This Article sheds a new light on the controversial issue whether employers should enjoy a market defense when confronted with claims of pay inequality. The conventional view that causation between group membership and adverse treatment is the essence of any discrimination claim enables employers to argue successfully that pay disparities are caused by market forces such as individual negotiation skills and not sex discrimination. The legitimization of the market defense contributes to the gender wage gap among employees performing equal work.
In this Article, I argue that in most cases market justifications for pay disparity in equal pay for equal work litigation should be rejected. I then propose an alternative model of gender discrimination, which is not based on causation. In Equal Pay Act (EPA) claims employers defending their disparate compensation decisions should be restricted to justifications relating only to individual ability and productivity.
Currently EPA jurisprudence is under-theorized. In a series of decisions the courts interpreted the EPA to emulate Title VII of the Civil Rights Act (1967) models of discrimination which are based on strong causation. This enabled employers to make use of the market defense to severe the causal link between sex and compensation. But in January 2009, Congress passed The Paycheck Fairness Act. This bill, amending the EPA, clarifies that the scope of the employer’s affirmative defense is narrow and does not include variants of the market defense.
Too often women encounter the argument that pay disparity is the outcome of market forces and not sex discrimination. Salary differentials are attributed to individual pay demands, bargaining effectiveness, external counteroffers and/or prior salaries. These are just a few examples of market justifications employers raise to explain why similar workers performing the same job are compensated differently. The market defense posits that as long as there is a wedge between employees’ reservation wages1 and their marginal productivity, wages can be set anywhere in between. Any variance in compensation between two similarly productive employees performing the same job is the outcome of differences in the division of the employment contractual surplus in two separate employment relationships.
As union membership continues to decline and the percentage of employees covered by collective bargaining agreements and other fairly rigid compensation schemes also declines, the majority of employees are confronted with the task of bargaining over wages on their own.2 Granting deference to market explanations in this individual bargaining setting will hinder the future quest for gender pay equality. This is because women, on average, tend to be less effective bargainers, and in some cases market justifications are a pretext to sex discrimination.
Under traditional discrimination theory, in order for a plaintiff to prevail in a compensation discrimination claim she needs to identify a causal link between membership in a protected group, i.e., being a women, and the outcome of lower compensation relative to a fellow male co-worker. If the employer argues successfully that market forces are the cause of this disparity, the fact a woman may be paid less than a man for performing the same work is not sufficient to prevail. The market justification is thus presented to overcome the causal link between gender and compensation.
This Article argues that, in most cases, market justifications for pay disparity in equal pay for equal work litigation should be rejected. This Article does not discuss the controversial theory of equal pay for comparable worth, according to which cross- occupational demands for equal pay are made based on a claim that the intrinsic worth of the compared occupations or jobs are equivalent, although the external market places different value on the jobs at question.3 My rejection of the market defense pertains only to individuals performing the same job.
The Article then takes on the more ambitious project of proposing an alternative model of gender discrimination, which is not restricted to causation. Anti-discrimination mandates outlaw employment practices that discriminate against women because of their sex. I argue that in the limited case of equal pay litigation, we should abandon this causation requirement. In Equal Pay Act (1963)4 claims, an alternative model of equality should be endorsed, which would restrict the employer’s ability to defend differential wages for equal work to cases where he can present evidence that individual ability or productivity considerations support the disputed pay disparity. In other words, the traditional causation model is based on an irrelevancy test. Discrimination occurs when an employment action is based on an irrelevant factor such as sex. The other paradigm, which I term “partial causation”, asks whether the decision-maker has confined himself to a check list of relevant factors.5 Whenever a female employee is compensated at a lower rate than a comparable male employee for the same work, it triggers the obligation to inquire whether this can be justified by one of the authorized grounds listed explicitly in the Equal Pay Act (EPA).
The distinction between the two models of discriminations parallels the distinction between two possible regimes that govern the employment relationship regarding job security and the ability of employers to fire their employees without cause. The “employment at will rule”, which is the default rule in most jurisdictions, allows an employer to terminate an employee for any reason - a good reason, a bad reason or for no reason at all.6 Specific exceptions to the at will rule are carved out7, but other then these explicit motivations, the employer is at liberty to base his decision on any other factor. The employment at will rule mirrors the causation model of discrimination. Parties to the employment relationship can opt out of the at will rule, and institute a termination regime based on “just cause”.8 Under a just cause rule, the employer’s discretion in termination decisions is limited by an identified list of authorized reasons to terminate employees, usually relating to performance, discipline and economic needs. A just cause regime is similar in structure to the proposal of identifying discriminatory practices by examining whether the employer was guided by one or more of the factors authorized for determining compensation.9
The gender wage gap is a complex phenomenon. There are at least three causation issues which remain unsettled in the literature discussing the gender wage gap. First, disagreement surrounds the identification of the variables responsible for the wage gap.10 Second, the relative significance of various contributing factors is disputed.11 Third, the underlying relationship between some factors and past and present societal discrimination is also often questioned.12 In this climate of empirical uncertainty adhering to a definition of discrimination which requires the plaintiff to articulate causality between the employer’s seemingly neutral and market guided compensation policies and sex, will frustrate most claims of pay discrimination. Broad interpretation of what constitutes a valid defense may undermine the EPA’s goal of eliminating unjustified wage disparities. Some defenses, especially variants of the market force defense, are actually discriminatory practices since they disadvantage women as a group.
The opportunity to interpret the EPA as restricting employer discretion to pay employees of opposite sexes disparate compensation to a checklist of authorized factors relating to productivity and ability was overlooked by the courts. Currently EPA jurisprudence is under-theorized. In a series of decisions the courts interpreted the EPA to emulate Title VII of the Civil Rights Act (1967)13 models of discrimination which are based on strong causation.14 This enabled employers to make use of the market defense to severe the causal link between sex and compensation. But in January 2009, Congress passed H.R. 12, The Paycheck Fairness Act.15 This bill, amending the EPA, clarifies that the scope of the employer’s affirmative defense is quite narrow, and in fact is limited to considerations closely related to individual ability and productivity. If this important legislation becomes law, the market defense will be eliminated altogether from the EPA framework of discrimination.
The Article proceeds as follows. Part II explores what employers assert when offering a market defense to discriminatory compensation claims, it then offers the normative analysis of why the market defense should be rejected. It draws on the growing body of research demonstrating gender differences in negotiation skills, self entitlement and competitive behavior. Part III explains how traditional models of discrimination, based on causation, are ill equipped to deal with pay disparities resulting from market behavior of individual employees. Part IV develops the argument that we should abandon the causation requirement for pay disparity claims, substituting it with a checklist of relevant factors which will govern compensation decisions. It then explains why the EPA was interpreted as embracing a causation model, although its structure clearly indicates a departure from traditional discrimination models. Part IV also discusses the Paycheck Fairness Act. If passed, this bill will settle the ambiguity surrounding the reach of the market defense.16 Part V concludes with a response to criticism voiced against eliminating the market defense from the EPA.
II. Why Are Some Market Justifications to The Gender wage Gap Discriminatory? A. What Are Market Justifications? A market explanation harnesses free market principles, such as supply and demand equilibrium, promotion of self interest and/or profit maximization to justify contested wage disparities. Early on employers turned to market justifications to distance themselves from accusations that animus toward women (and other protected groups) motivated their adverse decisions. Basically their defense culminated in employers asserting “it’s not me, it’s the market which forced me to treat women unfavorably”. In effect, two market theory assumptions were thus brought together. First, employers are usually not wage setters, but rather wage takers, following the market rate, dictated by the supply and demand for employees in the specific industry or profession. Second, efficient labor markets will gradually eliminate any irrational or animus-based discrimination. The concept that an employer can excuse sex based decision-making with rational and profit maximizing reasoning (as opposed to animus) was quickly dismissed by the Supreme Court. In Corning Glass Works v. Brennan17 women inspectors working the day shift demanded pay equal to that of male inspectors working the night shift. For other jobs there was one wage rate, applied both to day and night shifts. Corning argued that the wage differential between day and night shift inspection jobs resulted from the resistance of men to perform inspection work, work perceived by these workers as feminine. In order to fill the night shift, Corning had to offer higher compensation than what was offered to the women working the day shift. The Court determined that “the differential arose simply because men would not work at the low rates paid to women inspectors, and it reflected a job market in which Corning could pay women less than men for the same work. That the company took advantage of such a situation may be understandable as a matter of economics, but its differential nevertheless became illegal once Congress enacted into law the principle of equal pay for equal work.”18 After the Court held that the EPA prohibited compensation decisions that take into account the sex of workers even if such pay disparities are supported by market justification, a more subtle form of the market justification emerged. One variation focuses on the legitimacy of the employer attempting to extract as much of the contractual surplus as possible. In these cases, for example, an employer argues that there is no legal obligation to offer individual workers more than their initial pay demands, even if implementation of a wage scheme based on employee wage demands ultimately disadvantages women.19 Another strand emphasizes specific circumstances where employers are compelled by external market pressures to raise the compensation of one employee but not another. An example is if an employee presents a counter offer and threatens to quit if his current employer does not match it.20 Or temporal changes in the market wage in an industry or profession can result in disparate compensation of two equally productive employees, hired in two different time periods.21 The strength of these new versions of the market defense is that they purport to sanction neutral criteria that regrettably resulted in an individual female employee being paid less than a male co-worker,22 rather than being offered as a justification for an intentionally sex based compensation decision. Employers have utilized these versions of the market justification to overcome the causal link between sex and the level of compensation that a plaintiff must prove under the traditional theory of discrimination. This line of reasoning has gained acceptance, to varying degrees in the courts.23 In the next sub-sections the paper explores whether market explanations do in fact undermine the causal link between wages and gender. I set out to prove two assertions. First, despite being neutral on their face some market justifications adversely impact women as a group and therefore the causal link between gender and pay is still present. Second, in some cases employers claim their decisions are constrained by external market pressures, when in fact their decisions are not subject to such pressures. Rather their compensation schemes are a product of internal institutional policies and politics that are sometimes entangled with gender stereotypes and other sex based considerations. If this is an accurate description of labor market practices, market justifications as a normative matter should not be accepted as a legitimate defense in pay discrimination cases. Denying such defenses will also promote uniformity and coherence in the treatment of market based defenses that disadvantage women.
B. Market Justifications Feed and Perpetuate the Gender Wage Gap In their influential book “Women Don’t Ask: Negotiation and the Gender Gap”24 Linda Babcock and Sara Laschever present empirical evidence on the difference between the manner in which women and men engage in salary negotiations. The results of their research pose disturbing implications for the impact of allowing market justifications upon the gender wage gap. The essence of their findings is that women are more hesitant than men to initiate and pursue negotiations over wages. They are less likely than men to negotiate over initial wage offers when accepting a new job. Women are less likely than men to demand a raise and less likely to seek counteroffers to boost their current compensation. The different negotiation skills that women and men bring to the bargaining table impact the gender wage distribution in institutions that rely heavily on individual bargaining to set wages.
The book starts off by illustrating the disparity in starting salaries of graduates with a master’s degree from Carnegie Melon University. Male graduates received an initial average salary 7.6% higher than that of female graduates. The cause of the gendered deferential in the average wage was that 57% of men negotiate over the initial offer they received, while only 7% of the female graduates did so.25 This data shows that a majority of men used an initial salary offer as a starting point for negotiations, while most women simply accepted the initial offer.
The researchers point to the attainment of early childhood social skills as the explanation for the difference in negotiation skills. Boys are socialized to ask for things they want, and tend to believe they have control over the circumstances that shape their lives. Young girls are socialized to focus on the needs of others and to apply less control in altering their own situation. Therefore women are less inclined than men to come forward and demand raises, bargain over initial wages, or approach other employers in order to solicit competing job offers.26 Another line of research complementing the gender negotiation literature, investigates the correlation between competitive behavior and gender. Muriel Niedele has been conducting experiments demonstrating that women shy away from competitive environments, while men not only welcome them, but perform better when competing.27 Wage negotiations are commonly seen as a competition, a zero-sum game, and women tend to avoid it. Again, the variance in attitudes and performance in a competitive setting is linked to the development of early social skills and a disparity in the self-confidence women and men exhibit in their belief in the chances of winning the competition.28 These beliefs affect both one’s willingness to compete and performance in competition. The negotiation literature also emphasizes the importance of the negotiator’s self-confidence and optimism. There is correlation between self –confidence, optimism and the final outcome of the negotiation: if you expect more – you will get more.29 Tied to these findings are experimental studies, which reveal that women, absent external information, value the economic worth of their work less than men.30 In one of the first studies on self-perception of entitlement, participants were requested to perform a task. After completion of the task, participants were divided into two groups. The first group was asked to assign compensation to themselves for performing the task. The second group was asked to decide the compensation other participants received for the task. On average women in the first group assigned 19% lower compensation for themselves than men did, but when compensating other participants, women were slightly more generous than men, regardless of the sex of the participant they were compensating. There, however, was no difference in the participants’ evaluation of the quality of their own work.31 These results show that women undervalue their own work, but are able more objectively to assess the value of the work of others. Men, on the other hand, did not exhibit such a discrepancy. A follow up study included a third group that was required to assign compensation to themselves, like the first group, but were provided a bogus list with information on how much other participants paid themselves. Under these conditions, women adjusted their compensation upward to meet the rates included in the list.32 The results in this study suggest that lack of information about the going rate of compensation tends to depresses women’s wages and contribute to the gender wage gap33.
These empirical findings can also shed light on why women’s reservation wage is often lower than men’s. The conventional explanation emphasizes market discrimination34 and the greater variance in women’s attachment to the paid labor market.35 But lower self-valuation of one’s work, whether intrinsic or caused by lack of relevant and accurate information, depresses women’s reservation wage. There is a correlation between higher estimation of the market value of the job preformed and higher reservation wages36. Holding productivity constant, a lower reservation wage translates into a larger contractual surplus to be divided between the employer and employees. Even if the employer extracts an identical share of the surplus through bargaining, the employee will end up with lower wages than a co-worker whose reservation wage was initially set higher.
To bring to life the insights discussed so far, take for example two employees, Emma and Ben, with equivalent productivity of $100 per day. This sets the upper limit to their compensation at a rate of a $100 per day. Due to the constraints described above, Emma values her productivity at $80, while Ben estimates his productivity at $95. Emma sets her reservation wage, which is the minimum wage she is willing to work for, at $50 and Ben sets it at $60. The fact that Emma estimates her productivity lower than Ben does not necessitate that her reservation wage also be lower, she could be unwilling to work for less than $75 a day even if she estimates her productivity only at $85.37 The contractual surplus in Emma’s case is $50 and in Ben’s case it is only $40. This puts Ben at an advantage because even if in both cases bargaining will result in splitting the surplus evenly, Emma will end up with a wage rate of $75 per day and Ben with $80. Taking into account their subjective evaluation of their productivity, Emma estimates the contractual surplus at $30 compared to Ben’s estimate of $35. Splitting the employee perceived surplus evenly in both cases will result in $65 for Emma and $77.5 for Ben.
Incorporating the information about the systemic differences in the way women and men approach and handle wage negotiation it is fair to assume that Ben will be able to extract a higher share of the actual employment contractual surplus. Emma may not engage in bargaining at all, and if she does, she will be less effective. Emma will set her bargaining goals lower, in part because both her reservation wage and subjective estimation of her productivity is lower than Ben’s, and partly because she is not socialized to bargain for her own benefit and is averse to the competitive environment of wage negotiations. If this results in Emma extracting 10% of the surplus while Ben succeeds in extracting 20%, Emma will be compensated at a rate of $55 compared to Ben’s compensation of $72.
This hypothetical example illustrates that various factors pertaining to individual salary negotiation may contribute to gender wage disparity. None of these factors is connected to any objective measure of productivity, but are linked to gender, and adversely impacting women, thus meeting the causation requirement underlying traditional discrimination law, under disparate impact law.
C. Market Justifications May Serve as a Pretext for Discriminatory Behavior The strand of the market justification that emphasizes external market pressures relies on the assumption that both employers and employees are “price takers” in the sense that the external market determines the wage rate for industries and occupation. Under this theory the only discretion an employer exercises is determining how many employees they are willing to hire at the going wage. It is thus argued that when the employer is not the one who is actively setting wages, but simply following the market rate, he should not be held liable for the external valuation of worth.38 This intuition is probably one of the main causes for the failure of the comparable worth movement. In the late seventies to mid eighties comparable worth proponents advanced, unsuccessfully, the argument that when job segregation results in the depression of wages of female occupations compared to comparable worth male occupation this should be perceived as sex discrimination.39 But courts resisted expansion of discrimination law based on this theory, explaining that when an employer pays his workers according to the market going wage, wages are not actively set in a discriminatory manner. The employer is simply following the rules of supply and demand for the various jobs, even if the outcome is that female dominated occupations attain lower compensation levels than comparably worth male dominated occupations.40 The rejection of comparable worth rested partially on the strong belief that employers are wage takers for the various occupations for which they hire workers.41