The Government will no doubt claim in reply that broadening the factors EPA may consider to include costs necessarily expands its discretion, and thus heightens any nondelegation problems. That is not so, and this Court directly rejected such an argument in the NBC case, where it refused to narrow the Federal Communications Commission's authority under the "public interest" standard to the consideration of "technological objections to the granting of a license." 319 U.S. at 216. Such factors, the Court held, would be indeterminate in the context of radio regulation, given the evident purposes of the Communications Act of benefiting the public at large: "If the criterion of 'public interest' were limited to such matters, how could the Commission choose between two applicants for the same facilities, each of whom is financially and technically qualified to operate a station?" Id. at 216-17.
The same is true here; just as withdrawing relevant factors from the FCC's consideration would thwart protection of "public interest," so too would withdrawing consideration of costs and other regulatory effects of a NAAQS unduly limit the Administrator in achieving her mandate of protecting the public health.'6 Permitting the Administrator to consider costs does not increase her discretion; it simply prevents arbitrariness, and administrative charades, in selecting among alternative standards, and tethers her decisionmaking to statutory purposes, which is the overriding purpose of the nondelegation doctrine. And it promotes the accountability that is the animating reason for the nondelegation doctrine, see Loving, 517 U.S. at 757-58, by permitting meaningful judicial, congressional and public scrutiny of these enormously important administrative decisions.
By identifying indeterminate decisional factors as the only limits on its authority, EPA has arrogated to itself the arbitrary power to drive health risks all the way down to zero,
Thus, just as this Court in Iowa Utilities Board interpreted the local competition provisions of the Telecommunications Act of 1996 to "requir[e] the [FCC] to determine on a rational basis which network elements must be made available [to competitors], taking into account the objectives of the Act," 525 U.S. at 391-92, so too this Court should interpret section 109 to require EPA to determine on a rational basis which populations will be protected in setting NAAQS. Because the allocation of public health protections in managing risks is, in the face of scientific uncertainty, a socioeconomic and political decision, the consideration of economic factors must play some part in any rational decisionmaking by EPA. Lead
C. The Arbitrariness Of The Lead Industries Rule Is Evident In The PM And Ozone Rule-makings.
The incoherence of the decisional factors that the Government claims are intelligible principles" is manifest in the PM and ozone rulemakings. This incoherence is not the product of the Administrator's abuse of a proper discretion, but of a statutory misconstruction of the Act that yields no standards rationally related to the statutory task of setting NAAQS.
1. PM Rulemaking
In the PM rulemaking, the Administrator applied the decisional factors and purported to have set the PM2.5 standard by relying on studies with "statistical significance to the 95% confidence level." U.S. Br. 32. According to the Government, the Administrator picked the lowest concentration at which an epidemiological study showed a statistically significant correlation between PM and an adverse health effect, and added a margin of safety. Id. at 32-33.
If this were so, it would be about as arbitrary an approach to multibillion-dollar decisions regarding the public health as could be imagined. As an initial matter, it is at least ironic that EPA is trumpeting statistical significance as the dispositive factor even as it is criticized for downplaying that factor in other risk-management contexts. See Michael Gough & Steven Milloy, EPA's Cancer Risk Guidelines.' Guidance To Nowhere, Cato Policy Analysis No. 263 (Nov. 12, 1996), available at http://www.cato.org/pubs/pas/pa263.html. Furthermore, even if lack of statistical significance may be reasonable grounds for rejecting a study,18 it by no means follows that statistical significance can reasonably be the principal basis for regulatory action. First, statistical significance indicates that the results are not the product of chance, but it does not measure "the extent or importance of a difference" detected by the study or "the strength or importance of an association" between two variables (e.g., a pollutant exposure and a health effect). David H. Kaye & David A. Freedman, Reference Guide on Statistics, in Federal Judicial Ctr., Reference Manual on Scientific Evidence 331, 378-79 (1994). Second, statistical significance (which is often simply a function of sample size, id. at 379) assumes the validity of the study design and the randomness of the sampling, among other factors; worthless studies may generate statistically significant results. USGS Site, supra, at 1, 3-4. Finally, EPA's reliance on this single factor is absurd given its overall conclusion that the scientific evidence on PM is rife with "significant uncertainties," PM Rule, 62 Fed. Reg. at 38,655¢including the fundamental uncertainty as to whether particles of a certain size are causing any health effects, or whether those detected result instead from other factors (such as a specific chemical agent). See Br. for CrossPet'rs at 13-15. No regulator worth her salt would make decisions on this basis, and it is doubtful the Administrator did so here. This is more likely yet another science charade¢the invocation of technical jargon to support a decision made on other, undisclosed factors.19
19 It is noteworthy that EPA disregards good science when it conflicts with the agency's preferred policy, see Chlorine Cheniistr~' Council v. EPA, 206 F.3d 1286, 1291 (D.C. Cir. 2000) (criticizing "EPA's disregard of its own scientific findings"); Chemical Manufacturers Ass',i v. EPA, 28
The Govemment defends the ozone rulemaking with the ipse dixit that EPA "identified important and meaningful differences in the character of the scientific evidence regarding risks," U.S. Br. 33, but it is clear that EPA engaged in arbitrary linedrawing. For ozone, the evidence indicated that 41,000 children in nine urban areas would suffer moderate or severe pain on deep inspiration once or more per year at a concentration of .09 parts per million ("ppm"); 27,000 children would at the new NAAQS of .08 ppm; 22,000 would ai .08 ppm with a slightly lower average daily maximum; and approximately 9,000 would at .07 ppm. Similarly, 97,000 children would suffer large lung-function decreases of at least 20% at .09 ppm; 58,000 at the new NAAQS of .08 ppm; 43,000 at .08 ppm with a slightly lower average daily maximum; and approximately 12,000 at .07 ppm.20
20 Ozone Rule, 62 Fed. Reg. at 38,865; R.G. Whitfield, A Probabilistic Assessment of Health Risks Associated With Short-Terni Exposures to Tropospheric Ozone: A Supplement 2 1-22 (Jan. 1997) (6 Ozone JA 2325, 2352-53) (prepared for EPA); National Ambient Air Quality Standards for Ozone; Proposed Decision, 61 Fed. Reg. 65,716, 65,725 (1996) (0.07 ppm). The figures used for the .08 and .09 ppm standards are drawn from EPA's final rule, which did not repeat the risk analysis for the 0.07 ppm alternative standard that was included in the proposed rule. While the original and final risk assessments used slightly different methodologies, they show roughly similar disparities in health effects. See 61 Fed. Reg. at 65,725.
Congress fully intended for the Administrator to consider the cost-effectiveness of alternative standards in the face of scientific uncertainty. "The language of the Act does not withdraw such a [consideration)" from the Administrator's judgment, "and there is no evidence that Congress did not mean its broad language to carry the authority it expresses." NBC, 319 U.S. at 218. As the court below acknowledged, Pet. App. ISa, cost-effectiveness of a precautionary standard is an intelligible principle where the scientific evidence is indeterminate¢indeed, it is the generally accepted principle of risk management and line-drawing in the face of scientific uncertainty. See Hatch/Bliley Br. 7-8. Moreover, it is a principle which the judiciary can enforce to determine whether congressional will has been obeyed, Yakus, 321 U.S. at 426, and such considerations must be out in the open for the judiciary to determine whether the Administrator has substantially complied with the statute.21
CONCLUSION
On the grounds set forth above, the judgment below should be affirmed.
22 It is the function of this Court, not EPA, to interpret the statute to define intelligible principles. Chevron U.S.A. Inc. v. NRDC, 467 U.S. 837 (1984), is inapposite to nondelegation issues. Chevron is premised on congressional intent to delegate rulemaking power to the agency. See FDA v. Brown & Williamson Tobacco Corp., 120 S. Ct. 1291, 1314 (2000); Chevron, 467 U.S. at 844-45. A delegated power cannot be invoked to supply the principles that make the very delegation constitutionally permissible. The Constitution requires that the limiting principle must come from the statute itself, as construed by the judiciary.
LLOYD N. CUTLER CARTER G. PHILLIPS
C. BOYDEN GRAY ALAN CHARLES RAUL *
WILMER, CUTLER & PICKER1NG STEPHEN B. KINNAIRD
2445 M Street, N.W. SIDLEY & AUSTIN
Washington, D.C. 2003 1722 Eye Street, N.W.
(202) 663-6000 Washington, D.C. 20006
(202) 736-8000
Counsel for Amici Curiae
September 11, 2000 * Counsel of Record