Amicus Curiae Brief — International Union, United Automobile, Aerospace & Agricultural Implement v. Johnson Controls, Inc.

Supreme Court brief1991

Ask Donna

What actually matters in this document.

Text

gupreme Cour, US |

FILED

CY) JUN 1 280

No. 89-1215 JOSEPH F. SPANIOL, J

CLERK

IN THE

Supreme Court Of The United States

OCTOBER TERM, 1989

INTERNATIONAL UNION, UNITED AUTOMOBILE,

AEROSPACE AND AGRICULTURAL IMPLEMENT

WORKERS OF AMERICA, et al.,

Petitioners,

— Vvi—

JOHNSON CONTROLS, INC.,

Respondent.

BRIEF AMICUS CURIAE IN SUPPORT OF PETITIONERS

BY TRIAL, LAWYERS FOR PUBLIC JUSTICE

Arthur H. Bryant

(Counsel of Record)

Trial Lawyers for Public Justice

1625 Massachusetts Avenue, N.W.

Suite 100

Washington, D.C. 20036

(202) 797-8600

Lucinda M. Finley

SUNY Buffalo Law School

O’Brian Hall - Amherst Campus

Buffalo, NY 14260

(716) 636-2361

Attorneys for Amicus Curiae

Baimar Legz! Publishing Division, Washington, D.C., (202) 659-3610

TABLE OF CONTENTS

Seems We SUED AREILUS 2. ww te ee

SUMMARY OF ARGUMENT .............

II.

i Se eT a Wig we 6 6 < 6 4 's0 6

JOHNSON CONTROLS’ “FETAL PROTECTION

POLICY” HAS LITTLE TO DO WITH FETAL

PROTECTION AND A LOT TO DO WITH SEX

oe ee ee ee ee

A.

Johnson Controls’ Policy is Based on a Web of

Speculation About the Number of Women Who

Might Get Pregnant and Who Might Have

Children Adversely Affected by Lead. .....

Restricting Women’s Employment Opportunities

Because of The Cape :ity of Women to Become

Pregnant Has Long Been Discredited as Based

on Stereotypes That Trample on Women’s

ee ere a a ee ae

Johnson Controls’ Professions of “Moral

Concern” Are Severely Undermined by the

Amount of Risk to Worker and Consumer

Health That The Company Routinely Tolerates

as an Aspect of Lead Battery Manufacturing. . .

EXPANDING THE BFOQ DEFENSE TO IN-

CLUDE CONSIDERATIONS OF MORAL

CONCERNS AND SOCIAL COSTS, INCLUD-

ING POTENTIAL TORT LIABILITY, WILL

EVISCERATE TITLE VII AND UNDERMINE

THE SAFETY-ENHANCING GOALS OF

EG Fates a isp 8 4 4 6 6 6 6 8k 0

10

il

A. The Social Costs of Toxic Workplaces Are

So Open-ended That Consideration of

Them Would Overwhelm Title VII Trials

With Social Policy Judgments Better Left to

oe —

B. Concern About Potential Tort Liability Is an

Improper Component of a Title VII Defense. . .

III. THE ANIMAL STUDY DATA PRESENTED

BY PLAINTIFFS PRECLUDES THE ENTRY

OF SUMMARY JUDGMENT IN DEFENDANTS’

FAVOR. ....220200 68 5 5 ee

CONCLUSION ....+0220556 6 5) eee

TABLE OF AUTHORITIES

Cases:

Bichler v. Eli Lilly & Co., 55 N.Y.S. 2d 571 (App. Div.

TE A Ss woe et ce tt ee

eee

City of Los Angeles Dep’t of Water & Power v. Manhart,

os 5 kw te htt ee

Cleveland Board of Education v. LaFleur, 414 U.S. 632

wg on 5 6 8 et

EDF v. EPA, 510 F.2d 1292 (D.C. Cir. 1975) ......

Environmental Defense Fund v. EPA, 548 F.2d 998 (D.C.

elk 6 5 6 5 oe ee et ee

Ferebee v. Chevron Chemical Co., 552 F. Supp. 1293,

(D.D.C. 1982), aff'd, 736 F.2d 1529 (D.C. Cir.

1984), cert. denied, 469 U.S. 1062 (1984) .....

Hayes v. Shelby County Memorial Hospital, 726 F.2d

I gg 5 6 ce tte et ee

Hinkie v. United States, 715 F.2d 96 (3d Cir. 1983) .. .

In Re Agent Orange Products Liability Litigation, 611

Puma. Baas (e.w.N.Y.1965) ...........

In Re National Airlines, 434 F. Supp. 249 (S.D. Fla.

EE SES eee

Industrial Union Dep't v. American Petroleum Institute,

Ee

Mondelli v. United States, 711 F.2d 567 (3d Cir.

Eg bg we st et tt

20

14

¥, a3

14

20

20

21

4,15

16

16

14

20

16

iV

Muller v. Oregon, 208 U.S. 412 (1908) .........

Needham v. White Laboratories, 639 F.2d 394 (7th Cir.

re sree eas er es ee ee ae

Phillips v. Martin Marietta Corp., 400 U.S. 542

ie ere 2 ee eee er ee ee

Public Citizen Health Research Group v. Tyson, 796 F.2d

1479 DC. Cir. ISEB). cc cc eee

Rosenfield v. Southern Pac. Co., 444 F.2d 1219 (9th Cir.

er ee eae a ee ee a

Western Air Lines v. Criswell, 472 U.S. 400 (1985) . . .

Wright v. Olin, 697 F.2d 1172 (4th Cir. 1982) ...... |

Statutes:

Re es. gcc eee

Legislative History:

110 Cong. Rec. 13825 (June 15,1964) .........

H.R. Rep. No. 95-948, 95th Cong., 2d Sess. (1978) . . .

Sen. Rep. No. 95-331, 95th Cong., Ist Sess. (1977) . . .

Miscellaneous:

9 Occup. Safety & Health Rptr. (BNA) 339 (Current

Report, Sept. 13,1979) .. 2... 2.6 ee ee ees

Baer, J., The Chains of Protection (1978) ........

Becker, “From Muller v. Oregon to Fetal Vulnerability

Policies,” 53 U. Chi. L. Rev. 1219(1986) .....

6, 7, 11

20

14

20

3, 13

4,7,9

Vv

Finley, “The Exclusion of Fertile Women From the

Hazardous Workplace: The Latest Example of

Discriminatory Protective Policies, or a Legitimate,

Neutral Response to an Emerging Social Problem?,”

Proceedings of the N.Y.U. Thirty-Eighth Annual

Nat’! Conf’c on Labor, Ch. 16 (Mathew Bender &

Co. 1985)

Goodman, D. G., “Animal Testing of Carcinogens,” 2

Occupational Medicine: State of the Art Reviews

(Jan.- March 1987)

Gordis, L., Epidemiology and Health Risk Assessment

(1988)

Harrington, J., & Gill, F., Occupational Health

(1987)

McNamara, Eileen, “Factory and Fertility,” Boston Globe

Oct. 17, 1989

Office of Technology Assessment Task Force, Reproduc-

tive Health Hazards in the Workplace (1988)

Rall, “Relevance of Results from Laboratory Animal

Toxicological Studies”, in J. Last, ed. Public Health

and Preventive Medicine (12th ed)

2 oe. £6. 8 2 -@

Schuck, P., Agent Orange On Trial (1986)

Scott, Judith, “Keeping Women in Their Place”, in

W. Chavkin, Ed., Double Exposure (1984)

4,15

22

20, 21

20, 22

21

17

No. 89-1215

IN THE

Supreme Court Of The United States

OCTOBER TERM, 1989

INTERNATIONAL UNION, UNITED AUTOMOBIL ;:,

AEROSPACE AND AGRICULTURAL IMPLEMENT

WORKERS OF AMERICA, et al.,

Petitioners,

—_— ¥,=—

JOHNSON CONTROLS, INC.,

Respondent.

BRIEF AMICUS CURIAE IN SUPPORT OF PETITIONERS

BY TRIAL LAWYERS FOR PUBLIC JUSTICE

INTEREST OF THE AMICUS’

Trial Lawyers for Public Justice (“TLPJ”) is a national

public interest law firm dedicated to using the skills and ap-

proaches of plaintiffs’ trial lawyers to advance the public good.

Supported by and working through a nationwide network of

over 750 of the country’s outstanding trial lawyers, TLPJ brings

precedent-setting and socially significant cases designed to ad-

vance consumer and victims’ rights, environmental protection

and safety, civil rights and civil liberties, occupational health

and employee rights, and the protection of the poor and the

powerless.

” This brief amicus curiae is being filed with the consent of the parties.

The parties’ written consents are on file with the Clerk of the Court.

2

TLPJ is particularly interested in this case because it

presents the question of whether two battles TLPJ routinely

wages - - the battle against sex discrimination and the battle for

workplace safety - - will be treated as contradictory or, as

Congress mandated, complimentary. In addition, because the

majority below decided the issues presented without a trial, this

case also presents a question of critical importance to virtually

all victims and potential victims of toxic torts: whether courts

may simply disregard the results of animal studies and hold that,

unless and until enough people have been injured or kilied to

permit a statistically significant sample to be epidemiologically

studied. no inference of causation can properly be drawn.

SUMMARY OF ARGUMENT

While we agree with many of the arguments advanced by

the appellants and the other amici, there are three fundamental

flaws in the decision below that are particularly disturbing. We

write separately to stress these three critical, fundamental flaws.

First, both the majority opinion below and, to some extent,

Judge Posner’s dissent utterly distegard the fact that Johnson

Controls’ so-called “fetal protection policy” is nothing of the

sort. In reality, the policy is only tenuously related to fetal

protection and is based on long-discredited stereotypes that

trample on women’s autonomy. Moreover, the company’s

assertions of “mcral concern” for fetal health are severely

undermined by the risks to employee and consumer health that

Johnson Controls routinely creates and tolerates.

Second, both the majority opinion below and Judge

Posner’s dissent seriously err by suggesting that the bona fide

occupational qualification (“BFOQ”) defense is broad enough

to encompass an employer’s “ethical, legal, and business con-

cerns aboui the effects of its activities on third parties” or other

“social costs” like potential tort suits by future children of its

3

employees. The “social costs” related to toxic workplaces are

So open-ended that permitting such costs to be evaluated as part

of a BFOQ defense would overwhelm Title VII trials with social

policy judgments better left to Congress. Furthermore, on this

question, Congress has plainly spoken: “social costs” like

potential tort liability are not proper components of a Title VII

defense.

Third, regardless of whether its interpretation of Title VII

is correct, the majority opinion below improperly entered sum-

mary judgment in Johnson Controls’ favor because it did so in

the face of animal studies presented by plaintiffs that showed

that the ievels of lead exposures from which the company sought

to protect women (in the name of protecting fetuses) are also

dangerous to men. The majority’s wholesale rejection of animal

studies is legally and scientifically untenable. Plaintiffs’ animal

study evidence, without more, requires reversal.

INTRODUCTION

We believe that the Johnson Controls’ policy at issue in

this case, is, inescapably, overt, facial discrimination on the

basis of sex and capacity to become pregnant. Consequently,

according to the clear language and intent of Title VII, as

amended by the Pregnancy Discrimination Act (“PDA”), 42

U.S.C. 2000e(k), the only defense available to the employer is

that lack of capacity to become pregnant is a BFOQ; this defense

may only focus on the ability of substaritially all adult women

to perform the work. Western Air Lines v. Criswell, 472 U.S.

400 (1985). The arguments we advance in this brief, however,

are equally applicable to the business necessity defense.

Regardless of whether this Court finds the BFOQ or the business

necessity defense to be mandated by Title VII, the majority

below erred by granting summary judgment in defendant’s

favor.

4

I. JOHNSON CONTROLS’ “FETAL PROTEC-

TION POLICY” HAS LITTLE TO DO WITH

FETAL PROTECTION AND A LOT TO DO

WITH SEX DISCRIMINATION.

The majority opinion below and Judge Posner’s dissent,

like the decisions in Wright v. Olin, 697 F.2d 1172, 1189-90 (4th

Cir. 1982), and Hayes v. Shelby County Memorial Hospital, 726

F.2d 1543 (11th Cir. 1984) (general societal interest in having

enterprises be protective of health of unborn children provides

business necessity for policy excluding women from empioy-

ment), contort facial Title VII law and utterly disregard the plain

language of the PDA.! They do so because of well-intentioned

concerns about fetal health. To properly evaluate Johnson

Control’s exclusionary policy, as well as the arguments about

the relevance of moral concerns and social costs to Title VII

defenses, however, these concerns must be put into perspective.

The admirable concern that our toxic world be made safer

for the sake of future generations, when applied to this case,

merely obfuscates the fact that the connection between the

Johnson Controls policy and the health of babies or fetuses is

tenuous and speculative. What is not tenuous or speculative,

however, is the immediate harmful! impact of the policy on

existing hurnan beings, both male and female. Existing women

are denied well-paying jobs and advancement opportunities and

are branded with tre stigma of being regarded as marginal,

expendable workers. Existing men are virtually foreclosed from

working in tne jobs with the lowest levels of lead exposure, both

‘ Commentators who have analyzed Ulin and Hayes are uniformly critical

of their reasoning, assumptions, and troubling contortions of Title VII law

and precedent. See, e.g., Becker, “From Mujsler v. Oregon to Fetal Vul-

nerability Policies,” 53 U. Chi. L. Rev. 1219 (1986); Finley, “The Exclusion

of Fertile Women from the Hazardous Workplace: The Latest Example of

Discriminatory Protective Policies, or a Legitimate, Neutral Response to an

Emerging Social Problem?,” Proceedings of the N.Y.U. Thirty-Eighth An-

nual Nat’! Conf’c 9n Labor, Ch. 16 (Mathew Bender & Co. 1985).

%

5

because women have been transferred into these jobs and be-

cause men’s concerns about reproductive health are ridiculed at

Johnson Controls.?

A. Johnson Controls’ Policy is Based on a Veb of

Speculation About the Number of Women Who

Might Get Pregnant and Who Might Have

Children Adversely Affected by Lead.

Johnson Controls’ designation of its policy as a “fetal

protection policy” is a misnomer. The policy applies whether

or not a fetus is extant. The women to whom it applies are not

pregnant, and many of them have no plans to become pregnant,

or are long past their childbearing days, or are responsible users

of contraceptives. The policy does not protect fetuses other than

in the most indirect, attenuated sense. Rather, what it does do,

directly and immediately, is exclude existing adult women from

entire job categories and promotion opportunities, simply be-

cause women have the biological ability to become pregnant.

All of the professed “moral concern” about fetal health boils

down to the fc'lowing chain of reasoning: because women can

get pregnant, and because some of them might get pregnant, and

some of these who get pregnant might have high blood lead

levels, and then perhaps some of the children born to these

women might have some developmental problems, in order to

avoid this slight possibility, we must refuse to hire al] women

who are physically capable of bearing children.

This major leap - - from the fact that in utero lead exposure

can cause cognitive development and social adjustment

? Donald Penney, one of the named plaintiffs in this Case, charged in his

E.E.9.C. complaint that his ability to bid for low levei lead jobs is greatly

restricted by the Johnson Controls’ Fetal Protection Policy, since it gives

transfer preference to women. He also charged that, when he requested a

leave of absence in order to lower his blood lead levels because he wanted

to father a child, he was subjected to such a level of harassment as to

constitute constructive discharge. J. App. 66-69.

6

problems in children to the conclusion that all fertile, but not

necessarily pregnant, women must be barred from employment

- - is fraught with analytic and policy pitfalls.

B. Restricting Women’s Employment Opportunities

Because of The Capacity of Women to Become

Pregnant Has Long Been Discredited as Based

on Stereotypes That Trample on Women’s

Autonomy.

Justification of restrictions on women’s employment op-

portunities because of their capacity to bear children and atten-

dant concern about social cost and the well-being of future

generations is the most venerable and prevalent excuse for

sex-based emplovment discrimination. This sad fact was fre-

quently recognized by the proponents of the Pregnancy Dis-

crimination Act. See, e.g., Sen. Rep. No. 95-331, 95th Cong.,

1st Sess. 2-3 (1977); H.R. Rep. No. 95-948, 95th Cong., 2d Sess.

2-3 (1978). Indeed, the justifications offered by Johnson Con-

trols and by the majority below bear a startling resemblance to

the long-discredited turn-wf-ihe-century reasoning in Muller v.

Oregon, 208 U.S. 412 (1°98), in which Chief Justice Brewer,

writing for the Court, upheld a statute limiting employment of

women, out not men, because “healthy mothers are essential to

vigorous offspring, [and] the physical well-being of woman

becomes an object of public interest and care in order to preserve

the strength and vigor of the race.” Jd. at 421. And, the restric-

tion was necessary “largely for the benefit of all.” Jd. at 422.

Due to the recognition that excluding women in the name of

“protecting” them and future generations fails to treat women

as autonomous individuals and forever chains them to their

reproductive role, state legislation aimed at “protecting” women

out of jobs was swept aside by Title VII almost twenty years

ago. See, e.g., Rosenfield v. Southern Pac. Co., 444 F.2d 1219

(9th Cir. 1971); J. Baer, The Chains of Protection (1978).

7

It is odd that, in the face of judicial interpretations of Title

VII prohibiting state legislation that “protects” only women

from occupational hazards, the majority below would allow a

private employer to exclude women in the narne of protecting

potential offspring from the risks of only maternal, but not

paternal, employment. Becker, supra, 53 U. Chi. L. Rev. at 1220.

It is also distressing that, with the widespread adoption of fetal

protection policies in the last few years, id., women are still,

despite the PDA and the long-ago demise of protective legisla-

tion, being forced to fight for their right to employment over

claims that their childbearing capacity precludes that right. The

subtle shift from the reasoning in Muller that it was women who

were being protected, in the name of future generations, to

Johnson Controls’ reesoning that it is potential fetuses that musi

be protected, does no‘ mask the fact that it is still women who

are being excluded froin employmert.

The historical and contemporary exclusicn of women from

employment because of their childbearing capacity assumes that

women are more socially important because of their reproduc-

tive function, and that their value as workers is marginal. In

essence, women are reduced to their wombs. Moreover, women

are presumed to be less capable and conscientious decision-

makers about their reproductive desires and health than

employers. Thus, fetal protection policies such as the one

implemented by Johnson Controls invariably disrespect

women’s autonomy and individuality. Policies broadly ex-

clusive of all fertile women callously disregard actual

individuals’ plans about marriage and childbearing and fail to

contemplate that many working women are unlikely to have any

more children. Thus, these policies plainly contravene City of

Los Angeles Dep’t of Water & Power v. Manhart, 435 U.S. 702

(1978), in that they exalt a generalizable biological difference

between men and women into an excuse for not treating in-

dividual women as individuals.

8

Moreover, Johnson Controls arrogated to itself the judg-

ment that any woman who did exercise her autonomous right to

procreate despite her careful weighing of the risks of lead

exposure versus the risks of unemployment and her family’s

need for her income was acting irresponsibly. Prior to 1982,

Johnson Controis did not broadly exclude all women, because

the company recognized that “not ali women who can become

mothers wish to become mothers,” and that the decision whether

to work in certain jobs “should properly be made by the in-

dividual immediately involved.” J. App. 140. Yet Johnson

Controls regarded its pre-1982 voluntary program, under which

employees were informed of the risks and given counseling, 2

“failure”, because, throughout the company’s plants, eight

women became pregnant in six years. (There is no evidence in

the record that any of these women bore a child that showed any

adverse effects from prenatal lead exposure.) Rather than

regarding the occurrence of a pregnancy as a “failure,” as an act

engaged in by an irresponsible woman (in consultation with her

irresponsible partner), the occurrence of so few pregnancies

during such a long period of time at a large, nationwide company

indicates the success of the previous policy, and the respon-

sibility of autonomous women.

The insult to women’s dignity and autonomy perpetrated

by Johnson Controls is illustrated by the reactions of two women

over fifty who were demoted to lower paying, lower status jobs

at the Johnson Controls plant in Bennington, Vermont. In the

words of a Boston Globe reporter who interviewed them, “[t}he

company’s contention that the mandatory transfer to low lead

exposure jobs was for the protection of her future children sends

a rush of red from [Virginia] Green’s neck to her gray hairline,

prompting her to ask rhetorically whether women are not bright

enough to make sensible choices about their own reproductive

lives. It prompts [Elsie] Nason, now 57 ... [to say] ‘Ridiculous,

and degrading. I’ve been divorced for 16 years. I felt they’d

pointed a finger at me as if I was some shady lady going out and

9

getting pregnant. At my age! They took me off my job for

foolishness and I'l! be mad about it, and the smutty jokes I had

to hear from the men about an old lady like me still having her

period, until the day I die.’” Eileen McNamara, “Factory and

Fertility,” Boston Globe Oct. 17, 1989.

The majority below, however, was concerned that, given

the importance of jobs and adequate income to “better [her]

family’s station in life, it would not be improbable that a female

employee might somehow rationally discount” the risk to her

fetus if she became pregnant. 886 F.2d at 897. This reasoning

overlooks the fact that employers are just as likely to rationally

discount the risk to women and their families from unemploy-

ment or reduced wages, since, once the woman is out of the

workforce, any of her problems are simply no longer of any

concern to the employer. As explained by Professor Mary

Becker, “an employer faced with evidence that a group is

hypersusceptible ... is likely to make the decision to exclude that

group based, not on the costs and benefits to the workers and

their families, but on the employer’s need for members of that

group in its labor force. If the group is very marginal — as

women are in traditionally male jobs — the employer has every

incentive to exclude the group’s members even though the

heightened risks may be very iow ... and the benefits of allowing

the group to work (benefits in this case to both women and to

their children) are very high.” Becker, supra, 53 U. Chi. L. Rev.

at 1239-40. Title VII and the PDA were enacted to take away

from ¢.iployers the right to make such economic cost-benefit

calculations about a group of workers perceived as marginal

because of a characteristic — childbearing capacity — unrelated

to their ability to do the job.

Another major analytic flaw of Johnson Controls’ focus

only on women as the locus of concern about the health of future

generations is that the present policy totally ignores the male

capacity to impregnate. Men contribute to the health of future

10

generations as well as women, and lead is highly toxic to men’s

fertility and sperm motility. The definition of the policy at issue

in this case as a fetal protection policy means that the concern

about reproductive health remains a narrow one, considering

only one aspect of reproduction, only one way in which adverse

reproductive outcomes can occur. And it means that the

scrutiny will always remain on women, and on their role in

reproduction, to the detriment of men’s reproductive health.

While fears about developmentally impaired children can

loom large in a company’s mind, company decisionmakers are

less likely to think about the children who are desperately

wanted but never get conceived. They are less likely to think

about the miscarriages suffered by a wife of a male worker

because his sperm was damaged by lead exposure. But to the

families who must live with the consequences of lead, the

infertility or the pregnancy loss may be more of a tragedy, more

of an endangerment to physical and mental health, than a wanted

child who turns out to be less intelligent or capable than some

of her peers.

C. Johnson Controls’ Professions of “Moral Concern”

Are Severely Undermined by the Amount of

Risk to Worker and Consumer Health That The

Company Routinely Tolerates as an Aspect of

Lead Battery Manufacturing.

The channelling of all of the company’s concern about

reproductive health onto potential fetuses to the detriment of

existing men and women leads to the untenable conclusion that

only for fetuses — something that only women carry — shall

there be absolutely no risk. The company’s total intolerance for

fetal risk starts to look suspiciously like a justification for

discrimination when all of the other risks that it rationally,

economically tolerates are taken into consideration. Johnson

Controls countenances the high risk of cardiovascular damage

from blood lead levels as low as 10ug/dl to its existing male

——

ae

11

workers. It tolerates the risks to the environment and to the

health of future generations from the disposal of lead batteries

and other hazardous waste from the production process. It

accepts some risk of harm to consumers from defective or

exploding batteries. It tolerates risks to the children of workers

from the workers bringing lead residues home on their clothing.

It tolerates all sorts of risks to workers from innumerable con-

ditions in and around its plants, ranging from slippery floors to

repetitive motion injuries to cigarette smoke. seen from

the perspective of all of the risk that the company is willing to

accept, the “moral imperative” for excluding women because

some of them might some day choose to become pregnant starts

to look more and more like a disingenuous rationalization for

keeping women out of the workplace.

Against this background, which highlights the tenuous

connection between a refusal to hire all fertile women of a broad

age spectrum and protection of the health of future generations,

arguments that “moral concern,” “social cost,” “ethical con-

siderations,” or economic fears about future liability costs

should justify a radical departure from established law govern-

ing facial discrimination under Title VII and the PDA should be

greeted with skepticism. These arguments are no more tenable

a reason for barring women from employment opportunities

today than they were in the early 1970’s when Title VII was used

to discredit the outdated 1908 ideology of Muller v. Oregon.

Il. EXPANDING THE BFOQ DEFENSE TO

INCLUDE CONSIDERATIONS OF MORAL

CONCERNS AND SOCIAL COSTS, INCLUD-

ING POTENTIAL TORT LIABILITY, WILL

EVISCERATE TITLE VII AND UNDERMINE

THE SAFETY-ENHANCING GOALS OF

TORT LAW.

Both the majority and Judge Posner, dissenting below,

suggested that the BFOQ defense ought to be broad enough to

12

encompass, as Judge Posner put it, an employer’s concerns with

the “norma! operations” of a business, including “ethical, legal,

and business concerns about the effects of its activities on third

parties,” and such “social costs” as potential tort suits by future

children of employees. Such a wholesale expansion of the

BFOQ defense was squarely rejected by Congress, when the

Senate rejected a proposed amendment to Title VII that would

have allowed discrimination whenever it would be “more

beneficial to the normal operation of the particular business or

enterprise involved or to the good will thereof than the hiring of

an individual without consideration of his race, color, religion,

Sex, or national origin.” 110 Cong. Rec. at 13825 (June 15,

1964). This Court should not let concern about potential harm

to potential fetuses tempt it to open up what is supposed to be

the most narrow defense to facial discrimination.

A. The Social Costs of Toxic Workplaces Are So

Open-ended That Consideration of Them Would

Overwhelm Title VII Trials With Socia! Policy

Judgments Better Left to Congress.

ne At the end of this “social cost” path lies a tangle of

judicially unmanageable speculations, that, if indulged, would

swallow up Title VII’s ban on discrimination on the basis of

pregnancy and capacity to become pregnant. What social costs

Should employers and courts have to consider? And social costs

to whom? And how large, or how certain, must the social cost

be before it becomes a legitimate factor for consideration? If

employers may consider potential risks to fetal health, should

they not also have to consider potential risks to paternal health

and maternal health? What about the costs to women and their

children of unemployment or reduced wages and benefits?

What about the cost to a woman’s esteem and dignity from being

told that she cannot have a certain job simply because she has a

womb? What about the social cost of continuing to label women

as marginal workers, creatures more valued for their reproduc-

13

tive than their productive capacity? What about the social cost

of the perpetuation of some men’s stereotyped attitudes about

women’s productive capabilities and social role? What about

the social cost of perpetuating occupational segregation, and

with it, the wage gap between men and women?

Many of these important social costs flowing from exclud-

ing women from certain jobs can be just as scientifically docu-

mented as the risks to fetuses from lead exposure. Yet to have

to weigh all of these factors in assessing a BFOQ defense would

surely turn Title VII trials into legislative policy sessions,

evaluating the question of which social cost is of greater priority

— the immediate harms to current workers from discrimination,

or the possible future harm to nonexistent fetuses? Congress

has already answered this question, and has made the judgment

that a woman’s ability to do the job is the only relevant factor

for evaluaiing whether she may be excluded because of her

Capacity to become pregnant. 42 U.S.C. 2000e(k).

As the above questions illustrate, “social cost” analysis is

entirely open-ended. Thus, allowing “moral concerns” about the

social cost of the risk of materna! employment to fetal health to

enter into either the BFOQ or business necessity analysis will

swallow up Title VII’s prohibition of sex discrimination. There

is no logical line at which to end acceptance of professed

concerns for fetal health. Just as lead and other toxins may pose

a risk to fetuses, so may stress, fatigue, standing or sitting for

long hours, and exposure to the colds, viruses, and cigarette

smoke of other workers. So, should concern for fetal health

permit employers to refuse to hire women for jobs deemed to be

stressful, jobs that require standing long hours at assembly lines

or service counters or in restaurants? Should employers be able

to use concern for fetal health to justify forced leave for all

women after a certain month of pregnancy, because the stress

and fatigue risks of the job increase at that time? (This practice

was declared unconstitutional when practiced by a public agen-

14

cy in Cleveland Board of Education v. LaFleur, 414 U.S. 632

(1974).) Should taxi and bus companies be able to refuse to hire

women as drivers because, if they got pregnant and if they were

in an accident, it would be not only them but theiz fetuses that

might be injured? (It should be noted that the statistical risk of

getting in an accident given the number of hours spent on the

road by a professional driver is at least as high, if not higher,

than the risk of lead exposure to a fetus.) Should employers be

able to rely on child development studies that claim that very

young children develop better intellectually and emotionally

when given the constant attention of a primary parental caretaker

in the home to refuse to hire women with young children,

because those children will be better off with the mother at home

than with the mother employed? (This sort of policy was

declared to violate Title VII in Phillips v. Martin Marietta Corp.,

400 U.S. 542 (1971).) It does not take much of a march along

this “parade of horribles” to understand how great an incursion

on women’s employment opportunities the “social cost of pos-

sible fetal harm” defense would permit.

While an employee’s ability to perform the job safely has

been a legitimate BFOQ and business necessity consideration,

this factor properly includes only safety risks to co-workers and

existing third parties such as customers. It cannot include safety

risks even to actual fetuses. See, e.g., Burwell v. Eastern Air-

lines, 633 F.2d 361 (4th Cir. 1980), and Jn Re National Airlines,

434 F. Supp. 249 (S.D. Fla. 1977), in which courts upheld the

exclusion of pregnant flight attendants from flying during the

advanced stages of pregnancy because of evidence that late term

pregnancy interfered with the ability safely to serve passengers,

but rejected the argument that fetal safety also supported the

exclusion, because judgment about what risks to accept for one’s

future child were best left to the mother and were of no legitimate

concern to the airline. As mentioned in Part I of this brief, this

conclusion is the only one consistent with respect for the

autonomy and individuality of women.

15

After all, it is not the employer, but the woman and her

family, who has to bear both the consequences of job loss or

demotion and the consequences of any ensuing harm to a child.

Those whose lives will be most directly affected by an event are

the ones who should make the decision whether to bear the risk.

Finley, supra, at 16-27-28.

B. Concern About Potential Tort Liability Is an

Improper Component of a Title VII Defense.

Employers may assert that their businesses will be affected,

however, if harm occurs to a child of a female employee, because

that child may sue the employer in tort.4 Concern about tort

liability is essentially a cost concern, and cost has been emphati-

cally rejected as a defense to facial sex discrimination by this

Court in City of Los Angeles Dep’t of Water & Power v.

Manhart, supra. Even those courts that have allowed some

expansion of Title VII defenses because of concern about fetal

health have drawn the line at concern about future tort liability.

As the 11th Circuit explained in Haves v. Shelby County

Memorial Hospital, supra, 726 F.2d at 1553 n.15:

In today’s litigious society, the potential for litigation

rests in almost every human activity. For example,

every employer faces the risk that a pregnant

employee will encounter a workplace activity that

would not normally be hazardous to a nonpregnant

employee, but which could prove injurious to a

developing fetus. These hazards range from slippery

floors to uncontrolled cigarette smoke, asbestos,

> While Johnson Controls has not asserted fear of tort liability as a

justification for its policy, perhaps because liability is so speculative or

because this justification is so clearly precluded by existing law, Judge

Posner suggests that such fear should enter into BFQQ analysis. As an

organization expert in tort law and dedicated to seeing the safety-enhancing

goals of tort law respected, TLP3 is in a unique position to provide guidance

on this matter.

16

known and unknown carcinogenic materials used in

the workplace and a plethora of other hazards of

modern society. The employer is, of course, free to

protect itself from financially ruinous lawsuits by

purchasing insurance and by maintaining the degree

of care required by law.

Because tort liability is a danger pervading everything manufac-

turers such as Johnson Controls Jo, it can hardly stand as an

excuse for facial sex discrimination.

Johnson Controls faces a risk of tort liability for activities

ranging from the manufacture of batieries to the disposal of

batteries. The risk of future suits from subsequently born

children of female employees is actually one of the slightest tort

risks that it faces. The contemplation of these suits requires

several more speculative steps beyond the speculation about the

likelihood that a woman will get pregnant and bear a damaged

child. If that happens, it is speculative whether the child will

sue and, if so, whether the child will be able to mount sufficient

proof of injury caused by maternal lead exposure, given all the

other possible sources of lead exposure and possible causative

factors for impaired development. Johnson Controls faces at

least as great a risk of tort suits by children of male workers

alleging that they were harmed by paternal lead exposure.

Indeed, there have been several notable claims brought alleging

birth defects due to paternal exposure to a toxin. See, e.g. In Re

Agent Orange Products Liability Litigation, 611 F.Supp. 1223

(E.D.N.Y. 1985\(claims of children born to Vietnam Veterans

alleging birth defects caused by paternal exposure to Agent

Orange dismissed for failure to prove causation); Mondelli v.

United States, 711 F.2d 567 (3d Cir. 1983) (suit alleging that

paternal exposure to ionizing radiation during atomic testing

caused plaintiff's birth defects); Hinkie v. United States, 715

F.2d 96 (3d Cir. 1983)(same); 9 Occup. Safety & Health Rptr.

(BNA) 339 (Current Report, Sept. 13, 1979) (describing claim

17

brought by male worker that his occupational exposure to DBCP

caused deformities to his child); Judith Scott, “Keeping Women

in Their Place”, in W. Chavkin, Ed., Double Exposure 184-191

(1984) (describing claims of male railroad workers against Long

Island Railroad that their children were born with hip deform-

ities caused by paternal exposure to defoliants, and significant

number of incidents of children of male pesticide workers

fathering children with serious heart defects). In light of the

reality that tort liability can come just as much from the children

of male workers as those of female workers, it can hardly stand

as a defense to a refusal to hire women.

If fear of potential tort liability were to excuse sex dis-

crimination, such a result would undermine the goals of both

Title VII and tort law. It would give the dirtiest, least tech-

nologically efficient, least safety conscious employers a built-in

excuse for discriminating, because the more likely the prospect

of actual tort liability, the stronger the defense to employment

discrimination. Thus, incentives to provide as clean a

workplace as reasonably feasible, to search for improved tech-

nology to even further reduce lead exposure, and to push ahead

with the search for a lead-free battery would be greatly reduced.

After all, so long as women are marginal workers in the formerly

male-dominated lead battery manufacturing process, it will

always appear easier, and less costly for employers to exclude

women than to invest in technological improvements that might

better the health of everyone. For this basic reason, the theory

of law and economics must remain merely a fashionable

academic theory to explain employer actions, rather than be

turned into a loophole in Title VII law excusing employer

actions that keep women in the home or in “women’s work.”

One fundamental tenet of tort law is that fear of liability

is supposed to provide a deterrent to unsafe actions, and an

incentive to constantly improve safety. This goal would be

devastated if speculative tort liability could excuse a refusal to

18

hire women. The goal can only be accomplished if employers

cannot avoid the specter of tort liability by avoiding women.

Rather, employers must be encouraged, through the congruence

of Title VII and tort law, to face up to their potentia! liability by

making the workplace as safe as possible for women and men,

and for the offspring of both sexes.

The responsible employer concerned about the health of

future generation and tort costs has several viable options far

short of excluding women. The responsible employer can meet

all obligations imposed by the tort standard of care, can fully

warn and inform workers of both sexes of the reproductive risks,

and can continue to invest in reasonable technology and equip-

ment to further reduce occupational lead exposure. The respon-

sible employer can also allow workers of both sexes who wish

to procreate to transfer temporarily, at no loss of pay or benefits,

to positions that would enable them to lower their blood lead

levels. The intersection of tort law with Title VII requirements,

far from putting an employer between the proverbial rock and a

hard place, encourages employers to work vigilantly to improve

workplace safety.

iI. THE ANIMAL STUDY DATA PRESENTED BY

PLAINTIFFS PRECLUDES THE ENTRY OF

SUMMARY JUDGMENT IN DEFENDANTS’

FAVOR.

In order to enter summary judgment in Johnson Controls’

favor, the majority below did not simply misinterpret Title VII.

It also rejected plaintiffs’ substantial evidence of reproductive

harm to male workers by making an assertion that would as-

tound any reputable practitioner of occupational or environmen-

tal medicine. The majority asserted that “animal research

evidence does not present the type of solid scientific data

necessary for a reasonable factfinder to reach a nonspeculative

conclusion that a father’s exposure to lead presents the same

19

danger to the unborn child as that resulting from a female

employee’s exposure to lead.” 886 F.2d at 889-90.

This assertion is odd in light of the fact that the conclusion

about risk to potential fetuses of nonpregnant women was itself

so speculative. It is also curious in light of the fact that the level

of risk Johnson Controls deemed acceptable for women was

zero. But this assertion is also legally and scientifically un-

tenable, because it rejects out-of-hand a type of data routinely

and properly relied on by regulatory agencies in making risk

assessments and experts when assessing causation.

In deciding that the level of lead exposure approved by

OSHA was insufficient to protect a fetus if any of its women

employees were to become pregnant, Johnson Controls was

engaged in the same sort of risk assessment activity that

regulatory agencies themselves are authorized by Congress to

undertake. When OSHA set the lead standard and concluded that

there was no basis for excluding fertile women because men also

faced risks to their reproductive health at equal levels of ex-

posure, it relied in part on animal studies for assessing both male

and female mediated risks. Yet Johnson Controls presumed to

substitute its judgment for the relevant federal regulatory agency

that had engaged in painstaking factfinding by refusing even to

consider an entire type of data relied on by OSHA. Moreover,

the majority below compounded this error by presuming that

judges without scientific training did not even have to consider

the same types of evidence routinely relied on by scientists,

medical experts, and regulatory agencies.

The enterprise of risk assessment remains fundamentally

the same regardless of whether it is undertaken by an occupa-

tional or environmental medicine specialist, a regulatory agen-

cy, a jury, or an employer. Each risk assessor should have to

consider the same types of evidence, particularly a type of

evidence regarded as essential to the enterprise by medical

science. Indeed, this Court has approved the use by regulatory

20

agencies of animal studies, /ndustrial Union Dep’t v. American

Petroleum Institute, 448 U.S. 607 (1980), as has the U.S. Court

of Appeals for the District of Columbia Circuit on numerous

occasions, see, e.g., Environmental Defense Fund v. EPA, 548

F.2d 998 (D.C. Cir. 1976); EDF v. EPA, 510 F.2d 1292 (D.C.

Cir. 1975); Public Citizen Health Research Group v. Tyson, 796

F.2d 1479 (D.C. Cir. 1986). Dr. David Rall, the Director of the

National Institute of Environmental Health Sciences of the

National Institutes of Health, has observed that animal studies

provide evidence to link a chemical to a disease that “is critical

for the regulatory agencies that must make decisions to control

chemicals, but it [animal study evidence] is also important to the

physician and health care worker who needs to diagnose a

disease that may be caused by a chemical.” Rall, “Relevance of

Results from Laboratory Animal Toxicological Studies”, in J.

Last, ed. Public Health and Preventive Medicine 515 (12th ed).

Results of animal studies showing toxicity, carcinogenicity, and

adverse reproductive outcomes have also been deemed highly

relevant bases for concluding, in tort suits, that a manufacturer

reasonably should have known of the risks to humans and thus

should have warned humans. See, e.g., Needham v. White

Laboratories, 639 F.2d 394 (7th Cir. 1981); Bichler v. Eli Lilly

& Co.,55 N.Y.S. 2d 571 (App. Div. 1982) (DES cases in which

evidence from animal studies that DES could cross the placenta

and cause cancer and abnormalities in female offspring lead to

conclusion that manufacturers of the drug had a duty to warn

about these risks).

If animal studies are to be precluded out-of-hand, as was

done by the majority below, then the only public health alterna-

tive is for epidemiologists and experts in occupational and

environmental medicine to wait until enough people have been

injured or killed from toxic exposures that a large enough sample

exists for making a statistically significant epidemiological cal-

culation. This approach has been unanimously rejected by the

medical and public health community. L. Gordis, Epidemiology

21

and Health Risk Assessment 37 (1988); J. Harrington & F. Gill,

Occupational Health 99 (1987). It should similarly be rejected

by the courts. See, e.g., Ferebee v. Chevron Chemical Co., 552

F. Supp. 1293, 1301 (D.D.C. 1982), aff'd, 736 F.2d 1529 (D.C.

Cir. 1984), cert. denied, 469 U.S. 1062 (1984) (“A civilized

legal system does not require that much human sacrifice before

it can intervene.”)

Animal studies drawn from exposed animal populations

and epidemiological studies drawn from exposed human

populations are complementary, interdependent methods. An

epidemiologist will make an observation of an unexpected

health phenomenon in some humans, which will trigger animal

tests of the suspected agents. If these tests demonstrate toxicity,

then the epidemiologist will use this finding to devise a human

population study. Further animal studies may then be used to

strengthen a causal inference revealed by the human study.

Gordis, Epidemiology. “Without laboraiory studies of animals,

[scientists] could not estimate the substance’s risks under con-

ditions to which the human population may have been exposed,

but for which epidemiologists have not and perhaps cannot

control and for which epidemiological methods may lack re-

quired sensitivity.” P. Schuck, Agent Orange On Trial 236

(1986).

While animal studies are not a perfect predictor of effects

in humans, because of species and dose differences, they are

nonetheless relied upon in the field of environmental and oc-

cupational medicine, and by regulatory agencies, for the follow-

ing reasons:

One of the questions commonly raised about car-

cinogenicity studies is why are animal studies relevant

to humans. Some answers to this question include:

(1) Most known human carcinogens are also

animal carcinogens and many were first found to

be carcinogenic in animals;

22

(2) Animals have long been used in a variety of

toxicity tests and in most cases have been found

to be excellent predictors of human toxicity;

(3) Animals and humans have many of the same

metabolic pathways and responses to injury; and

(4) It is undesirable, to say the least, to use

humans as experimental animals when other

means exist to determine potential risk to humans.

D. G. Goodman, “Animal Testing of Carcinogens,” 2 Occupa-

tional Medicine: State of the Art Reviews -49 (Jan.- March

1987); see also Rall, supra at 516,519. These observations are

no less valid for reproductive hazards that can have teratogenic

effects on fetuses or mutagenic effects on sperm and eggs. See

Office of Technology Assessment Task Force, Reproductive

Health Hazards in the Workplace (1988).

The conclusion of the majority below that epidemiological

evidence derived from human populations is inherently much

more probative for determining risk to human beings than

animal studies overestimates the usefulness of epidemiological

studies just as it underestimates the importance of animal

studies. While animal studies present the problem of extrapolat-

ing from one species to another, epidemiological studies present

the problem of jumping from aggregate information about large

numbers of humans previously exposed to predictions about risk

to a particular individual with a different health level exposed

to different amounts under different conditions. The point is not

to disparage the usefulness of epidemiology — it is extremely

useful in making risk assessments and drawing causal inferences

— but to underscore that epidemiology is no more or less

relevant than animal studies for determining causation or assess-

ing risk. Any risk assessment that excludes data from animal

studies —such as the risk assessment of male mediated

reproductive harm engaged in by Johnson Controls and the

majority below — is not a scientifically sound risk assessment.

23

In deciding that the risk to women of adverse reproductive

outcomes was so much greater than the risk to men of adverse

reproductive outcomes from lead exposure that wo.nen should

be barred from jobs, Johnson Controls was functioning essen-

tially as a regulatory agency. It ought to have to rely on the same

kind of evidence that a regulatory agency would have to rely on

in making the assessment of whether there is some risk to men

from the current lead level in certain production jobs. The

standard of “some risk” to men is the appropriate one because,

by acting to exclude all fertile women regardless of childbearing

plans or likelihood, Johnson Controls has effectively set the risk

threshold for women at zero. See dissenting opinion of Judge

Easterbrook, below. Moreover, if Johnson Controls’ professed

concern for the health cf future generations is to be taken

seriously, then the company should have to consider a wider

variety of reproductive risks than those that operate on fetuses

through women. It should have to consider the risk of mal-

* formed sperm that might produce fetal harm or even moscar-

riage, or reduced sperm count leading to sterility in men As

pointed out in Part I, the company’s focus on apriman \ (ema e

mediated risk — the kinds of reproductive harm prese

a fetus has already been created — is inherent) oso e

This narrow focus fails to adequately protect repr

of men and women.

The company’s narrow definition of the types of reproduc

tive risks it will consider, coupled with its failure to consider

animal study evidence about male mediated harm, amounts to a

callous indifference to the reproductive health of men. __.!ure

to consider animal studies is tantamount to saying to men,

“Many of you will have to become sterile, many of you will have

to suffer through your wives’ miscarriages, many of you will

have to shoulder the burden of damaged children, until such time

as medical science produces more studies of men, and con-

clusive studies of men.”

24

The law Joes not and should not require such sacrifice.

The animal study evidence presented by plaintiffs, without

more, requires reversal of the decision below.

CONCLUSION

For the foregoing reasons, the judgment of the Seventh

Circuit should be overturned.

Respectfully submitted,

Arthur H. Bryant

(Counsel of Record)

Tria! Lawyers >: Public Justice

1625 Massachusetts Avenue, N.W.

Suite 100

Washington, D.C. 20036

(202) 797-8600

Lucinda M. Finley

SUNY Buffalo Law School

O’Brian Hail - Amherst Campus

Buffalo, N.Y. 14260

(716) 636-2361

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.

A word about cookies

We need a few to keep you signed in and the library working. The rest help us see which pages people use and where they get stuck. They stay off unless you say yes.