PREGNANCY AS A NORMAL CONDITION OF EMPLOYMENT: COMPARATIVE AND ROLE-BASED ACCOUNTS OF DISCRIMINATION.

• Author: Siegel, Reva B.
• Position: Cutler Lecture

TABLE OF CONTENTS INTRODUCTION 972 I. COMPARATIVE MODELS OF DISCRIMINATION AND PREGNANCY AS DIFFERENCE--BEFORE AND AFTER THE PDA 978 A. Pregnancy Discrimination Under the Constitution: The Uniqueness Trap 978 B. The PDA: Uniqueness Redux? 981 II. COMPARATIVE AND ROLE-BASED CONCEPTIONS OF DISCRIMINATION 987 A. Two Conceptions of Discrimination 987 B. Authority for the Social-Roles Inquiry Under the PDA 991 C. How the Social-Roles Framework Helps Identify Pregnancy Discrimination 993 III. YOUNG v. UPS 997 A. The Court's Decision 999 B. The Court's Decision in Context 1002 CONCLUSION: NEW SUPPORT FOR THE PDA AT FORTY 1005 Just defining pregnancy discrimination as sex discrimination does not tell us what it means to discriminate because of pregnancy. --Young v. UPS, Inc. (Scalia, J., dissenting). (1) INTRODUCTION

Nearly all countries except the United States offer women some form of paid maternity leave. (2) The United States requires employers of over fifty employees to provide their employees twelve weeks of unpaid leave for medical and family-care reasons. (3) But this benefit is of little use to many low-wage workers. (4) In the absence of paid leave, U.S. law asks working women to rely on antidiscrimination protections to retain their jobs during pregnancy--while Americans continue to debate what it means to discriminate because of pregnancy.

In this Lecture, I rejoin the long-running conversation about what it means to discriminate because of pregnancy. (6) In recent years, pregnancy discrimination has been defined almost exclusively through comparison. (6) But as I show, our understanding of discrimination, inside and outside the pregnancy context, depends on judgments about social roles as well as comparison. Conservatives as well as liberals appeal to social roles in arguing over discrimination. (7) More to the point, Congress and the Supreme Court have appealed to social roles in defining the wrongs of pregnancy discrimination.

In enacting the Pregnancy Discrimination Act (PDA), Congress repudiated employment practices premised on the view that motherhood is the end of women's labor force participation, and affirmed a world in which women as well as men would combine work and family--a world in which pregnancy would be a normal condition of employment. (8)1 show how this social-roles account helps make sense of disparate treatment and disparate impact claims of pregnancy accommodation under Young v. UPS, (9) the Supreme Court's most recent decision interpreting the PDA. (10) As the PDA turns forty, the nation increasingly recognizes the importance of accommodating pregnancy in the workplace.

In 1978, Congress amended Title VII of the 1964 Civil Rights Act to declare that discrimination against pregnant women was a form of sex discrimination prohibited by federal employment discrimination law. (11) In the Act's first decade, the Supreme Court enforced the PDA in ways that transformed employment practices. (12) However, in the ensuing decades, lower courts narrowed the PDA by rejecting disparate-impact claims and insisting on finding exact "comparators" before holding that the discharge of a pregnant woman was sex-based disparate treatment. Animating this search for perfect comparators was uncertainty about conceptualizing the exclusion of pregnant workers as discrimination--and concern about imposing on employers the actual or imagined costs of retaining and accommodating pregnant employees. (13)

The public did not accept these judicial decisions narrowly interpreting the PDA. With defeat of discrimination claims under the PDA increasingly common, advocates helped enact pregnant worker fairness acts in twenty-two states, the District of Columbia, and four cities, mandating the reasonable accommodation of pregnancy. (14) And, in 2015, the Supreme Court finally intervened, holding in Young v. UPS, under long-standing Title VII principles, discrimination claims can impose costs on employers, (15) that plaintiffs can bring disparate-impact as well as disparate-treatment claims of pregnancy discrimination, (16) and that disparate-treatment claims are properly analyzed in a framework that employs comparison and balancing, modes of analysis that the disparate-impact framework employs. (17)

Before the Court's judgment in Young, courts reasoned as if the wrong of pregnancy discrimination could be defined solely through techniques of comparison. Techniques of comparison do play a key role in antidiscrimination law. We often talk about discrimination as if it consisted solely in the differential treatment of persons with respect to some trait we deem irrelevant to an individual's ability to perform or contribute. But in this Lecture, as in my other work, I question the sufficiency of this account. Deciding when different treatment, or same treatment, is wrongful requires making a judgment about the larger social world in which the challenged practice occurs. At bottom, then, the wrong of discrimination concerns the social roles and relations it perpetuates. (18)

Yet the social-roles perspective is not just a matter of discrimination theory; it is a matter of law. As I show, both liberals and conservatives appeal to social roles as they argue about the proper application of antidiscrimination law and concern about social roles appears in disparate bodies of antidiscrimination law. (19) Most importantly for present purposes, a social-roles understanding of discrimination shaped Congress's decision to enact the PDA.

When Congress enacted the PDA to amend Title VII, it rejected the long-standing employer practice of firing women who became pregnant, and affirmed the importance of women as well as men supporting themselves and their families when they become parents. Congress amended federal employment discrimination law on the view that pregnancy is, or ought to be, a normal condition of employment. Substantial authority in Title VII case law, in the legislative history of the PDA, and in the case law interpreting the statute supports a social-roles approach to pregnancy discrimination. (20)

Because the social-roles account finds substantial authority in the PDA's legislative history and in Title VII case law, because the social-roles account can illuminate workplace dynamics that might otherwise pass unnoticed, and because the social-roles account is crucial in considering the distributive dimensions of pregnancy-discrimination claims, (21) I draw upon the social-roles account to guide enforcement of disparate-treatment and disparate-impact claims of pregnancy discrimination as the Court most recently addressed them in Young. (22)

In the United States, the limitations of the antidiscrimination framework have become sufficiently plain that members of both parties have endorsed proposals for paid leave. (23)

But paid leave will not supplant the need for antidiscrimination law. Antidiscrimination and welfare mandates are complements, not substitutes. Generously enforced, antidiscrimination laws shift costs from employees to employers, (24) although how completely the law does so will vary from case to case. Antidiscrimination mandates may well not be sufficient to enable certain women to work in workplaces built on the sex-role assumptions of separate spheres. Even so, antidiscrimination mandates still provide important social goods. For women who are able to keep working, antidiscrimination laws help secure an income stream at levels that wage-replacement laws generally do not. And antidiscrimination law has the potential to transform gender norms and expectations about the "ideal worker" (25) in ways that leave alone does not. (26)

Part I of this Lecture demonstrates how views about pregnant women in the workplace have evolved--and persisted--over the last century. In the years before passage of the PDA, judges employed practices of comparison to mark pregnant workers as "different" and to justify their exclusion from the workplace. As I show, these practices of interpretation continued under the PDA itself. Part II argues that practices of comparison depend on explicit or implicit judgments about roles. It demonstrates that appeals to role-based reasoning are commonplace in antidiscrimination law, illustrating this point with examples drawn from race-discrimination law, sex-discrimination law, and the legislative history and case law of the PDA itself. In prohibiting pregnancy discrimination as sex discrimination, Congress employed role-based as well as comparative reasoning. Congress affirmed that women as well as men could combine work and family, so that pregnancy would become a normal condition of employment. Part III then draws on this role-based account to analyze pregnancy accommodation claims within disparate-treatment and disparate-impact frameworks under the Supreme Court's 2015 decision in Young v. UPS.

As Part IV observes in conclusion, the Supreme Court's decision in Young--which invites new consideration of disparate treatment and disparate impact claims of pregnancy accommodation--reflects a growing popular consensus. Forty years after the PDA's enactment, nearly half the states have enacted pregnant worker fairness acts supporting reasonable accommodation of pregnancy in the workplace. As antidiscrimination law and evolving social norms come to inform one another, pregnancy may yet become a normal condition of employment.

1. COMPARATIVE MODELS OF DISCRIMINATION AND PREGNANCY AS DIFFERENCE--BEFORE AND AFTER THE PDA

   Practices of comparison can authorize as well as limit sex discrimination against women workers, as this Part briefly demonstrates. History illustrates that, standing alone, the comparative method is indeterminate and can be mobilized in the service of explicit or implicit role-based judgments. (27) History also illustrates that judgments about pregnancy as "unique" or "different" that preceded the PDA and prompted its enactment have resurfaced in case law enforcing the...