Abstract
Excerpted from: Laura T. Kessler, Employment Discrimination and the Domino Effect, 44 Florida State University Law Review 1041 (Spring, 2017) (320 Footnotes) (Full Document)
Who is responsible for gender, race, and other stubborn patterns of worker inequality? This question lies at the heart of all of the theories of liability under Title VII. From disparate treatment to sexual harassment, from affirmative action to disparate impact, the ultimate question is whether worker inequality is due to some unlawful action by employers, for which employers must be held accountable, or due to factors outside employers' responsibility or control. When an employer calls Greg Baker for a job interview rather than Lakisha Jones, is it because Greg's resume suggests he is better qualified or because Lakisha's name is African-American-sounding? When a casino fires a female bartender after twenty years of service for refusing to wear make-up in compliance with its new grooming policy, is her termination illegal sex discrimination or a legally permissible decision based on male customer preference to have their drinks served by feminine women with sex appeal? Similarly, if a retailer of teen apparel decides to brand its “Authentic American Clothing” around the concept of racial and other types of exclusion, is it responsible when it routinely steers Hispanics, Asians, and African Americans to stockroom jobs, or is this a legitimate profit-related practice? When an ambitious female junior investing partner in a Silicon Valley venture capital firm is not promoted, despite her investment successes, is it because she is a woman or because she is perceived as being ungrateful and difficult, and being a likeable “team player” is more important at the firm? At Wal-Mart, the most profitable retailer in the United States and the largest private employer in the world, women make up only thirty-three percent of management employees despite filling seventy percent of the retailer's national sales workforce. Women are also paid less than men in every region.
Is this because gender bias suffuses Wal-Mart's culture? Or can these patterns be explained by a lack of women who are qualified and interested in management positions at Wal-Mart and “left to their own devices most managers in any corporation--and surely most managers in a corporation that forbids sex discrimination--would select sex-neutral, performance-based criteria.” If an upscale restaurant has a reputation for not hiring female food servers, and this reputation discourages qualified women from applying for server positions, is this employment discrimination or the result of the women's personal choices? If a fire department uses a weightlifting test as its primary physical selection procedure, is the lack of women firefighters due to the fire department's hiring criteria or because the average man is stronger than the average woman? What if agility, balance, endurance, aerobic capacity, speed, and teamwork are as important to successful firefighter performance as upper-body strength? Should Title VII make a fire department liable for sex discrimination if it emphasizes upper-body strength over these other important qualities in its selection criteria? If a female postal service driver becomes pregnant and her doctor advises her not to lift more than twenty pounds, her employer forces her on unpaid leave, and she loses her medical insurance, is this sex discrimination or simply the employee's unfortunate problem, since she temporarily cannot meet the job's requirements? If an African-American dining services employee at a university is the subject of ongoing racial harassment by a white coworker, is the university vicariously liable for the harassment? Or, rather, is this behavior an unauthorized act of the white employee for which the university is presumptively not responsible unless the victim complains and the university negligently fails to respond? What if, fearing for her job, the victim does not complain at all, or she complains, but to the wrong person (for example, to a mid-level supervisor or a union representative who does not have the authority to discipline or fire the harasser)? Should this create liability or is the matter, again, not the employer's responsibility?
Each of these examples is drawn from a recent, real-world employment discrimination case. As they demonstrate, every instance of alleged employment discrimination can be conceptualized as a choice between an illegal “inside” cause of worker inequality and a legal “outside” cause of worker inequality. Economists and other social scientists describe this divide in terms of “demand side” and “supply side” explanations of worker inequality. Broadly, demand side explanations of worker inequality focus on discrimination occurring inside workplaces such as intentional discrimination, unconscious biases, and neutral policies and practices that systematically disadvantage workers protected by employment discrimination statutes. Supply side theories, in contrast, attribute inequality to workers' personal preferences, qualifications, and performance. Supply side factors include, for example, the absence of requisite job skills; differences in education, training, or motivation; culture and socialization; and choices that employees make in light of family obligations and other personal circumstances.
The major theories of employer liability under Title VII sharply differentiate between demand side and supply side causes of worker inequality. For example, Title VII disparate treatment claims are premised on the assumption that an adverse employment action is either because of an employer's illegal consideration of protected characteristics (such as race, sex, or national origin) or for a “legitimate, nondiscriminatory reason.” Within this analytical framework, there is no room to consider if discrimination may have negatively impacted the “legitimate, nondiscriminatory” basis for an employer's decision, such as an employee's job performance. That is, by its very definition, the legal concept of disparate treatment ignores the social structure in which prejudice, bias, and discrimination operate.
To be sure, some aspects of Title VII doctrine acknowledge that demand side and supply side explanations for worker inequality overlap and are difficult to neatly separate from one another. For example, the mixed-motive proof structure suggests that both demand side factors (i.e., discriminatory considerations of protected characteristics) and supply side factors (i.e., legal considerations of employee qualifications or performance) may concurrently play a role in an employment decision, with the ultimate inquiry focusing on which factor predominated the decision. Similarly, Title VII's disparate impact theory of liability recognizes that facially neutral employer policies or practices may so systematically and unjustifiably stack the deck against protected employees that liability for discrimination should attach. As such, the disparate impact theory recognizes that structural aspects of the workplace negatively affect individual workers. And the hostile work environment theory of liability defines unlawful discrimination to include a work environment severely and pervasively infected with discriminatory, offensive conduct, such as threats, intimidation, and ridicule, even in the absence of any formal personnel action, because of the exclusionary effects of such treatment.
It is easy to point to these examples and conclude that Title VII is at least reasonably sensitive to the interplay between demand side and supply side drivers of worker inequality. However, a close study of Title VII doctrine reveals a decidedly less positive picture. Courts routinely assume a sharp distinction between demand side and supply side explanations of worker inequality when analyzing and applying Title VII. For example, although the mixed-motive theory recognizes that both legal and illegal considerations may factor into an employment decision, the mixed-motive theory still assumes the absence of any causal relationship between the legal considerations (e.g., employee motivation, performance, qualifications) and illegal considerations (e.g., bias on the basis of protected characteristics). When considering systemic disparate treatment and disparate impact claims, courts often attribute stark racial and gender disparities in pay and workforce composition to external causes, such as the absence of diversity in the applicant pool, with little regard for the powerful role of employers in influencing the labor markets in which they operate. Finally, the Supreme Court has carved out a broad affirmative defense to employer liability for sexual harassment that, in practical effect, requires victims of harassment to report in virtually all circumstances or risk losing their claims. This doctrine neglects the power dynamics and economic vulnerabilities that lead victims not to report harassment. By defining discrimination and employee behavior as mutually exclusive phenomena, sexual harassment law, like Title VII's other theories of liability, ignores the social patterns of discrimination that shape the employees subject to them.
Mirroring the bifurcated approach in Title VII doctrine, much social science research and public discourse on employment discrimination defines and constructs the issue as a question of whether demand side or supply side phenomena are responsible for race, gender, and other identity-based patterns of worker inequality, with little attention to the causal interrelationships between demand side and supply side factors. Consider, for example, the recent public debate between Facebook CEO Sheryl Sandberg and former Princeton Professor Anne-Marie Slaughter about why women cannot rise to the top professionally. In her book, Lean In, Sandberg emphasizes the ways that women lower expectations for themselves in the workplace; she urges women to strive for leadership roles despite discriminatory challenges. Her analysis emphasizes individual women's responsibility for overcoming discrimination. In contrast, Anne-Marie Slaughter's Atlantic Monthly article and subsequent book on balancing work and family focus on governmental policies and structural features of workplaces that result in inequality for family caregivers (still primarily women in our society). Although the differences in Sandberg's and Slaughter's positions are perhaps more a matter of emphasis or degree, which in turn influences what each commentator sees as the appropriate remedy for gender-based economic inequality, the media has held up these two prominent influential women as opposing voices. There is a similar “either/or” framing in research and public discourse on the gender pay gap, with proponents of wage equity attributing the wage gap to discrimination and wage-gap deniers emphasizing supply side human capital factors, such as education, experience, and individual worker “choices.”
As these examples demonstrate, Title VII's major legal doctrines, as well as public debates about employment discrimination, regard the three prevailing explanations of worker inequality--individual employee choices and characteristics, biased decisionmaking, and structural features of the workplace--as distinct and independent phenomena. The result is that our country's most important federal employment discrimination law is oftentimes unable to redress employment discrimination as it actually manifests inside workplaces. Further, by failing to recognize the dynamic, interactive processes generating worker inequality, legal and political discourses on discrimination mask the pervasive and powerful role of institutions in creating inequality.
Many legal scholars have addressed the stubborn nature of discrimination and the often complex and nuanced ways that it manifests in the workplace. For example, many have discussed the unintentional and unconscious nature of much discrimination, emphasizing the mismatch between this reality and disparate treatment law. Others have examined the organizational context of work as a driver of inequality, focusing, for example, on how organizational practices, such as decentralized, subjective decisionmaking, the creation of non-diverse work groups, and other features of organizational design and culture may influence the occurrence of discrimination. Still other scholars have documented how employees may respond to discrimination with strategies aimed at dispelling stereotypes that may attach to their identities. Taken together, this substantial body of scholarship has led to considerable advances in our understandings of the dynamics of discrimination in the modern workplace. However, few scholars have sought to comprehensively theorize the interrelationships among all three drivers of inequality: bias, structure, and employee responses to these phenomena.
In this Article, I try to juggle all three balls at once, so to speak, that is, to re-theorize Title VII doctrine to account for the interplay between organizational structures and discriminatory bias, on the one hand, while also considering how employees commonly respond to these demand side forms of discriminatory exclusion. This analysis reveals that worker inequality is often the result of the interplay between supply side and demand side processes. That is, in a broader respect than has generally been appreciated, there is a dynamic relationship among individual employee characteristics and preferences, biased decisionmaking, and structural barriers to worker equality. For example, individuals' career aspirations and job performance are shaped by both biased employment decisions and the organization of work. Similarly, organizational arrangements can serve to exacerbate or dampen discriminatory bias. Biased decisionmaking and structural impediments to equality occur simultaneously and combine and interact in dynamic ways that are internalized by individual employees, affecting their “choices” and work performance. In this view, discrimination is not an act or set of acts (as contemplated by disparate treatment and systemic disparate treatment) or a neutral policy with discriminatory effects (as contemplated by disparate impact). Rather, discrimination is more like a chain reaction involving individual worker behavior, biased decisions, and the organization of work that, through a process of positive feedback, produces and amplifies inequality. I refer to this process as the “domino effect.”
In its most literal sense, the domino effect refers to the physics of a row of toppling dominos. However, the concept has come to be used in a variety of contexts either literally, to refer to an observed series of physical collisions, or metaphorically, to describe causal linkages within systems such as computer networks, global finance, or politics. The metaphorical meanings of the term have varied widely; at the most basic level, the idea denotes that a small event may have unanticipated, far removed effects. A broader conception, which I employ in this Article, is that a seemingly small and insignificant incident can mushroom into a much larger, comprehensive problem. As I will argue, in the employment context, relatively small and insignificant discriminatory acts, policies, or work structures oftentimes initiate a chain reaction resulting in substantial and materially adverse forms of worker inequality, such as unequal pay and status. In addition to highlighting this process, which has been underexplored in legal scholarship, a key contribution of this Article is to examine how employees' responses to discrimination are important to understanding the production of inequality.
It is important to note at the outset that I do not claim to definitively describe or predict workplace domino effects with certainty. Although social scientists have an ever-deepening understanding of the processes of stereotyping, prejudice, and discrimination, they have not arrived at any definitive theory. The problem of hierarchy and inequality in the workplace is multifaceted. Moreover, the precise character and manifestations of the domino effect are likely to differ across occupational and organizational contexts. Still, as I develop more fully below, social science research employing a wide range of methodologies in a wide range of work settings over a long period of time has consistently and reliably identified institutional and social processes by which inequality is created and maintained by organizations. This extensive body of research demonstrates that demand side and supply side drivers of worker inequality are not independent of one another. My objective is to begin a conversation. How might Title VII's major theories of liability be modified, and what might a larger social policy agenda look like, were we to reject the following two flawed premises of Title VII: First, that inequality is a result either of the characteristics and preferences of individual workers or biased decisionmaking and organizational-level systems of stratification; and second, that there is no causal relationship among these phenomena?
An immense reform agenda emerges when we consider the implications of the domino effect for Title VII. For example, the assumed independence of an employee's work performance from discriminatory employer actions in disparate treatment law becomes incoherent once we account for the domino effect. This Article represents an effort to provide the initial empirical and theoretical groundwork for the larger project. As a concrete starting point, this Article focuses on two related policy contexts: sex-based employment discrimination and worker inequality arising from work and family conflict. These are especially fruitful domains of legal concern to examine the workplace domino effect.
Lack of paid family leave, inflexible and unpredictable work schedules, insufficient paid sick leave, the absence of accommodation for the physical limitations of normal pregnancy, and long work hours are common features of American workplaces that make it difficult for employees who become pregnant or have significant family responsibilities to perform as ideal workers. At the same time, pregnancy and family care responsibilities can make an employee's sex and gender more salient in the workplace, triggering animus or bias by coworkers and managers. Once either or both of these processes are set in motion, a chain reaction often ensues. In many instances, what may have begun as inconsequential, isolated, or at least surmountable differences in employee availability or energy become the justification for differential treatment, whether it be differences in mentoring, training, and evaluation, for example, or more serious consequences, such as failure to promote or even the decision to terminate an employee. That is, structural barriers, bias, and employee responses to discrimination often combine and reinforce one another so as to produce substantial worker inequality.
Because these discriminatory dynamics are especially acute in the context of work and family conflict, this is a fruitful area of employment discrimination law to illustrate the operation of the domino effect. Specifically, I use this particularized form of gender discrimination as an example to illustrate how discrimination commonly plays out inside work organizations, the effects it has on individuals, and how it might be challenged. However, my extended focus on gender discrimination and work and family conflict is not meant to suggest that the domino effect is limited to this form of discriminatory exclusion. For this reason, many of the empirical studies and cases I discuss address race and other types of discrimination, and this Article's analysis should be useful to scholars and advocates working to remedy discrimination across identities and contexts. Its contributions operate on three levels: doctrine, theory, and methodology.
First, at its most basic level, this Article aims to assist plaintiffs' lawyers by distilling complicated processes of discrimination into a simple, familiar, concept that can be used to frame Title VII litigation. Most people-- including judges and jurors--have at some point in their lives witnessed the spectacle of dominos toppling in a cascade. The analytical framework introduced in this Article can also, ideally, persuade courts to update and reformulate Title VII doctrine to better account for the dynamics of contemporary discrimination.
Second, on a broader theoretical level, this Article demonstrates that the conceptual bifurcation of the causal mechanisms of worker inequality into supply side or demand side categories is, in and of itself, a political construct disguising the role of institutions and markets in producing inequality. Orthodox economic theory, struggle as it may, provides the basic template for this binary. According to this strain of economics, sustained observed differences in economic outcomes between groups are due to a deficiency in the group experiencing the inferior outcomes. Economists refer to the deficiency as one in human capital. Sometimes the deficiency is said to be associated with poor schooling opportunities, other times with culture, socialization, or motivation. But the thrust of the argument is to absolve organizational and market processes of a role in producing the differential outcome; the inherent deficiency is theorized to occur in pre-market or extra-market processes. This framework is woven into the very fabric of Title VII. Every major proof structure under Title VII is built on this template, setting up a conflict between alleged employee deficiencies, on the one hand, and rational employer decisions and efficient work structures, on the other. In this view, the trier of fact only needs to choose between mutually exclusive explanations for bad worker outcomes. As this Article demonstrates, this choice is overly simplistic, as even the most basic forms of discriminatory exclusion, such as individual disparate treatment, involve an interplay of demand side and supply side factors.
Finally, this Article offers a methodological innovation. I assert that a fruitful way to contest the pervasive influence of orthodox economic theory on employment discrimination law is to marshal the insights and theories from social sciences that take the “social” part of their mission seriously. Sociology, social psychology, and sociologically-grounded business management research on work organizations, in particular, are promising fields for challenging neoclassical economic foundations of employment discrimination law, because they focus on organizational and societal-level systems of social stratification. So many of Title VII's theories of liability and legal doctrines focus on the individual--that is, whether and to what extent the individual employee is to blame for his or her predicament--rather than the interplay between organizational structures and individual agency. Disciplines and methodologies that attend to the social dynamics inside work organizations and the institutional practices that shape employee behavior offer an antidote to the inordinate focus in Title VII doctrine on individual employees' education, qualifications, training, merit, performance, and personal choices.
Sociological and organizational-level understandings of worker inequality can also serve as an important supplement to scholarly work on unconscious bias in the workplace. In the past fifteen years, the science of implicit cognition has achieved a firm foothold in the legal field of discrimination law. Specifically, many legal scholars find special promise in a particular line of research in cognitive psychology that measures bias with the Implicit Association Test or “IAT.” The IAT assesses the existence and strength of racial, gender, and other biases by measuring “response latency,” for example, how long it takes to make a stereotype-consistent association, such as “women” and “crochet,” as compared with the time needed to make a stereotype-inconsistent association, such as “women” and “strong.” Scholars who promote the IAT emphasize the central role of unconscious bias in employer decisions.
Unconscious or “implicit” bias refers to prejudiced judgments that may affect our understandings, actions, and decisions. It is a type of cognitive shortcut that occurs when our brains make quick judgments and assessments of people and situations, informed by our background, cultural environment, and personal experiences. Many legal scholars see this brain science as having the potential to transform how we understand and address discrimination throughout the law, because so much discrimination law requires proof of intent.
Certainly, the science of implicit social cognition has been of some assistance in educating judges and policymakers about the nature and prevalence of bias, with important victories for this intellectual movement. Still, there are limitations to the utility of this science as a tool for achieving progressive legal change in the employment context. Most research in the field focuses on individual-level explanations of worker inequality and, therefore, may lack sufficient power to challenge conservative economic and political theories that similarly locate the cause of worker inequality inside the individual. In addition, deploying implicit bias research in law reform projects has had the unforeseen consequence of perpetuating the misconception that biased decisionmaking cannot be controlled, and consequently, the belief that employers cannot reasonably be held accountable for the resulting discrimination and inequality. Worse, it risks sending the message that stereotyping is okay, since the theory teaches that everyone has bias. This may make discrimination seem socially acceptable and lessen the motivation to avoid it. Given these risks and limitations, the project of achieving equality in the workplace for protected groups requires a more robust account of the interactions between bias and structural discrimination than advocates of implicit bias research in law sometimes propose. Sociology, social psychology, and related fields in law and society, such as new institutionalism and new legal realism, may help here, because these fields contribute to our understandings of how organizational- and societal-level systems of social stratification facilitate inequality. To succeed in developing a more integrated account, however, legal scholars must overcome the presumption that unconscious bias and structural contributors to worker inequality are unrelated to one another, thereby reinforcing the very same limiting frameworks reflected in legal doctrine and embraced by courts. Employment discrimination scholarship is roughly divided into two subfields, one concentrating on unconscious bias and the other on the institutional nature of discrimination. It would be beneficial to the field of employment discrimination law to develop an account of the interplay among different processes of discriminatory exclusion in the workplace. Attending to the dynamic interplay among individuals, organizations, and society in producing inequality is likely to lead to a better understanding and reduction of gender-based and other forms of employment discrimination.
Part II of this Article offers an extended fictional hypothetical involving work and family conflict, gender and sexuality discrimination, and sexual harassment to illustrate the workplace domino effect. Through a legal analysis of the factual problem presented, Part II then demonstrates how ill-equipped discrimination law is to identify and remedy the common domino-like processes that cause substantial worker inequalities.
Part III examines the prevailing conceptual frameworks that social scientists and courts use to explain and understand gender-based worker inequality. Specifically, Section III.A. reviews social science research on gender-based employment discrimination. This review demonstrates that social scientists tend to view women's work-related choices, gender bias, and structural features of the workplace as mutually exclusive phenomena. Section III.B. turns to the analytical frameworks that guide how employment discrimination cases are litigated, including disparate treatment, systemic disparate treatment, and disparate impact. It shows how these frameworks also, almost uniformly, assume that employee's choices and behaviors, employer bias, and exclusionary work structures are independent drivers of worker inequality.
Part IV sketches a more accurate, multidimensional account of the dynamic processes by which worker inequality is created and reinforced inside work organizations. This discussion draws from a variety of fields, primarily sociology and social psychology, but also from business management research on organizations, new institutionalism, feminist legal theory, and critical race theory. The research reviewed in Part IV demonstrates that various processes of discrimination, such as biased decisionmaking and structural impediments to equality, occur simultaneously and combine and interact in ways that amplify discrimination, oftentimes with the aid of a process of psychological internalization of its targets, resulting in tangible harm to employees.
Part V explores a number of interventions that follow from this Article's main empirical and theoretical contributions on the workplace domino effect. Specifically, Section V.A. discusses voluntary measures that employers can adopt to disrupt the feedback loops among processes of discrimination documented in this Article. These voluntary measures are evidence-based and therefore should be effective if there is a commitment to preventing and remedying discrimination. In Section V.B., recognizing that employers' commitment to antidiscrimination is oftentimes lacking without the risk of liability, I explore litigation strategies and logical revisions to several core doctrines in Title VII legal jurisprudence that would allow the law to better address the workplace domino effect.
In formulating solutions, I proceed from two working commitments: First, there is a grave mismatch between what we know from social science about how discrimination operates today and the model we inherited from fifty years ago, which does not account for the dynamic interaction among employee choices, bias, and structural features of the workplace that produce inequality. This mismatch goes well beyond the oft-discussed failure of Title VII to account for the unconscious nature of bias. Second, despite the major setbacks that Title VII has suffered in the past several decades, and plaintiffs' consequent difficulties proving employment discrimination, developing transformational analytic frameworks that can illuminate the social processes of inequality is an important and necessary project for employment scholars. Disparate treatment law, in particular, is a critical component of this work. Although disparate treatment has always served as the practical and conceptual core of Title VII, it is now all the more important to dedicate energy to reinvigorating individual claims, given the Court's apparent hostility to systemic claims of employment discrimination.
. . .
Title VII has labored too long under the weight of black and white thinking. A significant body of sociological research on how discrimination operates on the ground, inside workplaces, every day is now available to guide courts and policymakers. This research demonstrates that worker inequality often results from a series of discriminatory acts or conditions that combine and interact in ways that, over time, lead to large differences in employee status and pay due to their cumulative and mutually reinforcing nature. Unfortunately, the unwillingness to think rigorously about how discrimination occurs has had serious negative consequences. Stubborn patterns of discrimination exist across every industry and workplace setting in America. This situation will not change without a fundamental reconceptualization of Title VII so it may account for the domino effect and other contemporary forms of discriminatory exclusion.
Professor of Law, University of Utah S.J. Quinney College of Law.
