Amicus Brief — Industrial Union Dept., AFL-CIO v. American Petroleum Institute
Supreme Court brief1980
Ask Donna
What actually matters in this document.
Text
try
hs =
an
tr
0
Nos. 78-911, 78-1036
MAY 18 j979—*
IN THE
MICHABL RODAK, JR., CLERK
Supreme Court of the United States ——-——
OCTOBER TERM, 1978
INDUSTRIAL UNION DEPARTMENT, AFL-CIO,
Petitioner,
Vv.
AMERICAN PETROLEUM INSTITUTE, et al.,
Respondents;
RAY MARSHALL, SECRETARY OF LABOR,
Petitioner,
Ve
AMERICAN PETROLEUM INSTITUTE, et al.,
Respondents.
Qn Writ of Certiorari to the United States
Court of Appeals for the Fifth Circuit
BRIEF FOR NATURAL RESOURCES DEFENSE
COUNCIL, INC. AS AMICUS CURIAE
RICHARD E. AYRES
DAVID D. DONIGER
Attorneys for Natural Resources
Defense Council, Inc.
Mailing Address:
NATURAL RESOURCES
DEFENSE COUNCIL, INC.
917 15th Street, N.W.
Washington, D.C. 20005
(202) 737-5000
May 17, 1979
INDEX
INTERESTS OF THE NATURAL RESOURCES
DEFENSE COUNCIL...
QUESTIONS PRESENTED |
sa Snare AND STATEMENT OF THE
A. THE PROBLEM OF OCCUPATIONAL
RPRPEIREUID voi nicss)secerpnsccses:
B. THE OCCUPATIONAL SAFETY AND
HEALTH ACT...
THE BENZENE STANDARD . mee
THE LOWER COURT DECISION... |
SCIENTIFIC AND POLICY ISSUES RE-
LATED TO THE RELIABILITY OF
QUANTITATIVE RISK ASSESSMENT
AND QUANTITATIVE COST-BENEFIT
pe Reet) MARE Mn
1. The Best Available Scientific Know ledge
Permits Reliable Identification Of
Carcinogens, But Not Reliable Quan-
tification Of Risk ......... 3
a. The Basic Biochemical Process Of Can-
cer Causation Are Largely Un-
MIN Chote
b. Epidemiological Studies Are Capable
Of Identifying Carcinogens, But
Not The Magnitude Of Cancer Risks
ec. Animal Experiments Also Permit
Identification Of Carcinogens, But
Not The Degree Of Risk To Humans.
2. Techniques For Quantifying Economic
SS est NR ceraeaer) Fapeat ial Ma AA ES )
moO
3. Comparing Health And Economic Inter-
TRESS a eS aN SE Res or AEN bes ee ean sae Nee
Page
13
13
il
SUMMARY OF ARGUMENT
PX th) | ~, 5 See eOnEMee NEROMR TPE OR rey ogee fuce ewes
I. THE OCCUPATIONAL SAFETY AND
HEALTH ACT DOES NOT REQUIRE
UANTITATIVE JUSTIFICATION FOR
TANDARDS TO PROTECT EMPLOYEES
FROM CANCER-CAUSING CHEMICALS.
A. Conditioning Standards On Quantitative
Assessments Of Risk Conflicts With The
Congress’ Desire To Prevent Disease
Botore Ft TRG WRN acai osccdececsesie cs cceccovae hiss
1. The Primacy Of The Goal Of Pre-
venting Occupationally-Caused Dis-
2. Judicial Recognition Of The Primacy
Of Preventing Disease ........................
B. The Requirement That Standards Be
Feasible Does Not Mandate A Quan-
titative Demonstration That eir
Benefits Exceed Their Costs......................
1. The Legislative History Of The Term
“Feasible” Shows That Congress In-
tended Only A Limited Role For
Cost Considerations In Setting
Stans i eae
2. Prior Cases Have Concluded That
Costs Are To Be Accorded A Limited
Role In Setting Standards..................
C. The Lower Court Erred In Holding That
a epeor Cost-Benefit Analysis Is
andated By The Requirement That
Standards Be “Reasonably Necessary.”
1. The Legislative History Shows That
“Reasonably Necessary” Has No
Bearing On The Role Of Cost Con-
GIO ;... 5.68
Page
28
31
31
35
39
40
48
ili
2. “Reasonably Necessary” Merely Re-
quires A Rational Connection Be-
tween The Means And enna of
Standards ..
D. The Consumer Product Safety Act Does
Not Support The Lower Court’s Con-
struction Of The Occupational ase
and Health Act
1. The Two Acts Are Not Parallel
2. The Consumer Product Safety Act
Does Not Require Quantitative
Cost-Benefit Analysis.
II. THE SECRETARY ACTED WELL WITHIN
HIS DELEGATED AUTHORITY IN
DECLINING TO ADOPT tse eimanaien
RISK ASSESSMENT...
A. The Substantial Evidence Test Does Not
Permit The Court To Substitute Its
Judgment For The Secretary’s On The
Reliability Of Quantitative Risk
Assessments................. LARGE, ‘
1. Standard Setting Under The
Occupational Safety And Health
Act Involves A Mixture Of Fact-
Finding And Policy- Making
2. The Substantial Evidence Test Does
Not Prohibit The Secretary From
Making Policy Judgments .
B. The Secretary Reasonably Determined
That Quantitative Risk Assessment
Techniques Are Not Sufficiently Re-
liable To Be A Basis For Decision...
ee Ga a Oa
Page
49
50
55
63
67
iv
CITATIONS
Page
CASES:
Abbott Laboratories v. Gardner, 387 U.S. 1386
(GRRE aU TC as a AG eee treo AU Oe Pa a 62
AFL-CIO v. Brennan, 530 F.2d 109 (3rd Cir.
RAR RECRERSVTTA PGT PERI TY Wek tee ot IVa) Ee ee 12,45,46
American Iron & Steel Inst. v. OSHA, 577 F.2d
825 (3rd Cir. (1978), petition for certiorari
TOES eae le LT CAIER aod CORRAL PURELY fe 8 CRU 15,37,46,
47,54,60
American Petroleum Inst. v. OSHA, 581 F.2d 493
NN NI eer asi case ciintstanidiyeisieyoshemouetiaivenis 12,13,27,
28,30,37,
38,43,47,
50,54,60,
63,65,66,
67
Arkansas- Best Freight Systems v. Occ. Safety &
Health Rev. Comm'n. 529 F.2d 649 (8th Cir.
eS | ater eRe A Hebe Melannen Ee seen 47
Associated Industries of N.Y.S. v. United States
Dep't of Labor, 487 F.2d 342 (2d Cir. 1978).......... 57,60,61
Certified Color Mfrs. Ass’n v. Mathews, 543 F.2d
Be a ee IES sets fates sea nc ascansstypercecs lonvies 15,37
Citizens to Preserve Overton Park v. Volpe, 401
ee seen: 62
Consolo v. Federal Maritime Comm'n 383 U.S. 607
D_* Neg TEs EAR SR. OL Siemnatint Paltoan fy MARC Ate) eee Grates Rn 62
Environmental Defense Fund v. EPA, —F.2d—, 8
BERaee MOU CEP Cs RUUD i cesevcenepevcescseccsctsvecsncees 15,37,39
Environmental Defense Fund v. EPA, 510 F.2d
Be a It I ndash access iocy slacks cuxavastesdevcecesnt 15,37
Federal Communications Comm’n v: National
Citizens Committee for Broadcasting, 436 U.S.
1, aaNet Leman le via I Bed 0 OF Rr att ee eR 59
Florida Peach Growers Ass’n v. United States
Dep't of Labor, 489 F.2d 120, 129 (5th Cir. 1974) 60
Vv
Page
Industrial Union Dep’t, AFL-CIO v. Hodgson, 499
ne Ce, CR. Bete sa osviics sasissninedeccchprscactosaans 12,15,35,
44,4754,
57,58,59,
60,61,62,
63
National Nutritional Foods Ass’n v. Weinberger,
BES Fmt Geet CH Gey BIT gaa veci ck ccssstscicevecesvernc¥ess 61
Permian Basis Area Rate Cases, 390 U.S. 747
(> PUR ae ae mse a CE e dG Shep aon EROS W Eb Aa PilMerte pr 58
Society of the Plastics Industry v. OSHA, 509 F.2d
1301 (2d Cir. 1975), cert. denied 421 U.S. 992
| ARNO eed, tinal Oe cae ee aE AGI Mie MO ot, 15,35,45,
54,60
Synthetic Organic Chemical Mfrs. Ass'n v. Bren-
nan, 503 F.2d 1155 (3rd Cir. 1974) 0.00.0... ee
Universal Camera Corp. v. NLRB, 340 U.S. 474
i: BRAM Raine) Ment Me ananet ent WR eemo mec bE he 2 62
STATUTES:
Occupational Safety and Health Act of 1970
Section 3, 2) U.S.C. $651 .....:.......6csccbccses.. 5
Section S, 29 U.S.C. $668...........6.:..6:cc0000.:.. passim
Section 6, 29 U.S.C. §656............00.ccc. passim
Section 20, 29 U.S.C. §669 00000... 10
Section 21, 29 U.S.C. §670.........000. ce. 10
Section 23, 30 U.S.C. $671.....................0..... 10
Consumer Product Safety Act of 1972
Section 7, 15 U.S.C. §2056..............00000000. 50, 59
Section 9, 15 U.S.C. §2058 ......0...0..000. 50
Section 11, 15 U.S.C. $2060................0........ 59
Toxic Substances Control Act of 1976 ..............
Section 4, 15 U.S.C. §2608 0.000000... 14
Section 6, 15 U.S.C. §2605.................000000... 59
Section 19, 15 U.S.C. §2618................ Rave 59
Clean Air Act Amendments of 1977
Section 307, 42 U.S.C. §7607 ....0.000.0... 62
vi
REGULATIONS AND REGULATORY NOTICES:
Federal Regulations
GE rts, RR, PI hic facsbacasdecdtiscactadeeecst
Fe Es BEE iar ecctisccccedhan |
Federal Register Statements
43 Fed. Reg. 5918 (1978) ..........00000..00..00.
42 Fed. Reg. 54,148 (1977) 0.0.0.0...
42 Fed. Reg. 26,429 (1977) ................0.0068.
41 Fed. Reg. 46,742 (1976) .........e.,
41 Fed. Reg. 21,402 (1976) ............
36 Fed. Reg. 10,466 (1976) 00.0...
LEGISLATIVE HISTORY:
Occupational Safety and Health Act of 1970
Subcommittee on Labor of the Senate
Committee on Labor and Public Wel-
fare, Legislative History of the
Occupational Safety and Health Act
Oe Re a RE cin inh eae
S. Rep. No. 1282, 91st Cong., 2d Sess.
So, ERO TR Ke MAE 5 xlrea home de DLC oe 4s
| RASA aR RS j
Consumer Product Safety Act of 1972
H.R. Rep. No. 1153, 92nd Cong., 2d Sess.
RPE das lesolietc ccccehna neve acne aeviMepeerate
Page
32, 41,
42, 43,
48
4, 5, 6, 7,
vii
Page
Toxic Substances Control Act of 1976
H.R. Rep. No. 1341, 94th Cong., 2d Sess.
Be BHA NR Sirota be: IVs, Not RE EOE ee 51, 52, 59
Clean Air Act Amendments of 1977
H.R. Rep. No. 564, 95th Cong., lst Sess.
OMAN hic. eersen ed Da cabicnedoks oan dd Al Beis 62
MISCELLANEOUS:
Baram, Regulation of Health, Safety And
Environmental Quality And The Use of Cost-
Benefit Analysis, Final Report to the Adminis-
trative Conference of the United States
RRR CS OR RR Is a ce em pea aba Aa Oe ET 8 25
Boyland, The Correlation of Experimental
Carcinogenesis and Cancer in Man, 11 PRo-
GRESS IN EXPERIMENTAL TUMOR RESEARCH 222
ft” aber pierre ed AR C7 eS ai OLED Bene pat eee ele a 7
BRAUN, THE BIOLOGY OF CANCER (1974) 000000000000... 17
Bridbord, Decoufle, Fraumeni, Hoel, Hoover,
Rall, Saffiotti, Schneiderman & Upton, Esti-
mates of the Fraction of Cancer in the United
States Related to Occupational Factors (US.
IN I a ee ren ach mir, Wrens 7
Cairns, The Cancer Problem, SCIENTIFIC AMERI-
CAN November 1975, at 64....00000000000occocccn. 6
Cornfield, Carcinogenic Risk Assessment, 198
SOOEMPOEE Ge COG ED occ ccckincecssccccecscscocccsscssvsccevecevserss 22
COUNCIL ON ENVIRONMENTAL QUALITY, SIXTH
FADO AL, TEMPE CATO ooo. eh ci veb vssiccceesebeebicssivescess 6, 20
Crump, Hoel, Langley & Peto, Fundamental
Carcinogenic Processes and Their Implications
for Low Dose Risk Assessment, 36 CANCER RE-
UR IN ooo cadeccss lychee ss vinccccedenatiickdeses 17
DOLL, PREVENTION OF CANCER: POINTERS FROM
BUNUN ARIE Bs odccicebscacconcscsicndvessrecssssclonevcal 7
EPA Carcinogen Assessment Group, Final Re-
port on Population Risk to Ambient Benzene
Exposures (with cover memorandum dated
ME PR aio dlvevccctats oushiiniavideevlcestionst. 65
Viil
Epstein, Environmental Determinants of Human
Cancer, 34 CANCER RESEARCH 2425 (1974) .........
Fischhoff, Cost-Benefit Analysis and the Art of
Motorcycle Maintenance, 4—Porreyx—Aterrssss
EF Pong Lessors 7. UND...
FRAUMENI (Ed.), PERSONS AT HIGH RISK OF
CANCER, AN APPROACH TO ETIOLOGY AND CON-
GOLDSTEIN, ARONOW & KALMAN, PRINCIPLES OF
DRUG ACTION: THE BASIS OF PHARMACOLOGY
EE MII ated cies atoserk Anson th dcbepeeacea eens vas
Goodrich Cuts Cost of Meeting VCM Limits,
CHEMICAL WEEK, Dec. 10, 1975 ...............000000000.....
Greek, Vinyl Chloride May Force Shortages by
1977, CHEM. & ENGINEERING NEws, Aug. 11,
ARES AR A ee a eel enc, Sa Ae ale
HIATT, WATSON & WINSTEN (Eds.), ORIGINS OF
BRUMBAN COADIGIE CRITE Dons sn cncccscndjeccccescedehsccves(eee
Higginson, /mportance of Environmental Factors
in Cancer, in ENVIRONMENTAL POLLUTION AND
CARCINOGENIC RISKS 15, (C. Rosenfield & W.
Be I, IIE sack ates nestiesectccntrasa hte aeteilee
Hoover, Detection of Environmental Cancer Haz-
ards: Enidemiological Methods, 75 J. OF THE
MEDICAL Soc. OF N.J. 746 (1978) «0.000.
Infante, Wagoner, Rinsky & Young, Leukemia
in Benzene Workers, THE LANCET, July 9, 1977,
" e_ ep ADAIR RY oa rie tc Cnr SL ae Pts SU Ra Rear
INTERNATIONAL UNION AGAINST CANCER (UICC)
CARCINOGENICITY TESTING: A REPORT OF THE
PANEL ON CARCINOGENICITY OF THE CANCER
RESEARCH COMM’N (1969) .o0.0000.0000cccccccccccceeceee,
Kramer & Mutchler, The Correlation of Clinical
and Environmental Measurements For Work-
ers Exposed to Vinyl Chloride, 33 AM. INDUS.
HYGIENE ASS’N J. 19 (1972)...
Page
7,14, 18
14, 20
10
14
20
ix
Mancuso, Comments for Opening of Discussion o
“Neoplastic Effects,” 246 ANNALS OF THE N.Y.
ACADEMY OF SCIENCES 251 (1975) 0000000000...
Mantel & Schneiderman, Estimating “Safe” Lev-
els, a Hazardous Undertaking, 35 CANCER RE-
rte MUTAGENICITY, AND TERATOGENICITY
i REI AES CLERIC 9 Bal rh Sa EN 25 Piet [ea oe
NATIONAL ACADEMY OF SCIENCES, AN EVAL-
UATION OF THE CARCINOGENICITY OF CHLOR-
DANE AND HEPTACHLOR (1977) 2.0000.
NATIONAL ACADEMY OF SCIENCES, DECISION
MAKING FOR REGULATING CHEMICALS IN THE
ENVIRONMENT (1975) o0oo.o20oo0ccccccccccoccceccceececceeee.
NATIONAL ACADEMY OF SCIENCES, DECISION
MAKING IN THE ENVIRONMENTAL PROTECTION
PE, Sey Sai PS Fo wee st
NATIONAL ACADEMY OF SCIENCES, DRINKIN
WATER AND HEALTH (1977) 220000000000.
NATIONAL ACADEMY OF SCIENCES, PRINCIPLES
FOR EVALUATING CHEMICALS IN THE ENVIRON-
| | ENTE ALGAE eee IG St ea
NATIONAL ACADEMY OF SCIENCES, SACCHARIN:
TECHNICAL ASSESSMENT OF RISKS AND BEN-
RUE EAE AL er Ae Ay OO
OFFICE OF TECHNOLOGY ASSESSMENT (U.S. Con-
GRESS), CANCER TESTING TECHNOLOGY AND
I REE ce ef
PVC Rolls Out of Jeopardy, Into Jubilation,
CHEMICAL WEEK, Sept. 15,1976...
Rall, Problems of Low Doses of Carcinogens, 64 J.
WASH. ACAD. SCI. 63 (1974) 020000.
REPORT OF THE SECRETARY’S COMMISSION ON
PESTICIDES AND THEIR RELATIONSHIP TO
Page
14, 18
o2
23, 24
14, 18, 19
14
15, 22, 23
14
24
22
14
REPORT OF THE SURGEON GENERAL’S AD Hoc
COMMITTEE ON THE EVALUATION OF LOW LEV-
ELS OF ENVIRONMENTAL CARCINOGENS (1970) ..
ROWE, AN ANATOMY OF RISK (1977) ........... ees
Scalia & Goodman, Procedural Aspects of the
Consumer Product Safety Act, 20 U.C.L.A. L.
ee IIe pts coh cscs nahargh asl css
Schneiderman, Decoufle & Brown, Thresholds for
Environmental Cancer: Biological and Statisti-
cal Considerations,—ANNALS OF THE N.Y.
ACADEMY OF SCIENCES—(in press) ..........0000........
Schneiderman, Mantel & Brown, From Mouse to
Man--—Or How to Get From the Laboratory to
Park Avenue and 53rd Street, 246 ANNALS OF
THE N.Y. ACADEMY OF SCIENCES 237 (1975).......
Tomatis, Agthe, Bartsch, Huff, Montesand, Sa-
racci, Walker & Wilburn, Evaluation of the
Carcinogenicity of Chemicals: A Review of the
Monograph Program of the International
Agency for Research in Cancer, 38 CANCER
IEE T CROP UIED kvecckcovccecnsss adcanetbncesecaeswratsicss
Upton, Speech to American Cancer Society Sci-
ence Writers’ Seminar, Mar. 25-28, 1979............
URBAN ENVIRONMENT CONFERENCE (Eds.),
ENVIRONMENTAL CANCER: CAUSES, VICTIMS,
co Eh I teen Sete nena ely Pee Oana Sa
Weinstein, Yamaguchi, Gebert & Kaighn, Use of
Epithelial Cell Cultures for Studies on the
Mechanism of Transformation by Chemical
Carcinogens, 11 IN VITRO 130 (1975) 0000.
Weisburger, Chemical Carcinogens, in Tox-
ICOLOGY 333 (Casarett & Doull eds. 1975)...........
WORLD HEALTH ORGANIZATION, EVALUATION OF
THE CARCINOGENIC HAZARDS OF Foop ADDI-
TIVES, Technical Report No. 220 (1961) ...............
WORLD HEALTH ORGANIZATION, PREVENTION OF
CANCER, Technical Report No. 276 (1974) ..........
Page
14, 18
23
62
18
72
21
16
14, 20
IN THE
Supreme Court of the United States
OCTOBER TERM, 1978
No. 78-911
INDUSTRIAL UNION DEPARTMENT, AFL-CIO,
Petitioner,
Le
AMERICAN PETROLEUM INSTITUTE, et al.,
Respondents;
No. 78-1036
RAY MARSHALL, SECRETARY OF LABOR
Petitioner,
Wy
AMERICAN PETROLEUM INSTITUTE, et al.,
Respondents.
On Writ of Certiorari to the United States
Court of Appeals for the Fifth Circuit
BRIEF FOR NATURAL RESOURCES DEFENSE
COUNCIL, INC. AS AMICUS CURIAE
2
INTEREST OF AMICUS CURIAE
NATURAL RESOURCES DEFENSE COUNCIL
The Natural Resources Defense Council, Inc.
(“NRDC”)—a national environmental organization with
more than 44,000 members residing in all states and
territories, as well as abroad—has as one of its principle
objectives the prevention of environmentally-caused dis-
ease. The prevention of cancer caused by exposure to man-
made chemicals is among the highest of NRDC’s priorities,
in view of the fact that it is responsibly estimated that the
majority of the cases of this dreaded disease are caused by
such chemicals.
In pursuit of this objective, NRDC closely monitors
the development of federal cancer policies in the Depart-
ment of Labor, the Environmental Protection Agency,
and other federal agencies that are responsible for
protecting the public from exposure to cancer-causing
substances in the workplace and general environment.
Recently, for example, NRDC submitted detailed scien-
tific and policy comments on the proposed general cancer
policy of the Department of Labor. NRDC has also played
a similar role in the development of the general cancer
policies of the Environmental Protection Agency and the
Food and Drug Administration.
NRDC has also played an important part in a number
of administrative proceedings relating to specific toxic
substances with carcinogenic properties. NRDC . was
among the first to call to the government’s attention the
scientific evidence that synthetic fluorocarbons damage
the stratospheric ozone layer, which protects the world’s
people from solar radiation that causes skin cancer. This
evidence later become the basis for executive and legisla-
tive actions to limit nonessential uses of fluorocarbons.
NRDC also brought to the attention of the Food and Drug
Administration the potential cancer-causing properties of
3
certain kinds of plastic soft drink bottles, which that
agency has since decided should not be used.
In the Occupational Safety and Health Act, and a
number of other major laws, the Congress of the United
States has responded to the growing national concern over
the rapid increase in, and grim results of, exposure to
cancer-causing substances. NRDC and its members have a
substantial interest in preserving the integrity of these
laws, in view of the importance of environmental factors
in causing cancer. In addition, NRDC and its members
have an interest in the outcome of this case because the
basic legal issues presented here permeate the entire body
of the Congressional enactments to control the incidence
of cancer. This Court will provide an authoritative inter-
pretation of the Occupational Safety and Health Act,
which will affect as well the many other federal laws
designed to reduce the toll from cancer.
Regarding the issues presented here, NRDC believes
the Secretary properly interpreted the Occupational
Safety and Health Act when he concluded that it com-
mitted to his discretion the question whether quantifying
the magnitude of the risk of cancer-causing substances,
and balancing the costs and benefits of proposed regu-
lations, could be done with sufficient precision to provide
useful guidance in establishing the preventive standards
required by the Act. In addition, we believe the Secretary
properly exercised this discretion when he concluded that
these techniques were too immature to inform his judg-
ment meaningfully. As discussed in this brief, we agree
with the Secretary that quantitative risk assessment, and
cost-benefit analysis, are now too primitive to be thrust
into the prominent role envisioned by the court below. If
the objectives of the law are to be achieved, it is essential
that this Court uphold the Secretary’s authority to act to
prevent cancer on the basis of the best available evidence,
rather than forcing the public to wait for the development
d
of a degree of precision that is, at least at present, beyond
the capabilities of science.
All parties to both cases consented to the filing of a
brief by NRDC as amicus curiae. Copies of the letters of
consent are filed with this brief.
QUESTIONS PRESENTED
1. Whether the Occupational Safety and Health Act
provides authority for the Secretary of Labor to limit
employees’ exposure to carcinogenic chemicals to the ex-
tent feasible on the basis of qualitative evidence that they
cause cancer, rather than requiring him to withhold regu-
lation until he has information sufficiently detailed to
permit quantitative assessment of the magnitude of the
health risk, and until he performs a quantitative cost-
benefit analysis of contemplated standards.
2. Whether the Secretary acted within his discretion
when he determined that the techniques of quantitative
risk assessment and cost-benefit analysis would not use-
fully sharpen his judgment in establishing a standard to
prevent cancer induced by workplace exposure to benzene.
INTRODUCTION AND STATEMENT OF THE CASE
A. The Problem of Occupational Disease.
In 1970, Congress determined that the devastating
toll of deaths, illnesses and injuries caused in American
workplaces must be drastically reduced. Occupational
accidents alone then claimed more than 14,500 lives and
disabled at least 2.2 million people each year.’ Each year
chemicals and dusts in the workplace caused at least
1S. REP. No. 1282, 91st Cong., 2 Sess. 2 (1970) [hereinafter
cited as SENATE REPORT].
q
77
390,000 new cases of occupational illness.2 Congress con-
cluded that the engine of American prosperity had to be
made safer for those who operated it. To this end
Congress enacted the Occupational Safety and Health
Act,3 whose primary purpose is “to assure so far as possible
every working man and woman in the Nation safe and
healthful employment and places of employment.’’
The legislators set special emphasis on controlling the
careless use of toxic substances in the workplace, which are
causing cancer, respiratory and heart diseases, nervous
disorders, sterility and other killing and disabling dis-
eases. The Senate and House Committees cited chemical
dangers running the gamut from the “ancient” poisons
such as lead and mercury to the newest exotic products of
the Petrochemical Age:
|M Jaterials long in industrial use are only now
being discovered to have toxic effects. In addi-
tion, technological advances and new processes
in American industry have brought numerous
new hazards to the workplace. Carcinogenic
chemicals, lasers, ultrasonic energy, beryllium
metal, epoxy resins, pesticides, among others, all
present incipient threats to the health of work-
ers. Indeed, new materials and processes are
being introduced into industry at a much faster
rate than the present meager resources of
occupational health can keep up with. It is
estimated that every 20 minutes a new and
2 Id. at 3. The Committee stated: “In the field of occupa-
tional health, the view is particularly bleak, and, due to the lack
of information and records, may well be considerably worse than
we know.” Jd. at 2.
3PuB. L. No. 91-596, 84 Stat. 1591 (codified at 29 U.S.C.
§§ 651-678).
4 Id. §2(b), 29 U.S.C. §651(b).
6
potentially toxic chemical is introduced into in-
dustry. New processes and new sources of
energy present occupational health problems of
unprecedented complexity.5
Congress took especially seriously the developing edge
of scientific knowledge showing alarming trends in
occupationally induced disease:
Recent scientific knowledge points to hith-
erto unsuspected cause-and-effect relationships
between occupational exposures and many of the
so-called chronie diseases—cancer, respiratory
ailments, allergies, heart disease, and others. In
some instances, the relationship appears to be
direct: asbestos, ionizing radiation, chromates,
and certain dye intermediaries, among others,
are directly involved in the genesis of cancer. In
other cases, occupational exposures are impli-
cated as contributory factors. The distinction
between occupational and non-occupational ill-
nesses is growing increasingly difficult to de-
fine.6
Since 1970, these findings and predictions have been
amply borne out.
The chemical hazard most on the public’s and Con-
gress’ mind, then and now, is cancer. One American in
four will contract some form of cancer; one in five is
expected to die from it.” The concensus of scientific experts
is that the majority of cancer is caused by exposure to
5 SENATE REPORT, supra note 1, at 2.
8 Jd.
7 COUNCIL ON ENVIRONMENTAL QUALITY, SIXTH ANNUAL RE-
PORT 12 (1975); Cairns, The Cancer Problem, SCIENTIFIC AMERI-
CAN November 1975, at 66.
7
chemical substances.® At least a fifth—perhaps a third or
more—of all cancer in the United States is related to
chemicals in the workplace. A grim list of specific
occupational chemicals has been identified as carcinogenic
and continues to grow.'° Benzene, which causes leukemia,
is one such substance.
B. The Occupational Safety and Health Act
The Occupational Safety and Health Act authorizes
the Secretary of Labor to set standards to protect employ-
ees from toxic substances. The Act provides for three
kinds of standards for such substances. As an interim
measure, section 6 (a) of the Act directed the Secretary by
1972 to promulgate as federally enforceable standards any
existing “national consensus standards.’’'' These are safe-
ty and health standards developed before 1970, generally
by industry-funded organizations, and voluntarily agreed
to by individual companies.'2 In 1971 the Secretary
8 World Health Organization, Prevention of Cancer, Tech-
nical Report No. 276 (1974); Higginson, Jmportance of
Environmental Factors in Cancer in ENVIRONMENTAL POLLU-
TION AND CARCINOGENIC RISKS 15, 17 (C. Rosenfield & W. Davis,
eds. 1975); Boyland, The Correlation of Experimental Carcinoge-
nesis and Cancer in Man, 11 PROGRESS IN EXPERIMENTAL TUMOR
RESEARCH 222, 223 (1969); Epstein, Environmental Deter-
minants of Human Cancer, 34 CANCER RESEARCH 2425 (1974).
See also R. DOLL, PREVENTION OF CANCER: POINTERS FROM
EPIDEMIOLOGY (1967).
8 Bridbord, Decoufle, Fraumeni, Hoel, Hoover, Rall, Saf-
fiotti, Schneiderman & Upton, Estimates of the Fraction of
Cancer in the United States Related to Occupational Factors (U.S.
HEW 1978).
10 See Cole & Goldman, Occupation, in PERSONS AT HIGH
RISK OF CANCER, AN APPROACH TO ETIOLOGY AND CONTROL 167
(1975). [hereinafter cited as PERSONS AT HIGH RISK OF CAN-
CER].
1129 U.S.C. §655(a).
12 SENATE REPORT, supra note 1, at 5-6.
8
adopted some 400 such standards for air contaminants,
including the existing 10 parts per million (ppm) stan-
dard for benzene.’ These standards merely codified con-
temporary industry self-regulation. Generally they were
set at levels necessary to prevent only the acute and
chronic effects of moderately high exposures. They were
not designed to protect against the long-term adverse
effects of lower doses, which largely had not been in-
vestigated. This was the case for the benzene standard,
which was aimed only at some of the chemical’s non-
malignant effects.'4 The Senate and House Committees
made clear that national consensus standards were not
substitutes for more protective standards when further
research demonstrated the need.'5
Section 6(c) empowers the Secretary to set without
rule-making “emergency temporary standards” when
necessary to protect employees from “grave danger” due
to new or newly discovered hazards.'* These are effective
immediately upon promulgation, and must be replaced by
so-called “permanent” standards within six months,
which must be developed through normal rulemaking
procedures.'? In May 1977, The Secretary promulgated an
emergency temporary standard of one ppm for benzene
upon concluding that it causes leukemia. '®
1336 Fed. Reg. 10,466 (1971), codified in 29 C.F.R.
§1910.1000 table Z-1.
1443 Fed. Reg. 5919 (1978) (preamble to the Secretary’s
benzene standard).
1S SENATE REPORT, supra note 1, at 6; H.R. Rep. No. 1291,
91st Cong., 2d Sess. 17 (1970) [hereinafter cited as HOUSE
REPORT}. ;
16 Occ. Safety & Health Act §6(c)(1), 29 U.S.C. §655(¢) (1).
17 Id. §86(c)(2)-(3), 29 U.S.C. §655(c) (2)-(3).
18 42 Fed. Reg. 26429 (1977). The standard never became
effective. It was stayed by the Fifth Circuit, American Petro-
leum Inst. v. OSHA, No. 77-1516 (stay ordered, May 20, 1977),
and the case was rendered moot by the promulgation of the
final standard and the judicial challenge here under review.
9
The main vehicle for protecting employees from toxic
substances is the so-called “permanent standard,” estab-
lished under Section 6(b), the provision at issue in this
litigation.'9 This Act sets out the criteria for such stan-
dards in the detailed provisions of Section 6(b)(5). The
Secretary must set the standard for such a substance
which most adequately assures, to the extent
feasible, on the basis of the best available evi-
dence, that no employee will suffer material
impairment of health or functional capacity even
if such employee has regular exposure to the
hazard dealt with by such standard for the
period of his working life. Development of
standards under this subsection shall be based
upon research, demonstrations, experiments,
and such other information as may be appropri-
ate. In addition to the attainment of the highest
degree of health and safety protection for the
employee, other considerations shall be the latest
available scientific data in the field, the feasibil-
ity of the standards, and experience gained
under this and other health and safety laws.20
Section 6(b) provides for informal rulemaking, supple-
mented by an opportunity for a public hearing.2' Section
6(e) requires the Secretary to give a statement of reasons
with each standard.22 Section 6(f) provides for pre-
enforcement review of such standards in the Courts of
Appeals; the court must uphold the Secretary’s determi-
nations if they are supported by substantial evidence.23
19 29 U.S.C. §655(b).
20 Jd. §655(b) (5).
21 Id. §655(b)(1)-(4).
22 Id. §655(e).
23 Id. §655(f).
10
C. The Benzene Standard
The Secretary promulgated a section 6(b)(5) per-
manent standard for benzene in February 1978, after
extensive comment and hearings.24 The standard reduced
the permissible benzene exposure from 10 parts per mil-
lion (“ppm’’) to one ppm.?5 The Secretary accompanied the
standard with a detailed statement of the reasons for this
reduction. He based the requirement both on benzene’s
clearly demonstrated leukemogenicity and on its capacity
to cause a wide variety of non-malignant rc ei of the
blood-forming system. The most recent, Was by the
National Institute for Occupational Safety and Health,26
of employees in Ohio exposed to benzene in the manufac-
ture of Pliofilm.2?7 The study showed at least a five-fold
excess of leukemia over the unexposed population in these
workers.28 The Secretary concluded that because benzene
was a carcinogen, a safe level of exposure could not be
demonstrated. Finally, he concluded that the reduction to
one ppm would produce “appreciable” health benefits, in
terms of reduced risk of both cancer and other benzene-
induced disease.29
Facing a life-threatening substance for which no safe
level of exposure could be established, he determined that
under section 6(b)(5) he must reduce benzene exposure to
“the extent feasible.”’%0 After carefully estimating the
2443 Fed. Reg. 5918 (1978), 29 C.F.R. §1910.1028.
25 29 C.F.R. §1910.1028(c¢).
26The Occupational Safety and Health Act established
NIOSH as an institute in the National Institutes of Health,
under the Department of Health, Education and Welfare, and
gives it duties to perform research and to aid the Secretary in
setting standards. Occ. Safety & Health Act §§20-22, 29 U.S.C.
§§669-671.
27 Infante, Wagoner, Rinsky & Young, Leukemia in Ben-
zene Workers, THE LANCET, July 9, 1977, at 76.
28 Id.
22943 Fed. Reg. 5940 (1978).
30 Jd. at 5934, 5940-41.
11
total costs of compliance by the industries making and
using benzene, the Secretary concluded that a permissible
exposure level of one part per million was feasible, as, even
under the industries’ cost assumptions, “the economic
impact of the final benzene standard will not be such as to
threaten the financial welfare of the affected firms or the
general economy.’’3'
The Secretary rejected the claims—then raised under
the rubric of “feasibility” —that prior to setting a stan-
dard to protect employees from an undisputed carcinogen,
he must quantify the cancer risk reduction the standard
would achieve and show through a quantitative cost-
benefit analysis that the standard’s health benefits out-
weighed its economic costs. For reasons detailed below, he
concluded that techniques do not exist to quantify risks
with accuracy and reliability necessary to be of use to him
in making his decision.32 Moreover, he concluded that no
basis exists to determine the value of life and health in
dollars, so that no objective method of conducting a
quantitative cost-benefit analysis existed. Neither form
of quantitative analysis would “significantly clarify the
ultimate decision.’’33
The Secretary observed that the Occupational Safety
and Health Act does not require him to delay regulation of
known cancer-causing substances until precise quan-
titative knowledge of the number of cancers that will
occur at particular doses can be established.
Rather, he stated that he must act now to protect the
employees from known causes of killing and disabling
illness. As the House Report stated, he must act on the
basis of the best available evidence and must not be
“paralyzed by debate surrounding diverse medical opin-
ions.”34 When facing a life-threatening substance for
31 Jd. at 5939.
32 Jd. at 5940-41.
33 Jd.
34 HOUSE REPORT, supra note 15, at 18.
12
which no safe dose can be determined, the Secretary
determined that the Act directs him to reduce employee
exposure to the extent feasible. Citing prior cases,°5 he
concluded that a standard which closed whole industries or
forced a substantial fraction of employers in an industry
out of business would not be feasible. But short of this
drastic result, he concluded employees are entitled by the
Act to protection from such hazards.%6
D. The Lower Court Decision
The American Petroleum Institute and other trade
organizations and individual companies sought review of
the standard in the United States Court of Appeals for the
Fifth Circuit. The court vacated the standard in October
1978.37
The lower court decision held, for the first time, that
in setting a standard under section 6(b)(5) for an undis-
puted carcinogen, the Secretary must quantify the health
benefits the standard would provide and show through a
quantitative cost-benefit analysis that the benefits bear a
“reasonable relationship” to the costs. The court based
this holding not on the requirement that achieving the
standard be “feasible” as required by Section 6(b) (5), but
on Section 3(8), which defines an occupational safety and
health standard as a standard “which requires conditions,
or the adoption or use of one or more practices, means,
methods, operations, or processes, reasonably necessary or
appropriate to provide safe or healthful employment and
places of employment.”3* As additional bases for the
court’s conclusion that the Secretary must have quan-
35 Industrial Union Dep’t AFL-CIO v. Hodgson, 499 F.2d
467 (D.C. Cir. 1974); AFL-CIO v. Brennan, 530 F.2d 109 (8rd
Cir. 1975).
36 43 Fed. Reg. 5934 (1978).
37 American Petroleum Inst. v. OSHA, 581 F.2d 493 (5th
Cir. 1978).
38 Jd. at 502-03; 29 U.S.C. §652(8).
13
titative knowledge of the size of cancer risks at particular
doses before he may set a standard, the court pointed to
four phrases in Section 6(b)(5)—“the best available evi-
dence,” “research, demonstrations, experiments, and such
other information as may be appropriate,” “the latest
available scientific data in the field,” and “experience
gained under this and other health and safety laws.’’39
Finally, although the court conceded the Serretary
had acted rationally, it held the Secretary’s corciusion
that “appreciable” health benefits would flow from the
standard not to be supported by substantial evidence. The
court held that substantial evidence could be provided only
by a quantitative assessment of the risks.40
‘E. Scientific and Policy Issues Related to the Reliability
of Quantitative Risk Assessment and Quantitative
Cost-Benefit Analysis
A correct appraisal of the requirements of the Act
and the reasonableness of the Secretary’s actions depends
on a firm understanding of the techniques of risk assess-
ment and quantitative benefit analysis. Weighing the
costs and benefits of a proposed regulation of the sort
under review requires being able to quantify the magni-
tude of the health and economic interests at stake, and to
know the appropriate way to balance them.
1. The Best Available Scientific Knowledge Per-
mits Reliable Identification of Carcinogens, But
Not Reliable Quantification of Risk
The main tools for identifying chemicals that cause
cancer are studies of exposed human populations (epide-
39581 F.2d at 504; Occ. Safety & Health Act §6(b)(5), 29
U.S.C. §655(b) (5).
40581 F.2d at 503.
14
miology)4' and laboratory studies on animals.42 These
techniques provide a well-recognized basis for identifying
chemicals posing carcinogenic risk for humans. The vali-
dity of these techniques has been confirmed repeatedly for
two decades by expert scientists and scientific in-
stitutions.43 They have been accepted by Congress,44 the
41 Hoover, Detection of Environmental Cancer Hazards:
Epidemiological Methods, 75 J. OF THE MEDICAL Soc. OF N.J. 746
(1978) [hereinafter cited as Hoover]; Hoover, Cancer Epide-
miology, in ENVIRONMENTAL CANCER: CAUSES, VICTIMS, SOLU-
TIONS 4 (1978) [hereinafter cited as ENVIRONMENTAL CANCER];
Mancuso, Comments for Opening of Discussion of “Neoplastic
Effects,” 246 ANNALS OF THE N.Y. ACADEMY OF SCIENCES 251
(1975) [hereinafter cited as Mancuso]. See generally articles in
PERSONS AT HIGH RISK OF CANCER, supra note 10, and in
ORIGINS OF HUMAN CANCER, Book A (Hiatt, Watson, & Winsten
eds. 1977) [hereinafter cited as ORIGINS OF HUMAN CANCER].
42 CARCINOGENICITY TESTING, A REPORT OF THE PANEL ON
CARCINOGENICITY OF THE CANCER RESEARCH COMM’N, INTER-
NATIONAL UNION AGAINST CANCER (UICC) (1969); Saffiotti,
Identifying and Defining Chemical Carcinogens, in ORIGINS OF
HUMAN CANCER, supra note 41, Book C, at 1311.
43 E.g., REPORT OF THE SURGEON GENERAL’S AD Hoc Com-
MITTEE ON THE EVALUATION OF LOW LEVELS OF ENVIRONMENTAL
CARCINOGENS (1970), reprinted in Chemicals and the Future of
Man, Hearings before the Subcomm. on Executive Reorganization
and Government Research of the Senate Comm. on Government
Operations, 92d Cong., Ist Sess. 180 (1971). [hereinafter cited as
SURGEON GENERAL’S REPORT]; NATIONAL ACADEMY OF SCIENCES,
DRINKING WATER AND HEALTH 29-59 (1977) [hereinafter cited
as DRINKING WATER AND HEALTH]; OFFICE OF TECHNOLOGY
ASSESSMENT (U.S. CONGRESS), CANCER TESTING TECHNOLOGY
AND SACCHARIN (1977); NATIONAL ACADEMY OF SCIENCES, AN
EVALUATION OF THE CARCINOGENICITY OF CHLORDANE AND
HEPTACHLOR (1977) [hereinafter cited as NAS CHLOR-
DANE/HEPTACHLOR EVALUATION]; NATIONAL ACADEMY OF SCI-
ENCES, PRINCIPLES FOR EVALUATING CHEMICALS IN THE
ENVIRONMENT (1975); MINISTRY OF HEALTH & WELFARE (CAN-
ADA), THE TESTING OF CHEMICALS FOR CARCINOGENICITY,
MUTAGENICITY, AND TERATOGENICITY (1975); REPORT OF THE
SECRETARY’S COMMISSION ON PESTICIDES AND THEIR RELATION-
SHIP TO ENVIRONMENTAL HEALTH (U.S. HEW 1969). For an
early report reaching these conclusions, see World Health Or-
ganization, Evaluation of the Carcinogenic Hazards of Food
Additives, Technical Report No. 220 (1961).
44 See e.g., Toxic Substances Control Act, §4, 15 U.S.C. §2603,
which requires testing for carcinogenicity, among other effects.
15
agencies,*5 and the Courts of Appeals*¢ as a valid basis for
preventive regulation.
But at this time these techniques do not permit
reliable quantification of human risk, particularly at and
below levels to which employees are currently exposed.
Even where a qualitative causal relationship is clearly
visible, as is the case for benzene, reliable quantitative risk
estimates for humans cannot be made. For a given
substance the risk estimates from equally respected quan-
titative models—each based on different plausible
assumptions about how chemicals cause cancer—differ by a
factor of 5,000,000 or more.47
Facts such as these have led some of the nation’s
leading scientific experts in cancer cause and prevention
to conclude that quantitative risk assessment (and thus
cost-benefit balancing) are not useful aids to standard-
setting at this point. Dr. Arthur C. Upton, who is Director
of the nation’s anti-cancer program at the National
45 See e.g., the Secretary’s Proposed Carcinogen Policy, 42
Fed. Reg. 54,148 (1977), and the Environmental Protection
Agency’s Interim Carcinogen Policy, 41 Fed. Reg. 21,402 (1976).
46 See e.g., Environmental Defense Fund v. EPA, —F.2d—, 8
ELR 20765 (D.C. Cir. 1978) (PCBs, toxic water pollutants);
American Iron & Steel Inst. v. OSHA, 577 F.2d 825 (3rd Cir.
1978) (coke oven emissions); Society of the Plastics Industry v.
OSHA, 509 F.2d 1301 (2d Cir. 1975), cert. denied 421 U.S. 992
(1975) (vinyl chloride); Synthetic Organic Chemical Mfrs. Ass’n
v. Brennan, 503 F.2d 1155 (8rd Cir. 1974) (ethyleneimine);
Industrial Union Dep’t AFL-CIO v. Hodgson, 499 F.2d 467 (D.C.
Cir. 1974) (asbestos); Certified Color Mfrs. Ass’n v. Matthews,
548 F.2d 284 (D.C. Cir. 1976) (food color additives);
Environmental Defense Fund v. EPA, 510 F.2d 1291 (D.C. Cir.
1975) (pesticides aldrin/dieldrin).
47 NATIONAL ACADEMY OF SCIENCES, SACCHARIN: TECHNICAL
ASSESSMENT OF RISKS AND BENEFITS 3-72, Table 3-8 (1978)
[hereinafter cited as NAS SACCHARIN STUDY].
16
Cancer Institute, summarized their conclusions recently as
follows:
Based on current scientific knowledge it is
not possible to accurately calculate risks for most
substances. Various recently developed statisti-
cal methods attempt to extrapolate from animal
data to human risk, thereby computing an esti-
mate of how many persons in a given population
can be expected to get cancer from a given
dosage of a particular carcinogen. But the pro-
cess has severe limitations.
We have the scientific means to identify
substances which cause cancer in animals and
which are, therefore, potentia! carcinogens for
humans. There is as yet, however, no scientific
way to arrive at quantitative and precise risk
estimates for humans based on extrapolating
from animal data and, therefore no scientific
basis for accurately and confidently weighing
the benefits of a particular substance against its
risks as a potential carcinogen.48
The basic biochemical mechanisms of chemically-
induced cancer are as complicated as life itself. They are
not adequately understood, and the uncertainties will
undoubtedly be removed only through years of research,
depending largely on unpredictable, sereridipitous insight.
In the interim, however, we have the means to identify
qualitatively many carcinogenic chemicals.
a. The Basic Biochemical Processes of Cancer
Causation are Largely Unknown
The biological mechanism by which chemicals cause
cancer is not well understood. There is general agreement
48 Upton, Speech to American Cancer Society Science
Writers’ Seminar, Mar. 25-28, 1979, at 2-3.
17
that chemical carcinogenesis involves changes in the gen-
etic material of an individual cell, or small group of cells,
or in the mechanisms by which cells regulate growth and
differentiationg. These changes cause the affected cells to
grow without restraint, without performing the functions
of normal cells, and eventually threatening the life of the
host.49
These processes appear to be enormously complex.
Some chemicals appear to accomplish these changes di-
rectly. Most seem to need to be “activated” by being
metabolized by the body into the actually carcinogenic
substances. Still other chemicals apparently “potentiate”
or otherwise facilitate the action of others. Which chem-
icals operate in which of these manners is not always
known.‘
In addition, cancers usually appear to be the result of
a sequence of steps: a series of changes caused in the same
cells by repeated exposures to the same chemical, by a
sequence of exposures to different substances, or a complex
interaction of chemicals with genetic susceptibility, vi-
ruses or radiation. The theory of sequential steps suggests
that cancer may be caused even by past exposure to a
chemical long since eliminated from the body; that chem-
ical may have been responsible for an early step in a
process completed by other factors.5'
49 See BRAUN, THE BIOLOGY OF CANCER (1974); Weinstein,
Yamaguchi, Gebert, & Kaighn, Use of Epithelial Cell Cultures for
Studies on the Mechanism of Transformation by Chemical
Carcinogens, 11 IN VITRO 130 (1975).
50 See Weisburger, Chemical Carcinogens, in TOXICOLOGY 333
(Casarett & Doull eds. 1975) [hereinafter cited as Weisburger }.
51 Crump, Hoel, Langley & Peto, Fundamental Carcinogenic
Processes and Their Implications for Low Dose Risk Assessment,
36 CANCER RESEARCH 2973 (1976) [hereinafter cited as Crump
et al.}.
18
The complexity of the processes by which a chemical
gets from its point of entry (e.g., lungs, skin, digestive
system) to its point of damage adds further uncertainty.
These absorption, distribution, and excretion processes are
rarely more than partly known.®2 It also appears that
there are certain chemical reactions that can “detoxify”
some amounts of some carcinogens and perhaps certain
biological mechanisms that can repair or destroy some
damaged cells. These mechanisms do not appear to be
fully effective, however.‘
Moreover, the human population is genetically, geo-
graphically, and culturally heterogeneous. For this rea-
son, it is thought that people vary in sensitivity to a
particular carcinogen. It appears that genetic hetero-
geneity alone accounts for a substantial variation in
human response. Subgroups exposed to other substances
because of their diet, or place of residence or work, may be
especially sensitive to particular substances.*®4
These apparent properties of chemically-induced can-
cer have led to a virtual consensus among expert scien-
tists, committees, and institutions that “thresholds” for
carcinogens—i.e., doses for which no human cancer risk
exists—cannot be determined.55 While theoretically a
threshold dose may exist for a given person exposed to a
given substance in particular conditions, the enormous
variation in sensitivities and in past and future exposures
52 GOLDSTEIN, ARONOW, & KALMAN, PRINCIPLES OF DRUG
ACTION: THE BASIS OF PHARMACOLOGY 129-226 (2d ed. 1974).
53 See id. at 623-667. See also Weisburger, supra note 50.
54 DRINKING WATER AND HEALTH, supra note 43, at 39; see
generally articles in PERSONS AT HIGH RISK OF CANCER, supra
note 10.
55 See e.g., SURGEON GENERAL’S REPORT, supra note 43;
Schneiderman, Decoufle & Brown, Thresholds for Environmental
Cancer: Biological and Statistical Considerations, —ANNALS OF
THE N.Y. ACADEMY OF SCIENCES— (in press); DRINKING WATER
AND HEALTH, supra note 43, at 37-48, NAS CHLOR-
DANE/HEPTACHLOR EVALUATION, supra note 43.
19
to this and other substances is thought to preclude
determining reliable thresholds for whole populations.
For the same reasons, the size of the risk at particular
doses may vary greatly for different people. Even if the
size of the risk for a given person in given circumstances
could be demonstrated, one would not necessarily know
the size of the risk for members of a larger group.‘
b. Epidemiological Studies Are Capable of Identi-
fying Carcinogens, But Not the Magnitude of
Cancer Risks
Epidemiological studies take advantage of the
unfortunate fact that groups of people are already ex-
posed to a wide variety of chemicals. The most fruitful
studies are of occupational groups exposed to chemicals to
which the general public is not exposed, or only at much
lower levels. Studies of this type, for example, supplied
the principal basis for identifying benzene as a cause of
leukemia. The Infante study of workers in Ohio exposed
to benzene in the manufacture of Pliofilm and numerous
other studies discussed in the preamble to the benzene
standard’ demonstrate beyond doubt that benzene causes
leukemia.
But epidemiological studies are relatively insensitive.
Because of the small size of most occupational group
studies, an effect generally will not be detected unless it is
occurring in five to ten percent of the population. If the
effect is relatively common, such as lung cancer, and has
many chemical causes, an even greater response is needed
from one chemical in order to be detected. For the most
part, people are exposed to too many substances at un-
known doses for unknown periods to permit the drawing
56 DRINKING WATER AND HEALTH, supra note 43, at 39-47;
Crump et al., supra note 51.
57 43 Fed. Reg. 5918, 5925-32 (1978).
20
of statistically reliable conclusions from epidemiological
studies about the magnitude of the risk.5¢
The same factors that render epidemiological studies
insensitive as a means of identifying carcinogens make
them extremely unreliable for quantifying the risks of the
carcinogens they do identify.59 Some of the most impor-
tant factors are as follows.
First, sufficient latency periods often have not run to
reveal the full effect. Latency in humans ranges from 15-
40 years.® Second, dose information is often highly specu-
lative, especially for periods of two or more decades ago,
when exposure began. It is often difficult to determine
even which workers in a factory were exposed, let alone at
what level.6' Third, not all exposed persons have been
followed and their fate determined. Their illnesses and
causes of death, or even whether they still live, is not
determined. Since ill employees are the most likely to
have left employ, the failure to follow them can seriously
understate cancer incidence. Fourth, common cancers are
often not related to their causes. Lung cancer, for
example, appears to have many chemical causes. It is more
difficult to relate this cancer to a specific chemical than one
which is extremely rare and readily traced to a particular
source. ®2
58 See Hoover, supra note 41; ENVIRONMENTAL CANCER,
supra note 41.
59 See sources in note 58, supra.
60 COUNCIL ON ENVIRONMENTAL QUALITY, SIXTH ANNUAL
REPORT 23 (1975).
61 The benzene studies provide an example of this problem.
See discussion at 43 Fed. Reg. 5927. Sometimes doses two or
more decades past can be reconstructed only speculatively. For
an effort to do such reconstruction, see Kramer & Mutchler, The
Correlation of Clinical and Environmental Measurements For
Workers Exposed to Vinyl Chloride, 33 AM. INDUS. HYGIENE
Ass’N J. 19 (1972).
62 See sources in note 58, supra.
21
c. Animal Experiments Also Permit Identi-
fication of Carcinogens, But Not the Degree of
Risk to Humans
Lifetime studies on rodents are the major source of
data on the carcinogenicity of chemicals. The animals’
response to these chemicals is considered essentially sim-
ilar to that of humans, so that a substance carcinogenic to
one species is likely to be carcinogenic to the others.
Nearly all of the chemicals whose human carcinogenicity
has been proven epidemiologically have also been shown to
cause cancer in animals. Most of the 65,000 chemicals in
commerce pose no cancer risks. But there are at least
several hundred chemicals that have been identified as
carcinogens in animals, for which adequate epidemiologi-
cal studies are unavailable.®
But animal studies cannot reliably indicate the
magnitude of risks to people. To investigate directly the
effects of the doses of carcinogen commonly experienced
by people would require huge experiments. For example,
to demonstrate with 95 percent confidence that a realistic
dose of just one substance causes fewer than one cancer in
a million subjects would require a test involving at least
6,000,000 animals. Such ‘“mega-mouse” experiments
would be impractically expensive.®
Scientists attempting to estimate the risks from com-
monly encountered doses use mathematical models of
dose-response relationships to extrapolate from ex-
63 Tomatis, Agthe, Bartsch, Huff, Montesand, Saracci,
Walker & Wilburn, Evaluation of the Carcinogencity of Chem-
icals: A Review of the Monograph Program of the International
Agency for Research in Cancer, 38 CANCER RESEARCH 877 (1978);
Saffiotti, Jdentifying and Defining Chemical Carcinogens, in
ORIGINS OF HUMAN CANCER, supra note 41, at 1311.
64 Mantel & Schneiderman, Estimating “Safe” Levels, a
Hazardous Undertaking, 35 CANCER RESEARCH 1379, 13838
(1975).
22
perimental results on animals at higher doses. However,
reflecting the lack of knowledge of precisely what is
happening at and above the cellular level, the different
models yield widely divergent estimates of the risk associ-
ated with each low dose. While the major mathematical
models all agree with observed results, they can differ by a
factor of 100,000 on the size of the dose that creates a risk
of one cancer in a million animals.
Our ignorance of how to translate dose-response esti-
mates across species lines adds more uncertainties to risk
estimates. Not enough is known to determine confidently
the relative sensitivity of humans and each species or
strain of test animals. For example, people are bigger
than animals; there is uncertainty whether the proper
way to scale up doses is on the basis of relative body
weights or on the basis of relative surface areas of tissue.
Moreover, in contrast to human genetic heterogeneity
and diverse chemical exposures, test animals are inbred
for genetic homogeneity in order to facilitate uniform,
reproducible test results, and they are generally exposed
to only one substance at a time. Consequently, there may
be subgroups of the human population far more sensitive
than these inbred strains, and there may be interactions
between chemicals that the animal tests do not reveal.¢®7
Recent attempts by a committee of the National
Academy of Sciences to quantify the risks from saccharin
graphically illustrate the uncertainties. Using the results
of the recent animal studies, the various plausible models,
85 Cornfield, Carcinogenic Risk Assessment, 198 SCIENCE 693,
694 (1977).
86 Rall, Problems of Low Doses of Carcinogens, 64 J. WASH.
ACAD. ScI. 63 (1974). See also NAS SACCHARIN STUDY, supra
note 47, at 3-68 to 3-72 and table 3-8.
67 See DRINKING WATER AND HEALTH, supra note 43, at 35;
SCHNEIDERMAN, MANTEL & BROWN, From Mouse to Man—Or
How to Get From the Laboratory to Park Avenue and 58rd Street,
246 ANNALS OF THE N.Y. ACADEMY OF SCIENCES 237 (1975).
23
and the differing assumptions about how to cross species
line to humans, they attempted to quantify the effects of a
consuming one can of diet soda per day. The results
suggested that in each 50 million people there could be as
few as 0.0007 cancers per year or as many as 3640. This is
a range of error more than 5,000,000- fold.
2. Techniques for Quantifying Economic Impact
Forecasting the costs of controlling or avoiding ex-
posure to a carcinogen is also difficult. Adequate control of
carcinogens sometimes requires significant changes in
equipment and operating procedures. The precise cost of
improved measures to control exposure is difficult to
determine, and the uncertainty increases with the degree
of change. Moreover, as costs increase, consumers of the
dangerous chemical tend to use less of it. They shift to
safer chemicals or to other means of accomplishing the
purpose the dangerous chemical serves. The magnitude of
these effects and their economic value are extremely
difficult to predict. Over time consumers (whether in-
dustries or individuals) adjust and learn new ways of
satisfying their needs with less of a dangerous chemical or
with none at all. Thus the actual costs may turn out to be
smaller than first predicted after some time for in-
novation and adjustment.®9
Moreover, objective analysis of economic impacts is
hampered by the fact that often the parties with the most
accurate information on the costs and effects of con-
trolling a carcinogen have little incentive to make it
available.70 The industries making and using benzene, for
example, have the best access to technological and eco-
68 NAS SACCHARIN STUDY, supra note 47, at 3-72, table 3-8.
69 The uncertainties are illustrated by graphs in W. ROWE,
AN ANATOMY OF RISK 233-34 (1977).
70 See NATIONAL A.CADEMY OF SCIENCES, DECISION MAKING
IN THE ENVIRONMENTAL PROTECTION AGENCY 51-58 (1977)
[hereinafter cited as DECISION MAKING IN EPA]
24
nomic data to analyze the availability of substitutes and
likely consumer responses, yet they may not feel that
revealing it is in their commercial interest while a stan-
dard is under consideration.
An example illustrates the point. In 1974 the in-
dustries subject to the Secretary’s standard for vinyl
chloride stated that the one part per million standard for
that carcinogen was technologically impossible to meet
and that attempting to meet it would drive them into
bankruptcy. The industries urged their positions without
success before the Secretary, before the Second Circuit
Court of Appeals,’' and before this Court.72 Yet within a
month of the failure of their legal challenge, the com-
panies began to announce dramatic technological ad-
vances and to drop their cost estimates to a fraction of
their prior levels.73 Polyvinyl chloride plastic today re-
mains strongly competitive with other plastics despite
compliance with the standard.”4
3. Comparing Health and Economic Interests
There is no acceptable way of reducing health effects
to dollars to allow comparison with the costs of controlling
exposure. Economists have proposed a number of
methodological approaches to such an evaluation.’5 But,
71 See Society of the Plastics Industry v. OSHA, 509 F.2d
1301 (2d Cir. 1975), cert. denied 421 U.S. 992 (1975).
72 Id.
73 Greek, Vinyl Chloride May Force Shortages by 1977, CHEM.
& ENGINEERING NEws, Aug. 11, 1975, at 8, 9. See also Goodrich
Cuts Cost of Meeting VCM Limits, CHEMICAL WEEK, Dec. 10,
1975, at 59.
74 PVC Rolls Out of Jeopardy, Into Jubilation, CHEMICAL
WEEK, Sept. 15, 1976, at 34.
75 The various methods were surveyed and criticized by the
Secretary in the preamble to his standard for coke oven emis-
sions, 41 Fed. Reg. 46,750-51. For a concise summary of methods
and problems, see also DECISION MAKING IN EPA, supra note 70,
at 228-34.
25
none of the techniques can evaluate the questions of
fairness which arise when the risks and benefits fall to
different people, nor can they adequately evaluate the
present value of preventing disease in the future.’6 In the
end, they founder on the most basic ethical tenets, whic
turn away from consciously placing a monetary value on
human life.
SUMMARY OF ARGUMENT
The Occupational Safety and Healti: Act empowers
the Secretary of Labor to adopt highly protective stan-
dards in order to reduce the devastating toll of deaths,
illnesses, and injuries in American workplaces. The Act
places special emphasis on controlling exposure to chem-
icals that cause cancer and other fatal and disabling
diseases among employees. Section 6(b)(5) of the Act
directs the Secretary to set the standard for such a
substance
which most adequately assures, to the extent
feasible, on the basis of the best available evi-
dence, that no employee will suffer material
impairment of health or functional capacity even
if such employee has regular exposure to the
hazard ... for the period of his working life.7’
Congress recognized that with effects such as cancer,
the hazards of chemicals in the workplace can be identi-
fied well before their precise magnitude can be known.
Long latency periods, subtle cause-and-effect relation-
76 See generally Baram, Regulation of Health, Safety And
Environmental Quality And The Use of Cost-Benefit Analysis,
Final Report to the Administrative Conference of the United
States (1979); Fischhoff, Cost-Benefit Analysis and the Art of
Motorcycle Maintenance, -Bopreesteeatyrsts-64eeGe_
77 29 U.S.C. §655(b) (5). @ Policy Sevewces 177 61472)
26
ships, and complex interactions preclude having reliable
quantitative data until long after the availability of re-
liable qualitative information, In the interim, because
employees’ lives and health are at stake, the Act per-
mits—indeed requires—the Secretary to take action to
prevent cancer, on the basis of such qualitative informa-
tion, the best evidence then available.
By directing the Secretary to act on the basis of the
best available evidence, the statute commits to him the
determination of when, if ever, scientific knowledge has
progressed sufficiently to allow reliable quantification. As
long as he may reasonably determine such techniques are
not yet reliable, the Act neither requires the Secretary to
postpone taking protective action, nor demands that he
undertake a sham quantitative analysis.
In this case, the Secretary has reasonably determined
that quantitative risk assessment techniques do not re-
liably indicate the size of the cancer hazard. Accordingly,
he concluded he must establish a standard for benzene, an
undisputed carcinogen, using his best judgment about
what is needed to protect employees, on the basis of the
best qualitative evidence. He has not chosen to act blind-
folded; rather, he has determined that quantitative risk
assessment will not improve his vision.
Although the health of employees is the Secretary’s
paramount concern, Congress also recognized the need for
practical considerations in setting standards. Con-
sequently, the Act required the Secretary to set “feasible”
standards. There is, however, nothing to support and
much to refute the notion that the feasibility require-
ment, or any other provision of the Act, imposes on the
Secretary the obligation to show through a quantitative
analysis that the benefits of a standard outweigh its costs.
Congress recognized the severe ethical and methodological
limitations of such quantitative cost-benefit analysis. Not
the least of these are the usual absence of reliable quan-
27
titative information about risks, and the absence of any
acceptable way to place a dollar figure on the value of a
life.
Despite the requirements and poiicies of the law, the
lower court held that the Secretary may not adopt a
standard to reduce the cancer hazard from benzene, unless
he performs a quantitative assessment of the risk reduc-
tion to be gained, and provides a quantitative cost-benefit
analysis demonstrating that these benefits bear a “rea-
sonable relationship” to the costs to employers of com-
plying with the standard.78
To reach its desired result, the court had to turn the
terms and legislative history of the Occupational Safety
and Health Act on their heads. Holding that “the best
available evidence” and related phrases require quan-
titative risk assessment, the court imposed on them mean-
ings exactly opposite to what their legislative history
indicates.7? Contrary to the court’s conclusion, these
phrases require preventive action while risks are still
debated, rather than block it.
Recognizing it could find no support for quantitative
cost-benefit, 1h *the term “feasible” or its legislative his-
tory, the court poured this meaning into an innocuous
provision—“reasonably necessary” in the definition of
standards—that Congress never contemplated would bear
on the issue. The court also constructed a faulty parallel
with the Consumer Product Safety Act.% The parallel
would be of no help even if real, since the authors of the
latter Act explicitly rejected quantitative cost-benefit
analysis.
78 American Petroleum Inst. v. OSHA, 581 F.2d 493, 503
(5th Cir. 1978).
79 Td. at 504.
80 Jd. at 502-03.
28
In its zeal to require quantitative risk assessment and
cost-benefit analysis, the court bent the statutory
“substantial evidence” test for review into a barrier to
achievement of the Act’s protective policies. While the
court allowed that the Secretary’s decision was rational, it
held that without a quantitative risk assessment this was
not enough.*' To so read the substantial evidence test
would cancel the Secretary’s authority—clearly indicated
by the remainder of the Act—to act on the basis of
qualitative evidence when he reasonably determines that
it is the best available.
The Fifth Circuit’s decision conflicts with all other
cases reviewing the validity of standards promulgated
under the Occupational Safety and Health Act, from the
District of Columbia, Second, Third, and Eighth Circuits.
It would prevent the Secretary from effectively fulfilling
his statutory mandate to protect employees from
chemically -induced cancer and other killing and disabling
diseases. It would place requirements on the Secretary
that Congress did not impose, and that the Secretary
reasonably determined were inappropriate. For these rea-
sons, the decision should be reversed.
ARGUMENT
I. THE OCCUPATIONAL SAFETY AND HEALTH ACT
DOES NOT REQUIRE QUANTITATIVE JUSTIFICA-
TION FOR STANDARDS TO PROTECT EMPLOY -
EES FROM CANCER-CAUSING CHEMICALS
The Occupational Safety and Health Act particularly
emphasizes standards to protect employees from cancer
and other killing and disabling diseases caused by toxic
81 Jd. at 508.
29
chemicals and dusts in the workplace. Section 6(b)(5)
directs the Secretary to set the standard
which most adequately assures, to the extent
feasible, on the basis of the best available evi-
dence, that no employee will suffer material
impairment of health or functional capacity even
if such employee has regular exposure to the
hazard dealt with by such standard for the
period of his working life. Development of
standards under this subsection shall be based
upon research, demonstrations, experiments,
and such other information as may be appropri-
ate. In addition to the attainment of the highest
degree of health and safety protection for the
employee, other considerations shall be the latest
available scientific data in the field, the feasibi-
lity of the standards, and experience gained
under this and other health and safety laws.®2
Recognizing the long delays frequent between ex-
posure to carcinogens and illness, the subtle nature of
many cause-and-effect relationships, and the fact that
employees’ lives and health are at stake, Congress deter-
mined that a policy of minimizing exposure to these
substances as much as feasible was the only one consistent
with a prudent regard for public health. While obviously
it would be desirable if more exect quantitative data were
available on the risks of carcinogens and the costs of
controlling them, the Occupational Safety and Health Act
clearly reflects Congress’ conclusion that the protection of
public health from cancer-causing substances should not
await the development of techniques needed to provide it.
In effect, the Act represents a judgment that the potential
human and economic costs to our nation of providing too
little protection from this dread disease are too high to
warrant awaiting more precise information, even though
82 29 U.S.C. §655(b) (5).
30
the standards the Secretary adopts may entail substantial
expense, require technological innovation, or even cause
significant economic change within an mdustry.
The lower court has proposed an entirely different
approach from the precautionary one Congress adopted.
It would prohibit the Secretary from adopting standards
to protect employees from carcinogenic substances until
he provided quantitative estimates of the reduction in
risks associated with the proposed standard, and a quan-
titative analysis showing that the benefits outweighed the
costs. This position, contrary to the understanding of all
four of the other Courts of Appeal that have reviewed the
validity of standards adopted under the Act, is as unwise a
public health policy as it is contrary to the intention of the
legislators who wrote the Act.
In attempting to justify its proclivity for quan-
tification, the lower court could offer nothing more than a
“But see” citation to the contrary decisions of the other
circuits8? and a lame assertion that the issues in those
essentially identical cases were somehow different.® Sim-
ilarly, it was forced to turn the Occupational Safety and
Health Act inside out, disregarding the plain meaning
and legislators’ understanding of the terms Congress
chose to address the proper manner cf evaluating health
evidence and the role of economic factors in standard-
setting, while pouring its own notions of the role of costs
and risk assessment into terms not meant to contain
them.
At some point in the future, techniques for quan-
tifying the magnitude of risks from workplace carcino-
gens and comparing the costs and benefits of alternative
standards may be developed which are capable of increas-
83 American Petroleum Inst. v. OSHA, 581 F.2d 493, 504
(5th Cir. 1978).
84 Jd. at 505.
31
ing the precision of occupational safety and health stan-
dards. But to this point Congress has not been persuaded,
and the lower court has no power to substitute its own
conclusions for those of the legislature.
A. Conditioning Standards on Quantitative Assessments
of Risk Conflicts with the Congress’ Desire to Prevent
Disease Before It Develops.
Since cancer is a disease that resists a cure, pre-
venting it is one of the most basic purposes of the
Occupational Safety and Health Act. Thus the Act em-
powers—indeed requires—the Secretary to prevent ex-
posure to carcinogens before they take their toll in human
life. Since cancer is a disease that develops only after a
long latency period and one whose mechanisms are not
precisely known, prevention necessarily requires acting
before all the data are available to assess precisely the
magnitude of the risk. All too often, definite proof that a
substance causes cancer, let alone evidence to assess the
magnitude of its potency, can be found only after
unacceptable damage to public health has already oc-
curred—after some people have already contracted killing
or disabling disease and many more have been irreversibly
exposed.
It is abundantly clear in the Congressional debates
accompanying the passage of the Occupational Safety and
Health Act that Congress understood the import of these
considerations and intended that action should occur
without waiting until the numerical toll could be assessed.
1. The Primacy of the Goal of Preventing
Occupationally-Caused Disease
The legislators who wrote the Occupational Safety
and Health Act repeatedly stressed the seriousness of the
problem of occupationally-induced disease, and their
conclusion that rapid preventive action was both required
and practical. Section 6(b)(5) reflects this urgency
32
throughout, requiring that each standard “most ade-
quately assure|[ |,” to the degree feasible, “that no employ-
ee will suffer material impairment of health or functional
capacity,” even after life-long exposure.® It requires the
Secretary to seek “the highest degree of health and safety
protection for the employee.”%6
Moreover, the statute directs the Secretary to act on
“the best available evidence” and “the latest~Sctehtific
data in the field.’’®? By these phrases, the House Report
said, the legislators intended the Secretary to take pre-
ventive action promptly without waiting until the evi-
dence necessary to determine the precise magnitude of
the threat eventually becomes available: “The Committee
feels it is vital that when the Secretary sets an
occupational health standard he does so on the basis of the
best available professional evidence; it is not intended that
the Secretary be paralyzed by debate surrounding diverse
medical opinions.’’®8 By contrast, not a single Senator or
Representative even suggested an intent to require quan-
titative assessment of the risks before the Secretary
acted.
The legislative history of the Act shows that Congress
clearly. recognized the killing and disabling properties of a
growing number of chemicals and dusts once thought to
be safe, and the problems of latency and statistical sub-
tlety that impede early discovery of the magnitude of
risks.8° These factors added urgency to taking protective,
85 29 U.S.C. §655(b) (5).
86 Jd. The references to “employee” in the singular empha-
size the goal of protecting all employees, even those more
sensitive than average. See HOUSE REPORT, supra note 15 at 18.
87 29 U.S.C. §655(b) (5).
88 HOUSE REPORT, supra note 15, at 18 (emphasis added).
89 See e.g., Statement of Senator Saxbe, reprinted in SuB-
COMM. ON LABOR OF THE SENATE COMM. ON LABOR AND PUBLIC
WELFARE, LEGISLATIVE HISTORY OF THE OCCUPATIONAL SAFETY
AND HEALTH ACT OF 1970 338-39 (1971) [hereinafter cited as
LEG. HIST. }.
33
preventive action as soon as a substance’s dangers are
identified, without waiting for years until the precise
contours of the danger are fully researched.
The Senate Report on the bill emphasized the growing
pattern of discovering life-threatening dangers even in
familiar substances, as well as in the new chemicals of the
post World War II period:
[| M jaterials long in industrial use are only now
being discovered to have toxic effects. In addi-
tion, technological advances and new processes
in American industry have brought numerous
new hazards to the workplace. Carcinogenic
chemicals, lasers, ultrasonic energy, beryllium
metal, epoxy resins, pesticides, among others, all
present incipient threats to the health of work-
ers. Indeed, new materials and processes are
being introduced into industry at a much faster
rate than the present meager resources of
occupational health can keep up with. It is
estimated that every 20 minutes a new and
potentially toxic chemical is introduced into in-
dustry. New processes and new sources of
energy present occupational health problems of
unprecedented complexity.”
The Report took especially seriously the developing edge
of scientific knowledge on occupationally-induced disease:
Recent scientific knowledge points to hitherto
unsuspected cause-and-effect relationships be-
tween occupational exposures and many of the
so-called chronic diseases—cancer, respiratory
ailments, allergies, heart disease, and others. In
some instances, the relationship appears to be
90 SENATE REPORT, supra note 1, at 2.
34
direct: asbestos, ionizing radiation, chromates,
and certain dye intermediaries, among others,
are directly involved in the genesis of cancer. In
other cases, occupational exposures are impli-
cated as contributory factors. The distinction
between occupational and non-occupational ill-
nesses is growing increasingly difficult to de-
fine.9"
In addition, the Report observed that “the modern worker
encounters health hazards involving complex, often
synergistic interactions of numerous physical and chem-
ical agents. . . .’’92
The Committee cited the need to set standards forth-
with to protect employees from the chemicals, metals, and
dusts named above, as well as “ancient” poisons such as
lead and mercury. The Committee did not consider
quantification of the risk a precondition to regulation;
employee protection from identified killers should not be
postponed solely because their precise potency was uncer-
tain.9%
The House Report reiterated the same themes. It
cited the dangers of these same chemicals and the rapidly
growing knowledge of new dangers. Moreover, the Com-
mittee stated:
|WlJorkers are in daily contact with literally
hundreds of new chemicals and formulations
about which little is known. As occupational
health research becomes more _ sophisticated,
there is every indication that the toxicity of the
fumes, gases, and chemicals to which workers are
exposed will be discovered by researchers.%
91 Jd.
92 Jd. at 19.
93 Jd. at 2-4.
94 HOUSE REPORT, supra note 15, at 14-15.
% Jd. at 19.
35
As stated above, the House Committee emphasized that in
setting standards for toxic substances, the Secretary must
not be “paralyzed” by scientific uncertainties and differ-
ences of opinion.%* He must resolve the uncertainties in
favor of protecting the employees.
2. Judicial Recognition of the Primacy of Pre-
venting Disease
Except for the Fifth Circuit, the Courts of Appeals
that have reviewed OSHA standards for carcinogens have
all correctly emphasized the law’s requirement to take a
precautionary, preventive view of the evidence of health
harm. They have repeatedly upheld the Secretary’s au-
thority to set standards on the basis of qualitative evi-
dence that identifies a substance as a carcinogen. Only the
Fifth Circuit has sought to make quantifying the magni-
tude of risk a necessary prerequisite to protective action.
In Industrial Union Dep't, AFL-CIO v. Hodgson,%’ the
District of Columbia Circuit upheld the standard for
asbestos. The Secretary’s knowledge of the effects of
asbestos precisely paralleled his knowledge about benzene
in the case under review here: he had firm evidence that
asbestos ‘ai a carcinogen, although the precise effects of
particular doses was not known. The court observed that
the statute required the Secretary to base standards upon
conclusions drawn from well-established data and from
information “on the frontiers of scientific knowledge.”%
The court concluded:
[I]n this case the evidence indicated that reliable
data is not currently available with respect to the
precisely predictable health effects of various levels
of exposure to asbestos dust; nevertheless, the Sec-
96 Jd. at 18.
97 499 F.2d 467 (D.C. Cir. 1974).
98 Jd. at 474.
36
retary was obligated to establish some specific level
as the maximum permissible exposure. After con-
sidering all the conflicting evidence, the Secre-
tary explained his decision to adopt, over strong
employer objection, a relatively low limit in
terms of the severe health consequences which
could result from over-exposure. Jnasmuch as the
protection of the health of employees is the over-
riding concern of [the Act], this choice is doubtless
sound. .. .99
The Second Circuit faced the identical issue reviewing
the standard for viny! chloride. This chemical had been
shown to cause cancer in workers exposed to relatively
high doses (around 200 parts per million) and in animals
at lower doses (down to 50 ppm). In Society of the Plastics
Industry v. OSHA,’ the court rejected the claim that the
Secretary could not reduce exposures to one ppm without
evidence of human harm down to that level. The court
acknowledged that “though the factual finger points, it
does not conclude.” 'o' Nonetheless, ‘‘Under the command
of [the Act], it remains the duty of the Secretary to act to
protect the workingman, and to act even in circumstances
where existing methodology or research is deficient.” '°2
The Third Circuit upheld the Secretary’s standard for
ethyleneimine, another carcinogen, in Synthetic Organic
Chemical Manufacturers Ass'n v. Brennan,‘ on the basis
of two animal studies and the judgment of committees of
cancer experts that current knowledge showed no level of
exposure to a carcinogen can'€onsidered completely safe. 194
The court found no requirement that the Secretary be able
99 Jd. at 475 (emphasis added).
100 509 F.2d 1301 (2d Cir. 1975), cert. denied, 421 U.S. 992
(1975).
101 Jd. at 1308.
102 Jd.
103 503 F.2d 1155 (3rd Cir. 1974).
104 Jd. at 1157-61.
37
to quantify the risk. And in American Iron & Steel
Institute v. OSHA,‘ the court held that the Secretary had
ample evidence, from epidemiological studies of coke oven
workers and expert opinion on the issue of safe doses for
carcinogens, to conclude that coke oven emissions were
carcinogenic and no level of exposure could be considered
risk - free, 106
The lower court decision is starkly inconsistent with
those of its sister circuits. These courts have correctly
perceived that statutory phrases such as “the best avail-
able evidence”'°7 permit—even require—the Secretary to
take protective action when he discovers that a substan:e
is a carcinogen, without waiting for the development of
the capability to quantify the magnitude of the threat to
life. Yet the Fifth Circuit turned these phrases on their
heads, interpreting them to block preventive action. 10
This reading ignores their plain meaning and the legisla-
tive history indicating that the very purpose of such
105 577 F.2d 825 (3rd Cir. 1978).
106 Jd. at 831-32.
'07 Oce. Safety & Health Act § 6(b), 29 U.S.C. §655(b) (5).
‘08 American Petroleum Inst. v. OSHA, 581 F.2d 493, 504
(5th Cir. 1978). Besides “the best available evidence” and “the
latest scientific data in the field,” the court pointed to two other
phrases in section 6(b)(5): “research, demonstrations, ex-
periments, and such other information as may be appropriate,”
and “experience gained under other health and safety laws.” If
the last phrase indicates anything at all, it cuts against the
lower court, for the weight of experience under other laws
upholds the validity of the Secretary’s approach to carcinogen
regulation. See e.g., Environmental Defense Fund v. EPA, —
F.2d —, 8 ELR 20765 (D.C. Cir. 1978) (toxic water pollutants);
Certified Color Mfrs. Ass’n v. Mathews, 543 F.2d 284 (D.C. Cir.
1976) (food color additives); Environmental Defense Fund v.
EPA, 510 F.2d 1291 (D.C. Cir. 1975) (pesticides). As for
“research ...and such other information as may be appropri-
ate,” this phrase most fairly read envisions the use of just such
epidemiological studies as the ones on which the Secretary
relied.
38
phrases was to permit the Secretary to act to protect
employees from known dangers without being “paral-
yzed” by the lack of scientific certainty on the extent of
the hazard.'°9 Yet rather than address the inconsistency of
its position with the law and the other circuits, the court
lamely refused even to discuss the contrary cases.'1°
In sum, the Act and the legislative history empower
the Secretary not only to find facts about chemicals’
toxicity, but also to make judgments where necessary to
resolve scientific unknowns. The power to make such
judgments is a necessary consequence of the Act’s man-
date that the Secretary set standards for chemicals iden-
tified as carcinogens. To be sure, the Secretary cannot
regulate every existing substance simply because it might
be harmful; he must have a firm basis for identifying a
substance as dangerous, such as epidemiological or animal
studies. But where qualitative identification of risk is best
that can be done, Congress did not pose an impossible
barrier to regulation.
Shortly after the decision in this case, the District of
Columbia Circuit upheld toxic water pollutant standards
for polychlorinated biphenyls (PCBs), which have pro-
duced cancer in laboratory animals. The court stated that
when faced with claims of substantial economic impact,
the temptation to demand that the agency fur-
nish conclusive proof of carcinogenicity as sup-
port for the regulations is great. However, the
decision to delegate authority to an agency to
control suspected carcinogens is a legislative
judgment that is not open to question in this
court. Congress’ direction to EPA to protect
against incompletely understood dangers could
not be carried out if we were to adopt the proof
requirements advocated by industry petitioners.
109 HOUSE REPORT, supra note 15, at 18.
110 American Petroleum Inst. v. OSHA, 581 F.2d 4938, 505
(5th Cir. 1978).
39
What scientists know about the causes of
cancer is how limited is their knowledge. The
record in this case demonstrates that it may take
decades for human exposure to carcinogens to
result in cancer; in the meantime, the case for
inferring a cancer danger with respect to an
incompletely understood substance is vigorously
disputed. If regulation were withheld until the
danger was demonstrated conclusively, untold
injury to public health could result.'"'
The Fifth Circuit has succumbed to this temptation and
substituted its judgment for that of Congress.
B. The Requirement that Standards be Feasible Does Not
Mandate a Quantitative Demonstration that their
Benefits Exceed Their Costs.
While the purpose of the Occupational Safety and
Health Act was to decrease drastically the incidence of
workplace-induced disease, its authors recognized that
practicality imposed limits on the achievement of this
goal. Thus in section 6(b)(5) they limited the measures
that the Secretary might require to reduce exposure to
cancer-causing substances to those that were “feasible.”
While the use of the term feasibility obviously implies that
the Secretary must consider the economic cost of proposed
regulations to employers when he adopts workplace ex-
posure standards for carcinogens, the legislative history
demonstrated that Congress did not intend the consid-
eration of economic costs to impede vigorous action to
prevent occupationally-induced disease. In short, while
recognizing that a wholly risk-free working environment
might not always be practical, the legislators did not
111 Environmental Defense Fund v. FPA, —F.2d—, —, 8
ELR 20765, 20780 (D.C. Cir. 1978).
40
contemplate an unrealistic quantitative demonstration
that the benefits of a proposed standard outweighed its
costs to industry.
1. The Legislative History Of The Term “Feasible”
Shows That Congress Intended Only A Limited
Role For Cost Considerations In Setting Stan-
dards
Section 6(b)(5) directs the Secretary to set standards
‘that prevent impairment of employees’ health “to the
extent feasible.”''2 The legislative history shows that all
congressional discussion of the role of economic consid-
erations in setting standards was structured around this
feasibility requirement. Nothing in this provision or its
legislative history can be construed as requiring quan-
titative cost-benefit analysis.
Tracing the meaning of the feasibility requirement
begins with the bill reported by the House Committee in
July 1970.1'3 This bill, the so-called Daniels bill, would have
established the employees’ right to standards providing
fully safe workplaces without any qualification:
The Secretary, in promulgating standards
under this subsection, shall set the standard
which most adequately assures, on the basis of
the best available professional evidence, that no
employee will suffer any impairment of health or
functional capacity, or diminished life ex-
pectancy, even if such employee has regular
exposure to the hazard dealt with by such stan-
dard for the period of his working life.''4
The Senate Committee, which reported a bill several
months later, added the feasibility requirement to the
112 29 U.S.C. §655(b) (5).
113 H.R. 16785, reprinted in HOUSE REPORT, supra note 15,
at 1-14.
114 Jd. §7(a)(4).
41
Daniels formulation. As reported, Section 6(b)(5) at-
tained its nearly final form:
The Secretary, in promulgating standards
under this subsection, shall set the standard
which most adequately and feasibly assures, on
the basis of the best available evidence, that no
employee will suffer any impairment of health or
functional -capacity, or diminished life ex-
pectancy even if such employee has regular ex-
posure to the hazard dealt with by such standard
for the period of his working life. Development
of such standards shall be based on research,
demonstrations, experiments, and such other in-
formation as may be appropriate. In addition to
the attainment of the highest degree of health
and safety protection for the employee, other
considerations shall be the latest available scien-
tific data in the field, the feasibility of the stan-
dards, and experience gained under this and
other health and safety laws.''5
The feasibility requirement was added in committee
at the suggestion df Senator Javits, one of the principal
architects and managers: of the bill. He explained the
amendment in his individual views attached to the Re-
port:
As a result of this amendment the Secretary, in
setting standards, is expressly required to con-
sider feasibility of proposed standards. This is an
improvement in the Daniels bill, which might be
interpreted to require absolute health and safety
in all cases, regardless of feasibility, and the
Administration bill, which contains no criteria
for standards at all.''6
115. 2193 §6 (b)(5) (Oct. 5, 1970), reprinted in LEG. HIST.,
supra note 89, at 242 (emphasis added).
116 SENATE REPORT, supra note 1, at 58.
42
The Senate Report reiterated that “standards promul-
gated under section 6(b) shall represent feasible require-
ments. .. .”"1'7
The comments of Senator Javits in the Committee
Report,''® and a colloquy on the Floor between Senators
Williams and Dominick over the final fine-tuning amend-
ment to section 6(b)(5),''9 show that Congress recognized
completely risk-free workplaces might not always be
possible to achieve. The feasibility limit was intended to
free the Secretary from the duty to set absolutely safe
standards in those situations. Congress also recognized
that it would not make sense to protect employees by
eliminating their jobs in a wholesale manner. But these
legislators’ statements, and many others in the Senate
and House Reports and floor debates, also underscore that
the standards must drastically reduce the intolerably high
rates of work-related deaths, illnesses, and injuries. '20
Congress recognized that sometimes adequate protection
17 Jd. at 7.
18 Jd. at 58.
9 Amendment No. 1058, reprinted in LEG. HIST., supra
note 89, at 502-03 (colloquy between Senators Williams and
Dominick). In his original remarks on this subject, Senator
Dominick missed the point of Senator Javits’ feasibility amend-
ment in committee, which was to relieve the Secretary of the
obligation to set standards guaranteeing complete safety in all
cases. /d. at 480-82 (remarks of Senator Dominick.) But after
changing the Committee’s “most adequately and feasibly as-
sures” to the final version’s “most adequately assures, to the
extent feasible,” (thereby expressing feasibility syntactically as
a limitation but not changing meaning), Senators Dominick,
Javits, and Williams came to full accord. Jd. at 502-03.
120 See e.g., SENATE REPORT, supra note 1, 2-4; HOUSE
REPORT, supra note 15, at 14-16. See also, LEG. HIST., supra note
89, at 411-14 (remarks of Sen. Williams), 509-11 (remarks of
Sen. Yarborough), 512-13 (remarks of Sen. Muskie), 518 (re-
marks of Sen. Nelson), 522 (remarks of Sen. Cranston), 985
(remarks of Rep. Perkins), 1212 (remarks of Rep. Daniels).
43
would entail substantial expense and sometimes would
even result in closing some businesses.'2) As Senator
Eagleton stated, the costs of assuring such protection
must be borne as “necessary costs of doing business.” He
added: “| W Je can no longer permit profits to be depend-
ent upon an unsafe or unhealthy worksite.’’'22
_ The legislative history contains not a single statement
by any member of the House or Senate suggesting that
decisions should be reached through quantitative cost-
benefit analysis. Indeed, the legislative history reveals a
strong hostility to this approach, particularly to the idea
of putting a dollar price on human life. As Senator
Yarborough, then chairman of the full committee that
reported the bill, stated: “We are talking about people’s
lives, not the indifference of some cost accountants.’’'23
2. Prior Cases Have Concluded That Costs Are To
Be Accorded a Limited Role in Setting Stan-
dards
The cases reviewing standards prior to American
Petroleum Institute were unanimous that the requirement
that standards be feasible is the limit of economic concern
in the Act. The District of Columbia, Second, and Third
Circuits all upheld the Secretary’s “lowest feasible level”
approach to setting standards for carcinogens. In addi-
tion, the Third and Eighth Circuits specifically rejected
claims that the Act requires quantitative cost-benefit
analysis.
121 Both the Senate and House Committees contemplated
that some substances were simply too dangerous to remain in
use. See discussion of betanapthylamine, a very potent carcino-
gen, SENATE REPORT, supra note 1, at 4; HOUSE REPORT, supra
note 15, at 15. See also the statement by Senator Saxbe: “It is a
real question to me as to whether asbestos can be used at all,
whether it can be handled,” reprinted in LEG. HIST., supra note
89, at 338. ;
122 LEG. HIST. at 1150-51 (remarks of Sen. Eagleton).
123 Jd. at 510 (remarks of Sen. Yarborough).
44
In Jndustrial Union Dep't, AFL-CIO v. Hodgson,'24
reviewing the asbestos standard, the District of Columbia
Circuit stated: “There can be no question that [the Act]
represents a decision to require safeguards for the health
of employees even if such measures substantially increase
production costs.’’'25 The court recognized that feasibility
imposed a practical limit on standards. Citing Senator
Javits’ comments in the Committee Report, the court
stated:
The thrust of these remarks would seem to be
that practical considerations can temper protec-
tive requirements. Congress does not appear to
have intended to protect employees by putting
their employers out of business—either by
requiring protective devices unavailable under
existing technology or by making financial via-
bility generally impossible. '26
On the other hand, short of this drastic result the costs of
employee protection must be borne. When lives are at
stake, standards cannot be restricted to the lowest com-
mon denominator that the most recalcitrant employer will
eee, Accept;
Standards may be economically feasible even
though, from the standpoint of employers, they
are financially burdensome and affect profit
margins adversely. Nor does the concept of
economic feasibility necessarily guarantee the
continued existence of individual employers. It
would appear to be consistent with the purposes
of the Act to envisage the economic demise of an
employer who has lagged behind the rest of the
124 499 F.2d 467 (D.C. Cir. 1974).
125 Id. at 477.
126 Id. at 477-78.
45
industry and is consequently financially unable
to comply with new standards as quickly as other
employers. '27
The court opined that extreme burdensomeness might be
reason for phasing requirements in over some lead time. '28
But it did not suggest even extremely burdensome stan-
dards could never be imposed. Moreover, the court em-
phasized that the Act did not permit the Secretary to
delay standards “needlessly.’’'29
In Society of the Plastics Industry v. OSHA,’ review-
ing the standard for another carcinogen, viny! chloride,
the Second Circuit, Justice Clark writing, articulated the
technology -forcing character of the Act:
In the area of safety, we wish to emphasize, the
Secretary is not restricted by the status quo. He
may raise standards which require the devel-
opment of new technology, and he is not limited
to issuing standards based solely on devices al-
ready fully developed. '3'
In AFL-CIO v. Brennan,'32 the Third Circuit agreed
that standards may require technological innovation and
may alter the economic status quo. The court stated that
although standards must be feasible,
we agree with the Second Circuit in Society of
Plastics ... that, at least to a limited extent, [the
Act] is to be viewed as a technology-forcing
piece of legislation. Thus the Secretary would
127 Jd. at 478 (footnote omitted).
128 Jd.
129 Jd. at 479.
130 509 F.2d 1301 (2d Cir. 1975), cert. denied 421 U.S. 992
(1975).
131 Id. at 1309.
132 530 F.2d 109 (3rd Cir. 1975).
46
not be justified in dismissing an alternative toa
proposed health and safety standard as in-
feasible when the necessary technology looms on
today’s horizon. '33
The Third Circuit also held that imposing substantial costs
on industry was contemplated within the notion of
feasibility. It held that standards need not be limited to
the lowest common denominator that ali employers in an
industry will accept: “Congress did contemplate that the
Secretary’s rulemaking would put out of business some
businesses so marginally efficient or productive as to be
unable to follow standards otherwise universally feas-
ible.”’'34 Moreover, the court suggested that some work was
too dangerous to be permitted, especially if economic
consequences of doing without such activities were rela-
tively mild:
We do not question that there are industrial
activities involving hazards so great and of such
little social utility that the Secretary would be
justified in concluding that their total prohibi-
tion is proper if there is no technologically fea-
sible method of eliminating the operational haz-
ard, 135
In American Iron and Steel Institute v. OSHA,'36 the
Third Circuit reiterated this approach, upholding the
feasibility of the standard to control carcinogenic coke
oven emissions. The court held that the Act permitted the
Secretary to base the standard on the performance of the
best plant in operation and on improvements that could be
reasonably forecast in its performances, by using addi-
tional technology in use elsewhere or “looming on today’s
133 Jd. at 121 (citation and footnote omitted).
134 Jd. at 123 (footnote omitted).
135 Jd. at 121.
196 577 F.2d 825 (3rd Cir. 1978).
47
horizon.’’'37 The court upheld the Secretary’s conclusion
that the cancer hazard warranted the cost of the stan-
dard, which the employers admitted they could bear with-
out significant disruption.'%* Moreover, the court approved
the Secretary’s rejection of quantitative cost-benefit
analysis. '39
The Eighth Circuit also explicitly rejected quan-
titative cost-benefit ahalysis in Arkansas-Best Freight
Systems v. Occ. Safety & Health Rev. Comm’n.\4o The court
quoted with approval the discuss‘on of feasibility in J/n-
dustrial Union Dep't and added: “The Act does not
attempt to balance as neatly as does Arkansas- Best the
costs of compliance against the cost of employees’ in-
juries.”’"41 The court ‘held that the standard there in
question was within the Secretary’s legislative grant of
power “to protect the health of employees even though
increased production costs may result.’’'42
These four courts found nothing in the statute’s
notion of feasibility or any other provision of the Act to
require quantitative risk assessment or cost-benefit ana-
lysis. In this regard they stand in stark contrast to the
decision here under review. In this case the Fifth Circuit
did not even dispute the other courts’ understanding of
the notion of feasibility. Recognizing the hopelessness of
finding support for its quantitative requirements in this
term, the court merely brushed it away. In an act of pure
judicial legislation the court poured its quantitative
requirements into the unwilling vessel of the phrase
“reasonably necessary” in section 3(8). But this provision
offers the court no support either.
197 Id. at 833-35.
138 Jd. at 836-37.
139 Jd. at 836.
140 §29 F.2d 649 (8th Cir. 1976).
141 Td, at 654.
142 Jd. at 653.
48
C. The Lower Court Erred In Holding That Quantitative
Cost-Benefit Analysis Is Mandated by the Requirement
that Standards Be “Reasonably Necessary.”
Section 3(8)'43 supplies no additional requirements
regarding the role of economic considerations in setting
standards. Instead, it merely means that the measures the
Secretary includes in a standard must be rationally con-
ris to alleviating the hazard at which the standard is
aimed.
1. The Legislative History Shows That “Reason-
ably Necessary” Has No Bearing On The Role
Of Cost Considerations
In contrast to the extensive legislative history on the
feasibility requirement, section 3(8) received absolutely
no discussion. It was an innocuous, uncontroversial provi-
sion common to every bill introduced on occupational
safety and health in the 91st Congress.'44 Some of these
bills differed substantially on the role of economic consid-
erations. Yet all bills defined standards the same way.
Thus the legislators could not have understood this provi-
sion to bear at all on this issue, let alone to create a
quantitative cost-benefit analysis requirement.
The comments of Senator Javits, one of the legislators
most concerned to assure that the Secretary would consid-
er economic aspects of standards, make this clear. Despite
the fact that all bills contained the “reasonably necessary”
provision, Javits stated that without the feasibility
amendment the Daniels bill “might be interpreted to
require absolute health and safety in all cases,” and the
143 29 U.S.C. §652(8).
144 Each bill introduced into the 91st Congress defined
occupational safety and health standards as requirements either
reasonably necessary” or “reasonably necessary or appropri-
ate” to provide safe or healthful employment and places of
employment. See the bills collected in LEG. HIST., supra note 89.
49
Administration bill would “contain{ | no criteria for stan-
dards at all.”'45 If he had attached any importance to
section 3(8), Javits could not have reached this conclusion.
2. “Reasonably Necessary” Merely Requires A
Rational Connection Between The Means And
Objectives of Standards
Section 3(8) has a far more natural meaning than the
lower court wished to force upon it. By defining standards
as measures “reasonably necessary or appropriate’ to
provide safe or healthful workplaces, this provision merely
requires the Secretary to show that measures have a
rational connection to reducing the hazard at which the
standard is aimed. Thus, it prohibits the Secretary from
imposing patently unnecessary or irrelevant require-
ments, such as fire-fighting equipment for dealing with a
non-flammable substance. There can be no serious con-
tention that this lenient test of relevance requires the
Secretary to quantify cancer risk. This conclusion would
negate the precautionary approach so clearly indicated by
the remainder of the statute and its legislative history.
D. The Consumer Product Safety Act Does Not Support
The Lower Court’s Construction Of The ‘ -cupational
Safety and Health Act.
Unable to find any support for quantitative cost-
benefit analysis in the Occupational Safety and Health
Act or its legislative history, the court turned to a com-
pletely independent statute adopted two years later. The
court asserted that section 3(8) is identical to certain
provisions in the Consumer Product Safety Act,'4® which
the court further asserted require quantitative cost-
benefit analysis. Neither assertion is correct.
145 SENATE REPORT, supra note 1, at 58.
14615 U.S.C. §§2051-2081.
50
1. The Two Acts Are Not Parallel
Section 3(8) of the Occupational Safety And Health
Act defines standards as measures “reasonable necessary
or appropriate to provide safe or healthful employ-
ment.’’'47 Sections 7(a) and 9(c)(2)(A) of the Consumer
Product Safety Act (CPSA) limit standards under that
Act to requirements “reasonably necessary to eliminate or
reduce an unreasonable risk of injury”.48 Comparing
these provisions, the court stated: ‘We decline to construe
the precisely similar requirements of these two Acts
differently or to read words out of the OSHA legisla-
tion.’’'49 In fact, the provisions are not “precisely similar,”
and the court has read words into the Occupational Safety
and Health Act.
The court failed to comprehend that “unreasonable
risk,” not “reasonably necessary,” is the phrase which
establishes the obligation to consider costs in the CPSA.
This is clearly evident in the CPSA legislative history. 15°
The phrase “unreasonable risk” is pointedly absent from
the Occupational Safety and Health Act. So rather than
read “reasonably necessary” out of the Act, the court has
read “unreasonable risk” in.
2. The Consumer Product Safety Act Does Not
Require Quantitative Cost Benefit Analysis
Even if the CPSA were parallel to the Occupational
Safety and Health Act, it would provide the lower court no
support, because the CPSA does not require quantitative
cost-benefit analysis. The House Report on this statute
specifically rejected cost-benefit analysis as an element of
147 29 U.S.C. §652(8).
14815 U.S.C. §§2056(a), 2058(c)(2)(A).
‘49 American Petroleum Inst. v. OSHA, 581 F.2d 493, 502
(5th Cir. 1978).
150 See H.R. REP. No. 1153, 92d Cong., 2d Sess. 33 (1972)
(House Report on Consumer Product Safety Act).
51
determining “unreasonable risk.” The committee made
clear that the Commission must consider the traditional
elements of the tort law reasonableness test: the probabi-
lity and gravity of harm and the effect of regulation on
the product’s utility, cost, or availability.'5' The committee
then added the following explicit caveat: “There should be
no implication, however, that in arriving at its determi-
nation the Commission would be required to conduct and
complete a cost-benefit analysis prior to promulgating
standards under this Act.’’'52 Thus, even in enacting
“unreasonable risk” statutes Congress recognizes that
formal, quantitative techniques are not adequate to the
decision-making task.
The rejection of quantitative cost-benefit analysis for
determining “unreasonable risk” is by no means an anom-
aly of the CPSA. For example, Congress reached the same
conclusion in enacting the Toxic Substances Control Act'59
in 1976. The House Report stated:
During the hearings a number of witnesses rec-
ommended that the bill include a definition of
unreasonable risk. Because the determination of
unreasonable risk involves a consideration of
probability, severity, and similar factors which
cannot be defined in precise terms and is not a
factual determination but requires an exercise of
judgment on the part of the person making it, the
Committee did not attempt a definition of such
risk. In general, a determination that a risk
associated with a chemical substance or mixture
is unreasonable involves balancing the probabi-
lity that harm will occur and the magnitude and
severity of that harm against the effect of pro-
151 Jd.
152 Jd.
163 15 U.S.C. §§2601-2629.
52
posed regulatory action on the availability to
society of the benefits of the substance or mix-
ture, taking into account the availability of
substitutes...and other adverse effects which
such proposed action may have on society.
The balancing process described above does
not require a formal benefit-cost analysis under
which a monetary value is assigned to the risks
associated with the substance and to the cost to
society of proposed regulatory action on the avail-
ability of such benefits. Because a monetary value
often cannot be assigned to a benefit or cost, such
an analysis would not be very useful.'54
The fact that these “unreasonable risk” statutes ex-
plicitly reject quantitative cost-benefit analysis makes the
lower court’s attempt to use them to shoehorn quan-
titative analysis into the Occupational Safety and Health
Act even less plausible.
II. THE SECRETARY ACTED WELL WITHIN HIS
DELEGATED AUTHORITY IN DECLINING TO
ADOPT QUANTITATIVE RISK ASSESSMENT
The Occupational Safety and Health Act establishes
the basic criteria for standards for toxic substances: the
'54H.R. REP. No. 1341, 94th Cong., 2d Sess. 13-14 (1976)
(footnote omitted, emphasis added), The footnote quoted from
an influential study on chemical regulation and decision mak-
ing:
Highly formalized methods of benefit-cost analysis can
seldom be used for making decisions about regulating
chemicals in the environment. Thus the development of
such methods should not have high priority.
NATIONAL ACADEMY OF SCIENCES, DECISION MAKING For REGU-
LATING CHEMICALS IN THE ENVIRONMENT 44 (1975) (emphasis in
original).
53
goal of no impairment of health, the duty to take pre-
ventive action on the basis of the best available evidence,
and the constraint of feasibility.‘55 But within the bounds
of these criteria, the Act delegates to the Secretary
substantial responsibilities to determine which substances
shall be covered by standards and what requirements such
standards shall contain.
In this case, the Secretary determined from repeated
epidemiological studies that benzene causes leukemia in
humans. He also concluded that there was very little
margin, if any at all, between the benzene levels at which
the chemical had already been shown to induce leukemia
in humans and the existing 10 parts per million (ppm)
standard, which had been developed years before and was
aimed only at benzene’s non- malignant effects.
Relying on the best expertise among cancer scientists,
the Secretary determined that a safe level of human
exposure to a carcinogen cannot be demonstrated, and
that the size of the cancer risk at particular doses cannot
now be established with any useful precision. He con-
cluded, however, that a 90 percent reduction in exposure
would significantly reduce employees’ risks, and that such
a reduction was feasible. Consequently, he set the new
standard at one ppm.
The Secretary’s decision was well within the statutory
criteria that bound his delegated authority. His determi-
nations were supported by both the best available scien-
tific evidence and expertise. The basis for each of his
judgments was well represented in the record of the
decision.
In cases with essentially identical issues and records,
the three other Courts of Appeals to review the Secre-
tary’s carcinogen standards have respected the Secre-
165 Occ. Safety & Health Act §6(b)(5); 29 U.S.C.
§ 655(b) (5).
54
tary’s expertise and his mandate and upheld his deci-
sions.'5¢ On the issue of quantifying risk, they have re-
spected his expert judgment that the quantitative tech-
niques are not sufficiently precise or reliable to be useful in
determining standards. In fact, these courts have recog-
nized that the Secretary would exceed his authority if he
delayed acting until cancer risks could be reliably quan-
tified, since the statute mandates standards to protect
employees from life-threatening hazards on the basis of
qualitative evidence when that is the best available.
The lower court in this case, however, held that unless
the Secretary showed quantitatively the amount by which
the new benzene standard would reduce the employees’
cancer hazard, his decision could not be supported by
substantial evidence.'5? With this holding, the lower court
simply substituted its judgment for the Secretary’s on the
question whether benzene’s cancer risk at particular doses
can be reliably quantified. But the decision whether these
techniques are sufficiently reliable that they constitute an
appropriate basis for determining permissible exposure
levels to carcinogens is a judgment committed to the
Secretary, to which, if adequately supported and rational,
the courts must defer. The lower court’s drastic over-
estimation of the accuracy and reliability of the tech-
niques for quantifying risk classicly illustrates the dan-
gers of appellate judges’ venturing into areas requiring
detailed knowledge of a highly complex body of scientific
study and thought.
'86 Industrial Union Dep’t, AFL-CIO v. Hodgson, 499 F.2d
467 (D.C. Cir. 1974); Society of the Plastics Industry v. OSHA,
509 F.2d 1301 (2d Cir. 1975), cert. denied 421 U.S. 992 (1975):
Synthetic Organic Chemical Mfrs. Ass’n v. Brennan, 503 F.2d
1155 (8rd Cir. 1974); American Iron & Steel Inst. v. OSHA, 577
F.2d 825 (3rd Cir. 1978).
'S? American Petroleum Inst. v. OSHA, 581 F.2d 493, 503
(5th Cir. 1978).
55
A. The Substantial Evidence Test Does Not Permit The
Court To Substitute Its Judgment For The Secretary’s
On The Reliability Of Quantitative Risk Assessments
1. Standard Setting Under The Occupational Safe -
ty And Health Act Involves A Mixture Of Fact-
Finding And Policy-Making
The Secretary’s determination in setting standards
for carcinogens under section 6(b) (5) consist of both fact-
finding and policy-making. He begins with a core of well-
established facts about a chemical’s toxicity. In this
instance, epidemiological studies establish firmly that
benzene causes leukemia. But his inquiry does not end
with traditional fact-finding, because the solid facts do
not supply the complete picture. The Act mandates that
the Secretary take preventive action on the best available
evidence.'5® This means that he must take into aammEh
consideration the qualitative but not completely precise
evidence of harm, and, relying on the best scientific
expertise, resolve the uncertainties as to the magnitude of
the cancer risk in the manner that most protects employ -
ees’ health.
These determinations to resolve scientific uncer-
tainties are essentially policy judgments. The power to
make these policy judgments is a necessary consequence of
the Act’s mandate to protect employees from identified
hazards on the basis of the best available evidence. Re-
stricted to traditional fact-finding, the Secretary would
only be able to protect workers from the doses of carcino-
gens that already have been shown to kill,. despite the
strong but not completely certain belief that lower doses
are dangerous too. The Secretary necessarily must have
the most discretion to make such protective policy judg-
158QOcc. Safety & Health Act §6(b)(5), 29 U.S.C.
§ 655(b) (5).
56
ments with regard to the most serious health effects, such
as cancer, and in situations where difficulties in precisely
pinning down the risks are greatest. 159
The determinations that safe doses of carcinogens
cannot be identified, that lower doses are safer than
higher ones, and that risks cannot be reliably quantified
are mixed conclusions of fact and policy. They have a firm
base in observed facts, supplemented by scientific theory
of the processes of cancer. Ultimately, however, they rest
on the judgments guided by the policy of the Act to
resolve doubts about safety in the employees’ behalf.
2. The Substantial Evidence Test Does Not Prohi-
bit The Secretary From Making Policy Judg-
ments
The Occupational Safety and Health Act provides for
basically informal rulemaking of the kind traditionally
associated with policy-setting. Section 6(b) provides for
notice-and-comment procedures, supplemented with an
opportunity for a public hearing.‘ Section 6(e) requires
the Secretary to accompany standards with a statement of
reasons!®'
This statute was one of the first to combine informal
rulemaking with the substantial evidence test for judicial
review, which is specified by section 6(f).'62 Interpreting
159 The feasibility limitation does not detract from the duty
to take a precautionary view of the evidence. Rather, the
Secretary must take feasibility considerations into account only
after appraising the risk, on the basis of the best available
evidence. One must keep distinct the process of determining
what risk exists from the process of determining what risk will
be tolerated.
160 29 U.S.C. § 655(b)(1)-(4).
161 Jd. § 655(e).
162 Jd. § 655(f).
57
this “anomalous combination,”'®: the courts reviewing
standards set under section 6(b) (5), with the exception of
the Fifth Circuit, have correctly perceived their role as
having a dual aspect. With respect to factual issues, they
must assure that the Secretary’s determinations are sup-
ported by substantial evidence. But they have understood
that the test cannot be interpreted to bar the Secretary
from making policy judgments that the provision of
informal rulemaking and the substantive mandates of the
Act so clearly indicate he must make. Consequently, with
respect to issues ‘on the frontiers of scientific know-
ledge,”’'®* these courts have recognized that the Secre-
tary’s duty is to explain the reasoning he uses to resolve
uncertainties and to show that this reasoning is consistent
with the protective policies of the Act. The essential task
of the courts as to these issues is to assure that the
Secretary acted rationally.
The District of Columbia Circuit, Judge McGowan
writing, appropriately stated the Secretary’s task and the
court’s role in Industrial Union Dep't, AFL-CIO v. Hodg-
son. 165
From extensive and often conflicting evidence,
the Secretary in this case made numerous factual
determinations. With respect to some of those
questions, the evidence was such that the task
consisted primarily of evaluating the data and
drawing conclusions from it. The court can
review that data in the record and determine
whether it reflects substantial support for the
Secretary’s findings. But some of the questions
163 The combination was so characterized in Industrial
Union Dep’t, AFL-CIO v. Hodgson, 499 F.2d 467, 473 (D.C. Cir.
1974). The legislative history leading to this combination is
traced in Associated Industries of N.Y.S. v. United States Dep’t
of Labor, 487 F.2d 342, 347-50 (2d Cir. 1973) (Friendly, J.).
164 Industrial Union Dep't, 499 F.2d at 474.
165 Jd.
58
involved in the promulgation of these standards
are on the frontiers of scientific knowledge, and
consequently as to them insufficient data is
presently available to make a fully informed
factual determination. Decision making must in
that circumstance depend to a greater extent
upon policy judgments and less upon purely fac-
tual analysis. Thus, in addition to currently
unresolved factual issues, the formulation of
standards involves choices that by their nature
require basic policy determinations rather than
resolution of factual controversies. Judicial re-
view of inherently legislative decisions of this
sort is obviously an undertaking of different
dimensions.
Policy choices of this sort are not susceptible to
the same type of verification or refutation by
reference to the record as are some factual ques-
tions. Consequently, the court’s approach must
necessarily be different no matter how the stan-
dards of review are labeled. 66
The court sensitively summarized the Secretary’s obliga-
tion as follows:
What we are entitled to at all events is a careful
identification by the Secretary, when his pro-
posed standards are challenged, of the reasons
why he chooses to follow one course rather than
another. Where that choice purports to be based
166 Jd. at 474-75 (footnotes omitted). In one of the foot-
notes the court stated: ‘Where existing methodology or research
is deficient, the agency necessarily enjoys broad discretion to
attempt to formulate a solution to the best of its ability on the
basis of available information.” Jd. at 474 n. 18, citing Permian
Basin Area Rate Cases, 390 U.S. 747, 811 (1968).
59
on the existence of certain determinable facts,
the Secretary must, in form as well as substance,
find those facts in the record. By the same token,
when the Secretary is obliged to make policy
judgments where no factual certainties exist or
where facts alone do not provide the answer, he
should so state and go on to identify the consid-
erations he found persuasive. '§”
Judge McGowan’s view has been cited with approval
by this Court. Federal Communications Comm’n v. Na-
tional Citizens Committee for Broadcasting, 436 U.S. 775,
814 (1978).
Moreover, this view has been explicitly sanctioned by
Congress. Congress has adopted the combination of basi-
cally informal rulemaking and the substantial evidence
test in several subsequent health, safety, and
environmental protection statutes.'6® An example is the
1976 Toxic Substances Control Act;'®9 its legislative history
on judicial review expressly endorses the approach of
Industrial Union Dep’t..7 Congress has determined to
continue using this combination in laws dealing with
highly technical areas where there is substantial scientific
uncertainty, even for the experts. As the District of
Columbia Circuit recognized, the achievement of the Act’s
purposes is inseparable from and dependent on the Secre-
tary’s power to make policy judgments to resolve uncer-
tainties, and the Secretary’s actions must not be reviewed
in a manner that frustrates action “where facts alone do
167 499 F.2d at 475-76.
168 F.g., Consumer Product Safety Act, 15 U.S.C. §§ 2056,
2060(c); Toxic Substances Control Act, 15 U.S.C. §§ 2605(c),
2618.
169 Jd.
170 See H.R. Rep. No. 1341, 94th Cong., 2d Sess. 55-56 (1976)
(House Report). The Committee paraphrased Judge McGowan’s
statement with approval.
60
not provide the answer.’’'7' Other courts have adopted this
formulation of their review task, or have reached func-
tionally similar formulations that require the Secretary to
demonstrate the reasonableness or rationality of his policy
judgments but do not block him from accomplishing the
statute’s purposes. '72
Other than the Fifth Circuit, the courts have also
come to understand that, at least in these highly technical
subject areas, there is little, if any, meaningful difference
171 499 F.2d at 476.
172 Judge Friendly’s analysis in Associated Industries for
the Second Circuit is quoted in note 173, infra. Also for the
Second Circuit, Justice Clark adopted the District of Columbia
Circuit’s formulation. Society of the Plastics Industry v. OSHA,
509 F.2d 1301, 1303-04 (2d Cir. 1975), cert. denied 421 U.S. 992
(1975).
The Third Circuit has stated a five-part formulation of its
review function. The principle ones bearing on this issue are
whether the Secretary has adequately set forth his reasons,
whether his reasons reflect “consideration of factors relevant
under the statute,” whether he considered “presently available
alternatives,” and, if his determination is based “in whole or in
part on factual matters subject to evidentiary development,
whether substantial evidence in the record as a whole supports
the determination.” Synthetic Organic Chemical Mfrs. Ass’n v.
Brennan, 503 F.2d 1155, 1160 (3rd Cir. 1974). See also American
Iron & Steel Inst. v. OSHA, 577 F.2d 825, 830-31 (3rd Cir. 1978).
Even the Fifth Circuit’s formulation, although not its
practice, is consistent with these. After reviewing these ap-
proaches, the court stated: “[I]t seems clear that even with the
required substantial evidence test, our review basically must
determine whether the Secretary carried out his essentially
legislative task in a manner reasonable under the state of the
record before him.” American Petroleum Inst. v. OSHA, 581
F.2d 493, 497 (5th Cir. 1978), quoting from Florida Peach
Growers Ass’n v. United States Dep’t of Labor, 489 F.2d 120, 129
(5th Cir. 1974). In practice, this court substituted its judgment
for the Secretary’s expert conclusion that quantitative risk
assessments for carcinogens are too unreliable to be the basis of
standard-setting.
61
between the substantial evidence test and the arbitrary
and capricious test normally applied to rulemaking. The
tests have converged to a point where any remaining
theoretical distinction between them simply is not admin-
istrable.'73 The essential function of review under either
test is to assure that the agency has taken into account
173 See e.g., Associated Industries of N.Y.S. v. United States
Dep’t of Labor, 487 F.2d 342 (2d Cir. 1973) (Friendly, J.):
While we have felt constrained to determine and
sustain the applicability of the substantial evidence
test, it may well be that the controversy is semantic
in some degree, at least in the context of informal
rulemaking, and that it lacks the dispositional impor-
tance that respondents imply .... In WBEN, Ine. v.
United States ..., while we applied to F.C.C. rulema-
king the “arbitrary and capricious” standard ..., it is
hard to see in what respect we would have treated the
question differently if we had been applying a
“substantial evidence” test .... While we still have a
feeling that there may be cases where an adjudicative
determination not supported by substantial evi-
dence... would not be regarded as arbitrary and
capricious, ...in the review of rules of general appli-
cability made after notice and comment rulemaking,
the two criteria do tend to converge.
Id. at 349-50 (citations and footnotes omitted). See also
Friendly, “Some Kind of Hearing,” 123 U. Penn. L. Rev. 1267,
1318-14 (1975).
In Industrial Union Dep’t, AFL-CIO v. Hodgson, 499 F.2d
467 (D.C. Cir. 1974), the court stated that which of the two tests
applied would not “affect the rigorousness of our review to the
extent the Government seems to suppose, or that petitioners
purport to fear.” Jd. at 473. In National Nutritional Foods Ass’n
v. Weinberger, 512 F.2d 688 (2d Cir. 1975), a case arising under
the Food, Drug, and Cosmetic Act, Judge Lumbard asserted
outright that the two tests are identical: “In essence I think
that when an agency engages in substantive rule-making, it
abuses its discretion (or acts arbitrarily or capriciously) if its
actions are not supported by substantial evidence.” Jd. at 705
(concurring opinion of Lumbard, J.).
(footnote continued )
62
the information before it and has acted rationally. And
the fundamental rule under either test is that courts may
not substitute their judgment for the Secretary’s. Consolo
v. Federal Maritime Comm'n, 383 U.S. 607, 619-21 (1966);
Universal Camera Corp. v. NLRB, 340 U.S. 474, 491 (1950);
SEES Citizens to Preserve Overton Park v. Volpe, 401
U.S. 402, 415-16 (1971).
In Industrial Union Dep't,'74 the court illustrated the
application of these principles as follows:
For example, in this case the evidence in-
dicated that reliable data is not currently avail-
able with respect to the precisely predictable
health effects of various levels of exposure to
asbestos dust; nevertheless, the Secretary was
obligated to establish some specific level as the
(footnote concluded )
The Conference Report on the 1977 Clear Air Act Amend-
ments cites this trend, and Associated Industries in particular,
with approval. H.R. Rep. No. 564, 95th Cong., 1st Sess. 178
(1977). °
It has been suggested that the primary function of speci-
fying the substantial evidence test with informal rulemaking
may be to restrict the agency to a record. Scalia & Goodman,
Procedural Aspects of the Consumer Product Safety Act, 20
U.C.L.A. L. Rev. 899, 935, n. 188 (1973). But even this dis-
tinction between the tests is elusive, since some recent statutes
specifying the arbitrary or capricious test also provide for such a
record. See § 307(d) of the Clean Air Act, 42 U.S.C. § 7607(d).
Under either review test, such a record must reflect both the
factual matters at issue and the principles, hypotheses, and
policies by which the agency reasons to its decision.
Thus it may be time for this Court to re-examine the
assumption, expressed in Abbott Laboratories v. Gardner, 387
U.S. 186, 143 (1967), that the tests really do define different
depths of judicial scrutiny of agency decisions, at least in highly
technical, largely legislative areas such as these.
174 499 F.2d 467 (D.C. Cir. 1974).
63
maximum permissible exposure. After consid-
ering all the conflicting evidence, the Secretary
explained his decision to adopt, over strong em-
ployer objection, a relatively low limit in terms of
the severe health consequences which could re-
sult from over-exposure. Inasmuch as the
protection of the health of employees is the
overriding concern of [the Act], this choice is
doubtless sound, but it rests in the final analysis
on an essentially legislative policy judgment,
rather than a factual determination, concerning
the relative risks of underprotection as compared
to overprotection.'75
This is the conclusion the Fifth Circuit should have
reached.
B.. The Secretary Reasonably Determined That Quan-
titative Risk Assessment Techniques Are Not Suffi-
ciently Reliable To Be A Basis For Decision
In this case, the Secretary determined that tech-
niques for quantifying cancer risk cannot yet yield esti-
mates that are sufficiently reliable to be a basis for setting
the permissible level of exposure to benzene.'76 He con-
cluded that the exercise could not “significantly clarify the
ultimate decision.'7”? He stated the essential uncertainties
about how chemicals cause cancer which preclude reliable
quantification. He stated the problems that preclude
quantitative reliance on the epidemiological studies which
identify benzene as a leukemogen. There were large
175 Jd. at 475.
176 43 Fed. Reg. 5940-41 (1978). In addition, the Secretary
explicitly incorporated into the benzene preamble, id. at 5941,
his detailed consideration of the same issue in setting the coke
oven emissions standard, 41 Fed. Reg. 46,742, 46,750-51 (1976).
177
F, Fee. ee
64
uncertainties about the exact doses to which employees
studied had been exposed. It was lik«!y that the studies
had not recorded all the cases of leukemia caused by
benzene in these employees, because the full latency period
may not yet have run since exposure began, and because
all exposed employees have not been located and exam-
ined. There could be no quantitative estimates based on
animal studies even if these estimates could be made
reliably, because benzene’s carcinogenicity has not yet
been adequately shown in an animal test. Whether
benzene interacts with other chemicals and whether it
causes other cancers less readily identified and traced to a
source than leukemia could not be established. Also, any
reliable quantitative estimate would have to take into
account benzene’s effects other than cancer, some of which
can also be fatal, and about which there are many similar
quantitative uncertainties. '78
For these reasons, the Secretary rejected the back -of-
the-envelope calculations offered by industry witnesses
and declined to generate ones of his own.'79 This decision
was reasoned and well-supported. Indeed, the Secretary
could not have reached any other conclusion consistent
with the Occupational Safety and Health Act’s mandate to
take preventive action on the basis of the best available
evidence.
The Fifth Circuit opinion shows that it simply failed
to grasp how uncertain and imprecise the quantitative
risk estimates really are. The court appears to have been
misled into concluding that such techniques could pinpoint
the size of the danger, when in fact they cannot even
78 43 Fed. Reg. 5930-32, 5940-41.
79 Td. at 5940-41.
65
provide assurance of being within several orders of
magnitude of true risks. 18
In any event, it is not the court’s role to decide de
novo whether the Secretary’s conclusion was correct. The
court can only overturn his rejection of quantitative risk
180 American Petroleum Inst. v. OSHA, 581 F.2d 493, 503-05
(5th Cir. 1978). In a footnote, the court referred to two
particular items. /d. at 504 n. 23. The court cited a quantitative
risk assessment performed for benzene by a unit in the
Environmental Protection Agency. But this early version of the
EPA document lacked a passage present in the final draft,
which explains some of the heroic assumptions that the authors
made concerning the substantial uncertainties. This addition
undercuts the undeserved confidence the court placed in its
quantitative conclusions. See the final version, EPA Carcinogen
Assessment Group, Final Report On Population Risk to Ambient
Benzene Exposures (with cover memorandum dated January 10,
1979).
Moreover, while a unit at EPA produces such estimates for
informational purposes, the EPA Administrator has not
adopted the policy of.relying on these estimates in determining
standards or other regulations under any of the half dozen
statutes for control of chemical hazards that he administers.
Thus the lower court had no basis from EPA's actions to impugn
the Secretary’s judgment.
The court also cited the mere averment by an economic
consultant to the agency to the effect that he could do cost-
effectiveness analysis, which the court stated was “an analysis
which would have included some kind of risk quantification.”
This statement, by one who is not expert in the complex
scientific issues of cancer causation, was hardly a sound basis for
the court’s conclusion that reliable quantitative risk assessments
can be performed.
Beyond these specifics, the court appears simply to have
accepted the assurances of employer’s witnesses, who ignored
the massive uncertainties in the techniques. The lower court
adopted the term “rough” to describe the estimates it required.
Id. at 504. This choice of terms reflects its confusion on the
power of these techniques. If the court actually believed that
(footnote continued )
66
assessment if it can properly conclude that, under the
present state of knowledge and the policies of the Act, the
Secretary was irrational. In fact, the court conceded the
Secretary’s determinations were rational.'*' Thus, the
lower court’s decision cannot stand.
(footnote concluded )
“rough” estimates can come usefully close to the true risk, then
it simply failed to comprehend the irreducible uncertainties in
these techniques. If the court understood that “rough” esti-
mates really may be off by a factor of as much as 5,000,000 (see
text accompanying note 68, supra), then the court’s conclusion
that they are a useful basis for decisions was simply without
logical foundation. This use of the term “rough” distorts the
English language to the point of squeezing all meaning out of
the word.
181581 F.2d at 503.
67
CONCLUSION
The lower court has substituted its judgment for that
of Congress and the Secretary on the proper means to
protect employees from cancer and other serious diseases
caused by chemicals in the workplace. The court would
impose on the Secretary quantitative decision-making
requirements that Congress did not mandate, and that
the Secretary reasonably rejected. The decision would
preclude the Secretary from taking effective preventive
measures until a speculative future time when science has
solved the mysteries of carcinogenesis and policy-makers
have devised adequate quantitative means of making
basic societal decisions. With the knowledge that this day
is remote, Congress envisioned, and the Secretary has
practiced, a decision making process sensitive to the
qualities of the best available data and to the reality that
human lives and health are at stake. The lower court
would undo this. For these reasons, this Court should
reverse the decision below, and remand with instructions
for a decision consistent with the Occupational Safety and
Health Act as outlined here.
Respectfully submitted,
RICHARD E. AYRES
DAVID D. DONIGER
Attorneys for Amicus Curiae
Natural Resources
Defense Council
May 17, 1979
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.