Assessing Regulatory Costs and Benefits: Fifth Circuit Vacates OSHA Benzene Standard

Given that legislative and administrative controls on environmental pollution involve considerable economic sacrifices, a crucial underlying question is whether the prices paid for environmental protection by taxpayers, consumers, and industry are justified. Lawmakers have traditionally eschewed this difficult question, dismissing it as unanswerable. Within the last year or so, however, environmental, health, and safety regulations have been widely blamed for contributing to a uniquely pernicious and malignant economic malady: Inflation. This allegation had precipitated a nationwide reevaluation of the costs and, to a lesser extent, the benefits, of regulation. Critics of stringent regulation assert that such rules and to the cost of consumer goods and therefore to the inflation rate. Defenders rejoin that the protections offered by such regulations, in terms of human health and mortality as well as aesthetics, are more valuable to society than the economic costs and are thus noninflationary. The key concept in this debate is cost-benefit analysis. Environmental controls for which the benefits exceed the costs are considered worthy, while all others are considered unjustified.

Intensive scrutiny of the cost of environmental regulation is now being conducted in Congress¹ and in the executive branch. On October 31, 1978 the Carter Administration announced the formation of yet another panel of upper echelon officials whose prime task it is to review impending federal regulations for the economic effects.² It appears that these bodies will be less concerned with subjecting proposed regulations to cost-benefit analysis than with institutionalizing priorities under which resources will be applied first to rules which are most "cost effective." In theory this means that the agencies will first issue regulations which appear to provide benefits at the lowest cost, assuming that other important objectives are not defeated. Although environmentalists concede the noble ends of this theory, they are concerned that in practice the application of cost-benefit techniques or cost-effectiveness review may prove a guise for reducing the overall level of environmental protection.

A recent opinion by the United States Court of Appeals for the Fifth Circuit has been seen by many as a sign that the judiciary plans to add its voice and considerable muscle to the debate. In American Petroleum Institute v. Occupational Safety and Health Administration³ the court overturned a work place health standard promulgated by the Occupational Safety and Health Administration (OSHA) which reduced from 10 parts per million (ppm) to one ppm the permissible employee exposure level to airborne benzene, a widely used industrial solvent and known carcinogen. Focusing on the term "reasonably necessary" within the definition of "occupational safety and health standard" under the Occupational Safety and Health Act (OSH Act),⁴ the court interpreted that language to require the agency to estimate precisely the benefits and costs of proposed regulations and to show that there exists a "reasonable relationship" between the two. Finding that OSHA had failed to specify the risk posed by benzene in the one to 10 ppm range, and hence the benefit of reducing that risk, the court struck down the standard.

In view of OSHA's finding in the preamble to the final standard⁵ that even a rough estimate of the dangers of low ambient levels of benzene is impossible, its distress at this result is understandable. This problem is a familiar one to other federal agencies charged with regulating the vaguely understood hazards associated with toxic substances. But until the decision in American Petroleum Institute, their burden had been eased substantially by reviewing courts, which had been sensitive to the dearth of information endemic to the field and had given substantial deference to agency judgments.⁶

The Fifth Circuit's break with this pattern has implications far beyond the workplace, and indeed far beyond the Fifth Circuit. First, many environmental health laws contain provisions analogous if not identical to the "reasonably necessary" language within the OSH Act. There is no reason to believe that regulations based on such provisions would not be treated similarly if challenged in that court. Secondly, under most of those statutes venue in litigation over implementing regulations is proper in the circuit in which the first petition is filed. Where major corporations or industrial trade associations can prevail in the "race to the courthouse," they can assure that the case will be decided in the forum of their choice, which is more frequently becoming the Fifth Circuit.⁷ At the request of the Solicitor of Labor, the [8 ELR 10251] Solicitor General has petitioned the Supreme Court for certiorari.⁸

Standard Setting Under the OSH Act

In § 2 of the Occupational Safety and Health Act Congress briefly and forcefully set forth the purpose of the statute:

to assure so far as possible every working man and woman in the Nation safe and healthful working conditions….⁹

This goal is to guide the Secretary of Labor in issuing mandatory standards and in taking other measures to reduce occupational injuries and illnesses. Failure to comply with OSHA standards may subject an employer to civil and criminal penalties.¹⁰

In recognition of the extreme dangers posed by chemical toxins and the unique difficulties inherent in attempting to regulate them, a separate provision of the law provides special guidance in this area. When regulating toxics the Secretary is to choose the level of protection which most adequately assures that:

no employee will suffer material impairment of health or functional capacity even if such employee has regular exposure to the hazard dealt with by such standard for the period of his working life.¹¹

The paramount consideration is the "highest degree of health and safety protection for the employee."¹² The inclusion of such language in a provision dealing solely with toxic substance regulation suggests that Congress was not willing to accept the same level of danger from hazardous chemicals that it would in the case of conventional occupational hazards.

The widespread lack of conclusive information on the health effects of toxic substances led Congress to provide in § 6(b)(5) that standards under that section shall be based on the "best available evidence."¹³ It was not intended that regulation be postponed until a given substance has been proven fatal in the workplace, nor "that the Secretary be paralyzed by debate surrounding diverse medical opinion."¹⁴ As recognized in the cases, the agency has a duty to anticipate dangers and to draw inferences from the information that is available, even if incomplete.¹⁵

The "determinations" of the Secretary are to be judicially reviewed through application of the substantial evidence test. This test, which under the Administrative Procedure Act is applied only to findings issued in adjudications and rule makings "on the record,"¹⁶ is generally deemed to require a greater quantity of evidentiary support for factual conclusions that the "arbitrary and capricious" test, which would customarily apply to OSHA's informal rule-making proceedings. This appears to create a conflict within the statute. On one hand, OSHA is required to prevent risks even where proof of such risks may be unavailable or largely conjectural; at the same time, it is required on review to support its conclusions with evidence sufficiently probative to satisfy the substantial evidence test.

Despite some initial difficulty with this inconsistency,¹⁷ the courts have resolved it in favor of the statutory policy of protecting workers. They have recognized that as the agency's fact-finding responsibilities approach the frontiers of scientific knowledge decisions must be based less on undisputed fact and more on a mix of documentable scientific conclusions with subjective, inferential, educated guesses that incorporate legislative-type policy judgments.¹⁸ The method of review that has emerged is to distill factual findings from policy judgments: the former remain subject to strict "substantial evidence" review, while the latter are judged with "restraint" and "flexibility,"¹⁹ or under a standard indistinguishable from the [8 ELR 10252] "arbitrary and capricious test."²⁰ The Third Circuit has twice applied a unique formulation which essentially seeks to assure no more than that the Secretary's reasoning reflects consideration of "factors relevant under the statute."²¹ In sum, prior to American Petroleum Institute, the three circuits to have considered OSHA standards relating to toxic substances have consistently allowed the agency, when unable to provide missing links in an evidentiary chain, to substitute common sense judgment, consistent with the congressional intent to provide rigorous protection for workers.

Finally, absolute worker protection from toxic substances hazards is circumscribed by the requirements that such standards be "feasible."²² As used in the statute this term does not imply that the level of health protection provided by standards must be closely compared to the economic costs. Rather, the concept of feasibility establishes a cost ceiling above which a standard is deemed more burdensome to the affected industries than contemplated by Congress. A standard is not rendered infeasible despite the fact that the cost of compliance would bankrupt a limited number of firms in an industry,²³ or that the technology needed to comply does not exist.²⁴ The question of feasibility was not at issue in American Petroleum Institute.²⁵

The Benzene Standard

On February 10, 1978 OSHA issued a final standard reducing the permissible ambient exposure level of benezene from 10 ppm to one ppm averaged over an eight hour day.²⁶ Compliance with an extensive series of related measures was also required.²⁷ In essence, the setting of the standard at one ppm was based on the unanimous conclusion of several scientific studies²⁸ that at an ambient concentration of 25 ppm or greater benzene is carcinogenic or otherwise injurious to human health. The agency admitted that below that level the studies were inconclusive, and that "[i]t is not clear, what adverse health effects occur below 20 ppm."²⁹ Because the objective was to determine a level of safety rather than the lowest known level of danger, however, and because there was no evidence of a threshold level at which the chemical was nontoxic to humans, the agency relief on the "customary" factor of 10-100 times the lowest known harmful concentration in extrapolating a permissible concentration of one ppm.³⁰ It is thus a fair observation that OSHA had no hard evidence of the dangers extant at either 10 or one ppm, much less the difference between the two, and therefore had no way of calculating the benefit of the new standard in terms of reduced risk.

The costs of the standard, on the other hand, were relatively easy to quantify. The total estimated cost, aside from effects on market structure, competition, and the like, was in the vicinity of $500 million in first year costs, and $35 million in subsequent annual costs.³¹ The agency found compliance with the standard to be feasible because no party to the rule making had alleged either an inability to bear the anticipated costs or that any manufacturer or user would be threatened with closure.³² Aside from the question of feasibility the dollar costs did not factor into the setting of the standard. In a nutshell, once it had been determined that the benefits of the standard would exceed a floor of "appreciability" and that the costs of the standard fell short of a ceiling of infeasibility, no precise weighing of costs and benefits was conducted. OSHA denied that a scientific cost-benefit comparison was required under the Act or even possible. On the contrary, its position was that within these two parameters, any standard providing less than the maximum degree of protection for workers would be a violation of its statutory duty.

American Petroleum Institute v. OSHA

Petitions challenging the standard were filed in the Fifth Circuit by nine trade associations and corporations representing both manufacturers and users of the chemical. On the date of filing that court was in the latter stages of reviewing a consumer product safety standard promulgated by the Consumer Product Safety Commission (CPSC) requiring that swimming pool slides be affixed with warning signs. In that case, Aqua Slide 'N' Dive Corp. v. Consumer Product Safety Comm'n,³³ the court interpreted the Consumer Product Safety Act's (CPSA's) requirement that standards issued by the Commission be "reasonably necessary to prevent or reduce an unreasonable risk of injury"³⁴ to mean that the expected benefits of such standards must be assessed in light of the expected costs of compliance. Moreover, there must be a "reasonable relation" between the two. The court found that the CPSC failed to consider some of the likely costs of the standard, as well as to prove with substantial evidence that the standard would be as effective in reducing risks of injury as claimed by the agency. These defects rendered the agency unable to compare meaningfully the costs and benefits of the standard, and thus rendered the regulation invalid.

In American Petroleum Institute v. Occupational Safety [8 ELR 10253] and Health Administration³⁵ the court determined initially that since the purposes of the CPSA and the OSH Act were parallel, it would employ the same analytic approach and standard of judicial review that it had in Aqua Slide. In order to determine whether there existed a reasonable relationship between the costs and benefits of the benzene standard the court first examined the agency's cost estimate. The half billion dollar figure was apparently, but not expressly, found satisfactory.³⁶

Moving quickly to the agency's consideration of the standard's benefits, Judge Clark criticized OSHA's failure to attempt to measure the reduction in employee hazard accompanying the lowering of the standard from 10 ppm to one ppm. OSHA's view that the lower level would produce appreciable benefits was demand a "mere assumption" and found unsupported by substantial evidence. This was not to say that the agency's finding that the standard would likely result in benefits was unsupported. Differing scientific opinion over the existence of a safe threshold level of benzene was found to constitute substantial evidence of a danger at the 10 ppm level; further, the court upheld the agency's determination that lowering the permissible level would reduce that risk. Nevertheless, substantial evidence did not show that the anticipated reduction in risk would be "measurable," and without a measurable quantity of benefits the required cost-benefit comparison could not even be initiated, much less judicially examined.

Regulating In the Face of Uncertainty

Judge Clark's use of the term "measurable" in describing the minimum required showing of benefits was not accidental. The regulation fell because the agency did not measure, i.e., quantify, both the risk it was addressing and the effectiveness of its corrective action. The court read the requirement that standards be based on the "best available evidence" with emphasis on "evidence" rather than "available," and thus to preclude OSHA from supporting regulations with judgmental inferences based on incomplete scientific evidence. This holding contradicts the express conclusions of three other circuit courts, which have stated that OSHA need not document its determinations of risk with complete information.³⁷ Judge Clark recognized the conflict, but resisted a lawyer's instinct to reconcile these readings of the statute:

Congress provided that OSHA regulate on the basis of knowledge rather than the unknown. But see Society of Plastics Industries, Inc. v. Occupational Safety and Health Administration, …³⁸

The opinion does offer some justification for the fact that it is the first to infer a balancing obligation from the statute by noting that earlier cases had not addressed the term "reasonably necessary." In addition, Judge Clark explained, those decisions reviewed administrative records different from that supporting the benzene standard. Beyond that, however, the court simply declined to undertake a "critical analysis of what was established in those proceedings."³⁹

Critique

When viewed out of context, the rationale of American Petroleum Institute has a certain appeal. Regulations can logically be deemed unreasonable if they cannot be shown to yield more than they cost. Closer analysis of the decision and the relevant authority, however, casts considerable doubt on the strength of the opinion. The provision in which is located the "reasonably necessary" language from which the court deduced a duty to compare the costs and benefits of workplace standards, is a prefatory provision defining the term "occupational safety and health standard."⁴⁰ Assuming that Congress did in fact intend this provision to direct the Secretary of Labor to utilize cost-benefit analysis in the process of setting standards under the Act, it appears that § 6(b)(5), which provides separately for the promulgation of standards limiting exposure to hazardous chemicals, specifically requires that a different balancing process be used when setting these particular standards. Standards under this section must be set at the level "which most adequately assures … that no employee will suffer material impairment of health or functional capacity…."⁴¹ Even if a balancing of regulatory impacts is generally contemplated under the OSH Act, congressional intent seems to direct that when a toxic chemical standard is at issue health considerations are to be given exaggerated weight. Judge Clark never considered this view nor discussed the possible meaning of § 6(b)(5). The opinion treats all OSHA standards identically, and thus effectively reads § 6(b)(5) out of the statute.

Even more vulnerable to criticism may be the court's application of the substantial evidence test. On review of conventional OSHA standards at least one court has recognized that the substantial evidence test does not permit a judicial demand to the Secretary for a factual basis for every conclusion reached.⁴² But when dealing with the arcane science of toxic and carcinogenic substances, it has been the unanimous view of the courts and the authorities that decision makers must be accorded even greater leeway in forming judgments and setting limitations.⁴³ The extraordinarily low concentrations at which substances are injurious, combined with the inherent difficulty in establishing pathological cause-and-effect relationships, require that the scope of judicial review be largely confined to errors in procedure or application of policy. Section 6(b)(5) endorses this principle by providing that one limiting factor to be considered, as distinguished from data to be relied upon, in setting permissible concentrations of toxic chemicals is the availability of scientific data.Again, American Petroleum Institute effectively read this provision out of [8 ELR 10254] the statute, and summarily dismissed the judicial precedent.⁴⁴

The genesis of the court's error in its interpretation of the OSH Act appears to be the first analytical step taken. At the outset of the opinion Judge Clark announced that the subsequent analysis would track that of its earlier opinion in Aqua Slide 'N' Dive Corp. v. Consumer Product Safety Commission⁴⁵ which vacated a swimming pool slide regulation for noncomformance with therequirement of the Consumer Product Safety Act that standards be "reasonably necessary to reduce … an unreasonable risk of injury."⁴⁶ The court in American Petroleum Institute saw no legal difference between that language and the OSH Act requirement that standards be "reasonably necessary … to provide safe places of employment." Upon closer examination, however, there does appear to be a critical difference between the meaning of the two phrases.

A brief contrast of the CPSA and the OSH Act shows that they are organically different in terms of both ends and means. Whereas the OSH Act is designed to prevent risk of injury, the CPSA is designed to prevent unreasonable risk of injury. The term "unreasonable risk" is one which is well defined in the law of negligence⁴⁷ and was not idly inserted into the CPSA. The legislative history shows that the term was intended to incorporate a balancing requirement, but only of a certain kind of cost. It was expected that the process of standard setting would:

involve the Commission in balancing the probability that risk will result in harm and the gravity of such harm against the effects on the product's utility, cost, and availability to the consumer. …⁴⁸

That the focus should be on the costs to consumers is compelling: because the consumer ultimately bears the cost of regulation, those which are too restrictive will ultimately do more harm than good to the intended beneficiary of the law, and thereby defeat the statutory purpose.A common example is the carving knife. Despite that product's role in countless household accidents daily, a standard requiring dull knives would clearly impose a net loss on the consumer. Such risks are reasonable and, more importantly, consumers are willing to accept them.

The problem of occupational hazards is fundamentally different. Workers are not willing to accept risks unless compensated or unless the prevention of risk would cause greater harm, such as the loss of significant numbers of jobs through plant closure. Assuming a given standard, like the benzene standard, would have no such effects, i.e., is "feasible," there can be no balancing because there is no significant cost to the worker. OSHA standards which are "overly" restrictive levy an unjustifiable economic cost on society because the excessive costs are not absorbed by the beneficiaries of the standard, but they do not defeat the statutory purpose of protecting workers.

A crucial observation is that the duty to balance costs and benefits under the CPSA is derived solely from the term "unreasonable risk" and its connotations, not from the requirement that standards be "reasonably necessary." Thus, the Fifth Circuit in American Petroleum Institute may have been too hasty in inferring a balancing obligation from the "reasonably necessary" language in the OSH Act on the grounds thatthe requirements of the two laws are "precisely similar."

Effect of the Decision

The decision in American Petroleum Institute was announced at a time when the question of "costs and benefits" has become one of the most intensely debated issues of regulatory policy. The tendency among some federal officials has been to interpret the ruling as a judicial endorsement of the view that the net value of regulations should be more closely scrutinized. A more specific picture of its likely impacts, however, starts with the holding that OSHA standards must be supported by substantial evidence of their benefits. This new test is a hybridization of two distinct concepts: (1) that the "determinations" of the Secretary must be supported by substantial evidence, and (2) that the "reasonably necessary" language in the statutory definition of "standards" requires the agency to assess the benefits of its regulations. That the new hybrid may prove impossible to administer is suggested by the court's conclusions that the benzene standard was supported by substantial evidence of "some" benefits, but not of "discernible" benefits.⁴⁹ The essence of this test, however, seems to be simply that benefits must be quantifiably measurable and measured.

Assuming that the term "benefits" means the difference between the level of health risks which is currently extant or permissible and the residual level of risk once a proposed standard has been implemented, American Petroleum Institute requires that the dangers be quantified at each level and that the lower estimate be subtracted from the higher. This effectively prohibits OSHA from setting an exposure limitation below the level at which it can conclusively determine the degree of danger. If, after exhausting the most advanced measurement [8 ELR 10255] technology available, OSHA is unable to determine a safe exposure level for a substance, it will be caught between a rock and a hard place. Section 6(b)(5) forbids the setting of a standard at or above the level at which there is a material risk, yet American Petroleum Institute forbids the setting of a standard below the level of the agency's measurement capability.

It would not be suprising if the agency's response to this ruling were simply to produce quantified estimates of hazards where such estimates were only recently declared unavailable. In a footnote Judge Clark noted that the agency had declined to consider a study which claimed to be able to estimate the risks posed by ambient benzene at levels as low as one part per billion.⁵⁰ The agency had no doubt decided against using this type of information because the results of such predictions can be wildly imprecise or inaccurate.⁵¹ If this type of information would satisfy the Fifth Circuit's demand for quantified data, however, the agency may be able to circumvent this aspect of American Petroleum Institute.

While denying that OSHA must undertake an "elaborate cost-benefit analysis," the court stated that the agency:

does have to determine whether the benefits expected from the standard bear a reasonable relationship to the costs imposed by the standard.⁵²

Because the benzene standard was not vacated for failure to compare its costs and benefits, but for failure to support with substantial evidence an estimate of those benefits, the opinion provides no indication of the degree of scrutiny with which cost-benefit "comparisons" may in the future be reviewed. In Aqua Slide, however, the court gave some indication that its review of cost-benefit analyses may be rigorous indeed:

The only way to tell whether the relationship between the advantages and disadvantages of the [standard] is reasonable is to know exactly what those disadvantages are.⁵³

Thus, the court's requirement of a reasonable relationship may in fact be a call for exact estimates of costs and benefits. Leaving aside the staggering problem of quantifying in economic terms the benefits of environmental or health regulation,⁵⁴ this dictum may foretell the imposition of a very difficult burden of proof for agencies.

Unless the decision in American Petroleum Institute is substantially remolded by the Supreme Court, its impact on regulatory agencies could be substantial. To those agencies following broad statutory directives to assess the "reasonableness" of health standards,⁵⁵ or to "consider the costs" of their regulations,⁵⁶ the standard of review implicated in the decision looms as a threat to current decision-making processes insofar as it may require highly quantitative analyses of regulatory benefits. Where agencies are directed to implement health objectives without regard to costs,⁵⁷ or to set pollution levels to provide "an ample margin of safety,"⁵⁸ they may hope to establish an immunity from cost-benefit analysis obligations. Nevertheless, American Petroleum Institute may act to bar such agencies from limiting risks in the absence of solid scientific proof of danger.

Conclusion

The intuitively plausible result in American Petroleum Institute — that society must get what it pays for when "buying" health and environmental protection from the government — is at odds with the likelihood that the decision misinterpreted Congress' command in § 6(b)(5) of the OSH Act. To some extent, this disparity is due to the fact that Congress did not intend, in this provision of the OSH Act as well as in provisions of other environmental and health laws, to require agencies to assure an approximate balance between regulatory costs and benefits. It appears that Congress foresaw abstruse social benefits from stringent occupational health standards which by their nature are incompatible with the concept of cost-benefit review. One such benefit might be the assurance that society not be unjustly enriched through involuntary sacrifices of worker health and mortality. Similarly, the "ample margin of safety" required when regulating toxic air pollutants under the Clean Air Act⁵⁹ provides a certain confidence in the basic healthfulness of the natural environment which is valuable to every citizen but will never be capable of insertion into a cost-benefit calculation.

Congress may properly legislate to obtain such benefits; the fact is, however, that these statutory provisions were enacted with no clear understanding of their economic implications. Section 112 of the Clean Air Act,⁶⁰ which calls for an "ample margin of safety" from toxic pollutants, promises to impose economic burdens on industry so severe that the Environmental Protection Agency, in at least one case, has applied the provision in a manner requiring only that industrial emitters install the best available control technology.⁶¹ Congress has in [8 ELR 10256] hindsight amended another such measure, § 307 of the Federal Water Pollution Control Act, to effect a shift from health-based standards incorporating an ample margin of safety to technology-based standards.⁶² This reflects a congressional predilection to avoid, or at least postpone, the difficult and politically treacherous process of balancing health and environmental benefits against economic costs.

The current debate over regulatory costs and benefits which is consuming the energies of the executive and judicial branches belongs on Capitol Hill. Congress must recognize its responsibility to make these difficult but purely legislative policy decisions, and to ensure that its enactments explicitly instruct the bureaucracy and the courts on how to implement the choices it has made.

1. See, e.g., Hearings on the Cost of Government Regulation on the Consumer Before the Consumer Subcommittee of the Senate Committee on Commerce, Science and Transportation, 95th Cong., 2d Sess. (1978).

2. The original group is the Regulatory Analysis Review Group. The newly-formed group is the United States Regulatory Council, 14 WEEK. COMP. OF PRES. DOC. 1905 (Oct. 31, 1978). See also Exec. Order No. 12044, 43 Fed. Reg. 12661 (1978) (directing federal agencies to analyze more fully impacts of proposed and existing regulatory programs).

3. 581 F.2d 493, 8 ELR 20790 (5th Cir. 1978).

4. Section 3(8) of the Occupational Safety and Health Act, 29 U.S.C. § 652(8) provides in relevant part:

The term "occupational safety and health standard" means a standard … reasonably necessary or appropriate to provide safe or healthful employment or places of employment.

5. 43 Fed. Reg. 5940 (1978).

6. See discussion at text accompanying notes 17-21, infra.

7. Roughly five weeks after release of the opinion in American Petroleum Institute OSHA promulgated a standard limiting exposure to lead. 43 Fed. Reg. 54353 (1978). A trade association promptly filed suit in the Fifth Circuit, Lead Industries Association v. Occupational Safety and Health Administration, No. 78-3475 (5th Cir., filed Nov. 21, 1978). Interestingly, the United Steelworkers mounted a challenge just as promptly in the Third Circuit. In fact, the union's petition and the industries' petition were filed at the exact same minute. United Steelworkers of America v. Occupational Safety and Health Administration, No. 78-2452 (3d Cir., filed Nov. 21, 1978).

8. Marshall v. American Petroleum Institute, No. 78-1036, 47 U.S.L.W. __ (filed Dec. 26, 1978). The intervenor below also petitioned for certiorari. Industrial Union Department, AFL-CIO v. American Petroleum Institute, No. 78-911, 47 U.S.L.W. 3422 (filed Dec. 7, 1978).

9. 29 U.S.C. § 651(b).

10. 29 U.S.C. § 666.

11. 29 U.S.C. § 656(b)(5).

12. Id.

13. Id.

14. H.R. REP. NO. 91-1291, 91st Cong., 2d Sess. 18 (1970).

15. Society of the Plastics Industry, Inc. v. Occupational Safety and Health Administration, 509 F.2d 1301, 1309, 5 ELR 20157, 20160 (2d Cir. 1975) ("[I]t remains the duty of the Secretary to act to protect the workingman, and to act even in circumstances where existing methodology or research is deficient."); Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467, 474-475 n.18, 4 ELR 20415, 20419 n.18 (D.C. Cir. 1974) ("[W]here existing methodology or research in a new area is deficient, the agency necessarily enjoys broad discretion to attempt to formulate a solution to the best of its ability on the basis of available information." [citation ommited]). See also Ethyl Corp. v. Environmental Protection Agency, 541 F.2d 1, 28, 6 ELR 20267, 20281 (D.C. Cir. 1976) ("Where a statute is precautionary in nature, n57 the evidence difficult to come by, uncertain, or conflicting because it is on the frontiers of scientific knowledge, the regulations designed to protect the public health, and the decision that of an expert administrator, we will not demand rigorous step-by-step proof of cause and effect. Such proof may be impossible to obtain if the precautionary purpose of the statute is to be served.

See generally DAVIS, ADMINISTRATIVE LAW OF THE SEVENTIES 646-687, 674 (1976) ("If carcinogenicity of the chemicals in humans can be neither proved nor disproved by scientific evidence, the problem for rulemakers is not one of fact; it is one of making a legislative choice of policy in light of the absence of evidence…. [The agency] has to use its policy preferences when proof is lacking."); Doniger, Federal Regulation of Vinyl Chloride: A Short Course in the Law and Policy of Toxic Substances Control, 7 ECOLOGY L.Q. 497, 508-522 (1978); Note, Environmental Carcinogenesis: Regulating on the Frontiers of Science, 7 ENVT'L LAW 83 (1976).

57. Or, as with OSHA, mandatory in its command to act….").

16. 5 U.S.C. § 706(2)(E), ELR STAT. & REG. 41005.

17. Synthetic Organic Chemicals Manufacturers Ass'n v. Brennan, 503 F.2d 1155, 1157 (3d Cir. 1975); Associated Industries of New York State v. Department of Labor, 487 F.2d 342, 347-349 (2d Cir. 1973).

18. Society of the Plastics Industry, Inc. v. Occupational Safety and Health Administration, 509 F.2d 1301, 1304, 5 ELR 20157, 20158 (2d Cir. 1975); Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467, 474, 4 ELR 20415, 20419 (D.C. Cir. 1974); Associated Industries of New York State v. Department of Labor, 487 F.2d 342, 348 (2d Cir. 1973).

19. Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467, 476, 4 ELR 20415, 20420 (D.C. Cir. 1974).

20. Associated Industries of New York State v. Department of Labor, 487 F.2d 342, 349-350 (2d Cir. 1973).

21. American Iron and Steel Institute v. Occupational Safety and Health Administration, 577 F.2d 825, 830-831 (3d Cir. 1978); Synthetic Organic Chemicals Manufacturers Ass'n v. Brennan, 503 F.2d 1155, 1160 (3d Cir. 1975).

22. 29 U.S.C. § 655(b)(5) (1975). See generally, Economic and Technological Feasibility in Regulating Toxic Substances Under the Occupational Safety and Health Act, 7 ECOLOGY L.Q. 285 (1978).

23. Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467, 478, 4 ELR 20415, 20421 (D.C. Cir. 1974).

24. Society of the Plastics Industry, Inc. v. Occupational Safety and Health Administration, 509 F.2d 1301, 1309, 5 ELR 20157, 20160 (2d Cir. 1975).

25. 581 F.2d at 501, 8 ELR at 20793.

26. 43 Fed. Reg. 5918 (1978).

27. The other measures included a flat prohibition on dermal contact with liquid benzene and labeling and monitoring requirements.

28. 43 Fed. Reg. 5918-33 (1978).

29. 43 Fed. Reg. at 5925.

30. Id.

31. Id. at 5934-40.

32. Id. at 5940.

33. 569 F.2d 831 (5th Cir. 1978).

34. 15 U.S.C. § 2056(a). See also § 2058(c)(2)(A).

35. 581 F.2d 493, 7 ELR 20790 (5th Cir. 1978).

36. The court noted that the petitioners had not seriously challenged the figure. 581 F.2d 493, 503 n.22, 7 ELR 20790, 20794 n.22.

37. See authorities cited at note 15, supra.

38. 58 F.2d at 504, 8 ELR at 20794 (citation omitted).

39. 581 F.2d at 505, 8 ELR at 20795.

40. 29 U.S.C. § 652(8).

41. 29 U.S.C. § 655(b)(5).

42. AFL-CIO v. Brennan, 530 F.2d 109, 114-116 (3d Cir. 1975).

43. See note 15, supra.

44. The court's insistence on complete documentation of the benzene standard's effectiveness in reducing risk is even more perplexing when contrasted with another case decided by the court this year. At issue in D. D. Bean & Sons v. Consumer Products Safety Commission, 574 F.2d 643 (5th Cir. 1978), was the validity of a consumer product safety standard designed to increase the safety of matchbooks. With respect to that aspect of the regulation requiring the sandpaper striking surface to be placed on the reverse side of matchbooks, the opinion states that the CPSC had presented no studies or other empirical evidence showing that the requirement would have any effect in reducing the risk of injury. 574 F.2d at 649. Nevertheless, the court found the curative effects "patent" and upheld the standard, adding:

we do not think the Commission had to cite to empirical data in support of its finding that the particular requirements were likely to reduce the risk of injury.

45. 569 F.2d 831 (5th Cir. 1978).

46. 15 U.S.C. § 2056(a).

47. W. PROSSER, LAW OF TORTS § 31, at 145 (4th ed. 1971).

48. H.R. REP. NO. 1153, 92d Cong., 2d Sess. 33 (1972) (emphasis supplied). Accord, S. REP. NO. 749, 92d Cong., 2d Sess. 6, 14, 43-44 (1972). See generally Note, The Consumer Product Safety Act; Risk Classification and Risk Liability, 8 IND. L. REV. 846 (1975).

49. 581 F.2d at 503, 8 ELR at 20794.

50. 581 F.2d at 504 n.23, 8 ELR at 20794 n.23.

51. See discussion at 43 Fed. Reg. 5940-41.

52. 581 F.2d at 503, 8 ELR 20794.

53. 569 F.2d 831, 839.

54. The concept of cost-benefit analysis presumes that the benefits within the equation can be measured in a manner permitting meaningful consideration. The difficulty of quantifying the value of intangible benefits is beyond the scope of this comment, but it is clear that as increasing precision is required of and increasing reliance placed upon an agency's estimates of such benefits, the arbitrariness of the estimates and the complexity of judicial review will increase as well.

55. See, e.g., Toxic Substances Control Act § 6(a), 15 U.S.C. § 2605(a), ELR STAT. & REG. 41341; Federal Insecticide, Fungicide and Rodenticide Act, 7 U.S.C. § 136a(c)(5), ELR STAT. & REG. 41307.

56. See, e.g., Stfe Drinking Water Act, § 1412(a)(2), 42 U.S.C. § 3000g-1(a)(2), ELR STAT. & REG. 41131 (1978); Clean Air Act § 111(a)(1)(C), 42 U.S.C. § 7411(a)(1)(C), ELR STAT. & REG. 42216.

57. See, e.g., the Delaney Clause of the Food, Drug and Cosmetic Act, § 409(c)(3)(A), 21 U.S.C. § 348(c)(3)(A). ELR STAT. & REG. 41325.

58. See, e.g., Clean Air Act, § 112(b)(1)(B), 42 U.S.C. § 7412(b)(1)(B), ELR STAT. & REG. 42220.

59. Id.

60. Id.

61. The agency's adoption of this policy in the process of setting an emission standard for vinyl chloride was challenged in court, resulting in a settlement decree calling for stricter, though still technology-based, limitations. See Doniger, Federal Regulation of Vinyl Chloride: A Short Course in the Law and Policy of Toxic Substances Control, 8 ECOLOGY L.Q. 501, 565-588 (1978).

62. Clean Water Act of 1977 § 42, Pub. L. No. 95-217, 91 Stat. 1567.