# Petitioners Brief — American Textile Mfrs. Institute, Inc. v. Donovan > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/brief%3Amicro_IA40385006_2223%3A06 ## Record - **Collection:** Supreme Court brief - **Document type:** Petitioners Brief - **Published:** January 1, 1981 - **Citation:** 452 U.S. 490 ## Text ene ROR EES: No. 79-1429 (Consolidated with No. 79-1583) NOF 20 #0 IN THE MICHAEL RODAK, JR. CLERI Supreme Court of the United States OCTOBER TERM, 1980 AMERICAN TEXTILE MANUFACTURERS INSTITUTE, INC., MILLIKEN & COMPANY, ARKWRIGHT MILLS, SPARTAN MILLS, BLAIR MILLS, INC., HERMITAGE, INC., DAN RIVER, INC., CONE MILLS CORPORATION, MAYFAIR MILLS, SPRINGS MILLS, INC., RIEGEL TEXTILE CorR- PORATION, FIELDCREST MILLS, INC., AND WEST POINT- PEPPERELL, INC., é Petitioners, RAY MARSHALL, SECRETARY OF LABOR, UNITED STATES DEPARTMENT OF LABOR, et al., Respondents. On Writ of Certiorari to the United States Court of Appeals for the District of Columbia Circuit BRIEF FOR PETITIONERS ROBERT H. BoRK NEIL J. KING 142 Huntington Street Counsel of Record New Haven, Connecticut 06511 A. STEPHEN HUT, JR. ANDREW N. VOLLMER GREGORY B. TOBIN WILMER & PICKERING ~~ ' cgi Fs aa 1666 K Street, N.W. EDWARDS Washington, D.C. 20006 First National Bank Tower (202) 872-6000 Two Peachtree Street, N.W. Atlanta, Georgia 30383 Counsel for American Textile Manufacturers Institute, Inc. [Counsel for other Petitioners are listed on inside cover] ROBERT T. THOMPSON GARY S. KLEIN THOMPSON, MANN & HUTSON The Daniel Building Suite 2222 Greenville, S.C. 29602 Counsel for Milliken and Company JOSEPH K. MADDOX, Jr. P.O. Box 5784 Spartanburg, S.C. 29304 Counsel for Spartan Mills ROBERT T. THOMPSON GARY S. KLEIN THOMPSON, MANN & HUTSON The Daniel Building Suite 2222 Greenville, S.C. 29602 Counsel for Hermitage, Inc. SAMUEL K. ABRAMS BRIAN E. MoRAN BAKER & HOSTETLER 818 Connecticut Ave., N.W. Washington, D.C. 20006 H. J. ELAM, III NEIL W. KOONCE Cone Mills Corporation Greensboro, N.C. 27405 Counsel for Cone Mills Corporation DAN M. ByrbD, Jr. J. SPRATT WHITE P.O. Box 70 Fort Mill, S.C. 29715 Counsel for Springs Mills, Ine. ROBERT H. BoRK 142 Huntington Street New Haven, Connecticut 06511 Counsel for Fieldcrest Mills, Inc. November 20, 1980 THOMAS A. EVINS CLYDE H. HAMILTON BUTLER, MEANS, EVINS & BROWNE P.O. Box 451 Spartanburg, S.C. 29304 Counsel for Arkwright Mills ROBERT T. THOMPSON GARY S. KLEIN THOMPSON, MANN & HUTSON The Daniel Building Suite 2222 Greenville, S.C. 29602 Counsel for Blair Mills, Ine. HARLAN H. HUNTLEY ROGER L. TUTTLE 2291 Memorial Drive Danville, Virginia 24541 Counsel for Dan River, Ine. THOMAS A. EVINS CLYDE H. HAMILTON BUTLER, MEANS, EVINS & BROWNE P.O. Rox 451 Spartanburg, S.C. 29304 Counsel for Mayfair Mills FRED M. RICHARDSON Lovic A. BROOKS, JR. CHARLES A. EDWARDS CONSTANGY, BROOKS & SMITH 1900 Peachtree Center Building 230 Peachtree Street, N.W. Atlanta, Georgia 30303 Counsel for Riegel Textile Corporation RICHARD H. MONK, JR. C. POWERS DORSETT West Point-Pepperell, Inc. P.O. Box 71 West Point, Georgia 31833 Counsel for West Point- Pepperell, Inc. QUESTIONS PRESENTED FOR REVIEW 1. Whether Congress gave the Occupational Safety and Health Administration (“OSHA”) the unprecedented power to set enormously costly occupational health stand- ards in disregard of their economic impact, the sole limit being that the costs of a single standard must not be so great as to threaten the destruction of an entire industry. 2. Whether Congress intended to permit OSHA to adopt standards that consume enormous amounts of soc- iety’s limited industrial hygiene resources without regard to whether such standards would produce any significant or discernible reduction in a risk of material health impairment. ; 3. Whether OSHA has the authority to require that employers transfer and guarantee the wages and benefits of workers who are not suffering from any occupationally- related health impairment.* *In addition to the parties listed in the caption of the case, the following were parties to the proceeding in the court below: the National Cotton Council; the Cotton Warehouse Association; the American Cotton Shippers Association; the National Cottonseed Products Association; the American Federation of Labor & Con- gress of Industrial Organizations; the Industrial Union Depart- ment, AFL-CIO; and the Amalgamated Clothing and Textile Work- ers Union. The American Textile Manufacturers Institute, Inc. is a trade associaiton consisting of approximately 175 companies. (i) TABLE OF CONTENTS Page QUESTIONS PRESENTED FOR REVIEW ................ i RRR OO UPTO anno nn c rsnecansecsnnsctccescescee v Pe i 1 NTT id Soins vk cnedsich apie eleeaabebdeneiciohscaesiesotieahoeansces 2 CONSTITUTIONAL AND STATUTORY PROVI- I 2 STATMMENS OF THE CARR annonce cccsccccccccscie 2 A. The Nature of Byssinosis .................................... 3 B. The Proceedings Before OSHA ....................... 7 1. Health Effects Evidence ........0.2.0000000.00000...... 8 2. The ATMI Alternative 0.000000... 11 ’ 8. Evidence on the Economic Impact of the RR ARAMA Rie rt et 12 C. OSHA’s Adoption of the Standard _................. 14 D. The Decision of the Court Below ...................... 17 SUMMARY OF ARGUMENT Www. 19 POSER RRR MPL SANS SCPE OREO oe TR CRE SOS ee 21 I. THE COTTON DUST STANDARD IS PREDI- CATED UPON LEGALLY INSUFFICIENT CRITERIA FOR ASSESSING THE ANTICI- PATED ECONOMIC IMPACT AND REDUC- TION IN RISK THAT THE STANDARD IS EXPECTED TO PRODUCE ...W000 21 A. The Decision Below Mistakenly Permits OSHA To Adopt Health Standards That Are Subject to No Meaningful or Effective Cost- Related Limitations .0.0000000000..0oooo cc cccccceecceee 22 | PREVIOUS PAGE WAS BLANK iv TABLE OF CONTENTS—Continued i. OSHA Failed To Make a Responsible and Supportable Estimate of the Costs of Its Standard or To Apply Any Meaning- ful Criteria in Assessing the Standard’s Economic Impact -S.,........\......ccccccccccccceeees OSHA’s Construction of the Act—Under Which the Costs of a Standard Are Irrele- vant Short of the Point at Which They Threaten To Ruin an Entire Industry— Is Inconsistent with the Statutory Lan- guage and Legislative History .............. B. The Decision of the Court Below Should Be Reversed Because OSHA Did Not Show That the Standard Was Likely To Achieve a Reduction in Risk That Is Significant When Considered in Light of Its Costs _..... 1. OSHA Is Required To Show That a Rea- sonable Relationship Exists Between the Risk Reduction Benefits and the Costs of Dee PD Sibincteecie i ae SUNN." siblanhiglbdedidbctntshinbscbantsnipidubadhneiiitniesestneennnons | RUMP NERO ER LE OTe Te ERS The Judgment Below Must Be Reversed Because OSHA Did Not Show That the Standard Would Produce a Reduction in Risk That Is Significant in Light of Its 0 ETE SEARS 2 Mis eR are ARAL Rae Il. THE COURT BELOW ERRONEOUSLY CON- CLUDED THAT OSHA HAS AUTHORITY TO REQUIRE EMPLOYERS TO MAINTAIN THE WAGES AND BENEFITS OF EM- PLOYEES WHO ARE TRANSFERRED FOR REASONS UNRELATED TO HEALTH IM- TIE iikak bos niliind in) Dectiadciainccsnca)imocdtdvenicrshcapnsn GOIN | Ahi snssdcinucichca Rs lthicadat-cillith aden ssaeal Melisa APPENDIX Page 24 31 36 36 36 41 48 51 57 la Vv TABLE OF AUTHORITIES Cases: Page A.L.A. Schechter Poultry Corp. v. United States, ee es Ie CD Glas ee 23 AFL-CIO v. Marshall, 617 F.2d 636 (D.C. Cir. BE lesen cia aa a dr ia lea te passim Amalgamated Meat Cutters & Butcher Workmen v. Connally, 337 F. Supp. 787 (D.D.C. 1971) ...... 41 American Iron & Steel Institute v. OSHA, 577 F.2d 825 (3d Cir. 1978), cert. granted, 48 U.S.L.W. 3855 (U.S. July 2, 1980), cert. dis- missed, 49 U.S.L.W. 3145 (U.S. Sept. 10, 1980).. 4, 43 American Petroleum Institute v. OSHA, 581 F.2d 493 (5th Cir. 1978), aff'd sub nom. Industrial Union Department, AFL-CIO v. American Petroleum Institute, 48 U.S.L.W. 5022 (U.S. IE Ss, TD edict Ss Od eee 14, 27, 41 American Power & Light Co. v. SEC, 329 U.S. 90 | REN Per See A aA ae LER POR TR 22 Appalachian Power Co. v. Train, 545 F.2d 1351 S| ee ER. Cg A ee 44 Aqua Slide ‘N’ Dive Corp. v. CPSC, 569 F.2d 831 RN Ts II os scenchisiescnpamciniachiadsiescaice cane panei, 43 Associated Industries of New York State, Inc. v. Department of Labor, 487 F.2d 342 (2d Cir. RENTED colt sancaissebiskeieirsapihodtlecin Drciiltetluktia heats 34, 51 D.D. Bean & Sons v. CPSC, 574 F.2d 648 (1st Cir. | ee ERIE AURIS AMT AY bie RD ne ee 43 Dry Color Manufacturers Association v. Depart- ment of Labor, 486 F.2d 98 (3d Cir. 1978) ....... 4 Florida Peach Growers Association v. Department of Labor, 489 F.2d 120 (5th Cir. 1974) 34 . Forester v. CPSC, 559 F.2d 774 (D.C. Cir. 1977).. 43 General Electric Co. v. Southern Construction Co., 383 F.2d 135 (5th Cir. 1967), cert. denied, 390 os 6 OR ee eee 56 H & H Tire Co. v. Department of Transportation, 471 F.2d 860 (7th Cir. 1972) .........0.000000. oo... 44 vi TABLE OF AUTHORITIES—Continued Page Industrial Union Department, AFL-CIO v. Ameri- can Petroleum Institute, 48 U.S.L.W. 5022 (U.S. Se SRDS Coie SRE Fc i PR) Ab etre passim Industrial Union Department, AFL-CIO v. Hodg- son, 499 F.2d 467 (D.C. Cir. 1974) 2000000000... passim Kent v. Dulles, 357 U.S. 116 (1958) .................... 24 Marshall v. West Point Pepperell, Inc., 588 F.2d BU. PE PUD sa disdain sl ceseincceSieidllcchich'vadideecoe Laws 42 National Cable Television Association, Inc. v. United States, 415 U.S. 886 (1974) .................... 24 National Congress of Hispanic American Citizens v. Marshall, 626 F.2d 882 (D.C. Cir. 1979) ...... 46, 50 National Crushed Stone Association, Inc. v. EPA, 601 F.2d 111 (4th Cir. 1979), cert. granted, 48 U.S.L.W. 3535 (U.S. Feb. 19, 1980)... 44 National Wildlife Federation v. Andrus, 440 F. es See Ce. SOFT) onic 27 Panama Refining Co. v. Ryan, 298 U.S. 388 (| SAAN aA, EN Po OE MOOT ON ae 22 RMI Co. v. Secretary of Labor, 594 F.2d 566 (6th Ck, TSC A alte Ar Se OT Pa NSN IRE 42 Society of the Plastics Industry, Inc. v. OSHA, 509 F.2d 1801 (2d Cir.), cert. denied, 421 U.S. SRNR RTE SRSA Ceti ENA: eed See 4 Synthetic Organic Chemical Manufacturers Associ- ation v. Brennan, 506 F.2d 385 (3d Cir. 1974), cert. denied, 428 U.S. 880 (1975) ........................ 4 Synthetic Organic Chemical Manufacturers Associ- ation v. Brennan, 508 F.2d 1155 (3d Cir. 1974), cert. denied, 420 U.S. 973 (1975) ....................... 51 Taylor Diving & Salvage Co. v. Department of Labor, 599 F.2d 622 (5th Cir. 1979) ............... 55 Texas Independent Ginners Association v. Mar- shall, Nos. 78-2668, et al. (5th Cir. Nov. 14, og Ape SR FSIS RA ae ENE NT Re passim Turner Co. v. Secretary of Labor, 561 F.2d 82 (7th |. Ger SSR eta ane an Or Sa ee te 42 United Parcel Service of Ohio, Inc. v. OSHRC, 570 vase aoe fom Cie, 1978) .........:.......... 50 vii TABLE OF AUTHORITIES—Continued Page United States v. Rock Royal Co-Op, Inc., 307 U.S. RRR RIS Hee EE Rene Re ea cea 22 United Steelworkers of America, AFL-CIO-CLC v. Marshall, [1980] OSH Dec. (CCH) 30829, 8 OSH Cas. (BNA) 1810 (D.C. Cir. Aug. 15, ETRE Se NER de RY Teal SONU SUNT EPRI aOR passim Whirlpool Corp. v. Marshall, 445 U.S. 1 (1980)... 55 Yakus v. United States, 321 U.S. 414 (1944) ........ 41 Administrative Decisions and Reports: Castle & Cooke Foods, [1977-1978] OSH Dec. (CCH) 26225, 5 OSH Cas. (BNA) 14385 (1977), appeal docketed, No. 77-2565 (9th Cir. July 14, 5 gO EEE RI Soe RC Rey SS RUN RE 45 Continental Can Co., [1976-1977] OSH Dec. (CCH) 25250, 4 OSH Cas. (BNA) 1541 BEEING SMe ih TS EONS eh 8 Nea 44-45 Department of Labor Report to the Congress, Cotton Dust: Review of Alternative Technical Standards and Control Technologies (May 14, ) | ARRAS IANS 5 AER Ieee SSR IO PPE A PRS Ve a a eae 47, 49 Great Falls Tribune Co., [1977-1978] OSH Dec. (CCH) 26308, 5 OSH Cas. (BNA) 14838 (1977), appeal docketed, No. 77-2566 (9th Cir. July 14, BONO, Secu teeieaaegicinase an widiahachinantdensbiscerumissonstae 45 KLI, Inc., [1977-1978] OSH Dec. (CCH) 26935, 6 OSH Cas. (BNA) 1097 (1977) 0000.0... 45 Making Prevention Pay, Final Report of the Inter- agency Task Force on Workplace Safety and BRUM CUP NR BIPIOED oeheiak ccc Sdinpicbacccestecceiebsouse 40 Samson Paper Bag Co., [1980] OSH Dec. (CCH) I I iitah ith ia rbd le ee 45 West Point Pepperell, Inc., [1977-1978] OSH Dec. (CCH) 26136, 5 OSH Cas. (BNA) 1257 (1977), aff'd, Marshall v. West Point Pepperell, Inc., GOe Fame Oem COC CAR, ROTO) ..n.n Feu meee. SOOR C2978) ....................................... 46 48 Fed. Reg. 21888 (1978) ..................................... 46 43 Fed. Reg. 27350 (1978), amended, 48 Fed. Reg. 28473 (1978), 43 Fed. Reg. 35032 (1978), 43 Fed. Reg. 56898 (1978) ............0.0000 eee passim — 2 oe eee Chere) .................................... 28, 53 48 Fed. Reg. 54854 (1978) ..................................... 28 44 Fed. Reg. 58642 (1979)... 82, 45, 46 45 Fed. Reg. 53672 (Aug. 12, 1980) _............... 39 Legislative Materials: 115 Cong. Rec. 22517 (1969) 33 Senate Subcommittee on Labor of the Committee on Labor and Public Welfare, 92d Cong., 1st Sess., Legislative History of the Occupational Safety and Health Act of 1970 (Comm. Print pe ESE aR 6, 38, 34, 35 H.R. 16785, 91st Cong., 2d Sess. (1970), reprinted in Senate Subcommittee on Labor of the Com- mittee on Labor and Public Welfare, 92d Cong., Ist Sess., Legislative History of the Occupa- tional Safety and Health Act of 1970 (Comm. GARTER vc 55 S. Rep. No. 91-1282, 91st Cong., 2d Sess. (1970), reprinted in Senate Suhcommittee on Labor of the Committee on Labor and Public Welfare, 92d Cong., 1st Sess., Legislative History of the Occupational Safety and Health Act of 1970 a 6 Miscellaneous: Currie, OSHA, 1976 Am. B. Foundation Research TERE IE ARS RGA eg 82 J. Freedman, Crisis and Legitimacy ( | 23 Kasper, Cost-Benefit Analysis in Environmental Decisionmaking, 45 Geo. Wash. L. Rev. 1018 SSA 47 x TABLE OF AUTHORITIES—Continued Page Monthly Labor Review (March 1975) _............. 25 Monthly Labor Review (Oct. 1978) 25 National Academy of Sciences, Decision Making for Regulating Chemicals in the Environment ae oT PTE a CLL A TT OF OR 47 Oi, On the Economics of Industrial Safety, 38 L. & Contemp. Prob. 669 (1974) 47 Stewart, The Reformation of American Adminis- trative Law, 88 Harv. L. Rev. 1667 (1975) ...... 24 The Supreme Court, 1979 Term, 94 Harv. L. Rev. Pe IE ir a let Se passim Tumorigenic Citations Subfile of the Registry of Toxic Effects of Chemical Substances (Oct. 1979) (National Inst. for Occupational Safety RO APE SARL WESTER TC NDS, UIP eh Sy, 89 IN THE Supreme Court of the United States OCTOBER TERM, 1980 No. 79-1429 (Consolidated with No. 79-1583) AMERICAN TEXTILE MANUFACTURERS INSTITUTE, INC., MILLIKEN & COMPANY, ARKWRIGHT MILLS, SPARTAN MILLs, BLAIR MILLs, ING., HERMITAGE, INC., DAN RIVER, INC., CONE MILLS CORPORATION, MAYFAIR MILLS, SPRINGS MILLS, INC., RIEGEL TEXTILE Cor- PORATION, FIELDCREST MILLS, INC., AND WEST POoINT- PEPPERELL, INC., ‘ Petitioners, RAY MARSHALL, SECRETARY OF LABOR, UNITED STATES DEPARTMENT OF LABOR, et al., Respondents. On Writ of Certiorari to the United States Court of Appeals for the District of Columbia Circuit BRIEF FOR PETITIONERS OPINION BELOW The opinion of the United States Court of Appeals for the District of Columbia Circuit is reported at 617 F.2d 636 and is reprinted in the Appendix to the Petition for Certiorari (“Pet. App.”) at 19. The challenged regula- tion and the statement of reasons accompanying it appear 2 at 43 Fed. Reg. 27350 (1978), as amended at 43 Fed. Reg. 28473 (1978), 43 Fed. Reg. 35032 (1978), and 43 Fed. Reg. 56892 (1978). JURISDICTION The court of appeals entered judgment on October 24, 1979. Petitioners filed a timely petition for rehearing on November 7, 1979, which the court below denied on Janu- ary 11, 1980. Pet. App. at 103. Petitioners filed a timely petition for a writ of certiorari on March 14, 1980, in- voking jurisdiction under 28 U.S.C. § 1254(1) and 29 U.S.C. §660(a). This Court granted the petition by order dated October 6, 1980. CONSTITUTIONAL AND STATUTORY PROVISIONS INVOLVED Relevant sections of the United States Constitution and the Occupational Safety and Health Act of 1970, 29 U.S.C. §§ 651-78 (1976 & Supp. III 1979) (the “OSH Act” or “Act”) are set forth in the Appendix to this Brief. STATEMENT OF THE CASE Petitioners, the American Textile Manufacturers Insti- tute, Inc. (“ATMI”) and twelve individual textile manu- facturing companies, brought this action for judicial review of the Standard for Occ national Exposure to Cotton Dust (the “Cotton Dust Jtandard” or “Stand- ard”), promulgated by the Occupational Safety and Health Administration (“OSHA”) on June 19, 1978. 438 Fed. Reg. 27350 (1978), as amended at id. 28473, id. 350382, id. 56893. The Standard was promulgated pursuant to the OSH Act, Section 3(8) of which authorizes OSHA to adopt standards requiring “conditions, . . . practices, . operations or processes, reasonably necessary or ap- 3 propriate to provide safe or healthful employment and places of employment.” 29 U.S.C. § 652(8). Section 6(b) (5) of the Act provides that . in promulgating standards dealing with toxic mate- rials or harmful physical agents . . . , [OSHA] shall set the standard which most adequately assures, to the extent feasible, on the basis of the best avail- able evidence, that no employee will suffer material impairment of health or functional capacity ... . Id. § 655 (b) (5). OSHA adopted the Standard purportedly to protect workers in the textile manufacturing and other cotton handling industries from the risk of contracting byssino- sis, a respiratory condition that has been associated with exposure to cotton dust. Petitioners agree that, subject to the requirements and limitations of the Act, textile workers should be protected against serious and disabling effects of byssinosis. The dispute in this case centers on whether OSHA’s adoption of the Cotton Dust Standard conformed with the applicable statutory requirements and limitations—particularly as they relate to risk re- duction and cost considerations. Petitioners believe that the Standard must be set aside because OSHA misappre- hended the nature of the economic impact inquiry that the Act requires and imposed a wage guarantee obliga- tion that has no statutory basis. A. The Nature of Byssinosis Cotton dust is not a carcinogen, and byssinosis, the health impairment to which the Cotton Dust Standard is addressed, is not a form of cancer. Unlike cancer, bys- sinosis can be detected in its early stages when it is 4 reversible and nondisabling.' In these respects, the Cot- ton Dust Standard differs from the Benzene Standard reviewed by this Court and from all of the other toxic substance standards that had been reviewed by the courts of appeals prior to the decision of the court below.” The respiratory effects that characterize byssinosis have been categorized into four “grades” (14, 1, 2, and 3). 41 Fed. Reg. 56498, 56500-01 (1976). Grade % byssinosis is generally characterized by occasional chest tightness on Mondays; it is mild and reversible and dis- appears entirely upon removal from exposure.* Indeed, Grade % byssinosis cannot really be deemed a health impairment of any sort, since it involves essentially ‘The fact that byssinosis is not a malignant disease is also reflected by the fact that textile workers have not been found to exhibit an excess incidence of mortality as a result of their ex- posure to cotton dust. See Ex. 6, No. 76 at III-6. Citations to the record compiled by OSHA include references to exhibits (“Ex.”) and transcript testimony (“Tr.”). 2 See Industrial Union Dep’t, AFL-CIO v. American Petroleum Inst., 48 U.S.L.W. 5022 (U.S. July 2, 1980) (Benzene Standard) ; American Iron & Steel Inst. v. OSHA, 577 F.2d 825 (8d Cir. 1978), cert. granted, 48 U.S.L.W. 3855 (U.S. July 2, 1980), cert. dismissed, 49 U.S.L.W. 3145 (U.S. Sept. 10, 1980) (Coke Oven Emissions Standard) ; Society of the Plastics Indus., Inc. v. OSHA, 509 F.2d 1301 (2d Cir.), cert. denied, 421 U.S. 992 (1975) (Vinyl Chloride Standard); Synthetic Organic Chem. Mfrs. Ass’n v. Brennan, 506 F.2d 385 (3d Cir. 1974), cert. denied, 423 U.S. 830 (1975) (standard for 14 carcinogenic chemicals) ; Industrial Union Dep't, AFL-CIO v. Hodgson, 499 F.2d 467 (D.C. Cir. 1974) (As- bestos Dust Standard); Dry Color Mfrs. Ass’n v. Department of Labor, 486 F.2d 98 (3d Cir. 1973) (standard for 14 carcinogenic chemicals). 3 See 43 Fed. Reg. at 27354, col. 2; 41 Fed. Reg. at 56500, col. 3. Recent studies indicate that from 50-75% of those individuals re- porting any symptoms of byssinosis at all are classified in Grade 1/2. Pet. App. at 117, 125, 129. 5 “nebulous, transient feelings’;‘+ a person exhibiting Grade 44 symptoms is not sick, impaired or disabled.® Grade 1 byssinosis is characterized by chest tightness and/or breathlessness occurring only on Mondays after an absence from work of approximately two days. Grade 2 symptoms involve chest tightness and/or breathless- ness on Mondays and other davs. Like Grade 1%, Grades 1 and 2 are generally considered to be reversible.* Grade 3 byssinosis is characterized “by grade two Symptoms accompanied by evidence of permanent in- capacity from diminished effort tolerance and/or reduced ventilatory capacity.” Id. Although respected medical opinion suggests that even Grade 3 byssinosis may be reversible,’ other studies conclude that, in most cases, Grade 3 is not reversible. It is Grade 3 byssinosis—the advanced, chronic variety—that OSHA has described as being irreversible and disabling, constituting a form of chronic obstructive pulmonary disease. 43 Fed. Reg. at 27352. In sum, the term “byssinosis” standing by itself is a very imprecise description of health conditions that vary from a mild, reversible, nondisabling symptom to what in some instances may become a chronic, often irreversi- ble, and disabling disease. Although it is not known what component of cotton dust causes byssinosis, there is gen- eral agreement that some workers exposed to high enough concentrations of cotton dust will develop some symptoms 4Ex. 41 (Dr. Russell Harley); see also Tr. 1447 (Dr. W.K.C. Morgan); Tr. 1515 (Dr. Russell Harley); Tr. 813-14 (Dr. El Batawi). 5 Tr. 1779 (Dr. Charles F. Martin); Pet. App. at 157, 161 (Dr. Jack W. Whitworth). 641 Fed. Reg. at 56500; Tezxas Independent Ginners Ass’n v. Marshall, Nos. 78-2663, et al., slip op. at 1212 (5th Cir. Nov. 14, 1980) (copies of this recent opinion have been lodged with the Court). 7 Ex. 41 at 8 (Dr. Russell Harley). 6 of Grade % byssinosis and that, if exposure continues for long enough, the condition could progress, over time, to the more severe chronic and disabling stage. However, the overwhelming majority of persons.exposed to cotton dust do not exhibit any symptoms of byssinosis at all even after long exposure,® and a worker who has devel- oped the early mild symptoms of byssinosis will revert to a normal state if removed from exposure.” At the time that Congress was considering the OSH Act in 1970, concern was expressed about the occurrence of the disabling form of byssinosis among workers in certain operations within the textile industry."° However, Congress did not indicate concern about the mild, reversi- ble and nondisabling symptoms that characterize the lower grades of byssinosis. Moreover, the data before Congress in 1970 reflected experience with byssinosis during the early and mid-1960s when there were vir- tually no medical surveillance or other protective pro- grams in effect. As discussed below, medical surveillance (accompanied by respirator use and selective employee transfer) is an effective means of reducing the incidence of the chronic and disabling form of byssinosis. Since no such programs were operative in the 1960s (when cotton dust exposures were almost certainly higher than at present) ,'' experience from that period has very little 8 See, e.g., Pet. App. at 128a, 125; see also Texas Independent Ginners Ass’n v. Marshall, Nos. 78-2668, et al., slip op. at 1212. ® See, e.g., 48 Fed. Reg. at 27354, col. 3. 10S. Rep. No. 91-1282, 91st Cong., 2d Sess. 3 (1970), reprinted in Senate Subcomm. on Labor of the Comm. on Labor and Pub. Welfare, 92d Cong., 1st Sess., Legislative History of the Occupa- tional Safety and Health Act of 1970, at 143 (Comm. Print 1971) {hereinafter “Leg. Hist.’’]. 11 Until the latter part of the 1960s, neither the government nor any private standard-setting organization had placed a limit upon exposure to cotton dust. 43 Fed. Reg. at 27351, col. 1. As OSHA’s witness, Dr. Arend Bouhuys, observed, byssinosis has 7 relevance to whether OSHA’s Standard is necessary to prevent or reduce a significant health risk today. B. The Proceedings Before OSHA Within the textile manufacturing industry, cotton is handled in a series of successive operations, beginning with “opening” through “roving,” ” continuing with “spinning” through “warping,” and concluding in “slashing” and “weaving.” Before 1971, no mandatory exposure limit was gener- ally applicable to these operations, although a permissible exposure limit (“PEL”) of 1.0 milligram of total cotton dust (raw) per cubic meter of air (1.0 mg/m*) was established for federal contractors under the Walsh- Healey Act. 41 U.S.C. §§ 35-45 (1976 & Supp. III 1979). In 1971, OSHA adopted this limit as an established Fed- eral standard under Section 6(a) of the Act, 29 U.S.C. § 655(a); consequently, the PEL of 1 mg/m* applied to all operations in the textile manufacturing industry. See 43 Fed. Reg. at 27351, col. 1. That standard, how- ever, did not require medical surveillance. In December 1976, OSHA published a proposal to re- vise the standard by reducing the PEL, in all segments of the cotton industry, to 0.2 mg/m® of cotton dust as measured by a vertical elutriator. Testimony and been perceived as an important health hazard for United States cotton textile workers only since 1967. See Ex. 11 at 6. 12In the operations of “opening,” “picking,” “carding,” “draw- ing,” and “roving,” raw cotton is cleaned and prepared for spinning into yarn. 18 Tn the operations of “spinning,” “twisting,” “winding,” “spool- ing,” and “warping,” the prepared cotton is transformed into yarn and packaged for weaving or other processing. In “slashing” and “weaving,” the yarn is manufactured into a woven fabric. 1841 Fed. Reg. 56498 (1976). For the operations of “opening,” “picking,” and “carding,” there is some evidence that a limit of 8 written comments in the proceeding focused on the health need for the severe standard that OSHA had proposed, as well as on the standard’s technological feasibility and the economic impact that it would have on the affected industries and the economy as a whole. What follows is a brief description of the principal evidence on health effects and economic impact." 1. Health Effects Evidence For its health effects determination, OSHA placed principal reliance upon the so-called “Merchant Study” concluded in April 1971.17 Among textile workers not subject to a medical surveillance or respiratory protec- tion program, Merchant claimed to have found that the predicted prevalence rate of byssinosis (for Grades 14, 1 and 2 on an undifferentiated basis) was approximately 7 percent at exposures of 0.1 mg/m’, 13 percent at ex- posures of 0.2 mg/m* and 26 percent at exposures of 0.5 mg/m* in the operations of opening through warping."® Although these findings were subject to serious ques- 0.2 mg/m of vertical elutriated cotton dust may be approximately equivalent to a limit of 0.4 mg/m? of total cotton dust. Tr. 2396-97 (Hovan Hocutt). No such correlation between “vertical elutriated” and “total” dust has been established for any other operations. /d.; Tr. 2022-25 (Dr. John Neefus). 16 The question of the Standard’s technological feasibility per se is not before the Court at this time, and need not be considered except to the extent that it relates to the economic impact of the Standard. 17 See 43 Fed. Reg. at 27355; Exs. 38d, 6, Nos. 44 & 51, 1843 Fed. Reg. at 27355; Ex. 6, No. 51, Table 4, Charts for Groups. The “prevalence rate” is intended to relate the current exposure levels to the percentage of employees with specified symp- toms. It does not purport to show that the current exposures caused those symptoms. See Ex. 38d at 108; Ex. 26 at 2 9 tion, the significance of the findings, even when ac- cepted at face value, is easily overstated. First, the Merchant Study does not establish a dose- response relationship showing the percentage probability that workers exposed to specified levels of cotton dust will develop byssinosis.*° Rather, as Merchant admitted, his study purported to show only the prevalence rate of byssinosis (Grades %, 1 and 2) at a particular time among workers who already had reacted to unknown (and presumably higher) exposure levels in the past.?? Second, more than 50 percent of those who indicated any symptoms of byssinosis at all in the Merchant Study were recorded as Grade 14.” Finally, since none of the workers in Merchant’s study had participated in a medi- cal surveillance program, Merchant’s findings reveal nothing about how the prevalence of byssinosis (par- ticularly the higher grades) would vary from one ex- posure level to another, if the various exposure limits were combined with a program of medical surveillance, respirator use, and employee transfer in appropriate cases. Thus, even taken at face value, the Merchant Study 19The Merchant Study was shown to contain errors in data selection, methodology, and analysis that led to the prediction of “the highest possible value of predicted prevalence at any hypothe- sized dust level.” E.g., Ex. 55 at 2-9 (Dr. Moon W. Suh). Dr. Merchant himself testified: Clearly no inference from our study can be drawn as to the prevalence of byssinosis in the entire U.S. textile industry .... Tr. 1032. Even OSHA admitted that there were “some weaknesses in the Merchant data . . . and, consequently, .. . weakness of some correlation coefficients.” 48 Fed. Reg. at 27357, col. 3. 20 See Ex. 38d at 103. 21 Ex. 26 at 2. 22 Ex. 6, No. 51, Table 4, Groups 3 & 6; Tr. 1162-63 (Dr. Merchant). 10 must be interpreted very cautiously; it cannot blithely be applied to circumstances that differ substantially from those studied by Merchant. The Merchant Study cannot be taken at face value, however. Not only did it contain methodological flaws,” but also it was inconsistent with a number of more recent studies of larger worker populations, which indi- cated that the actual prevalence of byssinosis in the textile manufacturing industry was far below the levels sug- gested by Merchant. For example, a 1976 study conducted by Dr. Hans Weill, an OSHA witness at the hearing, found that among workers who had participated in a medical sur- veillance program, the prevalence of byssinosis was markedly low at exposure levels above 0.2 mg/m* and that even at levels of 0.5 mg/m? and above, it was sub- stantially lower than the rate reported by Merchant.” Dr. Weill’s findings were consistent with those of Dr. W.K.C. Morgan, who found that only 0.4% of the workers whom he studied in 1976 (workers who also had been subject to medical surveillance) reported any symp- toms of byssinosis—despite the fact that they were ex- posed to cotton dust levels ranging up to more than 1.0 mg/m*.** Studies reported by individual textile com- panies were in accord with the Weill and Morgan find- ings.”° 23 See note 19 supra. *4 Ex. 183A at 14-15, 44-45; Tr. 294-95. 25 Exs. 39, 39a, 40, 43, 47b. 26 Thus, studies of more than 10,000 textile workers by Burling- ton Industries revealed a prevalence rate (Grades 1/2, 1 and 2) of 4.5% in 1971 and 1.0% in 1976. Pet. App. at 108, 172-75 (Drs. Harold R. Imbus & Moon W. Suh). A 1973 study of more than 9,000 West Point-Pepperell textile workers exposed to dust levels ranging from less than 0.2 mg/m? to more than 1.0 mg/m* showed that only 1.34% of the workers reported byssinosis symptoms of any grade, and that 70.7% of these were Grade 1/2, with only two 11 These studies are significant in several respects. For one thing, taken as a group, they provide persuasive evidence that the byssinosis prevalence findings reported in the Merchant Study overstate the predicted level of byssinosis among textile workers.** This point takes on added significance in light of the fact that these other studies are more recent, more extensive, and thus more indicative of the current state of affairs in the industry. More importantly, these studies demonstrated that a medical surveillance program (combined with the se- lective use of respirators and transfer of employees in appropriate cases) is an effective means to achieve a very significant reduction in the occurrence of byssinosis (particularly the more serious grades) among textile workers. Indeed, OSHA itself recognized this point. See 43 Fed. Reg. at 27359, col. 3. R 2. The ATMI Alternative Based upon this experience, ATMI and individual tex- tile companies urged OSHA to consider and adopt an alternative to the OSHA proposal. The alternative pro- posed less stringent exposure limits than OSHA’s stand- ard, combined with an active and effective medical sur- veillance program that would identify employees having the mild reversible symptoms of byssinosis.2® Such em- ployees could then be protected against any risk of dis- workers (in the entire study population of more than 9,000) exhibiting Grade 2 symptoms. Pet. App. at 124-125, 157-161 (Dr. Jack W. Whitworth). Similarly, a 1973 study of more than 6,600 Cone Mills textile workers found that approximately 3% of the workers reported symptoms of Grades 1/2 and 1 (more than 70% of these being Grade 1/2), with no workers reporting symptoms of Grade 2 byssinosis. Pet. App. at 126-129 (Dr. Charles F. Martin). 27 See Comments of the Council on Wage and Price Stability, Ex. 111 at 16-18. °8 See Ex. 160 at 86-89. 12 abling or irreversible lung damage by selective use of respirators or transfer to low exposure areas, as med- ically appropriate for the individual concerned. On the basis of the experience that had accumulated since 1970, ATMI argued that this alternative would provide effec- tive and cost-efficient protection for textile workers against any significant risk of material lung damage from exposure to cotton dust. 3. Evidence on the Economic Impact of the Standard Two principal analyses addressed the issue of eco nomic impact. One, the Hocutt-Thomas study, was pre- sented by the principal suppliers of dust control equip- ment to the textile industry. It concluded that textile manufacturers would incur capital costs of $550 million to achieve much less stringent exposure limits than those established in OSHA’s Standard. 43 Fed. Reg. at 27372, 27380.% The second economic analysis, prepared by OSHA’s outside contractor, Research Triangle Institute (“RTI”), estimated that the textile industry would incur capital costs of $1.1 billion in 1974 dollars to achieve the Standard’s exposure limit of 0.2 mg/m? in the operations of opening through warping alone.” RTI expressed considerable concern about the ability of many textile firms to finance the capital expenditures that it estimated would be required to reach the PEL set by the standard. According to RTI, 2°The PELs for which the Hocutt-Thomas cost estimate was developed were 0.2 mg/m* in opening through roving, 0.5 mg/m* in spinning through warping, and 1.0 mg/m* in slashing and weav- ing. Pet. App. at 152-56. 3° This consisted of $984.4 million in opening through spinning, see Pet. App. at 189; 48 Fed. Reg. at 27371, and $127.7 million in winding through warping, see Ex. 6, No. 76 at V-28. RTI esti- mated the total capital cost of the proposed standard for all indus- try segments to be $2.7 billion. Pet. App. at 139. 13 unless the compliance capital investments are spread over a substantial number of years [RTI envisioned an eight-year period] all the companies engaged heavily in cotton yarn production will face severe difficulty in financing compliance capital for the 0.2 mg/m* exposure limits and the difficulty will be more intense for smaller companies." In RTI’s view, implementation of the Standard would be difficult for many firms and would result in adjust- ments in the structure of the textile manufacturing industry.” Although RTI did not believe that the se- vere impact of the Standard would threaten the existence of the cotton textile industry as a whole, it reached this conclusion only by adopting (without any supporting analysis) what it characterized as the “extreme assump- tion” that the costs of the Standard could be passed on fully in product prices. Pet. App. at 144; 48 Fed. Reg. at 27370. RTI noted that if the opposite assumption of no price increases were made, all industry sectors would have “negative ‘profits after taxes’” under a 0.2 mg/m? standard. Pet. App. at 148. Other witnesses echoed this concern about the ability of the industry to finance the required capital expenditures.” 31 Ex. 6, No. 76 at VI-48. These difficulties would be occasioned because, as RTI noted, profits in the textile industry have been low since the mid-1950s, averaging about one-half of the profit percent- age for United States manufacturers as a whole. Jd. at C-27. This poor profit performance, RTI observed, had led to a decline of nearly 37% in textile industry investment in plant and equipment between 1947 and 1974, a period in which plant and equipment expenditures by U.S. manufacturers as a whole had steadily increased. Id. at C-21, Table C-12. 82 Tr. 561-63 (Dr. LeSourd). 33 Thus, John P. Figh of the Chase Manhattan Bank testified that the standard would threaten the viability of a number of firms in the textile industry, particularly the enum firms. Pet. App. at 191-95. 14 C. OSHA’s Adoption of the Standard On June 19, 1978, OSHA adopted the final Standard. The Standard establishes a PEL of 0.2 mg/m? of cotton dust as measured by a vertical elutriator in the textile manufacturing operations of opening through warping (yarn manufacturing) and a PEL of 0.75 mg/m? in slash- | ing and weaving.” In all cases, the applicable PEL is to be achieved solely through the use of engineering and work practice controls, and such controls are to be implemented fully within four years.* The Standard also imposes a host of ancillary requirements, including exposure monitor- ing, medical surveillance, work practice restrictions and provisions requiring employers to transfer and to guarantee the earnings, rights and benefits of employees who are unable to wear respirators in circumstances where the Standard requires their use.*® OSHA took the same approach to setting exposure limits for cotton dust as it has in setting standards for carcinogens, including benzene. Finding that for cotton dust, as for carcinogens, there is no absolutely safe level of exposure, OSHA concluded that the Act requires it to set the exposure limit at a level that will “assure maximum benefit . . . constrained only by the limits of feasibility.” 43 Fed. Reg. at 27378, col. 3. As inter- 34 29 C.F.R. § 1910.1043(c) (1979), Pet. App. at 3. In other cov- ered segments of the cotton handling and processing industry, the PEL is 0.5 mg/m*. Id. 85 29 C.F.R. § 1910.1043(e), (g), (m) (1979), as amended at 45 Fed. Reg. 12416 (1980), Pet. App. at 5-6, 9-10, 17-18. 36 See, e.g., 29 C.F.R. § 1910.1043(d), (f), (g), (h) (1979), Pet. App. at 3-13. 37 43 Fed, Reg. 5918 (1978), set aside, American Petroleum Inst. v. OSHA, 581 F.2d 493 (5th Cir. 1978), aff’d sub nom. Industrial Union Dep’t, AFL-CIO v. American Petroleum Inst., 48 U.S.L.W. 5022 (U.S. July 2, 1980). 15 preted by OSHA, this means that the exposure limit must be set at the lowest level that is technologica!ly achievable, as long as the costs of implementing such technology will not threaten the existence of the affected industry as a whole. 43 Fed. Reg. at 27378. Moreover, as in the case of benzene, OSHA specifically rejected the contention that it must assess the benefits of the Cotton Dust Standard in light of its costs or con- sider the relative health effectiveness and costs of al- ternative approaches.** For cotton dust, as for benzene, once OSHA had determined that the Standard was “with- in the financial capability of the covered industries,” it considered its economic analysis to be complete. Echoing word for word the approach that it took in the Benzene proceeding, the agency declared “that the benefits of the proposed standard are likely to be appreciable,” and asserted that, having made such a determination, OSHA is not obligated to-carry out further exer- cises toward more precise calculations of benefit which would: not significantly clarify the ultimate decision.” OSHA acknowledged that the early, mild stages of byssinosis are reversible, 43 Fed. Reg. at 27352, 27354, that these early, nondisabling effects progress to the chronic state only with further exposure, id. at 27354, and that “a properly managed [medical surveillance] program should identify affected workers well before the onset of chronic obstructive pulmonary disease,” id. at 27359, col. 3. Nonetheless, OSHA summarily rejected 8843 Fed. Reg. at 27379, col. 3 (Cotton Dust); 43 Fed. Reg. at 5941, col. 2 (Benzene). 3943 Fed. Reg. at 27379, col. 8 (Cotton Dust); 43 Fed. Reg. at 5941, col. 2 (Benzene). 40 43 Fed. Reg. at 27379, col. 3 (Cotton Dust); 43 Fed. Reg. at 5941, col. 2 (Benzene). 16 ATMI’s proposed alternative, apparently because of its belief (1) that even the mild, nondisabling and reversible symptoms of byssinosis constitute “material impairment of health” within the meaning of the OSH Act, and (2) that the most stringent approach (providing the maxi- mum degree of protection possible against even mild, reversible symptoms) must always be adopted in any single health standard, so long as compliance with that one standard will not threaten the financial viability of the entire industry, 43 Fed. Reg. at 27378, col. 2.* OSHA’s treatment of the Standard’s cost impact was enigmatic. As noted above, the only estimate of the capi- tal cost of a 0.2 mg/m® standard in yarn manufactur- ing operations was the $1.1 billion figure presented by OSHA’s contractor, RTI. However, OSHA rejected the RTI cost estimate out of hand. Instead, OSHA decided that an appropriate estimate of the capital cost of the Standard for the textile industr; would be the $550 million estimate developed by Hocutt and Thomas for a less stringent standard with higher exposure limits. OSHA was unable to demonstrate (and, indeed, made no real effort to demonstrate) how the Hocutt-Thomas fig- ures could be transmuted into an accurate estimate of the costs of an entirely different standard having much more stringent exposure limits. Relying on RTI’s “extreme assumption” that the com- pliance costs of the Standard would be passed on in prod- uct prices, OSHA then concluded that the Standard would be economically feasible, since capital costs of $550 mil- lion would not threaten the economic viability of the *1 In light of the approach followed by OSHA, it is not surprising that the preamble to the Cotton Dust Standard, like the preamble to the Benzene Standard, makes no reference whatsoever to Section 3(8) of the OSH Act and no finding that the Standard is “reason- ably necessary or appropriate to provide safe or healthful employ- ment and places of employment.” 17 industry as a whole. 43 Fed. Reg. at 27377-78; Pet. App. at 144. OSHA reached this conclusion despite its recognition that some employers would be forced to shut down and that the financial impact of the Standard would be felt more acutely by smaller firms, thus enhanc- ing a trend toward concentration within the industry. 43 Fed. Reg. at 27378, col. 2. D. The Decision of the Court Below In all significant respects, the court below upheld the approach to setting occupational health standards that OSHA had followed in the Cotton Dust proceeding. The court agreed that, under the Act, OSHA must adopt the standard that provides the maximum degree of protection against any risk of health impairment, as long as that standard is technologically feasible and not financially ruinous to the industry as a whole. Pet. App. at 53-54, 617 F.2d at 655. According to the court, nothing in the OSH Act requires any further assessment of the health protection benefits of the Standard in light of its costs, Pet. App. at 71, 617 F.2d at 663, or of the relative risk. reduction that might be achieved by less costly alternatives. The court acknowledged that the early symptoms of byssinosis are reversible, Pet. App. at 53, 617 F.2d at 655, and did not find that these symptoms constitute “material impairment” of health within the meaning of the Act. Pet. App. at 50-51 n.83, 617 F.2d at 654 n.83. Nonetheless, believing that OSHA’s more stringent (and far more costly) approach would provide some additional protection beyond ATMI’s proposed alternative, the court declined to require OSHA to attempt to assess the extent of this additional protection or to consider the benefit of such additional protection in light of the costs. Id, 18 The only economic question that the court below deemed relevant was whether the Standard would have such a devastating cost impact as to make financial viability impossible for the industry as a whole, effectively “put- [ting] the industry out of business.” Pet. App. at 69, 617 F.2d at 662.° The fact that a number of plants and small companies might be forced to close, thereby chang- ing the market structure of the industry, did not “sig- nify economic infeasibility’ in the court’s view, Pet. App. 67-68 & n.148, 617 F.2d at 661-62 & n.148, par- ticularly since OSHA had found an existing trend toward increasing concentration in the industry.** Moreover, in finding the Standard to be economically feasible, the court, like OSHA, relied on RTI’s “extreme assumption” that the costs of compliance would be passed on in prod- uct prices and did not require OSHA to present any evi- dence to support that assumption. Pet. App. at 67 n.145, 617 F.2d at 661 n.145. Finally, the court did not require OSHA to base its economic impact analysis upon any study that purported to assess the costs of the Standard actually adopted. Rather, the court sustained OSHA’s decision to use the Hocutt-Thomas cost estimate, even though that estimate addressed the costs of a much less stringent standard. The court thus permitted the agency to choose between two cost estimates, both of which it had found to be inac- curate, and to adopt a Standard the costs of which were not assessed in any study that OSHA believed to be valid. 42 At the same time, the court conceded that “the actual test for economic feasibility has yet to be fully developed by the courts.” Pet. App. at 67, 617 F.2d at 661. *8 Curiously, the court held that OSHA had not established the economic feasibility of its Standard as applied to the cottonseed oil industry, despite the fact that there too OSHA had pointed to similar factors independent of the Standard that would contribute to anticipated shutdowns in the industry. Pet. App. at 90-91, 617 F.2d at 671-72. 19 SUMMARY OF ARGUMENT I, This case calls upon the Court to decide what deter- minations OSHA must make regarding the economic impact of its health standards. OSHA’s approach treats the costs of its standards as being totally irrelevant until they reach some undefined point at which the standard, if implemented, would cause the destruction of an entire industry. This interpretation of the Act leaves OSHA with virtually unfettered authority to impose enormous costs upon American employers. Only a few months ago, this Court rejected OSHA’s claim to such extravagant authority. Industrial Union Department, AFL-CIO v. American Petroleum Institute, 48 U.S.L.W. 5022 (U.S. July 2, 1980) [hereinafter “Benzene’’]. In that case, OSHA claimed that it need take no ac- count of the magnitude of the health risks being ad- dressed. In this case, OSHA asserts that it need take no account of costs short of the point where they threaten the ruination of an entire industry. If OSHA’s position were accepted, the agency would enjoy precisely the “un- precedented power over American industry,” 48 U.S.L.W. at 5032, that this Court held to be contrary to Congress’ intent in Benzene. The present case illustrates the inadequacy of OSHA’s approach. OSHA concluded that the Cotton Dust Stand- ard would be economically feasible without relying on any study purporting to assess its costs. Instead, OSHA took the estimated cost figure of a less severe standard and, without warrant, treated it as the cost of ‘its own far more stringent Standard. Relying on an unsupported “extreme assumption” regarding price-demand elasticity, it then concluded that a standard having such costs would not threaten the financial viability of the industry as a whole. The economic limitation that OSHA purported to apply here is, in reality, no limitation at all. The qualified 20 and pragmatic language of the OSH Act reflects a con- gressional expectation that standard setting under the Act would be a more balanced exercise, in which the cost impact of regulatory action would play a more prominent role. Because the position of OSHA and the court below shows an utter disregard for the more prag- matic and balanced approach that Congress intended, and because it raises serious questions of an overbroad delegation of legislative authority, it must be rejected. The judgment below also must be reversed because OSHA made no attempt to ascertain whether the bene- fits of its Standard (in the form of a claimed reduction in health risk) bear a reasonable relationship to the attendant costs. Both in its overall objectives and its more specific directives, the Act reflects a congressional intent that OSHA spend society’s limited industrial hy- giene resources wisely, so that employees can be pro- tected, as far as possible, against the wide variety of health and safety risks in the workplace. This goal cannot be achieved if vast expenditures are mandated to achieve negligible reductions in risk. Consequently, OSHA must assess the significance of the reduction in risk expected from its standards in light of the costs of achieving that reduction. The failure of OSHA to perform such an assessment is particularly egregious in the present case, because the petitioners had presented a far less costly alternative that promised to provide workers with health protection com- parable to that expected from OSHA’s far more burden- some Standard. Cotton dust, unlike benzene, is not a carcinogen, and byssinosis, the health impairment with which it is associated, is detectable in its early stages, when it is nondisabling and reversible. Consequently, the only material health impairment involved in the pres- ent case is much more easily averted than the malignant condition that was involved in Benzene. In these circum- stances, OSHA’s failure to perform a careful and re- 21 sponsible assessment of health protection benefits in light of the respective costs of the two alternatives requires reversal. II. The court below also erred in sustaining a provision of the Standard that requires employers to transfer and guarantee the wages of an employee who for any reason is unable to wear a respirator in circumstances where the Standard requires its use. This wage guarantee obliga- tion—which is not tied to the existence of any health im- pairment—exceeds the authority that Congress conferred upon OSHA. The statute contains no express grant of such authority, and the legislative history of the OSH Act and provisions of related legislation demonstrate that Congress did not intend to confer such authority by implication. ARGUMENT I. THE COTTON DUST STANDARD IS PREDICATED UPON LEGALLY INSUFFICIENT CRITERIA FOR ASSESSING THE ANTICIPATED ECONOMIC IM- PACT AND REDUCTION IN RISK THAT THE STANDARD IS EXPECTED TO PRODUCE. The decision below permits OSHA to regulate in a manner that is essentially oblivious to considerations of cost, except in the very most extreme circumstances. The hallmark of OSHA’s approach is a financial impact limi- tation that is wholly illusory and provides no constraint whatever on agency discretion. In Part I.A below, we show that this approach is both constitutionally question- able and at odds with the intent of Congress. OSHA is required to make a responsible prediction, supported by substantial evidence, of the economic impact that its standard will have and to explain why it believes that a standard having such an impact is feasible. 22 Moreover, as we show in Part I.B below, OSHA must responsibly assess the costs and risk reduction benefits of its standards in order to determine that employers are not being forced to incur enormous costs that bear no reasonable relationship to the health benefits expected to be achieved. A. The Decision Below Mistakenly Permits OSHA To Adopt Health Standards That Are Subject to No Meaningful or Effective Cost-Related Limitations. As its actions in the Benzene and Cotton Dust pro- ceedings demonstrate, OSHA appears to believe that Con- gress left it free to impose enormously costly require- ments in the context of a single health standard, as long as compliance with that one standard would not ruin an entire industry.** This position reflects OSHA’s belief that Section 3(8) of the Act, which provides that occupa- tional health standards should be “reasonably necessary or appropriate to provide safe or healthful employment,” imposes no substantive restrictions on its standard-setting activities, and that Section 6(b) (5) mandates the crea- tion of absolutely risk-free workplaces, subject only to technological limitations and the most attenuated eco- nomic constraints. So construed, the Act imposes no meaningful or effec- tive cost-related limitations upon OSHA’s standard- setting authority. In such circumstances, industry can- not know what OSHA may do,* and effective judicial review is impossible.** In part because such an uncon- fined delegation of legislative power would raise constitu- “4 See Benzene, 48 U.S.L.W. at 5030; Pet. App. at 67-69, 617 F.2d at 661-62. 45 See United States v. Rock Royal Co-Op., Inc., 307 U.S. 533, 574 (1939); Panama Refining Co. v. Ryan, 293 U.S. 888, 415 (1935). : 46 American Power & Light Co. v. SEC, 329 U.S. 90, 106 (1946). 23 tional problems,** this Court recently rejected OSHA’s approach to standard setting in Benzene. As the plurality stated : {W]e think it is clear that the statute was not de- signed to require employers to provide absolutely risk-free workplaces wherever it is technologically feasible to do so, so long as the cost is not great enough to destroy an entire industry. 48 U.S.L.W. at 5031. The plurality continued: “[I]t is unreasonable to assume that Congress intended to give the Secretary the unprecedented power over American industry that would result from the Government’s view of §§ 3(8) and 6(b) (5).” Td. at 5032. In Benzene, the plurality limited OSHA’s standard- setting authority to situations in which a “significant risk of material health impairment” would exist in the absence of the standard. Jd. at 5030. But that threshold requirement standing alone will do little, if anything, to curb OSHA’s power over American employers if the only cost-based restraint on OSHA’s authority is that any single standard must not destroy an entire industry. In effect, the “sweeping delegation of legislative power” that the Benzene plurality avoided with respect to health risks will be presented with respect to costs.** The Act can and should be construed to avoid this result.*® 47 See U.S. Const. art. I, $1; A.L.A. Schechter Poultry Corp. v. United States, 295 U.S. 495 (1935). *8In such circumstances, OSHA would be making the “crucial policy choices” that the Constitution reserves for Congress. The Supreme Court, 1979 Term, 94 Harv. L. Rev. 75, 242, 249 (1980). 49 See Benzene, 48 U.S.L.W. at 5032. Where, as here, adminis- trative regulation presents a “sharp break with our traditions,” J. Freedman, Crisis and Legitimacy 84-85 (1978), this Court has strictly scrutinized the statutory grant claimed to authorize the action in question and has upheld the action only if authorized 24 Just as OSHA was required by the Benzene case to show that its standards address a significant health risk, the agency must also be required to identify and assess the economic impact of its standards on the basis of criteria that set some meaningful and effective limits on its standard-setting authority. As we now show, the economic feasibility test that OSHA and the court be- low applied here imposes no such limits on the agency’s authority and departs from what Congress intended. Ac- cordingly, the judgment of the court below must be reversed. 1. OSHA Failed To Make a Responsible and Sup- portable Estimate of the Costs of Its Standard or To Apply Any Meaningful Criteria in Assess- ing the Standard’s Economic Impact. A fundamental and dispositive consideration in this case is that OSHA did not support the Cotton Dust Standard with any study or analysis purporting to assess its costs. One cost analysis, a $1.1 billion estimate pre- pared by OSHA’s contractor, RTI, was in the record, but OSHA rejected it out of hand.” Apparently OSHA be- clearly and unmistakably. See, e.g., Kent v. Dulles, 357 U.S. 116, 128-30 (1958); National Cable Television Ass’n, Inc. v. United States, 415 U.S. 386, 342 (1974); Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667, 1697 (1975) (advocating “a policy of narrow construction of statutory dele- gations’’). 5° OSHA’s principal objection was its contention that approxi- mately 30% of the capital costs estimated by RTI were for con- trolling dust from equipment processing synthetic fibers in cotton blend mills. 43 Fed. Reg. at 27370. In fact, the Standard requires controls on such equipment, as long as that equipment is located in a work area where cotton is handled or processed, even though on other machines. See 29 C.F.R. § 1910.1043 (b) (1979), Pet. App. at 2 (definition of “cotton dust” and “lint-free respirable cotton dust’’). OSHA also was disturbed by RTI’s assumption that all companies were presently in compliance with the previous standard. 43 Fed. Reg. at 27370. In fact, while some firms were already below the limit set by the previous standard, many others had not yet achieved that limit. Jd. The record indicated that the industry would have 25 lieved that its Standard would be prohibitively expen- sive (and thus not economically feasible) if it cost as much as RTI estimated.*' But the RTI estimate was the only analysis directed to the exposure limits OSHA ac- tually adopted. Even OSHA recognized that if the RTI estimate were rejected, something else was needed. To fill the gap, OSHA treated the Hocutt-Thomas estimate of a far less stringent and less costly proposal as an appropriate estimate of the capital cost of the Standard it ultimately adopted. This treatment, accomplished in two steps, is utterly without justification. First, OSHA announced its belief that the Hocutt- Thomas capital’ cost estimates of $550 million for ex- posure limits substantially above those ultimately adopted in the Standard were “overstatements.” But the agency provided no rational explanation for this conclusion. Thus, OSHA observed that compliance with the final standard would be less costly than it would have been for the proposed standard. 43 Fed. Reg. at 27372, col. 3. While accurate, this observation sheds no light at all on why the Hocutt-Thomas capital cost estimate is likely to have been overstated, since the Hocutt-Thomas estimate did not relate to either the proposed or final standard. The balance of OSHA’s rationale is equally to incur more than $143 million in capital costs (above and beyond what RTI had estimated) just to comply with the previous stand- ard. Ex. 69 at 3-6 (L. K. Fitzgerald). *1In fact, the RTI estimate of $1.1 billion was, if anything, unduly low, since it was presented in 1974 dollars and should have been increased by 40% (to $1.54 billion) to account for inflation in machinery and equipment costs between 1974 and June 1978 when the Standard was adopted. This increase is re- flected in the Wholesale Price Index for Machinery and Equipment, which rose from 139.4 to 195.1 between 1974 and June 1978. Monthly Labor Review 114 (March 1975) ; id. at 94 (October 1978). RTI considered this Index to be a reliable guide to price increases for textile plants and equipment. See Ex. 6, No. 76 at C-20, Table C-12. 26 unpersuasive. The fact is that OSHA was engaging in the sheerest speculation when it characterized the Hocutt- Thomas capital cost estimate as an “overstatement.” But only at this point did the real speculation begin. For OSHA then proceeded to announce that it would treat the Hocutt-Thomas estimate of the capital costs of a much less stringent standard as an appropriate esti- mate of the capital costs required to achieve the much more stringent limits required under its Standard. 43 Fed. Reg. at 27373. There is a certain elegance and sim- plicity in this leap of logic; however, there is absolutely no basis to support it. 52 Thus OSHA speculated that Hocutt and Thomas “may have included some equipment which is used exclusively for syn- thetics.” 43 Fed. Reg. at 27372, col. 3 (emphasis added). Apart from the fact that this was merely a surmise on ©SHA’s part, the fact is that the Standard as adopted requires controls on most such equipment. See note 50 supra. OSHA also predicted that new production equipment would reduce the number of ma- chines on which the Hocutt-Thomas estimate was based. No record support was cited for this prediction. And, even if it were true, the cost of such equipment, much of which would not be pur- chased in the absence of the Standard, would have to be added to the cost of compliance. In addition, OSHA complained that Hocutt and Thomas had not accounted for technological improvements in the four-year compliance period. No citation to the record supported this statement; nor was “technology” defined. Moreover, this vague generalization was flatly refuted by Hocutt’s testimony that, for the next seven years, he saw little new dust control technology that would make compliance less expensive for the operations of spinning through weaving. Tr. 2388-93. Finally, OSHA stated that the Hocutt-Thomas estimate included some costs necessary to come into compliance with the pre-existing exposure limits. Such costs were properly included by Hocutt and Thomas, since they are part of the total costs that would have to be incurred to comply with a new standard. See United Steelworkers of Amer- tea, AFL-CIO-CLC v. Marshall, [1980] OSH Dec. (CCH) 30329, 30418, 30429, 8 OSH Cas. (BNA) 1810, 1902, 1918-14 (D.C. Cir. Aug. 15, 1980) (MacKinnon, J., dissenting); Comments of the Council on Wage and Price Stability, Ex. 111 at 10. °8 For example, Hocutt and Thomas estimated that the capital costs to achieve a 0.5 mg/m? limit in spinning would be only $17.8 27 What really occurred is quite simple. Confronted with one estimate (RTI) that demonstrated a lack of eco- nomic feasibility by any test and another (Hocutt- Thomas) that did not address the Standard actually adopted, OSHA proceeded to select the Hocutt-Thomas figure as being the less objectionable of the two, since it was more compatible with the result OSHA wished to achieve. 43 Fed. Reg. at 27373. But, even if Hocutt and Thomas did overestimate the costs of the less string- ent proposal that they had analyzed, only by the most remarkable coincidence would the amount of that over- estimate be equal to the additional costs required to at- tain the far more stringent limits of the Standard OSHA actually adopted. Such an approach to the determina- tion of costs cannot be sustained, for it dispenses with substantial evidence and amounts to regulation by spec- ulation.** The OSH Act requires something better.® million. See Pet. App. at 152-156 (Hovan Hocutt). By contrast, RTI estimated that achieving a 0.2 mg/m’ limit in spinning would cost $553.3 million. (This is derived by applying a process-by- process ratio to RTI’s final total capital cost estimate for the operations of opening through spinning. See Ex. 6, No. 76 at I-15, V-17.) Thus, the only estimates in the record indicated that the Hocutt-Thomas figures would have to be increased by at least $535 million if it were to be applied to achieving a 0.2 mg/m? limit in spinning alone. OSHA itself appeared to recognize that substantial costs beyond those estimated by Hocutt and Thomas would be required to achieve a 0.2 mg/m limit in spinning. Thus, OSHA stated that a “major effort” would be necessary to attain this limit and observed that its decision to set a higher exposure level in weaving would per- mit the industry to “focus more resources on spinning operations” in order to make this “major effort.” 43 Fed. Reg. at 27367, col. 2. 54 See Texas Independent Ginners Ass’n v. Marshall, Nos. 78-2663, et al., slip op. at 1216 & n.27; cf. National Wildlife Fed’n v. Andrus, 440 F. Supp. 1245, 1253 (D.D.C. 1977) (weighing of benefits and costs is impossible “where there is only speculation .. . concerning the impact upon the environment’). 55 29 U.S.C. § 655(f). See Benzene, 48 U.S.L.W. at 5034 (plural- ity opinion) ; American Petroleum Inst. v. OSHA, 581 F.2d at 497; 28 OSHA’s determination that the Standard would not threaten the economic viability of the industry as a whole was as flawed as its estimate of what the Standard would cost. In critical respects, this determination rested on what RTI characterized as the “extreme assumption” that the compliance costs of the Standard would be passed through in the price of cotton products.” RTI made this assumption (and OSHA and the court below accepted it)®’ without any support, analysis or explana- tion—even though, as RTI recognized, the Standard would reduce profits and return on investment in the textile industry to very low or negative levels if the in- creased costs could not be passed on in product prices. Pet. App. at 143. Industrial Union Dep’t, AFL-CIO v. Hodgson, 499 F.2d at 474. The extreme to which OSHA will go if permitted to regulate as it has in the present case is also well illustrated in the Lead Standard proceeding. 43 Fed. Reg. 52952 (1978); 43 Fed. Reg. 54354 (1978). In that case, OSHA proposed a PEL of 100 ug/m*, and the record contained several studies estimating the cost of achieving such a standard. However, the final standard estab- lished a PEL of only 50 ug/m?, and OSHA stated that compliance with the final standard “would not result in undue economic hard- . ship” to the affected industries. Jd. at 54496, col. 3. OSHA reached this conclusion even though it “did not undertake a formal analysis of cost of compliance with the 50 ug/m*? PEL,” and even though it admittedly lacked sufficient information to make a “meaning- ful quantification of cost” for any standard stricter than the one it had proposed. Jd. Instead, it simply utilized the compliance cost esti- mate prepared for the proposed 100 ug/m® standard as the basis for determining the economic feasibility of the 50 ug/m® standard that it actually adopted. On review, the D.C. Circuit held this to be a sufficient basis for establishing economic feasibility. United Steelworkers of America, AFL-CIO-CLC v. Marshall, [1980] OSH Dec. (CCH) at 30394-95, 30398-401, 30402-03, 8 OSH Cas. (BNA) at 1877-78, 1881-83, 1885-87. 56 See pages 13, 16-17 supra. 5743 Fed. Reg. at 27370; Pet. App. at 67 & n.145, 617 F.2d at 661 & n.1465. 29 In fact, there was little basis for RTI’s “extreme assumption.” To the contrary, the evidence of record suggested that competitive constraints (including inroads made by foreign competition) would preclude the passing on of full compliance costs in product prices,5® and neither RTI nor OSHA presented evidence or analysis that would support a contrary conclusion. As the Fifth Circuit recently observed in setting aside the application of OSHA’s Cotton Dust Standard to the ginning industry, the Act requires that OSHA’s findings be based on sub- stantial evidence rather than “on assumptions without an adequate evidentiary basis.” Texas Independent Ginners Association v. Marshall, Nos. 78-2663, et al., slip op. at 1220. Even accepting this extreme assumption, however, RTI and others expressed great concern about the ability of many textile firms to raise the capital needed to in- stall controls sufficient to achieve a 0.2 mg/m? exposure limit. If the controls were phased in over an eight- year period (rather than the four years ultimately al- lowed), RTI expected that the Standard would precipi- tate some plant closures. OSHA joined in this con- clusion, acknowledging that the Standard would result in the shutdown of a number of employers and would cause greater concentration in the industry.” Indeed, OSHA observed that RTI had probably underestimated 58 The record showed that cotton’s share of total domestic fiber sales dropped from over 70% in 1947 to only 30% in 1974, largely as a result of foreign competition and the availability of synthetic yarns. Pet. App. at 147-48. And this competition was shown to be continuing, with imports growing at a rate of about 6-7% a year for products covered by multilateral trade agreements. /d. at 148, 150-51; see also Tr. 2565-71 (Mr. John P. Figh). 59 Ex. 6, No. 76 at VI-125-29; Tr. 561-63 (Dr. LeSourd). 6043 Fed. Reg. at 27378, col. 2. In fact, the impact necessarily would be more severe than OSHA presumed, since RTI had assumed that the capital costs would be incurred over an eight- year period. The Standard shortened this to four years, but OSHA took no account of the added economic impact that would result. 30 the financial impact that the Standard would have on smaller firms, since compliance with the Standard by such firms would require an investment of more capital per unit of production than in the case of larger firms. 43 Fed. Reg. at 27378, col. 2. In sum, OSHA’s estimate of the costs of the Standard reflects bald assertions and outright speculation; its as- sessment of the economic impact of the Standard was based on an extreme and unsupported assumption that the agency applied to these speculative costs; and it acknowl- edged that, even so, the Standard would result in plant closures and further concentration in the industry. How much of the industry would be forced to close and what impact this would have on employment and competitive conditions are matters that OSHA failed to address in any concrete manner. Instead, on the basis of the fore- going findings, OSHA declared the Standard to be “gen- erally feasible.” Id. The court below upheld this de- termination, stating that “changes in the market struc- ture of the industry do not signify economic infeasibil- ity,” so long as “the standard will not put the industry out of business.” Pet. App. at 68-69 & n.148, 617 F.2d at 662 & n.148. ATMI submits that this approach effectively imposes no constraint whatsoever on OSHA’s standard-setting authority. Rather, it allows OSHA to exercise enormous, unstructured and “unprecedented power over American industry.” Benzene, 48 U.S.L.W. at 5032 (plurality opinion).*' As we show below, the language of the OSH Act and its legislative history demonstrate that Congress 61 The lengths to which OSHA and some courts may go under the approach followed in the recent case is illustrated by the Lead Standard proceeding. In that case, the D.C. Circuit found that OSHA’s Lead Standard would be economically feasible for the battery manufacturing industry even if it forced 200 small producers out of business, since OSHA could still find that com- petition would survive, through the existence of 30 firms that already controlled 90% of the market. United Steelworkers of America, AFL-CIO-CLC v. Marshall, [1980] OSH Dec. (CCH) at 30403, 8 OSH Cas. (BNA) at 1886-87. 31 intended that OSHA standards would have to satisfy a more exacting test of economic feasibility than was ap- plied in this case. 2. OSHA’s Construction of the Act—Under Which the Costs of a Standard Are Irrelevant Short of the Point at Which They Threaten To Ruin an Entire Industry—Is Inconsistent with the Statutory Language and Legislative History. As this Court recognized in Benzene, the language of the OSH Act is not focused solely upon the promotion of worker health and safety to the exclusion of other concerns.” The mere fact that compliance costs are not “great enough to destroy an entire industry” does not mean that a standard comports with the requirements of the Act. 48 U.S.L.W. at 5031 (plurality opinion). To the contrary, “Congress did not intend OSHA to re duce each significant hazard without regard to economic consequences . . . short of serious dislocation.” Jd. at 5038 n.5 (Powell, J., concurring). The overall objective of the Act, as set forth in Sec- tion 2(b), is to assure safe and healthful working con- ditions, “so far as possible.” 29 U.S.C. § 651(b). And, pursuant to Section 3(8), an occupational health standard may require only such conditions, practices, and opera- tions as are “reasonably necessary or appropriate to pro- 62 The qualified and pragmatic language of the OSH Act stands in sharp contrast to the unqualified language of such statutory provisions as the Endangered Species Act of 1973, 16 U.S.C. § 1536(a) (2) (Supp. III 1979), and the Delaney Clause of the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 848(c) (3) (A) (1976). In relevant part, the Endangered Species Act unquali- fiedly prohibits the funding of projects that would “jeopardize the continued existence of any endangered species . . . or result in the destruction or adverse modification of habitat of such species .-’ The Delaney Clause states unequivocally that “no additive shall be deemed to be safe if it is found to induce cancer when ingested by man or animal.” 32 vide safe or healthful employment and places of employ- ment.” As noted by the plurality in Benzene, a work- place can be “safe” within the meaning of Section 3(8), even though it is not entirely risk-free. 48 U.S.L.W. at 5031.% Similarly, as Professor Currie points out: What is “appropriate” [within the meaning of Sec- tion 3(8)] may be what is desirable in light of feasibility or cost; what is “reasonably necessary” may be what is both necessary and reasonable.™ Like Sections 2(b) and 3(8), the standard-setting provisions of Section 6 are qualified and relative in nature. Thus, Section 6(b)(5) provides that occu- pational health standards are to protect—not abso- lutely—but only “most adequately” against the health hazard to which the standard is directed. The require- ments that protection be provided only “to the extent feasible” and only against “material impairment” are further pragmatic limitations. This language reflects the inevitable fact that there are some risks that it will not be “feasible” to eliminate or reduce and, as the Benzene plurality recognized, 48 U.S.L.W. at 5033, others that must be disregarded as not “material,” if employees are to receive the most adequate overall protection against all workplace hazards. The legislative history of Section 6(b) (5) confirms the qualified nature of the statutory language. Thus, Senator Javits, the author of the original Administration bill (S. 2788), took the position that OSHA standards Should be “feasible” in the sense of being “reasonable” and “practical” as well as technologically achievable. His *3 Cf. Currie, OSHA, 1976 Am. B. Foundation Research J. 1107, 1134 (1976) (the terms “safe or healthful” allow “room for some play”). EPA also has recognized that “use of the term ‘safety’ [does not] necessarily imply a zero-risk concept.” 44 Fed. Reg. 58642, 58660, col. 3 (1979). 64 Currie, supra note 63, at 1134. 33 original bill would have required the submission to an independent board of what Senator Javits described as “a report on the feasibility of the proposed standards.” © “Feasibility” for the purpose of such standard setting was defined as including “the technical feasibility, rea- sonableness and practicality of such standard.” ® This “feasibility report” was to be considered by the inde- pendent board, along with other record evidence, in formulating its standards.” The Senate Committee rejected the concept of an inde- pendent board but agreed that “standards promulgated under Section 6(b) shall represent feasible require- ments.” Leg. Hist. at 147. Accordingly, the term “feas- ibly” was added to Section 6(b) (5) by the Committee and remained in the version passed by both Houses, thereby satisfying the concern of Senator Javits and others that standards be reasonable and practical. The substitution of the phrase “material” for “any” impairment of health responded to a similar concern that OSHA might seek to establish a risk-free utopia regardless of cost. Leg. Hist. at 480 (remarks of Sen- 65 115 Cong. Rec. 22517, col. 3 (1969). 66S. 2788, 91st Cong., Ist Sess. § 4(c) (1969), reprinted in Leg. Hist. at 38-39. 87 Id, § 4(c) (2), (3), reprinted in Leg. Hist. at 39-40. 68 Senator Dominick supported Senator Javits’ attempt to include a feasibility limitation. However, he was concerned that the place- ment of the term “feasibly” in the Committee bill might still be read to require that an occupation be banned if there remained “some risk of injury, impaired health, or life expectancy.” Leg. Hist. at 367. He attempted to cure this problem by relocating the feasibility requirement to the position that it occupies in Sec- tion 6(b) (5) as enacted. See Benzene, 48 U.S.L.W. at 5083 n.53. As Senator Dominick explained: “What we were trying to do in the bill . . . was to say that when we are dealing with toxic agents or physical agents, we ought to take such steps as are feasible and practical to provide an atmosphere within which a person’s health or safety would not be affected.” Leg. Hist. at 502. 34 ator Dominick). Substituting “material” for “any” was part of an effort to achieve a more “balanced” bill, so that, in the words of Senator Saxbe, OSHA health stand- ards would not amount to “harassing measures that would, in effect, limit production in areas that are not necessarily going to increase safety.” Id. at 321.” The legislative compromise that resulted in the imposi- tion of the “substantial evidence” requirement in Sec- tion 6(f) of the Act, 29 U.S.C. § 655(f), also is indica- tive of Congress’ concern that enormous costs not be imposed to produce insignificant benefits.” This com- promise, as the Fifth Circuit has observed, accommodated the Senate’s desire for administra- tive convenience and flexibility by accepting informal rulemaking procedures, but placated House concern for protection of employers from arbitrary bur- dens imposed by a massive federal bureaucracy, [by] adopting the substantial evidence test for judicial review.” In short, the legislative history of the OSH Act reflects Congress’ belief “that it was enacting a fair and reason- able bill that balanced the needs of workers and indus- 6? As noted in a recent comment, the legislative history of the amendment that added the word “material” to Section 6(b) (5) suggests “that risks that could not be eliminated cost efficiently were also meant to be excluded from OSHA regulation.” The Supreme Court, 1979 Term, 94 Harv. L. Rev. at 247. 7 Unless the “substantial evidence” test were substituted for the more lenient “arbitrary and capricious” standard, many in Congress feared that OSHA would be free to impose unjustified burdens. See Leg. Hist. at 343-44 (remarks of Senator Cook). As part of the compromise, OSHA was permitted to act through in- formal rulemaking procedures. 1 Florida Peach Growers Ass'n v. Department of Labor, 489 F.2d 120, 128 (5th Cir. 1974) (citing Associated Indus. of New York State, Inc. v. Department of Labor, 487 F.2d 842 (2d Cir. 1973) ). 35 try.” It clearly shows the congressional concern that some meaningful and effective economic limitations be placed upon OSHA’s standard-setting authority. With- out such limitations, it would be impossible to ensure that the legislation was balanced or that OSHA standards would be “neither oppressive nor needlessly costly and without unjustified harassment of management.” ™ The open-ended affordability test applied by OSHA and the court below—under which costs are irrelevant short of the point at which destruction of an entire industry is threatened—does not comport with the intent of Con- gress. Rather, OSHA must present a responsible prediction, supported by substantial evidence, of what its standard will cost and what impact it will have on such factors as production, employment, competition, and prices.“ And the agency must explain in a cogent manner—on the basis of intelligible criteria—why it concludes that a standard having such an economic impact is “feasible.” OSHA’s performance in the present case falls far short of satis- fying such a test. For that reason alone, the judgment below must be reversed. Moreover, as shown below, the Cotton Dust Standard must be set aside because OSHA made no attempt to determine whether the Standard would produce a reduction in risk that is significant when considered in light of the attendant costs. ™ The Supreme Court, 1979 Term, 94 Harv. L. Rev. at 248. 8 Leg. Hist. at 1091 (remarks of Congressman Randall). This concern for balanced legislation was a consistent theme. See, €.9., id. at 320-21 (remarks of Senator Saxbe) ; id. at 488 (remarks of Senator Williams); id. at 1148 (remarks of Senator Javits) ; id. at 1150 (remarks of Senator Eagleton). ™ See Texas Independent Ginners Ass’n v. Marshall, Nos. 78- 2663, et al., slip op. at 1228 (“OSHA’s estimate of the anticipated cost and expected benefit of proposed regulations are factual find- ings that must be supported by substantial evidence in the record”), 86 B. The Decision of the Court Below Should Be Re- versed Because OSHA Did Not Show That the Standard Was Likely To Achieve a Reduction in Risk That Is Significant When Considered in Light of Its Costs. In Part I.A above, we showed that the extravagantly expansive concept of OSHA’s standard-setting power em- braced by the court below cannot be maintained as a constitutional or statutory matter. In this part, we discuss an alternative construction of the Act that more faithfully reflects the congressional intent that society’s limited industrial hygiene resources be spent wisely —so that vast sums are not squandered on measures that produce no significant health protection for workers. Because OSHA acted in disregard of this congressional expectation, the Cotton Dust Standard must be set aside. 1. OSHA Is Required To Show That a Reasonable Relationship Exists Between the Risk Reduction Benefits and the Costs of Its Standards. The language of the OSH Act, as construed by this Court in Benzene, requires OSHA to show that its stan- dard will produce a significant reduction in the risk of material health impairment. Such a showing must reflect an assessment of the costs and benefits of the standard (and of available alternatives) sufficient to permit OSHA to determine that the risk reduction benefits of the stan- dard bear a reasonable relationship to its costs. This construction is consistent with the weight of pertinent judicial and administrative precedent under the OSH Act and analogous health and safety legislation. a. The OSH Act as Construed in Benzene In Benzene, the plurality construed the qualified and pragmatic language of the OSH Act to reflect Congress’ intent that OSHA not be permitted to make exorbitant 37 and excessive demands upon the limited resources of American employers in a quixotic effort to eliminate all incremental risks from the workplace. To give effect to these pragmatic congressional concerns, the Benzene plu- rality limited the exercise of OSHA’s standard-setting authority to situations where the agency shows that its standard can achieve “significant benefits” in the form of a reduction in the risk of material health impairment. 48 U.S.L.W. at 5082. As Mr. Justice Powell put it, OSHA must show that its standard would “significantly reduce the hazard.” /d. at 5037 (Powell, J +» concurring). Requiring OSHA to show that its standard is likely to produce a significant reduction in the risk of material health impairment is the necessary and logical comple- ment to the requirement that the agency demonstrate, in the first instance, that its standard is addressed to a significant health risk. It would make no sense to per- mit OSHA to mandate the expenditure of vast sums that will result in no substantial reduction in the risk of material health impairment. Thus, the requirement that OSHA “consider differences in degrees of significance rather than simply a total elimination of all risk,” id. at 5031 n.48 (plurality opinion), is a critical element in the overall statutory scheme. If OSHA were free to disregard the extent of risk re- duction that its standards would achieve, the agency, as the Benzene plurality noted, would possess “the power to impose enormous costs that might produce little, if any, discernible benefit.” Jd. at 5032. A standard that consumes enormous resources to produce a negligible re- duction in risk is not “reasonably necessary or appropri- ate to provide safe or healthful employment,” as required by Section 3(8) of the Act, or “feasible” within the meaning of Section 6(b) (5). By contrast, a standard that produces even a very small reduction in a signifi- cant risk of material health impairment may well satisfy 38 the requirements of Sections 3(8) and 6(b) (5) if its cost is relatively low.” The critical point is that, pursuant to the rationale of the Benzene plurality, OSHA cannot validly adopt a health standard without showing both that the standard addresses a significant risk of material health impair- ment and that the standard is expected to achieve a significant reduction in that risk. And the determina- tion of whether a reduction in risk is “significant” must reflect an assessment of the costs of achieving it. Other- wise, OSHA would be free to impose on employers the arbitrary burdens—costing much and accomplishing little —that Congress clearly meant to avoid. “It seems un- likely that Congress . . . intended to require massive expenditures to obtain small benefits whenever the expenditures could be made without bankrupting in- dustries.” *° If a reduction in risk were automatically deemed sig- nificant as long as the costs of achieving it would not produce massive dislocation, OSHA could not possibly discharge its overriding statutory obligation to protect workers “so far as possible.” For, under such an ap- proach, OSHA could require an industry to incur com- pliance costs up to the very limit of “affordability” (wherever that point might lie in OSHA’s judgment) in order to achieve a negligible reduction in the risk ad- dressed by a single health standard, while other more substantial risks would perforce remain unaddressed for lack of resources. OSHA itself should be keenly aware 75 For example, a standard that is expected to reduce the number of deaths in the workplace from ten per year to nine per year might well be deemed to satisfy the requirements of Sec- tions 3(8) and 6(b) (5) if it cost only $100,000, but might not be thought to satisfy the requirements of those statutory provisions if it cost $1 billion. In the former instance, the reduction in risk could be deemed significant in light of the costs, whereas in the latter instance it might not. See The Supreme Court, 1979 Term, 94 Harv. L. Rev. at 247. 76 Id. at 248-49, 39 of this problem, since it already has identified more than 200 substances that it considers to be candidates for regulation as workplace carcinogens pursuant to its Can- cer Policy.” In sum, OSHA must demonstrate that its standards are reasonably necessary or appropriate to achieve a sig- nificant reduction in the risk of material health impair- ment. Such a determination must reflect an assessment of costs and risk reduction benefits sufficient to permit OSHA to conclude that there is a reasonable relationship between the two.’* Unless such an assessment is made, OSHA’s standards inevitably will result in a serious misallocation of resources and a lower effective level of safety than could be achieved un- der standards set with reference to the comparative benefits available at a lower cost. Id. at 5038 (Powell, J., concurring). This does not mean that OSHA must engage in a rigidly formal cost-benefit calculation that places a dollar value on employee lives or health.” Assessing the bene- fits of a standard in light of its costs does not require that a value be assigned to worker life or health; nor does such an assessment predetermine the decisions that OSHA must make. Rather, as the court below recog- nized, it permits the agency to engage in a “ ‘systematic ™7 See 45 Fed. Reg. 538672 (Aug. 12, 1980). In addition, the National Institute for Occupational Safety and Health has pub- lished a list of 2,590 substances in its Tumorigenic Citations Subfile of the Registry of Toxic Effects of Chemical Substances (Oct. 1979). See also Benzene, 48 U.S.L.W. at 5032 (noting “that there are literally thousands of substances used in the workplace that have been identified as carcinogens or suspect carcinogens”). 78 See The Supreme Court, 1979 Term, 94 Harv. L. Rev. at 249 (OSH Act should be construed to require a showing “that the benefits of a regulation bear a reasonable relationship to its costs’’). ™ See Texas Independent Ginners Ass’n v. Marshall, Nos. 78- 2663, et al., slip op. at 1222 n.44. ee ee 40 analysis and evaluation of alternative courses of action.’ ” Pet. App. at 69-70 n.152, 617 F.2d at 662 n.152 (citation omitted). And it allows the agency (and the public) to ensure that society’s limited industrial hygiene resources are not being squandered or misallocated in a fashion that produces very little health benefit and that provides workers with less overall protection than might otherwise be possible.” The nature of the exercise in which OSHA must en- gage was adumbrated in Benzene. First, OSHA must make a responsible determination of the costs and risk reduction benefits of its standard. Pursuant to the requirement of Section 6(f) of the Act, this determina- tion must be factually supported by substantial evidence in the record. The subsequent determination whether the reduction in health risk is “significant” (based upon the factual assessment of costs and benefits) is a judg- ment to be made by the agency in the first instance.* Although a court might well be reluctant to second guess the agency’s judgment on the question of significance, the mere fact that OSHA is obligated to make such a determination, based upon an assessment of costs and benefits, will have a salutary effect on the agency’s de- cisionmaking process. For one thing, engaging in such an assessment will lead OSHA to reject control strategies the costs of which 8° Of course, life and health-protective resources are not un- limited, and compliance with OSHA standards is not the only cost industry must bear. As noted in Making Prevention Pay, Final Report of the Interagency Task Force on Workplace Safety and Health VII-15 (Dec. 14, 1978) (footnotes omitted) : The cost of complying with OSHA is only part of the cumu- lative impact of Federal regulation on individual firms and their industries. There is growing concern that these cumula- tive impacts are costing jobs and draining substantial invest- ment from new efforts at industrial innovation, while adding perhaps 3/4 of 1% to the yearly rate of inflation. 81 See Benzene, 48 U.S.L.W. at 5034, 5035 n.62 (plurality opin- ion). 41 are grossly disproportionate to their benefits. It also will cause uhe agency to give some consideration to the impact that its standard would have on the industry’s ability to address other, more significant health and safety risks in the future—including risks that OSHA anticipates will be addressed in standards that are at least in the planning stage.” Finally, forcing OSHA to engage in such an ass -sment and to explain the de- termination it has reached provides a potential legisla- tive check on what might otherwise amount to the exer- cise of virtually untrammeled authority. It will allow for correction, through the political process, of actions that are deemed by the Congress to be extreme, unwar- ranted, and inconsistent with congressional intent."* b. Judicial and Administrative Precedent While the courts of appeals have not been unanimous in requiring that OSHA standards reflect a careful as- sessment of risk reduction benefits in light of costs, the weight of relevant precedent decidedly favors such a construction of the Act. Before the decision below, only one case had considered explicitly the nature of the eco- nomic impact assessment that is required in order for OSHA to comply with Section 3(8) as well as Section 6(b) (5) of the Act. American Petroleum Institute v. OSHA, 581 F.2d 493 (5th Cir. 1978). That decision, re- viewed by this Court in Benzene, construed Sections 3(8) and 6(b) (5) as requiring OSHA to assess the expected health protection benefits of its standards in light of the cost burdens to be imposed and to show that the benefits 82 A contrary reading of the Act “could force the depletion of an industry’s resources in an effort to reduce a single risk by some speculative amount, even though other significant risks remain unregulated.” Jd. at 5088 & n.7 (Povell, J., concurring). 83 See Amalgamated Meat Cutters & Butcher Workmen v. Con- nally, 337 F. Supp. 737, 758-59 (D.D.C. 1971) (3-judge court) (Leventhal, J.); cf. Yakus v. United States, 321 U.S. 414, 426 (1944). 42 bear a reasonable relationship to the costs. 581 F.2d at 504. A different panel of the Fifth Circuit recently va- cated the Cotton Dust Standard as it applied to the ginn- ing industry because, among its other failings, OSHA did not show that such a reasonable relationship existed in that case. Texas Independent Ginners Association v. Marshall, Nos. 78-2668, et al., slip op. at 1224-25. The Courts of Appeais for the Sixth and Seventh Cir- cuits have interpreted the term “feasible” to reach a similar result. Thus, in RMI Co. v. Secretary of Labor, 594 F.2d 566, 573 (6th Cir. 1979), the Sixth Circuit rejected the notion that controls required by an OSHA standard are “necessarily . . . economically feasible merely because the employer can easily (or otherwise) afford them.” Rather, according to the court, the ex- penditures required by a standard are economically feas- ible only if OSHA demonstrates that there is “reasonable assurance that there will be an appreciable and cor- responding improvement in working conditions.” Jd. Ac- cordingly, the Sixth Circuit insisted that OSHA weigh the costs of compliance against the benefits expected to be achieved thereby in order to determine whether the proposed remedy is economically feasible. Id. In Turner Co. v. Secretary of Labor, 561 F.2d 82, 86 (7th Cir. 1977), the Seventh Circuit construed the term “feasible” in the same way, concluding that required controls could not be deemed feasible without determining “whether the health benefits to employees . . . justify the cost.” Such a construction of the term “feasible,” the court noted, was “in accordance with the clear in- tent of Congress and the purpose of the Occupational Safety and Health Act.” Jd. at 85.% Thus, the decision * In Marshall v. West Point Pepperell, Inc., 588 F.2d 979, 981 n.3 (5th Cir. 1979), the Fifth Circuit observed that the Turner decision “certainly comports with the plain import of the term feasible.” 43 below is inconsistent with the weight of the most directly relevant judicial precedent.® It is also inconsistent with judicial interpretations of similiar language in other statutes. For example, courts have construed the provision of the Consumer Product Safety Act authorizing the adoption of rules “reasonably necessary to eliminate or reduce an unreasonable risk of injury” * to require the agency to assess the expected benefits in light of the burdens to be imposed by the standard.” Prior to the Benzene decision, OSHA took the position that this construction of the Consumer Product Safety Act did not apply to the OSH Act, since the Con- sumer Product Safety Act is addressed only to “unrea- sonable risks,” while, in OSHA’s view, the OSH Act re- quired the elimination of all risks of material health im- pairment, whether or not they are so significant as to be unreasonable. However, this position was rejected in Benzene, where the OSH Act was construed as limiting OSHA to addressing only significant risks in its stand- ards. Consequently, the distinction that OSHA and the court of appeals, see Pet. App. at 71-72 n.159, 617 F.2d 85 Admittedly, other courts have taken a more restrictive ap- proach to the concept of economic feasibility. See Americar Iron & Steel Inst. v. OSHA, 577 F.2d 825 (3d Cir. 1978), cert. granted, 48 U.S.L.W. 3855 (U.S. July 2, 1980), cert. dismissed, 49 U.S.L.W. 3145 (U.S. Sept. 10, 1980); Industrial Union Dep’t, AFL-CIO v. Hodgson, 499 F.2d 467 (D.C. Cir. 1974). However, prior to the decision below, only the Third Circuit in American Iron & Steel Institute had limited economic feasibility to an affordability test in an action brought by employers to challenge the validity of an OSHA standard. Another panel of the D.C. Circuit has since followed the con- struction of the Act adopted by the court below. United Steel- workers of America, AFL-CIO-CLC v. Marsheil, [1980] OSH Dec. (CCH) 30329, 8 OSH Cas. (BNA) 1810 (D.C. Cir. Aug. 15, 1980). 8615 U.S.C. § 2058(c) (2) (A) (1976). 87 Aqua Slide ‘N’ Dive Corp. v. CPSC, 569 F.2d 831 (5th Cir. 1978). Accord, D.D. Bean & Sons v. CPSC, 574 F.2d 643 (1st Cir. 1978) ; cf. Forester v. CPSC, 559 F.2d 774, 789 (D.C. Cir. 1977). 44 at 663 n.159, drew between the Consumer Product Safety Act and the OSH Act can no longer be maintained.™ The decision below is also at odds with the long- standing position of the Occupational Safety and Health Review Commission (“OSHRC’’), the agency charged by Congress with responsibility for deciding enforcement proceedings under the Act.” In Continental Can Com- pany, [1976-1977] OSH Dec. (CCH) 25250, 25256, 4 OSH Cas. (BNA) 1541, 1547 (1976), OSHRC inter- preted the term “feasible” to require that “all the rele- vant costs and benefits [be] weighed.” * OSHRC has ad- 88 Absent an express bar, other health and safety statutes also have been read to require consideration of risk, benefit and cost evidence. Thus, the section of the National Traffic and Motor Vehicle Safety Act providing for the adoption of “reasonable, prac- ticable and appropriate” traffic safety standards, 15 U.S.C. § 1392 (f) (3) (1976), has been construed as requiring the Secretary of Transportation to “identif[y] some of the costs associated with the proposal and determine[] that these costs are overridden by rea- sonably predictable benefits.” H & H Tire Co. v. Department of Transportation, 471 F.2d 350, 356-57 (7th Cir. 1972) (Stevens, J., concurring opinion adopted by the court) (footnotes omitted). Similarly, the Federal Water Pollution Control Act provision directing that EPA’s 1983 effluent limitation guidelines reflect “the best available technology economically achievable,” 33 U.S.C. § 1811(b) (2) (A) (1976 & Supp. III 1979), has been held to re- quire consideration by EPA of “the benefits derived from the application of its effluent reduction requirements.” Appalachian Power Co. v. Train, 545 F.2d 1351, 1361 (4th Cir. 1976). Accord, National Crushed Stone Ass’n, Inc. v. EPA, 601 F.2d 111, 122-23 (4th Cir. 1979), cert. granted, 48 U.S.L.W. 3535 (U.S. Feb. 19, 1980). 89 See Section 12 of the Act, 29 U.S.C. § 661; Leg. Hist. at 462. * In Continental Can, OSHA couched its argument in the very terms embraced by the court below in the present case, claiming that Section 6(b)(5) “requires employers to expend funds for each and every health hazard whether life threatening or not with- out limit so long as the expenditures for each hazard can be borne without putting the employers’ financial condition in jeopardy.” [1976-1977] OSH Dec. (CCH) at 25256, 4 OSH Cas. (BNA) at 45 hered to this view in subsequent decisions.” As these decisions indicate, there is nothing novel in Suggesting that an agency should develop and consider information relating to the expected benefits and costs of its proposed actions. To the contrary, the experience of numerous federal agencies (including OSHA itself) demonstrates that assessing the costs and benefits of regulatory action is both a commonplace and extremely useful exercise. Perhaps the best example of this is the experience of the Environmental Protection Agency (“EPA”), which routinely conducts risk assessments for carcinogenic substances, an exercise that, in EPA’s view, represents “a significant step toward the objective of achieving real benefits in improved public health while avoiding the burdens of undesirable regulatory action.” 41 Fed. Reg. 21402, 21402-03 (1976). More recently, under Section 112 of the Clean Air Act, 42 U.S.C. § 7412 (Supp. III 1979), EPA proposed a policy for identifying and regulating airborne substances posing a risk of cancer (the “Air Cancer Policy”). 44 Fed. Reg. 1547. OSHRC rejected OSHA’s position as being “inconsistent with the intent of Congress” and observed: Clearly, employers have finite resources available for use to abate health hazards. And just as clearly if they are to be made to spend without limit for abatement of this hazard their financial ability to abate other hazards, including life threat- ening hazards, is reduced. Id. *1 See Castle & Cooke Foods, [1977-1978] OSH Dec. (CCH) 26325, 5 OSH Cas. (BNA) 1485 (1977), appeal docketed, No. 77- 2565 (9th Cir. July 14, 1977); Great Falls Tribune Co., [1977- 1978] OSH Dec. (CCH) 26308, 5 OSH Cas. (BNA) 1443 (1977), appeal docketed, No. 77-2566 (9th Cir. July 14, 1977); West Point Pepperell, Inc., [1977-1978] OSH Dec. (CCH) 26136, 5 OSH Cas. (BNA) 1257 (1977), aff'd, Marshall v. West Point Pepperell, Inc., 588 F.2d 979 (5th Cir. 1979); KLI, Inc., [1977-1978] OSH Dec. (CCH) 26935, 6 OSH Cas. (BNA) 1097 (1977). Most recently, in Samson Paper Bag Co., [1980] OSH Dec. (CCH) 380038, 30044 (1980), the three OSHRC Commissioners 46 58642 (1979). Section 112 of the Clean Air Act requires EPA to establish national emission standards for hazard- ous air pollutants at a level that “provides an ample margin of safety to protect the public health.” 42 U.S.C. § 7412(b) (1) (B). If anything, this language would ap- pear to be less qualified and more singleminded in its focus on health protection than the language of Sections 3(8) and 6(b) (5) of the OSH Act. Yet EPA has stated that Section 112 was not intended to require “the com- plete elimination of all risks,” but rather permits the agency to “consider other social and economic factors [including the cost of reducing the risks further] in determining whether an ample margin of safety is pro- vided by a given control level.” 44 Fed. Reg. at 58661, col. 1. Other agencies, including the Food and Drug Administration and the Consumer Product Safety Com- mission, also utilize risk estimation procedures to guide their regulatory decisions.” OSHA itself performs cost and benefit assessments. As noted in Benzene, 48 U.S.L.W. at 5031, OSHA en- gages in cost-benefit analysis in determining priorities for regulation under Section 6(b) (5), an exercise that OSHA believes is essential to ensure that it allocates its resources in a reasonable and efficient manner.” If such an analysis is appropriate in order to ensure that OSHA’s resources are not misallocated, its use surely is vital to were divided, but two Commissioners agreed to remand the case for further proceedings consistent with Continental Can. See id. at 30045 (Commissioner Barnako) ; id. at 30046 (Chairman Cleary). 2 For example, under Section 406 of the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 346 (1976), FDA conducted a risk assessment to assist it in setting a tolerance level for aflatoxin in consumer peanut products, 43 Fed. Reg. 8808 (1978), and the Consumer Product Safety Commission developed a quantitative risk estimate for benzene in connection with its proceedings for regulating that substance. 43 Fed. Reg. 21838, 21845 (1978). 93 See also National Congress of Hispanic Am. Citizens v. Mar- shall, 626 F.2d 882 (D.C. Cir. 1979). 47 ensure that the much greater societal resources com- mitted under OSHA standards are not misallocated. Moreover, eleven months after the Standard was prom- ulgated, OSHA prepared and submitted to Congress a lengthy and detailed cost-effectiveness/cost-benefit analy- sis for the Cotton Dust Standard itself. As can be seen from the foregoing discussion, the “ex- perience gained under the [OSH Act] and other health and safety laws,” 29 U.S.C. § 655(b) (5), further con- firms the importance and usefulness of performing a responsible cost and benefit assessment in the setting of occupational health standards.* Because OSHA failed to perform such an assessment here, it could not deter- mine that the Cotton Dust Standard is reasonably nec- essary or appropriate to achieve a reduction in risk that is significant in light of its enormous costs. The failure of the court below to require OSHA to perform such an assessment and to make such a determination requires reversal. * The Report was not part of the rulemaking proceeding, was not submitted for industry comment, was not before the court of appeals, and, as OSHA acknowledges, “does not reflect the decision- making process used to promulgate the final standard.” Department of Labor Report to the Congress, Cotton Dust: Review of Alterna- tive Technical Standards and Control Technologies, at v (May 14, 1979) [hereinafter “Cotton Dust Report to Congress”]. Accord- ingly, the Report cannot be relied upon by CSHA to support the Cotton Dust Standard. While ATMI does not agree with many of the assumptions employed, the methods followed, or the conclusions reached in the Report, the Report still demonstrates that OSHA can do a better job of assessing costs, benefits and available alter- natives than it did in the Cotton Dust proceeding. *5 The usefulness of cost and benefit assessments in regulatory decisionmaking is discussed in an extensive literature. See, e.g., National Academy of Sciences, Decision Making for Regulating Chemiculs in the Environment 44-45 (1975) ; Kasper, Cost-Benefit Analysis in Environmental Decisionmaking, 45 Geo. Wash. L. Rev. 1013 (1977); Oi, On the Economics of Industrial Safety, 38 L. & Contemp. Prob. 669 (1974). 48 2. The Judgment Below Must Be Reversed Because OSHA Did Not Show That the Standard Would Produce a Reduction in Risk That Is Significant in Light of Its Costs. Under the approach followed by OSHA and the court below, it was irrelevant that the Cotton Dust Standard might yield health protection benefits not significantly greater than those that could be realized from a much less costly standard. Consequently, OSHA did not assess the relative risk reduction benefits of its Standard and of the alternative proposed by ATMI, to see whether its more stringent (and far more costly) Standard was jus- tified in light of the relative benefits and respective costs. For that reason, the judgment below must be reversed. If the court below had imposed such a requirement, however, it could not reasonably have found that OSHA satisfied it here, since the agency never correctly iden- tified the costs or the benefits of its Standard, let alone of the ATMI alternative. The critical flaws in OSHA’s cost estimate already have been described at some length, see pages 24-30 supra, and need not be revisited. It will be useful, however, to comment briefly on the in- adequacy (indeed, the virtual nonexistence) of OSHA’s assessment of risk reduction benefits. OSHA devoted only one page of the Standard’s 45- page preamble to a discussion of benefits. 43 Fed. Reg. at 27378-79. That discussion reveals almost nothing about the risk reduction benefits that OSHA’s Standard might be expected to produce in comparison to the ATMI alternative.” *6In any event, it is analytically flawed. For example, OSHA arbitrarily raised the number of employees in yarn manufacturing by almost 60% from 126,000 to 200,000. No support was offered for this capricious rejection of figures taken directly from a United States Department of Labor survey. Ex. 6, No. 76 at III-1, 8, 28. Moreover, OSHA applied Merchant’s predicted prevalence rate to 49 First, OSHA made no attempt to analyze the impact that a medical surveillance program (combined with se- lective respirator use and employee transfer in appropri- ate cases) would have on the development of byssinosis at any level of exposure. Yet OSHA acknowledged that a program of medical screening (combined with such administrative measures) is likely to have a significant impact in reducing the incidence of byssinosis among textile workers.” And the President’s Council on Wage and Price Stability suggested that the benefits of a med- ical surveillance or respirator program might well be greater than the benefits that would be achieved by re- ducing permissible exposure levels through engineering dust controls.** Second, OSHA made no attempt to differentiate be- tween the various grades or symptoms of byssinosis in evaluating the potential benefits of its Standard. Since the court below did not decide that the nondisabling and reversible symptoms of byssinosis constitute “material impairment of health,” Pet. App. at 52-58, 617 F.2d at 655, any reduction in the risk of developing this mild form of byssinosis should not be included as a potential health benefit within the meaning of the OSH Act. As noted by the Council on Wage and Price Stability, “if the entire yarn manufacturing industry, even though Merchant had acknowledged that his study cannot be used to draw inferences for the entire textile industry. See note 19 supra. *7 See page 15 supra; 48 Fed. Reg. at 27359, col. 8; Cotton Dust Report to Congress 31, 49. *8 Comments of the Council on Wage and Price Stability, Ex. 111 at 29. As the Council pointed out, medical screening combined with respirator use and employee transfer where appropriate “might provide more cost-effective solutions to the byssinosis problem.” Jd. at 30. For that reason, it urged OSHA to examine the “cost effec- tiveness of each provision [i.e., dust control, medical surveillance, respirator use and employee transfer] . . . independently and in concert with the other provisions in order to ensure the most pro- tection for the least cost.” Jd. 50 eliminating ‘material impairment’ and not occupational discomfort is OSHA’s main concern, then standard de- velopment should focus on the irreversible harmful health effects,” °° Even if prevention of the mild, low-grade symptoms of byssinosis were a cognizable health benefit under the Act, however, it surely is a much less significant benefit than the prevention of chronic and disabling byssinosis. As the Council on Wage and Price Stability observed, any meaningful assessment of benefits must reflect these differences in the health impact of the different grades of byssinosis, for “a system that simply sums all degrees of health problems from the serious to the inconsequential omits important information and may be misleading.” ' Yet OSHA’s offhand discussion of benefits treats all forms of byssinosis the same. Thus, OSHA’s discussion of benefits in the Cotton Dust proceeding is at best a formality that does not begin to approach the kind of careful and systematic assessment that would permit the agency to determine whether the Standard is likely to result in a significant reduction in risk as compared to ATMI’s much less costly alterna- tive." Instead, OSHA treated cotton dust like a car- cinogen, ignoring the fact that byssinosis, in its early *? Comments of the Council on Wage and Price Stability, Ex. 111 at 15. 100 7d. at 15, 27, 29. Cf. National Congress of Hispanic Am. Citizens v. Marshall, 626 F.2d 882, 889 (D.C. Cir. 1979) (approving the decision of OSHA to consider “the nature and severity of the hazardous exposure” as well as the number of employees to be bene- fited in setting priorities for the promulgation of occupational health and safety standards). 101 Cf, United Parcel Serv. of Ohio, Inc. v. OSHRC, 570 F.2d 806, 812 (8th Cir. 1978) (overturning an order enforcing a safety standard because of the failure “to give consideration to less rigor- ous requirements” having less severe economic consequences). 51 stages is nondisabling and reversible and that a program of medical surveillance is likely to have a dramatic im- pact in reducing the number of serious cases of byssinosis that might otherwise occur.’” As a result, the record in this case contains no supportable determination that the reduction in risk that might result from the Cotton Dust Standard is significant when considered in light of the costs. For that reason, the judgment of the court below must be reversed.!* Il. THE COURT BELOW ERRONEOUSLY CON- CLUDED THAT OSHA HAS AUTHORITY TO RE- QUIRE EMPLOYERS TO MAINTAIN THE WAGES AND BENEFITS OF EMPLOYEES WHO ARE TRANSFERRED FOR REASONS UNRELATED TO HEALTH IMPAIRMENT. Under the Standard, respirators must be provided and used in various circumstances—for example, during the period in which engineering controls are being installed or where such controls are not sufficient to reduce ex- 02 43 Fed. Reg. at 27858-59; Comments of the Council on Wage and Price Stability, Ex. 111 at 6-7, 18, 15, 27, 29. 8 In fact, OSHA did not even establish that textile workers would be exposed to a “significant risk of material health impair- ment” if the ATMI alternative were adopted. For that reason alone, the Cotton Dust Standard should be set aside, as the Benzene Standard was in similar circumstances. The judgment below also should be reversed because, although ATMI “opposed on substantial grounds” the strict numerical lim- its adopted by OSHA and proposed a much less costly alternative, OSHA made no serious attempt to “explain and support” its rejec- tion of the ATMI alternative in favor of the much more burden- some approach of the Standard. Associated Indus. of New York State, Inc. v. Department of Labor, 487 F.2d 842 (2d Cir. 1978) ; ef. Synthetic Organic Chem. Mfrs. Ass’n v. Brennan, 508 F.2d 1155, 1160 (3d Cir. 1974), cert. denied, 420 U.S. 973 (1975) (OSHA must consider presently available alternatives) ; Industrial Union Dep't, AFL-CIO v. Hodgson, 499 F.2d at 475 (OSHA must explain why it “chooses to follow one course rather than another’ ) 52 posures below the applicable PEL.’* Respirators also must be provided for use during various maintenance, repair and cleaning activities, or whenever an employee requests a respirator, whether or not the PEL is ex- ceeded. ! Pursuant to subsection (f) (2) (v) of the Standard, if a physician determines that an employee is unable to wear a respirator when the Standard requires its use, the employee must be given the opportunity to transfer to another position, having a dust level at or below the PEL, that is available or later becomes available.’ Such transfers are to be made even though the employee shows no symptom of byssinosis and even though his inability tc wear a respirator is not attributable to past employment. Moreover, employers must insure that an employee who is so transferred “suffers no loss of earnings or other employment rights or benefits as a result of the trans- fer.” **% This guarantee against any loss of earnings, rights or benefits appears to be subject to no limitation in terms of duration or otherwise. The wage guarantee provision of the Cotton Dust Standard raises a relatively narrow question for review in this case: Whether OSHA has authority to require employers to maintain the earnings, benefits, and rights of employees who are transferred without any demon- stration that they are suffering a work-related health impairment or even that their inability to wear respira- tors is related to exposure to cotton dust or other aspects of their employment.’” 104 29 C.F.R. § 1910.1043(f) (1979), Pet. App. at 6-9. 106 29 C.F.R. § 1910.1043 (f) (2) (v) (1979), Pet. App. at 8-9. 106 Jd. 107 On the facts of the present case, there is no need to address the separate question whether OSHA can impose a medical removal- wage retention requirement that is more specifically related to health impairment. This latter question is raised by OSHA’s Lead 53 No one has suggested that the OSH Act contains an express grant of authority for the imposition of a wage guarantee provision,’’* and the court below did not pre- tend otherwise. The court did conclude, however, that the provision was impliedly authorized by Section 3(8) of the Act. See Pet. App. at 96, 617 F.2d at 674. The court reasoned that, in the absence of a wage guarantee provision, employees might “refrain from disclosing ac- tual health impairments from the dust exposure,” be- cause they might be reluctant to risk “disadvantageous transfers, or even risk losing their jobs.” Pet. App. at 97, 617 F. 2d at 674-75. The court apparently misconceived the nature of the employee transfer-wage guarantee provision of the Cot- ton Dust Standard. Under the Standard, employee trans- fers are not made when the worker is found to have suffered “actual health impairments from the dust ex- posure.” To the contrary, the fact that an employee has suffered actual health impairment from dust exposure does not entitle him to a transfer and wage guarantee. A transfer is required only when an employee is found to be unable to wear a respirator, whether or not he ex- hibits any symptoms of even the mildest form of bys- Standard, 29 C.F.R. § 1910.1025(d), (j), (k) (1979), as amended at 44 Fed. Reg. 50338 (1979), id. at 60981, id. at 68828, which was at issue in United Steelworkers of America, AFL-CIO-CLC v. Mar- shall, [1980] OSH Dec. (CCH) 30829, 8 OSH Cas. (BNA) 1810 (D.C. Cir. Aug. 15, 1980). As OSHA recognized, the broader ques- tions of mandatory medical removal and wage retention posed in the Lead Standard proceeding are not presented by the Cotton Dust Standard, 48 Fed. Reg. at 27387, col. 8, and the court below did not address those questions. See Pet. App. at 95-96 n.238, 617 F.2d at 674 n.2388. Consequently, a decision regarding the wage guarantee provision of the Cotton Dust Standard would not neces- sarily be applicable to the substantially different medica] removal- rate retention provision of the Lead Standard. 108 See 43 Fed. Reg. at 52976. 109 The fact that health impairment does not entitle a worker to the wage guarantee is consistent with OSHA’s description of the 54 sinosis.‘° Thus, the court below appears to have upheld the validity of the Cotton Dust wage guarantee provision under a health impairment rationale that does not apply to the Standard. In any event, reading Section 3(8) as an authoriza- tion for OSHA to impose the wage guarantee provision under the rationale advanced by the court below proves too much, for it would i [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/brief%3Amicro_IA40385006_2223%3A06. Public record. Not legal advice.