29 ELR 10653 | Environmental Law Reporter | copyright © 1999 | All rights reserved

Run Over by American Trucking Part I: Can EPA Revive Its Air Quality Standards?

Editors' Summary: In the first of two Articles, Professor Craig N. Oren examines the recent "blockbuster" opinion in American Trucking Ass'n v. United States Environmental Protection Agency, in which a panel of the D.C. Circuit Court of Appeals remanded EPA's air quality standards for ozone and particulate matter. The author begins with a discussion of the statutory and regulatory framework and describes the court's holding. From there, the Article moves to a detailed explanation of the reasons why the oft-articulated concerns about possible revival of the delegation doctrine are overstated. Professor Oren asserts that American Trucking should, instead, properly be viewed as a finding that the EPA was arbitrary and capricious in setting the standards because the Agency's decisions were insufficiently explained. The Article suggests that while American Trucking signals the need for reexamination of the criteria for establishing ambient air quality standards, a number of obstacles, perhaps insurmountable, stand in the way of meaningful reform. Expressly incorporating cost considerations would, the Article notes, complicate regulatory analyses. Moreover, overt consideration of cost would arguably be inconsistent with the statutory goal of public health protection. Nor is the alternative approach suggested by Judge Williams in American Trucking, based on an Oregon Medicaid program, likely to succeed in the CAA context. Professor Oren argues that the agency responses that occurred subsequent to the Lockout/Tagout and Vinyl Chloride decisions are, likewise, of doubtful assistance. Perhaps, the Article contends, the time has come for courts to provide the EPA increased deference in the setting of air quality standards. Although American Trucking likely does not mark the resurrection of the delegation doctrine, Professor Oren concludes that the decision will undoubtedly significantly complicate the process of developing ambient air quality standards.

Professor of Law, Rutgers (The State University of New Jersey) School of Law—Camden. The author wishes to thank all those who assisted him.

[29 ELR 10653]

The question is not one of science, the real question is one of judgment.

EPA Administrator Carol Browner, December 1996 (proposing new air quality standards for ozone and particulate matter).¹

I think it is not a question of judgment, I think it is a question of science.

EPA Administrator Carol Browner, March 1997 (defending the same proposed standards).²

Last May, in American Trucking Ass'n v. U.S. Environmental Protection Agency,³ a panel of the D.C. Circuit Court of Appeals refused to uphold new air quality standards for ozone and particulate matter promulgated by the U.S. Environmental Protection Agency (EPA), and instead remanded them to the Agency for further consideration.⁴ [29 ELR 10654] The panel's decision has proven quite controversial. EPA Administrator Browner, whose standard-setting choices were remanded, calls the court's result "extreme, illogical and bizarre."⁵ Professor Cass Sunstein, a prominent administrative law scholar, criticizes the decision as part of a trend toward "conservative activism" by courts and declares that the court "struck down a provision of the Clean Air Act, as interpreted by the Environmental Protection Agency."⁶ Commentators have speculated that the reasoning used by the court revives the nondelegation doctrine last seen in the 1930s,⁷ and could lead to invalidation of other environmental regulations.⁸ Others have suggested that the decision requires that EPA consider costs openly in setting air quality standards.⁹

There can be no denying that the decision is a "blockbuster."¹⁰ But the reasons for this are not those usually indicated. The invocation of the "nondelegation" doctrine of the court's decision may represent rhetorical flourish as much as an attempt at doctrinal exhumation. Instead, American Trucking should be read as indicating that the courts will continue to take a hard look at the quality of an agency's explanation for its actions, despite academic criticisms that this leads to "ossification" of rulemaking. Indeed, the decision would be sounder if it were articulated expressly on this basis. From a Clean Air Act (CAA) standpoint, the decision complicates EPA's responsibility to obey the Act's command to set air quality standards to protect public health and welfare. For that reason, the decision will further slow revision of the standards to take account of current scientific data.

There is a second important aspect to American Trucking that has been ignored thus far in the commentary about the case: the decision holds that the CAA denies EPA the authority to implement any tightened air quality standard for ozone.¹¹ Thus, even if EPA can explain its revised air quality standard for ozone to the court's satisfaction, the Agency cannot implement the new standard without new legislation from Congress. Given attitudes on Capitol Hill toward EPA and environmental protection, the decision strikes a body blow at efforts to remedy our nation's most serious air quality problem. This aspect of the decision will be discussed in a forthcoming Article.

Background

Understanding American Trucking requires an appreciation of what was at stake and what the court did. Since 1970, the CAA has authorized EPA to establish and revise periodically national ambient air quality standards—standards that prescribe maximum permissible levels for pollutants in the outside air to which the public has access.¹² These air quality standards are crucially important both in determining the amount of health and welfare protection that the nation will seek to achieve and in driving much of the regulatory structure of the Act; the tougher the air quality standards, then the more demanding the emissions control measures that are required of cars, industry, and other sources of air pollution. Hence, revising air quality standards is quite contentious; these particular standards brought about intervention from the White House to decide heated interagency disputes.¹³ As a result, the air quality standards are altered only infrequently, despite the Act's command that the standards be reviewed every five years.¹⁴ EPA's revision represented only the second alteration in the ozone and particulate matter (PM) standards since they were originally set in 1971, and environmental groups had to bring law suits to bring about even this revision.¹⁵

Under both of the new standards, the number of counties classified as in nonattainment—that is, in violation—of air quality standards would drastically increase, and greater emission reduction efforts would be required of areas already in nonattainment.¹⁶ In both cases, meeting the new standards would necessitate further controls on emissions of volatile organic chemicals and nitrogen oxides (NO_x); these compounds not only react in the presence of sunlight to form [29 ELR 10655] ozone and other photochemical oxidants,¹⁷ but also contribute to the formation of fine particles,¹⁸ which are a special focus of the standards for PM. The PM standards would also mean controls on emitters of gases that can be transformed into fine particles. Literally millions of sources contribute to these emissions. Implementation of the standards also raises regional issues of the kind that embittered the debate over acid rain for so long; the centerpiece of EPA's strategy to implement the new ozone standard is a 22-state program for control of NO_x whose burden would fall largely on utilities and other large sources of NO_x in the Midwest.¹⁹ Thus, it is not surprising that litigation was brought to overturn the air quality standards. What is startling is that, for the first time since the early 1970s, a challenge by industry to an ambient air quality standard proved successful.²⁰

The Court's Holding

American Trucking did not, to be sure, accept all of the theories used to challenge the standards. For instance, the court, speaking through Judge Douglas Ginsburg, rejected the claim that EPA must comply with the procedural and substantive requirements of the Unfunded Mandates Reform Act²¹ and the Regulatory Flexibility Act (as amended by the Small Business Regulatory Enforcement Fairness Act)²² in setting the ambient air quality standards.²³ These holdings in effect exempt the establishment of EPA's air quality standards from the most recent spate of "regulatory reform" legislation. Similarly, Judge Ginsburg rejected the theories that EPA must consider costs in setting air quality standards²⁴; that EPA must file an environmental impact statement under the National Environmental Policy Act²⁵; that EPA must consider the health consequences of the economic effects of implementing the new standards²⁶; and that EPA must show a biological mechanism for explaining why epidemiological studies show a correlation between concentrations of PM and hospitalization and death.²⁷ These victories for the Agency should not be minimized; without them, the task of revising the standards would be immensely complicated. Yet the plaintiffs' success gives pause. While the court does not make it impossible for EPA to revive its standards, the court sends a strong signal that the court believes the Agency needs to re-think its approach.

Most attention has focused on the court's holding, written by Judge Stephen Williams and joined by Judge Ginsburg, that the ozone and PM standards must be remanded because EPA failed to articulate an intelligible standard for setting them.²⁸ To quote the court:

Here it is as though Congress commanded EPA to select "big guys," and EPA announced that it would evaluate candidates based on height and weight, but revealed no cut-off point. The announcement, though sensible in what it does say, is fatally incomplete. The reasonable person responds, "How tall? How heavy?" … The agency rightly recognizes that the question is one of degree, but offers no intelligible principle by which to identify a stopping point.²⁹

This theory initially appears to resuscitate the doctrine, invoked by the U.S. Supreme Court to invalidate statutes for the first and last time in the 1930s, that the courts ought to limit the extent to which Congress delegates law-making authority to administrative agencies.³⁰ But this seems unlikely. As American Trucking itself notes, current Court decisions do not apply the doctrine, despite Justice William Rehnquist's attempts in the 1980s to revive it.³¹ Moreover, [29 ELR 10656] this case does not seem a particularly outstanding example of a vague delegation; as Judge David Tatel notes in his dissent, the Supreme Court has upheld statutes allowing regulation under far less specific standards.³² For instance, the Court has upheld regulation by the Federal Communications Commission (FCC) of cable TV under a statute allowing FCC regulation "as public convenience, interest or necessity requires."³³ The choice whether to regulate an entire industry that was unknown when the original authorizing statute was passed seems far more fundamental, and to involve far greater issues of notice to the affected industry, than the choice between, say, an 0.09 part per million (ppm) ozone standard and a 0.07 ppm standard, as important as the latter is. Perhaps most fundamentally, there is an obvious difference between the delegation doctrine and American Trucking. The delegation doctrine invalidates the legislature's enactments; by contrast, American Trucking is expressly directed only at the Agency's interpretation of the statute.

The Lockout/Tagout Connection

Judge Williams suggests in American Trucking that the court's delegation analysis is in accord with his opinion for the D.C. Circuit in the first of the Lockout/Tagout cases.³⁴ Those cases concerned challenges to Occupational Safety and Health Administration (OSHA) regulations that require industry to prevent equipment from going into operation while it is undergoing maintenance. OSHA had interpreted the relevant portion of the OSH Act to allow the agency, upon finding that a workplace practice poses a significant risk, to require steps that would minimize the risk so long as the steps are economically and technologically feasible.³⁵ Such a reading of the statute, according to the court, gave the agency virtually unfettered discretion in deciding how strictly to regulate and therefore raised "a serious nondelegation issue"³⁶; the court therefore rejected OSHA's view as contrary to the statute and instead remanded the rule so that OSHA could develop a new interpretation.

Lockout/Tagout I indicates there is less to the court's use of the delegation doctrine than initially appears. Its approach is not particularly antiregulatory; broad agency power of the kind criticized by the court could as easily be used to set lax as stringent standards. Thus, in American Trucking, the court worried aloud that EPA could potentially allow concentrations just below the level of London's killer fog.³⁷ More importantly, the analogy to Lockout/Tagout I indicates that the court is not trying to impose an especially stringent test on the EPA. On remand, OSHA interpreted the statute to require the agency, once it identifies a significant risk, to establish a safety standard that provides "a high degree of worker protection"; according to the court in the final Lockout/Tagout opinion, this sufficiently cabined the agency's discretion to satisfy the nondelegation doctrine and to allow the agency to promulgate the same standard that had been upset by Lockout/Tagout I.³⁸ It is hard to imagine that the agency ever thought seriously of controlling at some other level, and so the court's invocation of the delegation doctrine proved to be more bark than bite.

There is another sense in which the Lockout/Tagout doctrine is less drastic than initially appears. Usually, when a court finds administrative action to be defective, the court vacates the action. In Lockout/Tagout, by contrast, the court allowed the agency to keep the regulation on the books while the agency prepared a new explanation³⁹; the court reasoned that it was "most unclear" that correction of the defects in the agency's reasoning would result in a different standard.⁴⁰

Similarly, the American Trucking panel vacated only in part the air quality standards that it rejected.⁴¹ The failure to vacate is especially remarkable because there has been considerable debate in the D.C. Circuit about whether the Administrative Procedure Act allows a court to remand without vacating.⁴² This course offers the agency a key advantage; the agency need not go through a complete new rulemaking, but instead can simply adhere to its original rule and issue a supplemental statement (usually after a fresh round of public comment) that fixes the defects the court found in its explanation. This is, in fact precisely what OSHA did in Lockout/Tagout. By remanding without vacating, American Trucking may be inviting a similar approach by EPA. Observers should therefore not be surprised if EPA attempts to adhere to the standards that were remanded by American Trucking rather than initiating a new rulemaking.

It is not clear why the American Trucking panel decided not to vacate. One reason, acknowledged by the court, is that the ozone standard is in any case unenforceable under the court's reading of the CAA.⁴³ The court may also have been troubled by the consequences of nullifying the new standards,⁴⁴ particularly since EPA had already revoked the previous [29 ELR 10657] ozone standard in many areas.⁴⁵ But the choice to remand without vacating may also shed light on the underlying rationale of the case.

Typically, the D.C. Circuit remands without vacating when the court believes that the agency action is consistent with the statute, but requires a more careful explanation.⁴⁶ Indeed, American Trucking, like Lockout/Tagout, should be seen as resting largely on the court's desire for a better explanation from the agency. This is even more true of American Trucking than of Lockout/Tagout. The latter case professes to represent an attempt to save a statute from invalidation through a narrowing construction. Lockout/Tagout can therefore be seen as an application of the Benzene case,Industrial Union Department, AFL-CIO v. American Petroleum Institute,⁴⁷ in which the Court read the same statutory provision narrowly, in part to avoid a delegation issue.⁴⁸ But American Trucking is not parallel. Unlike Benzene or Lockout/Tagout, the court in American Trucking never identifies the agency statutory construction that it believes poses a delegation issue.⁴⁹ Rather, the court criticizes "EPA's formulation of its policy judgment"⁵⁰—suggesting, again, that the court's difficulty is less with the Agency's reading of the statute than with the quality of its explanation for how it exercised its statutory authority.

Instead, the connection between American Trucking and the delegation doctrine can be traced to Amalgamated Meat Cutters & Butcher Workmen of North America, AFL-CIO v. Connally,⁵¹ Judge Harold Leventhal's noteworthy opinion for a three-judge court upholding President Nixon's establishment of wage-price controls in the early 1970s. In rejecting the plaintiffs' claims that the Economic Stabilization Act,⁵² the statute under which President Nixon ordered the controls, gave an unconstitutional "blank check" to the Executive Branch to regulate as it pleased, the court observed:

Another feature that blunts the "blank check" rhetoric is the requirement that any action taken by the Executive under the law … must be in accordance with further standards as developed by the Executive. This requirement, inherent in the Rule of Law and implicit in the Act, means that however broad the discretion of the Executive at the outset, the standards once developed limit the latitude of subsequent executive action.⁵³

American Trucking seems to view this language as adopting Professor Kenneth C. Davis' proposal in the 1970s that courts use the delegation doctrine to ensure development of administrative standards.⁵⁴ This reading is questionable. Amalgamated Meat Cutters does not purport to hold that the delegation doctrine obligates an agency to develop the standards. Rather, the court is instead saying that it is unnecessary to invoke the delegation doctrine because the statute itself, as well as principles of law apart from the delegation doctrine, demand the development of standards that control administrative action. The court therefore was giving a reason not to invoke the delegation doctrine, rather than attempting to give the doctrine new meaning. In addition, as American Trucking itself concedes, one of the delegation doctrine's key functions—to ensure that Congress itself makes the basic policy decisions—is not fulfilled by interpreting the doctrine to require that agencies themselves develop standards.⁵⁵

Arbitrary and Capricious?

In fact, the principal assurance of the Rule of Law has proven not to be the delegation doctrine itself, but rather judicial review of agency decisionmaking.⁵⁶ The courts have become increasingly more vigilant in policing the quality of agency explanations to ensure that agencies have not been arbitrary and capricious. Judge Leventhal himself helped to develop this trend when he wrote on behalf of the D.C. Circuit that agencies must show they have taken a "hard look" at the problem before them.⁵⁷ The Court summarized this standard in Motor Vehicle Manufacturers Ass'n v. State Farm Mutual Automobile Insurance Co.⁵⁸ as requiring that an agency "articulate a satisfactory explanation for its action including a rational connection between the facts found and the choice made"⁵⁹; or, to put it more concisely, that an agency demonstrate from its explanation that it has engaged in "reasoned decision-making."⁶⁰ Such a requirement vindicates Amalgamated Meat Cutters' concern about maintaining the Rule of Law because an agency's explanation will, at [29 ELR 10658] least implicitly, create standards to guide future agency action; an agency will need to either follow these standards, or explain the need for departing from them.⁶¹

American Trucking can readily be explained as deciding that the Agency was arbitrary and capricious in setting the ozone and PM standards in that EPA did not sufficiently explain its decisions. The court's critique of EPA's reasoning is illustrative. The court notes that EPA defends its PM standards on the basis that "there is greater uncertainty that health effects exist at lower levels than the level of the standards"; but, the court responds, such an explanation could just as easily justify a standard of zero as the standard the Agency picked.⁶² The same is true, according to the court, of EPA's choice of an ozone standard.⁶³ In both instances, the court is essentially instructing EPA that the quality of its explanation is insufficient to satisfy the court that reasoned decisionmaking took place.

The precedent most on point for American Trucking is therefore American Lung Ass'n v. U.S. Environmental Protection Agency⁶⁴ decided just a year earlier by a panel that included two of the three judges who later participated in deciding American Trucking. American Lung concerned a challenge by environmentalists to EPA's refusal to set a short-term air quality standard for sulfur dioxide (SO₂). Such a standard would have protected asthmatics living near coal-fired power plants and other large sources of SO₂ from brief bursts of emissions that can cause bronchoconstruction—shortness of breath, coughing, wheezing, chest tightness, and sputum production—during moderate exercise. EPA concluded that the effects on asthmatics occurred and that they were significant, but that the effects did not warrant establishment of a national ambient air quality standard. The court held that EPA had failed to explain why it believed that the effects caused by short-term bursts do not constitute a public health problem; the court therefore remanded the Administrator's decision.⁶⁵ American Trucking is similar in its result, if not its reasoning. An instance in which the Agency has not expressed a reason for its decision is analytically not greatly different from one, like in American Lung, in which the Agency's statements seem contradictory.

Explaining the Court's Decision

Agency decisions, even in important matters, are struck down as arbitrary often enough that there would have been nothing exceptional about a decision invalidating EPA's standards on this ground. Indeed, the court's decision can be seen as rejection of those commentators who have suggested that the arbitrary and capricious test is ossifying agency rulemaking.⁶⁶ It may be significant, then, that the court chose instead to couch its decision in terms of delegation. Expressing the conclusion in terms of delegation rather than in terms of insufficient explanation may serve to emphasize the court's determination to compel EPA to do a better job of explaining its policy determinations. It is as though the court believes that, if the arbitrary-and-capricious rationale of American Lung failed to get the Agency's attention, then a bigger hammer is needed.⁶⁷ Moreover, invalidating an agency decision as arbitrary and capricious implies criticism only of the agency. But a decision based on delegation invokes at least the shade of the cases invalidating statutes in the 1930s; it implies that Congress should consider whether it has sufficiently delineated the agency's authority. Indeed, the American Trucking panel specifically suggests that EPA might wish to seek congressional action.⁶⁸ In addition, the court, by referring to the lack of an intelligible standard, may have been expressing a sense that EPA's explanations of its decisions in setting these standards have, for all of their length, a rather ad hoc quality; the Agency sets forth a large number of factors that influenced its decision, but does not cite one as decisive. While such an approach is often followed in judicial opinions, the court may fear that an agency can hide its real rationale with such an approach.⁶⁹

There is a final importance to articulating American Trucking in terms of delegation rather than the arbitrary and capricious test: doing so gave the court an easy way of distinguishing earlier cases in which the D.C. Circuit had upheld revisions by EPA of the same standards on explanations [29 ELR 10659] that resembled those at issue in American Trucking.⁷⁰ The court needed only to say that the nondelegation issue had not been raised in those cases.⁷¹ In this way, the court was relieved of the need for explaining how the revisions at issue in American Trucking differed from previous changes to the ambient air quality standards. But the court's move carries a corresponding disadvantage: it implies that the court was questioning the vitality of earlier decisions upholding revisions. Moreover, the court, by not distinguishing the earlier cases on their facts, invites the criticism that factually similar cases are being decided differently.⁷²

One answer to this criticism is to attempt the very factual distinction that the court avoided drawing, and to argue that EPA's decisionmaking was in fact inferior in this particular rulemaking to the previous revisions of the same air quality standards. It would be arrogant in a brief article to reach that conclusion about a rulemaking that spanned several years and several thousand pages of record, particularly when one judge of the three-judge American Trucking panel was able to find a reasoned explanation for the Agency's major decisions.⁷³ Still, the court seems correct in finding troubling signs in this rulemaking. One such sign was the Agency's failure to explain clearly its decision to set separate standards for fine and coarse particles, and yet include fine particles in the definition of coarse particles. Doing so seemed to mean that fine particles would be counted twice in determining if areas are in attainment.⁷⁴ Judge Tatel, the one member of the American Trucking panel who dissented from the court's delegation discussion, had no difficulty finding EPA arbitrary in making this decision.

A second danger sign is that EPA appears to have been inconsistent in applying the statute. On the one hand, EPA stated that it based its standard-setting decisions on health considerations alone.⁷⁵ On the other hand, EPA seemed to be making use of nonhealth considerations when doing so helped the Agency support its decisions. The coarse particle issue is an example. Here EPA tried to justify its decision to include fine particles within coarse particles on the "pragmatic" grounds that present air quality monitoring does so. This amounts to an attempt to use nonhealth considerations to justify a standard that is to be explained only on health grounds. The same is true of the ozone standard. EPA rejected a tougher standard than the one it adopted partly on the grounds that the tougher standard would be close to background levels of ozone.⁷⁶ But this consideration is impermissible under EPA's own hypothesis that nonhealth considerations are irrelevant. Instead, EPA must either explain why natural concentrations do not pose a health threat—certainly the mere fact that the concentrations are natural does not justify such a conclusion—or perhaps argue that the Act denies it the authority to regulate natural background levels.⁷⁷ Conversely, EPA, although properly not requiring quantifiability in studies that show health effects from ozone, appears to have required such quantifiability in studies asserting that man-made ozone concentrations help protect health.⁷⁸

[29 ELR 10660]

There are more overarching problems with the Agency's decision. Administrator Browner's remarks, as quoted at the beginningof this Article, are indicative of a fundamental confusion within the highest levels of EPA. Administrator Browner seems to believe that either science or policy is the primary consideration in standard-setting. In fact, both are equally important. On the one hand, the decision of who should be protected, and what effects they should be protected against, is an ethical decision, not a scientific one.⁷⁹ The Supreme Court recognized this point as long ago as its Benzene decision.⁸⁰ For this reason, the majority in American Lung is on sound ground in suggesting that EPA cannot justify its standards simply by reference to the advice of the Agency's Clean Air Scientific Advisory Committee (CASAC).⁸¹ On the other hand, the Agency needs to know the facts, as best they can be determined, so that it has a basis on which to make policy decisions. Scientific knowledge is therefore a sine qua non. CASAC's endorsement may not be enough to vindicate a standard, but its disagreement would probably sink one. Indeed, as Professor Peter Strauss has recently stressed, the legitimacy of delegating issues such as the ambient air quality standards to agencies like EPA rests on the need for expertise to assess scientific and other complex information.⁸² By initially seeming to deprecate the role of science, Administrator Browner may have inadvertently opened the door for challenges to the Agency's decisions.

Is the System Broken?

Even if these attempts to distinguish this particular revision fail, a broader argument can be used to defend the court from the charge of inconsistency: to urge that the time has come for a reevaluation of the criteria for setting ambient air quality standards. American Trucking's demand for an intelligible principle appears to be a call for such a reevaluation. Unfortunately, searching for an intelligible principle is unlikely to be successful.

Since the 1970s, a variety of scholars have argued that there is no sensible way to obey the statute's command that ambient air quality standards be set without regard to cost.⁸³ There are two reasons for this. One is scientific uncertainty about what effects are occurring and to whom. Given this uncertainty, any decision to set a standard trades off the cost of implementing the standard against the possibility that the standard may be efficacious. Second, and more fundamentally, available scientific data does not allow the identification of thresholds at which effects begin to occur.Instead, current knowledge suggests that, at any level of pollution, some effect may be happening to someone somewhere. The standard-setter must decide who should be protected and what effects should be protected against. Any such decision necessarily involves at least an implicit weighing of costs versus benefits.

The PM and ozone standards well illustrate these points.⁸⁴ Since 1971, when EPA set the original ambient air quality standards for PM, the Agency has become increasingly persuaded that these standards should concentrate on small, respirable particles. In 1987, EPA revised the PM standards to regulate only particles 10 microns or smaller (known as PM[10]); this meant that industry could no longer meet the standards simply by controlling larger particles, such as road dust.⁸⁵ New evidence indicates that the Agency should focus special attention on particles that are 2.5 microns or smaller (PM[2.5]). Data show a correlation between concentrations of fine particles and increases in mortality and hospital admissions. For instance, EPA staff believes that fine particles are associated with as many as 1,000 deaths per year in Philadelphia County, even though the area meets the 1987 air quality standards for PM.⁸⁶ Other researchers agree that a correlation exists between PM concentrations and mortality and morbidity, but assert that factors other than particulate levels are responsible for it.⁸⁷ One researcher has even suggested that variations in hospital admissions among days of the week may be responsible for the observed effects.⁸⁸ One particular problem is that, since there is no air quality standard for fine particles and hence no required monitoring of them, there is doubt about what levels of particles people are exposed to. Consequently, the chair of CASAC endorsed setting a standard in part so that data would be gathered that could be used to judge whether such a standard is necessary and what its level should be.⁸⁹ In other words, EPA was in [29 ELR 10661] the position of setting a PM standard to obtain the data to decide to set such a standard.⁹⁰

Any standard-setting decision therefore involves the recurrent and difficult question of what to do in the face of scientific uncertainty. To further complicate matters, current data on PM do not show any threshold; rather, any exposure to fine particles appears to impose some extra chance of death or illness. Thus, EPA must decide how much risk it is willing to accept in a situation in which the Agency cannot accurately gauge the size of the risk; the Agency must, as it were, decide how much risk of a risk to accept.

Ozone presents even more vexing problems. Here there is controversy not only about what effects ozone causes, but also about which effects the air quality standards ought to protect against; as the Agency put it in promulgating the standard, "The core issue … is who is to be protected and from what."⁹¹ Individuals who are sensitive to ozone—active children, outdoor workers, those with preexisting respiratory disease as well as some otherwise normal individuals⁹²—suffer a continuum of effects that range from shortlived discomfort when breathing deeply to persistent uncontrollable cough, and severe discomfort on exercise or deep breath, that last longer than a day and are likely to limit activity.⁹³ These effects are responsible for 10 to 20 percent of all summertime emergency room visits related to respiratory problems,⁹⁴ but only about I percent of hospital admissions for asthma.⁹⁵ While probably everyone would agree that some of the effects are health effects, drawing the line is clearly a daunting task; since there is no threshold level at which no effects exist, any standard will allow some effects.

Other considerations make the line-drawing even more difficult. Ozone may cause harm beyond its noticeable effects. An individual who is exposed to ozone for three or four days will experience lessening effects. As a result, the individual may exercise or go outdoors, and potentially suffer even more serious consequences.⁹⁶ Another difficulty is that asthma is apparently becoming more common—even though air pollutant concentrations have been dropping—and appears to be concentrated among the poor and the non-white. According to the Centers for Disease Control in Atlanta, the incidence of acute asthma attacks among children has doubled in the past decade, even though highly effective medications have been developed. Asthma is the most common cause of hospitalization among children—five million hospitalizations each year—and deaths among children with asthma rose by 78 percent from 1980 to 1993. The disease is concentrated in heavily populated urban areas. A recent study in New York City shows that hospitalization rates for asthma are far higher in poorer, minority areas than in affluent areas.⁹⁷ Setting the ozone standard therefore involves difficult issues of assuring environmental justice in a context in which it is not clear how much of the health burden on the minority population is being caused by ozone, and in which at least some of the costs of compliance might fall on efforts to bring economic prosperity to the inner city.⁹⁸

To make matters worse, there is controversy about the significance of the reductions in health effects brought about by a tighter standard. EPA believes, for instance, that a standard of 0.09 ppm would allow over 1 percent of out-door children, or 41,000 children, to experience on 220,000 occasions per year moderate or severe pain while breathing deeply. The 0.08 ppm standard actually set by EPA results in this effect occurring 120,000 times a year to 27,000 children, or 0.9 percent of outdoor children.⁹⁹ There is no obvious way to decide whether this reduction in the affected population is worthwhile. One must choose whether to focus on the modest percentages or the larger absolute numbers. In this instance, focusing on percentages, as laypeople are apt to do, results in a lower evaluation of risk than an expert might reach.¹⁰⁰ The uncertainty in these estimates makes analysis even more difficult.¹⁰¹ Clearly, as CASAC advised EPA, "there is no 'bright line' which distinguishes any of the proposed standards … as being significantly more protective of public health."¹⁰²

[29 ELR 10662]

EPA inevitably must therefore consider costs in standard-setting to help decide how stringent to make the standards. Indeed, EPA decisionmakers have admitted that they examine cost data when deciding on the levels of the standards.¹⁰³ For this reason, it might seem obvious that EPA be required to justify its decisions in terms of cost.¹⁰⁴ In that way, a process of cost consideration that goes on behind the scenes would become visible for public comment and reaction.¹⁰⁵ Yet, as American Trucking reiterates, the Act's text and legislative history make clear that EPA may not consider costs in standard-setting.¹⁰⁶

Nonetheless, the exclusion of costs from the setting of air quality standards can be defended even if it is inevitable that costs be taken into account in setting the standards.¹⁰⁷ The exclusion of costs from the standard-setting language of the Act can be explained as a means, albeit perhaps crude, of minimizing as much as possible the influence of cost calculations upon EPA's standard-setting decisions and upon judicial review of those decisions. Consider, for example, EPA's previous revision of the ozone standard: Administrator Douglas Costle's decision in 1979 to promulgate an ambient standard for ozone of 0.12 ppm rather than the 0.10 standard urged by environmentalists or the 0.16 or 0.20 standard preferred by industry. Insofar as Costle chose to disregard the costs of implementation, the absence of cost language in the statute helped to protect him from public criticism and from judicial reversal for improperly ignoring costs.¹⁰⁸ To put the point in administrative law terms, the Administrator's decision, to the extent it proceeds in the face of costs, is "committed to agency discretion by law" and hence unreviewable.¹⁰⁹ Considering the many difficulties of accurately determining the cost of attaining a given standard,¹¹⁰ much less intelligently weighing those costs against the benefits of attainment,¹¹¹ Congress' decision to insulate the Administrator from cost-based arguments is an understandable way of assuring that standards be set.

The exclusion of costs from the statute, while expanding Administrator Costle's discretion to be protective, also limited his ability to decide not to be protective. In rejecting the environmentalist 0.10 position, Costle had to frame his arguments in terms of the weakness of the scientific evidence for effects at very low level or the lack of public health significance of the effects that do exist. Consideration of cost may be implicit in considering whether a given effect is significant; but the Administrator who does not wish to safeguard health to the utmost is forced to address broader concerns than cost. Moreover, precisely because cost considerations cannot be invoked as a rationale for the decision, the Administrator is discouraged from putting much weight on cost. This acts to limit, though not eliminate, the Administrator's ability to decline to be protective.¹¹²

"Intelligible Principles"

Expressly incorporating cost considerations would militate against public health protection and complicate analysis. EPA might, alternatively, proceed by trying to devise some kind of decision rule for deciding how to set ambient air quality standards that does not involve cost considerations. Judge Williams, writing in American Trucking, suggests that the Agency should construct a formula comprised of the probability of the effect, the population that would be affected, and the effect's severity, for deciding the level of the standards.¹¹³ To judge the effect's severity, the court advises EPA to follow Oregon's example in designing its Medicaid program for poor people and use both polling data and expert opinion to determine what conditions should be covered.¹¹⁴

The court is careful to refrain from holding that EPA must use this approach, stating that the Agency "may well find a completely different method for securing reasonable coherence."¹¹⁵ Lockout/Tagout itself illustrates that suggestions like this from Judge Williams need not be followed by the Agency. There, Judge Williams argued that OSHA could use cost/benefit analysis as its decision rule, and included a brief paean to the virtues of that technique, quoting Benjamin [29 ELR 10663] Franklin to suggest that it amounts to everyday common sense.¹¹⁶ Nevertheless, Judge Williams wrote the opinion upholding OSHA when it later decided not to use cost-benefit analysis.¹¹⁷

The references in American Trucking to Oregon's Medicaid program should therefore be taken merely as suggestions. This is fortunate, because the Oregon technique does not seem workable. It is not new for the D.C. Circuit to recognize that more than one factor is relevant in deciding whether to regulate an environmental risk; the landmark decision in Ethyl Corp. v. U.S. Environmental Protection Agency similarly held that, in deciding whether to regulate, EPA must consider both the magnitude of an effect and the likelihood that the effect will take place.¹¹⁸ Indeed, American Trucking reinforces Ethyl Corp.'s teaching that EPA need not show that it is more likely than not that harm would occur.¹¹⁹ But the difficulty of building a structure as formal as the court suggests is daunting.

In saying so, I do not wish to disparage the Oregon experience. A recent analysis indicates that advocates of environmental protection and of citizen participation might well draw comfort from what happened in Oregon. In practice, little rationing occurred; rationing (or "prioritization", as it was euphemistically known) accounted for only 2 percent of the Medicaid health savings in Oregon in the first five years of operation.¹²⁰ Moreover, in practice, Oregon's rationing decisions have tended to be made not on a cost-benefit basis, but are rather influenced by political pressures and administrative judgments.¹²¹ American Trucking is therefore incorrect in its apparent assumption that the Oregon experience will lead to greater rationality. In fact, the primary importance of prioritization was that it created a way for the public to be heard on health care decisions; in this way, prioritization helped to build public confidence in the political process, leading eventually to expansion of efforts to cover health care expenses.¹²² For this reason, even low-income groups have become supporters of the Oregon prioritization effort.¹²³ Similarly, a public debate on what precisely amounts to a health effect could create some measure of consensus on how far air pollution control should go, or could at least aid public understanding.¹²⁴ In addition, public debate in Oregon about where to draw the line between covered and uncovered services meant that legislators who wanted to be restrictive had to go on record about which procedures they wanted not to cover; the consequence was to make it more difficult to oppose coverage.¹²⁵

Yet the Oregon experience has its limitations as a model for the setting of ambient air quality standards. It is one thing to create a debate within a state like Oregon with a well-established political culture of citizen participation, and quite another to do the same thing nationally. EPA's best-known effort to create such a debate—Administrator William Ruckelshaus' attempt in the 1980s to involve the citizens of Tacoma, Washington, in the debate over how to balance the environmental benefits of restricting arsenic emissions at a nearby ASARCO smelter with economic costs—met with a mixed reception.¹²⁶ Moreover, the decision facing EPA is much more complex even than that facing Oregon. That state's program concerns only the question of whether effects will be covered when they occur; the state did not need to consider the probability of whether an effect occurs or the size of the population that experiences the effect. Taking these factors into account would make an already difficult process virtually unruly.

It might be argued that Lockout/Tagout furnishes an alternative model of a workable principle; as we have seen, OSHA was able to devise a satisfactory decision rule in Lockout/Tagout.¹²⁷ But, here too, EPA is facing a more complex task. In Lockout/Tagout, all parties agreed that the agency could not impose a regulation that required more than was feasible.¹²⁸ The only question, then, was how OSHA should decide whether to stop short of feasibility. By contrast, EPA's ambient air quality standards can be sufficiently stringent to require technology-forcing measures that go beyond what is currently available.¹²⁹ The problem, therefore, is not merely one of establishing how lax the Agency should be, but also of determining the extent to which it should technology-force industry; even more than in Lockout/Tagout, the Agency has the authority to "dictate the vitality and even survival of whatever segments of American business it might choose."¹³⁰

Instead, the task facing EPA somewhat resembles what the Agency had to do in response to the Vinyl Chloride decision, Natural Resources Defense Council v. U.S. Environmental Protection Agency.¹³¹ There the D.C. Circuit declared that EPA, in setting standards for hazardous air pollutants (those that pose a risk of cancer or similar effects), had to first decide what is "safe" and then decide how much additional protection was necessary to assure an ample margin of safety. On remand, EPA considered a number of approaches and ultimately opted for one based on attempting to reduce risk to the individual who is most exposed to emissions of the pollutant to no more than one in one million.¹³²

[29 ELR 10664]

It might seem the same could be done here. But the Vinyl Chloride precedent is of doubtful value in this context. First, EPA has little experience in administering its post-Vinyl Chloride approach. Congress, in the CAA Amendments of 1990, altered the statute to require that EPA issue technology-based standards for emitters of hazardous air pollutants, and use risk-based regulation—for which the post-Vinyl Chloride approach was designed—only in a second stage that has not yet occurred.¹³³ Second and more fundamentally, setting ambient air quality standards is even more complicated than establishing benchmarks for hazardous air pollutants; the former involves not only deciding what is an unacceptable level of risk and what amount of scientific evidence of the existence of that level must be shown, but also, as we have seen, the vexing issue of deciding what effects are the kind of effects against which EPA should seek to protect.

An Alternative Approach

Insisting that EPA try to enunciate a clear decision rule might therefore amount simply to making the Agency chase a will-of-the-wisp. Instead, it might be preferable to treat the case as a garden variety arbitrary-and-capricious decision, and to read it as requiring the Agency to write a clearer rationale that is free of internal inconsistency. This does not militate against the Rule of Law; we are all used to the notion that some legal decisions are too complex to reduce to rule. Judge Williams, for instance, would surely agree that he cannot construct an exact formula for deciding when an agency decision is arbitrary or capricious or whether an agency's decisionmaking principle is intelligible; it is difficult to see why the courts ought to demand a greater degree of coherence from agencies than courts themselves are able to deliver. Moreover, demanding a firm decision rule seems inconsistent with the language of the CAA. The Act calls for EPA to set air quality standards that, "in the judgment of the Administrator" protect public health.¹³⁴ Of course, these words do not give EPA unfettered discretion. But they do represent, in part, a recognition that standard-setting decisions involve policy considerations.¹³⁵ In addition, and more importantly for presen purposes, they indicate Congress' recognition that standard-setting cannot be reduced to formula, but rather incorporate the same degree of discretion that a good judge brings to deciding a difficult case.

Hence the arbitrary-and-capricious doctrine seems a preferable basis for the court's opinion. By making clear that this is the basis, the court could give EPA needed guidance about how to respond. The court should instruct EPA that its task is to explain the rationale for its decisions. If, for instance, EPA wishes to retain the new 0.08 ppm standard for ozone, the Agency needs to explain why it views 0.09 as inadequate and 0.07 as more stringent than necessary. As lawyers would say, EPA needs to distinguish its preferred levels from other possibilities. The Agency should, as has already been suggested, speak the language of public health: that is, explain why it believes a given effect does or does not qualify as a health effect (for instance, by explaining the impact of an effect on the sufferer's quality of life) and what amount of evidence the Agency believes is needed to justify the conclusion that a pollutant is causing a health effect.

Yet assigning EPA even this task gives pause. The arbitrary-and-capricious doctrine is one of several that have made agency rulemaking into an obstacle course over the past two decades. American Trucking arguably illustrates that the time has come for the courts to give the Agency more deference in making such difficult decisions as setting air quality standards.

This contention has merit. By suggesting above that the Agency not be required to justify its decisions in cost terms, I am in effect proposing that at least a portion of the Agency's decisionmaking be insulated from judicial review. But one should be hesitant to endorse a substantial relaxation of arbitrary-and-capricious review. In a nation that considers accountability of governmental officials to be fundamental, no review is not a feasible alternative to judicial review. Instead, less judicial review probably would mean more involvement by the elected branches. That would in turn make agency decisions seem—and become—more political, thereby undercutting the agency expertise on which, as Professor Strauss stresses, the legitimacy of agency decisionmaking depends. Moreover, agency rules would become even more hostage than they are now to the political climate. If agency rulemaking becomes seen as simply an exercise of political will—as resulting, in this case, from the inability of the new standards' opponents to mobilize the two-thirds vote needed to override an expected veto by President Clinton of legislation nullifying the standards—it will be all the easier for some future, less environmentally concerned President to ordain an opposite result.

Yet the consequences of continuing aggressive judicial review should be understood. No matter how American Trucking is read, the decision makes the process of developing ambient air quality standards more complex. For this reason alone, revision of these standards—either to make them more lenient or more stringent—will probably become even less frequent than in the past. This involves a trade off. The less frequently EPA revises the standards, the more stable the requirements for states and pollution sources, and the greater the opportunity for those concerned to make themselves heard about a potential revision. On the other hand, less frequent revision makes it more difficult to keep the standards in accord with the latest scientific knowledge. Since industry has, at times, been as active in seeking revision of the standards as environmental groups,¹³⁶ this trade off is not simply a matter of one's political position. How to balance these considerations is, to use Administrator Browner's phrase, the real question of judgment the reader must decide.

1. Air Pollution: Agency Announces Proposals to Toughen Regulations for Ozone, Particulate Matter, 27 Env't Rep. 1571 (Nov. 29, 1996).

2. Air Quality Standards: Science-Driven Ozone, PM Proposals Will Be Finished by July 19, EPA Says, 27 Env't Rep. 2068 (Feb. 14, 1997).

3. 175 F.3d 1027, 29 ELR 21071 (D.C. Cir. 1999). Judges Stephen Williams and Douglas Ginsburg joined all portions of the court's opinion; Judge David Tatel dissented in part. Responsibility for writing the opinion was divided among the three judges on the panel. Id. at 1032 n*, 29 ELR at 21071.

4. The remanded standards are codified at 40 C.F.R. § 50.7 (particulate matter) and 40 C.F.R. § 50.10 (ozone) (1998). EPA's justification may be found at National Ambient Air Quality Standards for Particulate Matter, 62 Fed. Reg. 38652 (July 18, 1997) and National Ambient Air Quality Standards for Ozone, 62 Fed. Reg. 38856 (July 18, 1997) [hereinafter NAAQS for Ozone].

5. Air Quality Standards: Court's Decision on Ozone, PM Rules Called "Extreme, Illogical" by Browner, 30 Env't Rep. 158 (May 28, 1999).

6. Cass R. Sunstein, The Courts' Perilous Right Turn, N.Y. TIMES, June 2, 1999, at 25A.

7. Linda Greenhouse, An Arcane Doctrine Surprisingly Upheld, N.Y. TIMES, May 15, 1999, at A15.

8. Air Quality Standards: Legal Doctrine Used to Strike Down EPA Rules Could Apply to Other Actions, 30 Env't Rep. 104 (May 21, 1999).

9. EPA, Lawyers, Scholars Take Measure of "Nondelegation" Theory in Ozone Ruling, 67 U.S.L.W. (Analysis & Perspective) 2739 (1999) (quoting C. Boyden Grey, one of the attorneys for the successful challengers).

10. Richard Lazarus, National Air Standard Ruling a Blockbuster, ENVIL. F., July/Aug. 1999, at 8.

11. American Trucking, 175 F.3d at 1049, 29 ELR at 21078 ("As the petitioners argue, because the 1990 amendments extended the time for nonattainment areas to comply with the [previous] ozone NAAQS, they must preclude the EPA from requiring ateas to comply either more quickly or with a more stringent ozone NAAQS.").

12. See 42 U.S.C. §§ 7408-7409, ELR STAT. CAA §§ 108-109. (Unless otherwise specified, all future statutory citations are to the CAA.) These provisions were originally added by the Clean Air Act Amendments of 1970, Pub. L. No. 91-604, § 4(a), 84 Stat. 1676, 1678-80.

13. John C. Cushman, Clinton Sharply Tightens Air Pollution Regulations Despite Concern Over Costs, N.Y. TIMES, June 26, 1997, at Al (recounting involvement by Vice President Gore and President Clinton). This is not a first; the Carter Administration became involved in the previous revision of the ozone standard in 1979. See MARC K. LANDY ET AL., THE ENVIRONMENTAL PROTECTION AGENCY: ASKING THE WRONG QUESTIONS 72-73 (1990).

14. 42 U.S.C. § 7409(d), ELR STAT. CAA § 109(d); see Environmental Defense Fund v. Thomas, 870 F.2d 892, 19 ELR 20660 (2d Cir.), cert. denied, 493 U.S. 991 (1989) (holding that this provision imposes on EPA a nondiscretionary duty to decide every five years whether to revise the ambient air quality standards, but rejecting the argument that the provision requires EPA to actually change the standards).

15. U.S. EPA, REVIEW OF NATIONAL AMBIENT AIR QUALITY STANDARDS FOR OZONE: ASSESSMENT OF SCIENTIFIC AND TECHNICAL INFORMATION: OAQPS STAFF PAPER [hereinafter OZONE STAFF PAPER] 5-7 (1996); EPA, REVIEW OF NATIONAL AMBIENT AIR QUALITY STANDARDS FOR PARTICULATE MATTER: POLICY ASSESSMENT OF SCIENTIFIC AND TECHNICAL INFORMATION: OAQPS STAFF PAPER [hereinafter PM STAFF PAPER] II-4 (1996).

16. Air Quality Standards: Number of Counties Violating Air Rules Would Double Under Changes, EPA Says, 28 Env't Rep. 550 (July 18, 1997) (estimating that the number of counties designated as non-attainment for ozone would increase from 106 to 250, and for PM from 41 to 150). On the other hand, the new ozone standard may be easier for some areas to meet than the old one because the new standard focuses on the eight-hour average for ozone instead of one-hour concentrations that might simply be "spikes." Mary Nichols, Introduction: The New National Standards for Ozone and Particles: Modest Public Health Progress or Public Policy Meltdown?, II TUL. ENVTL. L.J. 181, 184 (1998).

17. NATIONAL RESEARCH COUNCIL, OZONE-FORMING POTENTIAL OF REFORMULATED GASOLINE 12 (1999).

18. PM STAFF PAPER, supra note 15, at IV-3a. EPA developed the two standards together largely because of this overlap. See National Ambient Air Quality Standards for Ozone and Particulate Matter, 61 Fed. Reg. 29719, 29720 (1996). For detailed discussion, see Implementation of New or Revised Ozone and Particulate Matter (PM) National Ambient Air Quality Standards (NAAQS) and Regional Haze Regulations, 61 Fed. Reg. 65763, 65767-769 (Dec. 13, 1996).

19. See Finding of Significant Contribution and Rulemaking for Certain States in the Ozone Transport Assessment Group Region for Purposes of Reducing Regional Transport of Ozone, 63 Fed. Reg. 57356 (Oct. 27, 1998); see also Air Quality Standards: NO_x Cuts of 1.1 Million Tons Annually Sought in Rule on Transported Pollution, 29 Env't Rep. (BNA) 1093 (Oct. 2, 1998). Shortly after remanding the new air quality standards, the D.C. Circuit stayed implementation of the 22-state program. Michigan v. EPA, No. 98-1457 (D.C. Cir. May 25, 1999). EPA has suggested withdrawing the program as part of a compromise settlement with midwestern states, Air Quality Standards: Agency Proposes Control Plan to States in Dispute Over Regional Ozone Transport, 30 Env't Rep. (BNA) 637 (July 30, 1999), but apparently negotiations among the states on a compromise have fallen through. Richard Perez-Pena, Interstate Deal for Clean Air Falls Apart, N.Y. TIMES, Sept. 2, 1999, at Bl.

20. The only previous invalidation of an ambient air quality standard occurred in Kennecott Copper Corp. v. EPA, 462 F.2d 846, 2 ELR 20116 (D.C. Cir. 1972) (remanding the secondary ambient air quality standards for sulfur dioxide (SO₂) because of EPA's failure to explain them adequately). For examples of industry defeats in attempts to overturn EPA's ambient air quality standards, see Lead Indus. Ass'n v. EPA, 647 F.2d 1130, 10 ELR 20643 (D.C. Cir.), cert. denied, 449 U.S. 1042 (1980) (upholding EPA's ambient air quality standards for lead); American Petroleum Inst. v. Costle, 665 F.2d 1176, 11 ELR 20916 (D.C. Cir. 1981), cert. denied, 455 U.S. 1034 (1982) (upholding the 1979 revision of the ozone standards); Natural Resources Defense Council v. EPA, 902 F.2d 962, 20 ELR 20891 (D.C. Cir. 1990), modified, 921 F.2d 326 (D.C. Cir. 1991), cert. denied, 498 U.S. 1082 (1991) (upholding the 1987 revision of the PM standards).

21. 2 U.S.C. § 1501 et seq. (Supp. II 1996).

22. 5 U.S.C. § 601 et seq. (Supp. II 1996).

23. American Trucking, 175 F.3d at 1043-45, 29 ELR at 21075-76.

24. Id. at 1040, 29 ELR at 21074.

25. Id. at 1041-43, 29 ELR at 21074-75.

26. Id. at 1040, 29 ELR at 21074.

27. Id. at 1055-56, 29 ELR at 21080-81.

28. Id. at 1033-40, 29 ELR at 21071-74. This was the only portion of the decision from which Judge Tatel dissented. The challengers also prevailed on two other theories: that the Agency should have considered the health benefits from ozone in setting the standards for that pollutant, id. at 1050-52, 29 ELR at 21078-79, and that EPA acted arbitrarily in including fine particles in its definition of coarse particles, id. at 1052-55, 29 ELR at 21079-81. Judge Williams and Judge Tatel, respectively, wrote for the court on these two issues; the court was unanimous in both respects. EPA's petition for rehearing en banc specifically requests reconsideration of the first holding, but not the second. Air Quality Standards: EPA Seeks Rehearing of Court Decision Rejecting Ozone and Particulate Matter Standards, 30 Env't Rep. (BNA) 413 (July 2, 1999). Evidently the Agency believes that it might be possible to disaggregate fine particles from coarse particles.

29. American Trucking, 175 F.3d at 1034, 1037, 29 ELR at 21071, 21073.

30. See Schechter Poultry Corp. v. United States, 295 U.S. 495 (1935); Panama Ref. Co. v. Ryan, 293 U.S. 388 (1935); Carter v. Carter Coal Co., 298 U.S. 238 (1936). The last of these cases "marked the third and final time the Supreme Court found a statute unconstitutional on delegation grounds." PETER L. STRAUSS ET AL., GELLHORN AND BYSE'S ADMINISTRATIVE LAW: CASES AND COMMENTS 88 n.9 (9th ed. 1995).

31. American Trucking, 175 F.3d at 1037, 29 ELR at 21073.

32. Id. at 1057, 29 ELR at 21081.

33. United States v. Southwestern Cable Co., 392 U.S. 157 (1968).

34. International Union, United Automobile, Aerospace & Agricultural Implement Workers of America v. Occupational safety & Health Admin., 938 F.2d 1310 (D.C. Cir. 1991) (Lockout/Tagout I); International Union, United Automobile, Aerospace & Agricultural Implement Workers of America v. Occupational safety & Health Admin., 976 F.2d 749 (D.C. Cir. 1992) (Lockout/Tagout II); International Union, United Automobile, Aerospace & Agricultural Implement Workers of America v. Occupational Safety & Health Admin., 37 F.3d 665 (D.C. Cir. 1994) (Lockout/Tagout III).

35. To quote the court, this interpretation gave the agency "an asserted power, once significant risk is found, to require precautions that take the industry to the verge of economic ruin (so long as the increment reduces a significant risk …), or to do nothing at all. All positions in between are evidently equally valid." Lockout/Tagout I, 938 F.2d at 1317.

36. Id. "OSHA's proposed analysis would give the executive power untrammelled power to dictate the vitality and even survival of whatever segments of American business it might choose." Id. at 1318. This statement seems excessive since, as the court itself concedes, OSHA was not claiming the power to impose measures that are economically or technologically infeasible.

37. American Trucking, 175 F.2d at 1037, 29 ELR at 21073 ("As our prior discussion should have indicated, here EPA's formulation of its policy judgment leaves it free to pick any point between zero and a hair below the concentrations yielding London's Killer Fog.").

38. Lockout/Tagout III, 37 F.3d at 669.

39. Lockout/Tagout II, 976 F.2d at 749.

40. Lockout/Tagout I, 938 F.2d at 1325.

41. American Trucking, 175 F.3d at 1056, 29 ELR at 21081. The opinion reserved the question of whether to vacate the standard for fine particulate matter; the panel has since decided not to do so. American Trucking Ass'n v. EPA, No. 97-1440 (D.C. Cir. June 18, 1999).

42. STRAUSS, supra note 30, at 338-39, 444-45.

43. American Trucking, 175 F.2d at 1056, 29 ELR at 21081.

44. Air Quality Standards: Court Refuses to Vacate PM[2.5] Standard, Leaves EPA Rule Open to Future Challenge, 30 Env't Rep. (BNA) 360 (June 25, 1999).

45. See, e.g., Identification of Ozone Areas Attaining the I-Hour Standard and to Which the I-Hour Standard Is No Longer Applicable, 63 Fed. Reg. 31014 (June 5, 1998).

46. Patricia M. Wald, Regulation at Risk: Are Courts Part of the Solution or Most of the Problem?, 67 S. CAL. L. REV. 621, 638 n.72 (1994).

47. 448 U.S. 607, 10 ELR 20489 (1980).

48. Id. at 646, 10 ELR 20500. Here, too, the agency was able eventually to re-promulgate the rule without substantial change. See Occupational Exposure to Benzene, 52 Fed. Reg. 34460 (Sept. 11, 1987).

49. Moreover, as Judge Tatel notes in dissent, the statutory provisions in this case seem more specific than those involved in Lockout/Tagout. American Trucking, 175 F.3d at 1058, 29 ELR at 21081-82.

50. See supra note 37.

51. 337 F. Supp. 737 (D.D.C. 1971).

52. Pub. L. No. 91-379, 84 Stat. 799 (1969).

53. Amalgamated Meat Cutters, 337 F. Supp, at 758.

54. KENNETH C. DAVIS, DISCRETIONARY JUSTICE: A PRELIMINARY INQUIRY 57-59, 220-21 (1970) ("The courts … can save the non-delegation doctrine [by continuing] their requirement of meaningful standards, except that when the legislative body fails to prescribe the required standards the courts can allow the administrators to prescribe them within a reasonable time.").

55. American Trucking, 175 F.3d at 1037, 29 ELR at 21073.

56. It is quite possible that Judge Leventhal, by referring to "further standards," anticipated that the courts would adopt Judge Henry Friendly's suggestion that courts compel agencies to engage in rule-making rather than case-by-case adjudication. HENRY J. FRIENDLY, THE FEDERAL ADMINISTRATIVE AGENCIES: THE NEED FOR BETTER DEFINITIONS OF STANDARDS (1962). That hope was dashed when the Supreme Court unanimously reversed Judge Friendly's own attempt to require agency rulemaking. See National Labor Relations Bd. v. Bell Aerospace Co., 416 U.S. 267, 290-95 (1974), rev'g Bell Aerospace Co. v. National Labor Relations Bd., 475 F.2d 485 (2d Cir. 1973) (opinion by Friendly, J.). Rulemaking, to be sure, has become a much more important aspect of administrative law than when Judge Friendly and Judge Leventhal wrote. But that occurred less because of judicial compulsion of rulemaking than because the environmental statutes of the 1970s would have been impossible to implement without rulemaking.

57. Greater Boston Tel. Corp. v. Federal Communications Comm'n, 444 F.2d 841 (D.C. Cir. 1970).

58. 463 U.S. 29 (1983).

59. Id. at 42 (internal quotations and citation omitted). This was the decision remanding the U.S. Department of Transportation's revocation of the requirement that new cars contain automatic seatbelts or airbags.

60. Id. at 52.

61. See id. at 42 ("If Congress established a presumption from which judicial review should start, that presumption … is … against changes in current policy that are not justified by the rulemaking record").

62. American Trucking, 175 F.3d at 1036, 29 ELR at 21072-73.

63. To quote the court:

Consider EPA's defense of the 0.08 ppm level of the ozone NAAQS. EPA explains that its choice is superior to retaining the existing level, 0.09 ppm, because more people are exposed to more serious effects at 0.09 than at 0.08…. In defending the decision not to go down to 0.07, EPA never contradicts the intuitive proposition, confirmed by data in its Staff Paper, that reducing the standard to that level would bring about comparable changes….

[EPA's explanation is that] effects are less certain and less severe at lower levels of exposure. This seems to be nothing more than a statement that lower exposure levels are associated with lower risk to public health.

Id. at 1035, 29 ELR at 21072.

64. 134 F.3d 388, 28 ELR 20481 (D.C. Cir. 1998).

65. The court, speaking through Judge Tatel, wrote:

The link between this conclusion and the factual record as interpreted by EPA—that "repeated" exposure is "significant" and that thousands of asthmatics are exposed more than once a year—is missing. Why is the fact that thousands of asthmaties can be expected to suffer atypical physical effects from repeated five-minute bursts of high-level sulfur dioxide not a public health problem? Why are from 180,000 to 395,000 annual "exposure events" (the range indicated by the 1994 studies) or some fewer number (as suggested by the industry studies) so "infrequent" as to warrant no regulatory action? Why are disruptions of ongoing activities, use of medication, and hospitalization not "adverse health effects" for asthmatics? Answers to these questions appear nowhere in the administrative record.

American Lung, 134 F.3d at 392, 28 ELR at 20483. EPA has not yet responded to the remand, although it promised shortly after the decision that it would propose a response in the summer of 1999, and take final action by December 2000. Proposed Settlement; SO₂ NAAQS Remand, 63 Fed. Reg. 30493 (June 4, 1998).

66. See, e.g., Thomas O. McGarity, Some Thoughts on "Deossifying" the Rulemaking Process, 42 DUKE L.J. 1385 (1992); Richard J. Pierce, Seven Ways to Deossify Agency Rulemaking, 47 ADMIN. L. REV. 453 (1995).

67. On the other hand, American Lung was decided only after EPA had already released the standards invalidated in American Trucking.

68. American Trucking, 175 F.3d at 1040, 29 ELR at 21074.

69. Compare Mark H. Grunewald, The NLRB's First Rulemaking: An Exercise in Pragmatism, 41 DUKE L.J. 274, 281-82 (1991) (expressing similar concerns about the use of adjudication as a means of making agency policy).

70. See American Petroleum Inst. v. Costle, 665 F.2d 1176, 11 ELR 20916 (D.C. Cir. 1981), cert. denied, 455 U.S. 1034 (1982) (upholding the 1979 revision of the ozone standards); Natural Resources Defense Council v. EPA, 902 F.2d 962, 20 ELR 20891 (D.C. Cir. 1990), modified, 921 F.2d 326 (D.C. Cir. 1991), cert. denied, 498 U.S. 1082 (1991) (upholding the 1987 revision of the PM standards). For background on and critiques of these revisions, see LANDY ET AL., supra note 13, at 49-88 (ozone standard); Joseph H. Feller, Non-Threshold Pollutants and Air Quality Standards, 24 ENVTL. L. 821 (1994).

71. American Trucking, 175 F.3d at 1037, 29 ELR at 21073.

72. See Lazarus, supra note 10.

73. American Trucking, 175 F.3d at 1059-60, 29 ELR at 21082-83 (partially dissenting opinion by Judge Tatel).

74. EPA's PM Staff Paper seems to suggest that this "double-counting" would not be an important problem because areas in violation of the coarse particulate standard would focus on reducing large particles—in other words, that areas with high levels of coarse particles would not be affected by the double-counting—see PM STAFF PAPER, supra note 15, at VII-17, but the explanation is not particularly clear.

75. See NAAQS for Ozone, supra note 4, at 38878-83.

76. Id. at 38863. For a suggestion that natural levels are usually far less than this level, see Curtis Moore, The Impracticality and Immorality of Cost-Benefit Analysis in Setting Health-Based Standards, 11 TUL. ENVTL. L.J. 187, 193 (1998).

77. Perhaps section 302(g) offers the solution. This provision defines the term "air pollutant" as "any air pollution agent … which is emitted into or otherwise enters the ambient air." 42 U.S.C. § 7402(g), ELR STAT.CAA § 302(g). Windblown dust is therefore an air pollutant—since it enters the air from the earth—but perhaps not ozone that is naturally formed in the atmosphere. EPA could therefore argue that it does not have statutory authority to set an air quality standard aimed at such ozone, since EPA may only set ambient air quality standards for "air pollutants," see 42 U.S.C. § 7409(a)(1), ELR STAT. CAA § 109(a)(1) (requiring EPA to promulgate air quality standards for "each air pollutant"). This approach would allow EPA from taking exposure to natural ozone into account in setting the air quality standard for ozone, just as EPA takes into account exposure to non-air sources of lead in setting the air quality standard for that pollutant. See National Primary and Secondary Ambient Air Quality Standards for Lead, 43 Fed. Reg. 46246, 46253 (Oct. 5, 1978) (explaining how EPA took non-air sources of lead into account in setting the ambient air quality standards for lead).

78. American Trucking, 175 F.3d at 1052, 29 ELR at 21079. The court does not give a citation for its assertion that EPA rejected on grounds of insufficient quantification the studies that purported to show a protective effect, but apparently the rejection was contained in EPA's response to comments. See Gary E. Marchant, Turning Two Blind Eyes: The EPA's Failure to Consider Costs and Health Disbenefits in Revising the Ozone Standard, 11 TUL. ENVTL. L.J. 261, 275 (1998).

The court's holding is based on the premise that the Agency must consider the health benefits of an air pollutant in setting the air quality standards for that pollutant. EPA has requested rehearing en banc on that premise. Air Quality Standards: EPA Seeks Rehearing of Court Decision Rejecting Ozone and Particulate Matter Standards, 30 Env't Rep. (BNA) 413 (July 2, 1999). The Agency's position that it need not do so seems counterintuitive; reducing concentrations that aid health does not seem in accordance with the Act's goal of protecting public health. But it is not clear that the issue is significant here. The particular health claim here rests on the suggestion that ozone at the earth's surface, like ozone in the stratosphere, protects the earth from the sun's cancer-causing ultraviolet radiation. One draft study done for EPA estimates that reductions in ozone concentrations to meet a tighter ozone concentration would result in 700 additional non-melanoma skin cancers per year. Air Quality Standards: EPA Should Revise Estimates on Ozone in Response to Court Rulings, Analyst Says, 30 Env't Rep. (BNA) 504 (July 16, 1999). There are a million such cancers diagnosed each year, says the American Cancer Society's web site http://www.cancer.org (visited Sept. 7, 1999), and 1,900 of them—or about 2 percent—are fatal. An extra 700 cases annually, therefore, would mean 1 to 2 additional cancer deaths each year. The additional cases are readily preventable. To quote the Society's web site,

The most important way to lower the risk of non-melanoma skin cancer is to reduce unprotected exposure to the sun and other sources of ultraviolet light. The simplest way to keep from getting too much ultraviolet light is to stay out of the sun and stay in the shade whenever possible.

By contrast, the health effects from breathing ozone are not preventable, except by staying indoors on high ozone days. This may not be possible either for adults who work outside or for children who live in unair-conditioned homes. Effects on them would seem to deserve much more weight than the additional skin cancer cases.

79. To quote Bertrand Russell, "Unlike religion, [science] is ethically neutral; it assures men that they can perform wonders, but does not tell them what wonders to perform. In this way it is incomplete." B. RUSSELL, A HISTORY OF WESTERN PHILOSOPHY 494 (1945).

80. Industrial Union Dep't, AFL-CIO v. American Petroleum Inst., 448 U.S. 607, 655 n.62, 10 ELR 20489, 20501 (1980) ("We recognize that [the agency's] determination that a particular level of risk is 'significant' will be based largely on policy considerations.").

81. American Trucking, 175 F.3d at 1035, 29 ELR at 21072.

82. Peter Strauss, Presidential Rulemaking, 72 CHI. KENT L. REV. 965 (1997). For this reason, Strauss criticizes the Clinton Administration's tendency, displayed in this rulemaking, to present the President as the deciding official.

83. R.S. MELNICK, REGULATION AND THE COURTS: THE CASE OF THE CLEAN AIR ACT (1983); David Currie, Federal Air-Quality Standards and Their Implementation. 1976 A.B.F. RES. J. 365, 367-69. See Feller, supra note 70.

84. See also J.W. ANDERSON, RESOURCES FOR THE FUTURE, REVISING THE AIR QUALITY STANDARDS (Feb. 4, 1997). A symposium of articles on the standards appears in 11 TUL. ENVTL. L.J. 181-441 (1998). Curtis Moore, one of the writers in the symposium, eloquently depicts the health effects of both pollutants. See Moore, supra note 76, at 195-98.

85. For background on revision of the PM standard, see Lucinda Minton Langworthy, EPA's New Air Quality Standards for Particulate Matter and Ozone: Boon for Health or Threat to the Clean Air Act?, 28 ELR 10502, 10503 (Sept. 1998).

86. PM STAFF PAPER, supra note 15, at IV-25 to IV-27.

87. See Suresh H. Moolvagkar & E. Georg Luebeck, A Critical Review of the Evidence on Particulate Air Pollution and Mortality, 7 EPIDEMIOLOGY 420 (1997); John F. Gamble & Jeffrey Lewis, Health and Respirable Particulate (PM[10]) Air Pollution: A Causal or Statistical Association, 104 ENVTL. HEALTH PERSP. 838 (1996); but see HEALTH EFFECTS INSTITUTE, PARTICULATE AIR POLLUTION AND DAILY MORTALITY: ANALYSES OF THE EFFECTS OF WEATHER AND MULTIPLE AIR POLLUTANTS (1997) (seeking to rebut these assertions).

88. See Statement of Dr. Ronald E. Wyzga to the Clean Air, Wetlands, Private Property and Nuclear Safety Subcommittee of the Committee on Environment and Public Works, United States Senate, February 5, 1997, http://www.house.gov/science/maassel_5-7.html (visited Aug. 16, 1999).

89. To quote one analysis of the proposed standard:

Ironically, CASAC's endorsement of a fine PM standard reflects concerns about the lack of PM[2.5] data, according to [CASAC chairman Dr. George] Wolff. Although some members of the panel supported setting a PM[2.5] standard based on the available evidence, the consensus of CASAC was to endorse a standard to get the ball rolling on data-gathering efforts, Wolff said. Lack of data was seen as the critical issue for most of the panel, he said.

Wolff said it is important for EPA to keep in mind that CASAC will have a new chance to review data on PM[2.5] before states begin to implement a new standard…. The standard is not likely to take hold until about 2003, Wolff said. CASAC … is scheduled to conduct another review in 1999….

"We're going to have another shot at this with hopefully more information … before anything happens," Wolff said.

Alec C. Zacaroli, Revising the PM. Ozone Standards: Major Questions Remain on Major Projects, 27 Env't Rep. (BNA) 929, 931 (Aug. 23, 1996).

90. This is reflected in EPA's implementation strategy, which postponed any regulatory action until the PM standards have been again revised. See Implementation Plan for Revised Air Quality Standards, 62 Fed. Reg. 38423, 38427-429 (July 18, 1997).

91. See NAAQS for Ozone, supra note 4, at 38867.

92. Id. at 38859.

93. OZONE STAFF PAPER, supra note 15, at 67-72.

94. Zacaroli, supra note 89 (summarizing EPA's criteria document for ozone).

95. See OZONE STAFF PAPER, supra note 15, at 130 (showing that in New York City, the current ozone standard is believed responsible for about 200 ozone admissions out of 14,800).

96. OZONE STAFF PAPER, supra note 15, at 71.

97. Holcomb B. Noble, Far More Poor Children Are Hospitalized for Asthma, Study Shows, N.Y. TIMES, July 27, 1999, at Bl. In East Harlem, where the average household income is $ 18,732, the rate of hospitalization was 222.28 per 10,000, per year. In high-income areas in the city (such as the Rockaways, where the average household income is $ 103,125), the rate was zero. Id. See also OZONE STAFF PAPER, supra note 15, at 39-40 (discussing incidence of asthma).

98. See Air Quality Standards: Minority Groups Divided Over Air Proposals; Environmental Justice, Development at Issue, 28 Env't Rep. (BNA) 303 (June 13, 1997).

99. See NAAQS for Ozone, supra note 4, at 38865.

100. See Richard L. Hasen, Efficiency Under Informational Asymmetry: The Effect of Framing on Legal Rules, 38 U.C.L.A. L. REV. 391, 397 (1990) ("People may be more willing to drive across town to save $ 5 on a $ 10 item than to save $ 5 on a $ 100 item"); see Roger G. Noll & James E. Krier, Some Implications of Cognitive Psychology for Risk Regulation, 19 J. LEGAL STUD. 747, 753 (1990) (labeling this effect as "ratio difference" and defining it as "thinking of changes in magnitude in percentage terms and treating a given percent of a big number as equivalent to the same percent of a small number."). By contrast, experts tend to focus purely on absolute numbers. See Clayton P. Gillette & James E. Krier, Risk, Courts, and Agencies, 138 U. PA. L. REV. 1027, 1071-72 (1990). On the other hand, CASAC seems to have done its evaluation in terms of percentages. See OZONE STAFF PAPER, supra note 15, at app. G (reprinting Letter from George T. Wolff, Chair, Clean Air Scientific Advisory Committee, to Carol M. Browner, Administrator, EPA (Nov. 30, 1995) [hereinafter Wolff Letter]) ("For example, the differences in the percent of outdoor children … responding between the present standard and the most stringent proposal … are small and their ranges overlap for all health endpoints.").

101. See NAAQS for Ozone, supra note 4, at 38864 ("There are significant uncertainties in such quantitative estimates.").

102. OZONE STAFF PAPER, supra note 15, at app. G. See Wolff Letter, supra note 100.

103. See Clean Air Act Oversight—Part 3: Hearings Before the Senate Comm. on Environment and Public Works, 97th Cong. 244 (statement of David Hawkins that "theAdministrator [in revising ambient air quality standards during the Carter Administration] had before him cost calculations"). Hawkins served as EPA Assistant Administrator for Air, Noise and Radiation for the whole of this period.

104. See Langworthy, supra note 85, at 10507. It could also be argued that it is especially odd for EPA not to consider costs in setting ambient air quality standards now that the D.C. Circuit has decided in the Vinyl Chloride case that EPA is to consider costs in setting standards for hazardous air pollutants. See Natural Resources Defense Council v. EPA, 824 F.2d 1146, 17 ELR 21032 (D.C. Cir. 1987) (en banc). But this argument can be turned around; it may simply be that Vinyl Chloride is incorrect in its attempt to distinguish standards for hazardous air pollutants from ambient air quality standards.

105. See C. Boyden Gray, The Clean Air Act Under Regulatory Reform, 11 TUL. ENVTL. L.J. 235 (1998).

106. See supra note 24.

107. Craig N. Oren, Prevention of Significant Deterioration: Control-Compelling Versus Site-Shifting, 74 IOWA L. REV. 1, 72 (1988).

108. See American Petroleum Inst. v. Costle, 665 F.2d 1176 (D.C. Cir. 1981), cert. denied, 455 U.S. 1034 (1982) (upholding the standard).

109. The phrase comes from the Administrative Procedure Act, 5 U.S.C. § 701(a), available in ELR STAT. ADMIN. PROC. at 21; see also Cass R. Sunstein, Agency Inaction After Heckler v. Chaney, 52 U. CHI. L. REV. 653 (1985).

110. See Moore, supra note 76, at 198-204.

111. For accounts of controversy over estimates of costs and benefits, see Langworthy, supra note 85, at 10504-05. For a recent extended critique, see David M. Driesen, The Societal Cost of Environmental Regulation: Beyond Administrative Cost-Benefit Analysis, 24 ECOLOGY L.Q. 545 (1997); see also David M. Driesen, Should Congress Direct the EPA to Allow Serious Harms to Public Health to Continue? Cost-Benefit Tests and NAAQS Under the Clean Air Act, 11 TUL. ENVTL. L.J. 217 (1998). On the problem of reducing costs and benefits to a common metric, see CASS R. SUNSTEIN, FREE MARKETS AND SOCIAL JUSTICE 70-108 (1997).

112. It can be countered that the device of excluding costs entirely from the statute is an overly simplistic means of minimizing the role of economic factors. George Eads, The Confusion of Goals and Instruments: The Explicit Consideration of Cost in Setting National Ambient Air Quality Standards, in TO BREATHE FREELY 222, 230-231 (M. Gibson ed., 1985); David W. Barnes, Back Door Cost-Benefit Analysis Under a Safety-First Clean Air Act, 23 NAT. RESOURCES J. 827, 856-57 (1983) There is some force to this objection. Telling the Administrator not to consider costs artificially distinguishes between the standard with no benefits, and the standard which has benefits that are offset by its costs: while both standards have an equal lack of utility, the Administrator will be able to justify promulgating the latter standard. On the other hand, consider the difficulties of drafting statutory language that would accomplish the same results without explicitly barring cost consideration. Congress would be driven either to attempt to write a rigid formula for considering costs—probably impossible—or to provide the Administrator virtually untrammelled discretion.

113. American Trucking, 175 F.3d at 1039, 29 ELR at 21073-74.

114. Id. at 1039 n.5, 29 ELR at 21074 n.5.

115. Id. at 1039-40, 29 ELR at 21073-74.

116. Lockout/Tagout I, 938 F.2d at 1318-21.

117. Lockout/Tagout III, 37 F.3d at 669-70.

118. See 541 F.2d 1, 18-20, 6 ELR 20267, 20272 (1976) (en banc) (upholding EPA's decision to regulate lead because of concern about its possible effects on children).

119. American Trucking, 175 F.3d at 1039, 29 ELR at 21073-74.

120. Lawrence Jacobs et al., The Oregon Health Plan and the Political Paradox of Rationing: What Advocates and Critics Have Claimed and What Oregon Did, 24 J. HEALTH POL. POL'Y & L. 161, 164 (1999).

121. Id. at 168.

122. Id. at 172.

123. Howard M. Leichter, Oregon's Bold Experiment: Whatever Became of Rationing?, 24 J. HEALTH POL. POL'Y & L. 147, 151 (1999).

124. See LANDY ET AL., supra note 13, at 78-82 (urging such a debate).

125. Jacobs et al., supra note 120, at 174.

126. See William D. Ruckelshaus, Risk in a Free Society, 14 ELR 10190 (May 1984); Mr. Ruckelshaus as Caesar, N.Y. TIMES, July 18, 1983 (editorial); Gregory Call, Arsenic, ASARCO, and EPA: Cost-Benefit Analysis, Public Participation and Polluter Games in the Regulation of Hazardous Air Pollutants. 12 ECOLOGY L.Q. 567, 598-603 (1985).

127. See supra note 38.

128. Lockout/Tagout I, 938 F.2d at 1316.

129. See Union Elec. Co. v. EPA, 427 U.S. 246, 6 ELR 20570 (1976) (holding that states may include technology-forcing measures in their plans to attain ambient air quality standards).

130. Lockout/Tagout I, 938 F.2d at 1317.

131. 824 F.2d 1146, 17 ELR 21032 (D.C. Cir. 1987) (en banc).

132. National Emission Standards for Hazardous Air Pollutants, 54 Fed. Reg. 38044 (Sept. 14, 1989). For discussion of the rulemaking and the alternatives considered by the Agency, see Gary E. Marchant & Dawn P. Danzeisen, "Acceptable Risk" for Hazardous Air Pollutants, 13 HARV. ENVTL. L. REV. 535 (1989).

133. See Pub. L. No. 101-549, § 301, 104 Stat. 2399, 2532 (1990) (rewriting section 112 of the Act, 42 U.S.C. § 7412, ELR STAT. CAA § 112; U.S. EPA, RESIDUAL RISK: REPORT TO CONGRESS, EPA-453/R-99-001, at 9 (1999); Bradford C. Mank, What Comes After Technology: Using an "Exceptions Process" to Improve Residual Risk of Hazardous Air Pollutants, 13 STAN. ENVTL. L.J. 263 (1994).

134. 42 U.S.C. § 7409(b)(1), ELR STAT. CAA § 109(b)(1).

135. Cf. Ethyl Corp. v. EPA, 541 F.2d 1, 20, 6 ELR 20267, 20272 (D.C. Cir. 1976) (en banc) (holding that, in deciding whether to regulate lead in gasoline, EPA had to make a policy judgment concerning the relative risks of underprotection and overprotection).

136. See Consolidated Coal Co. v. Costle, 483 F. Supp. 1003, 10 ELR 20296 (S.D. Ohio 1979) (litigation by coal company seeking order obligating EPA to revise ambient air quality standard for sulfur oxides).

29 ELR 10653 | Environmental Law Reporter | copyright © 1999 | All rights reserved