Does the Occupational Safety and Health Act require the Secretary of Labor, in promulgating a toxic material or harmful physical agent standard under § 6(b)(5), to determine that the costs of the standard bear a reasonable relationship to its benefits?

Congress enacted the Occupational Safety and Health Act of 1970 to assure safe and healthful working conditions for every worker. In 1978, the Secretary of Labor, acting through OSHA, promulgated a standard limiting occupational exposure to cotton dust, which causes byssinosis ("brown lung" disease), a potentially disabling respiratory disease. At least 35,000 active and retired cotton mill workers suffered the most disabling form of the disease. The Standard set permissible exposure limits of 200 micrograms per cubic meter for yarn manufacturing, 750 for slashing and weaving, and 500 for other processes. It relied on engineering controls and work practice controls, with respirators during a four-year compliance period. It also required employers to transfer workers unable to wear respirators to lower-dust jobs with no loss of earnings or employment rights. OSHA estimated total compliance costs of $656.5 million and found the Standard technologically and economically feasible. Cotton industry representatives challenged the Standard, arguing that OSHA had to show a reasonable relationship between its costs and benefits, and that its feasibility findings lacked substantial evidence. Respondents were the Secretary of Labor and two labor organizations.

OSHA issued the Cotton Dust Standard on June 23, 1978, after notice-and-comment rulemaking and three hearings. Industry petitioners challenged the Standard in the Court of Appeals for the District of Columbia Circuit under § 6(f) of the Act. The Court of Appeals upheld the Standard in all major respects, holding that the Act does not require a cost-benefit comparison and that OSHA's feasibility determinations were supported by substantial evidence. It also upheld the wage guarantee provision. It remanded only the cottonseed oil industry portion of the Standard. The Supreme Court granted certiorari to resolve the cost-benefit question left open in Industrial Union Dept. v. American Petroleum Institute and to decide the related issues.

(Brennan, J.)

No. The Act does not require cost-benefit analysis for standards under § 6(b)(5). The Court affirmed in part and vacated in part.

The plain meaning of "feasible" is "capable of being done." Section 6(b)(5) directs the Secretary to set the standard that most adequately assures that no employee will suffer material impairment of health, limited only by what is achievable. Congress itself defined the relationship between costs and benefits by placing worker health above all considerations except those making its attainment unachievable. A cost-benefit requirement would substitute a different balance for the one Congress struck. Congress uses explicit language when it wants cost-benefit analysis, as in the Flood Control Act of 1936 and the Outer Continental Shelf Lands Act Amendments of 1978.

Section 3(8)'s "reasonably necessary or appropriate" language does not engraft a cost-benefit requirement onto § 6(b)(5). Even if § 3(8) alone contemplated balancing, § 6(b)(5) imposes a specific feasibility requirement for toxic materials and harmful physical agents, and the general language cannot countermand it. Reading in a cost-benefit test would nullify the feasibility requirement and violate the rule that all parts of a statute be given effect. The legislative history confirms that Congress meant "feasible" and nothing else, and that it knew the Act would impose substantial compliance costs and regarded them as a cost of doing business.

Substantial evidence supported OSHA's cost estimate. It reasonably adopted the Hocutt-Thomas figure after rejecting the RTI estimate, and it was limited by the industry's refusal to supply confidential data. Substantial evidence also supported the economic feasibility finding, since RTI found even a $2.7 billion standard would not seriously threaten the industry as a whole.

The wage guarantee provision is vacated. OSHA never explained it as related to health protection, and post hoc rationalizations of the agency or the parties cannot supply the statement of reasons that § 6(e) and (f) require. The Act does not authorize OSHA to remedy general unfairness unrelated to health and safety.

Dissent (Stewart, J.) OSHA failed to support its cost estimate with substantial evidence, so the judgment should be reversed without reaching the cost-benefit question. Economic feasibility requires a reasonably clear idea of cost. OSHA rejected RTI's estimate of the final Standard as a gross overestimate. The only other estimate, Hocutt-Thomas, addressed a far less stringent and less costly standard for nearly all processes. OSHA treated that estimate as reliable for the stricter final Standard without rationally explaining why. This is unsupported speculation, not substantial evidence. The agency categorically misconceived the evidence needed to support its regulation, and the Court of Appeals failed to correct the error.

Dissent (Rehnquist, J., joined by Burger, C.J.) Section 6(b)(5) is an unconstitutional delegation of legislative authority to the Executive Branch. The phrase "to the extent feasible" is no standard at all. Congress faced a hard policy choice among mandating, permitting, or prohibiting cost-benefit analysis, and it passed the choice to the Secretary by using vague language that masked a fundamental disagreement. The Court's "feasibility analysis" has no fixed meaning, and nothing in the statute limits feasibility to technological and economic considerations. The decision whether to disregard the statistical possibility of future deaths in light of economic costs is a quintessentially legislative choice for elected representatives. He invokes the delegation doctrine because Congress failed to choose among plausible interpretations, not merely because several exist, which answers the majority's footnote suggestion.