chapter 17

TITLE VII OF THE CIVIL RIGHTS ACT OF 1964 AND RELATED LAWS

The Civil Rights Act is a far-reaching law that grew out of the civil rights movement of the 1950s and 1960s. Title VII of the act prohibits discrimination on the basis of sex, race, color, religion, or national origin in any employment condition, including hiring, firing, promotion, transfer, compensation, and admission to training programs. Title VII was amended in 1972, 1978, and 1990. The EEOC is responsible for Title VII enforcement. As we saw earlier in Exhibit 17.8, the cost to employers of equal employment opportunity (EEO) issues broadly (including Title VII) was about $2 billion over a 5-year period, taking into account both EEOC enforcement activity and private class action lawsuit settlements. In addition to Title VII, other key EEO statutes, the 1967 Age Discrimination in Employment Act (ADEA) and the 1990 Americans with Disabilities Act (ADA) also prohibit discrimination based on age and disability, respectively. Compliance with the ADEA is typically a key concern when companies use workforce reduction programs. The ADEA pertains not only to age-related differences in pay and employment outcomes, but in addition, it was amended in 1990 to include the Older Workers Benefit Protection Act (OWBPA), which has detailed rules regarding how separation agreements (e.g., an early retirement incentive) involving older workers are used. As one example, at least 21 days must be given to consider the agreement. Title VII cases of pay discrimination typically focus on differences in pay, promotions, pay raises, and performance reviews. Race-based differences in these areas were at the center of litigation against Merrill Lynch and gender-based differences were the issue at Bank of America. These companies settled with the plaintiffs for $160 million and $39 million, respectively.57 Of course, these cases are very costly for the company not only in terms of legal expenses, but also in terms of unfavorable publicity, employee relations, and the allocation of time away from the core business. Organizations that can successfully be proactive in maintaining compliance increase their chances of avoiding such litigation. These types of settlements are the result of class action lawsuits. A class action lawsuit is "any civil case in which parties indicated their intent to sue on behalf of themselves as well as others not specifically named in the suit at some point prior to the final resolution of the matter."58 Employment-related class action lawsuits such as those that pertain to EEO/discrimination can be very costly to employers because they can include large numbers of potential plaintiffs (employees, former employees, potential employees). These potential plaintiffs can potentially be included as part of the class without contacting them in advance. Instead, they can be given the opportunity to opt out in the event the lead plaintiffs prevail in court or reach a settlement with the employer. (FLSA lawsuits differ.)59 Such settlements typically include back pay (pay that would have received absent discrimination) for each member of the class. The bigger the class, the bigger the potential cost to the employer. Clearly, a key issue in class Page 646actions is the definition of the class. The easier it is to include large numbers of potential plaintiffs, the more potential liability the employer faces and the bigger the potential payoff for plaintiffs (and their attorneys). Rule 23 of the Federal Rules of Civil Procedure and 29 U.S.C. § 216(b) provide the requirements for forming a class and one requirement is a commonality of interests.60 A 2011 U.S. Supreme Court decision in Walmart Stores, Inc. v. Dukes, et al. made it more difficult for plaintiffs to certify a class of potential plaintiffs. The Supreme Court "reversed a class certification decision that joined the claims of 1.5 million female salaried and hourly employees who held any number of positions across Walmart's 3,400 stores because the plaintiffs could not articulate a common question that was capable of common answers as to the entire class. The Supreme Court ruled that the plaintiffs were required to demonstrate that Walmart operated under a 'general policy of discrimination,' which it concluded was 'entirely absent' [in the evidence presented in the case]."61 Some experts believe the Walmart ruling has already resulted in employers more aggressively challenging class action lawsuits and similarly in employers being less willing to settle such lawsuits to avoid litigation.62 However, there is also evidence that plaintiffs and their attorneys are using new approaches to satisfy Rule 23 class action requirements and it is perhaps noteworthy that the Merrill Lynch settlement of a class action discrimination lawsuit for $160 million (see above) came after the Walmart decision.63 Another added impediment to bringing a class action lawsuit is a new strategy by employers to not only require employees to agree to resolve their individual employment discrimination complaints via arbitration (versus legal action), but to also require them to agree to give up their right to pursue their complaints as part of a class action lawsuit. (In return for waiving such a right, the employer must provide something in return referred to as "consideration.") Employers prefer arbitration because it is faster and cheaper, especially so if it can be used in lieu of costly class action litigation. At present, the law is not clear on whether such waivers are legal. The National Labor Relations Board has ruled that workers cannot be required to give up their right to be part of a class action lawsuit. However, another view is that a 2011 Supreme Court ruling (in a nonemployment context), AT&T Mobility v. Concepcion, can be applied to employment law to permit use of such waivers. Companies such as Sears, Nordstrom, Uber, and Haliburton are doing so, and more generally the use of such waivers is reported to have increased dramatically since the 2011 Supreme Court decision.64 Although class action lawsuits may provide more challenging to mount for plaintiffs, the passage in 2009 of the Lilly Ledbetter Fair Pay Act is expected to further increase the compliance challenge for employers. The statute of limitations for filing a claim of discrimination is within 180 days (300 days in states with their own equal employment opportunity agencies) of the date of the alleged discriminatory employment practice. Lilly Ledbetter's claim was made after she left her job as a supervisor in a tire plant and was based on the lasting effects of compensation decisions she alleged to be discriminatory that were made as much as 19 years earlier, far outside the 180-day period. In 2007 the Supreme Court ruled (Ledbetter v. Goodyear Tire & Rubber Company) that such decisions could not be litigated because they were outside the statute of limitations. However, the 2009 Act overturns this rule, instead stating that discrimination occurs—and starts the 180/300-day time period for filing Page 647a claim—"each time a discriminatory paycheck is issued, not just when the employer makes an adverse pay-setting decision."65 According to the EEOC, "The Act restores the pre-Ledbetter position of the EEOC that each paycheck that delivers discriminatory compensation is a wrong which is actionable under the federal EEO statutes, regardless of when the discrimination began."66 It has been argued that "employers will likely be called upon to defend against actions and decisions made by retired managers and supervisors that occurred years, and even decades, ago."67 Court cases have established two theories of discrimination behavior under Title VII: (1) disparate treatment and (2) disparate impact. Disparate Treatment - Disparate or unequal treatment applies different standards to different employees: For example, asking women but not men if they plan to have children. In Japan, for example, women college students continue to report that recruiters ask them different questions than are asked of male college students. The mere fact of unequal treatment may be taken as evidence of the employer's intention to discriminate under U.S. law. In the pay context, an example would be requiring a woman to have higher performance than a man to be promoted to a higher-paying job. Disparate impact - Practices that have a differential effect on members of protected groups are illegal, unless the differences are work-related. The major case that established this interpretation of Title VII is Griggs v. Duke Power Co., which struck down employment tests and educational requirements that screened out a higher proportion of blacks than whites. Even though the practices were applied equally—both blacks and whites had to pass the tests—they were prohibited because (1) they had the consequence of excluding a protected group disproportionately and (2) the tests were not related to the jobs in question. Under disparate impact, whether or not the employer intended to discriminate is irrelevant. A personnel decision can, on its face, seem neutral, but if its results are unequal, the employer must demonstrate that the decision is work-related. The two standards of discrimination—disparate treatment versus disparate impact—remain difficult to apply to pay issues, since pay differences are legal for dissimilar work. It is still not clear what constitutes pay discrimination in dissimilar jobs in the United States.68

PAY DISCRIMINATION AND DISSIMILAR JOBS

In 1981, the Supreme Court, in Gunther v. County of Washington, determined that pay differences for dissimilar jobs may reflect discrimination. In this case, four jail matrons in Washington County, Oregon, claimed that their work was comparable to that performed by male guards. The matrons also were assigned clerical duties, because guarding the smaller number of female prisoners did not occupy all of the work time. Lower courts said the matrons had no grounds because the jobs did not meet the equal work requirement of the Equal Pay Act. But the Supreme Court stated that a Title VII pay case was not bound by the definitions in the Equal Pay Act. While the Supreme Court did not say that Washington County had discriminated, it did say that a claim of wage discrimination could also be brought under Title VII for situations where the jobs were not the same. Unfortunately, the Court did not say what might constitute evidence of pay discrimination in dissimilar jobs. The case was returned to a lower court for additional evidence of discrimination and was eventually settled out of court. So if jobs are dissimilar and if no pattern of discrimination in hiring, promotion, or other personnel decisions exists, then what constitutes pay discrimination? Courts have ruled on the use of market data as well as the use of job evaluation. We will look at both of these possible standards in turn. Evidence of Discrimination: Use of Market Data - In a landmark case regarding the use of market data, Denver nurse Mary Lemons claimed that her job, held predominantly by women, was illegally paid less than the city and county of Denver paid jobs held predominantly by men (tree trimmers, sign Page 651painters, tire servicemen, etc.). Lemons claimed that the nursing job required more education and skill. Therefore, to pay the male jobs more than the nurses' jobs simply because the male jobs commanded higher rates in the local labor market was discriminatory. She argued that the market reflected historical underpayment of "women's work." The court disagreed. The situation identified by Lemons—pay differences in dissimilar jobs—did not by itself constitute proof of intent to discriminate. The courts continue to uphold use of market data to justify pay differences for different jobs. Spaulding v. University of Washington developed the argument in greatest detail. In this case, the predominantly female faculty members of the Department of Nursing claimed that they were illegally paid less than faculty in other departments. They presented a model of faculty pay comparisons in "comparable" departments that controlled for the effects of level of education, job tenure, and other factors. They asserted that any pay difference not accounted for in their model was discrimination. But the courts have been dubious of this statistical approach. As the late Carl Sagan used to say, "Just because it's a light doesn't make it a spaceship." Far better to define discrimination directly, rather than concluding that it is "whatever is left." The judge in the Spaulding case criticized the statistical model presented, saying it "unrealistically assumed the equality of all master's degrees, ignored job experience prior to university employment, and ignored detailed analysis of day-to-day responsibilities." Without such data, "we have no meaningful way of determining just how much of the proposed wage differential was due to sex and how much was due to academic discipline." "Market prices," according to the judge, "are inherently job-related." It is interesting to ask whether this judge's level of confidence in the objectivity of "the market" is justified. As you recall from Chapter 8, a lot of judgment goes into the wage survey process.77 Which employers constitute the "relevant market"? Does the relevant market vary by occupation? Do different market definitions yield different wage patterns? Clearly, judgment is involved in answering these questions. Yet the courts have thus far largely neglected to examine those judgments for possible bias. Evidence of Discrimination: Jobs of Comparable Worth - Comparable worth: overview= A second approach to determining pay discrimination on jobs of dissimilar content hinges on finding a standard by which to compare the value of jobs. The standard must do two things. First, it must permit jobs with dissimilar content to be declared equal or "in some sense comparable."78 Second, it must permit pay differences for dissimilar jobs that are not comparable. Job evaluation has become that standard.79 If an employer's own job evaluation study shows that jobs of dissimilar content are of equal value to the employer (equal total job evaluation points), then isn't failure to pay them equally proof of intent to discriminate? That was the issue considered in AFSCME v. State of Washington, where the state commissioned a study of the concept of comparable worth (discussed later in this chapter) and its projected effect on the state's pay system. The study concluded that by basing wages on the external market, the state was paying women approximately 20 percent less than it was paying men in jobs deemed of comparable value to the state. The state took no action on this finding, alleging it could not afford to adjust wages, so the American Federation of State, Page 652County, and Municipal Employees (AFSCME) sued the state. The union alleged that because the state was aware of the adverse effect of its present policy, failure to change the policy constituted discrimination. But an appeals court ruled that the state was not obligated to correct the disparity. Even though the state had commissioned the study, it had not agreed to implement the study's results. Therefore, the employer had not, in the court's view, admitted that the jobs were equal or established a pay system that purported to pay on the basis of comparable worth rather than markets. Rather than appeal, the parties settled out of court. The state revamped its pay system and agreed to make more than $100 million in "pay equity" adjustments. Since this case, many public employers have undertaken "pay equity studies" to assess the "gender neutrality" of pay systems. In states and cities which enacted comparable-worth legislation for public employees, the results of these studies are used to adjust pay for jobs held predominantly by women. In other places the results become part of the give and take of collective bargaining. We return to issues of pay equity and comparable worth later in this chapter where we discuss the mechanics of implementing these policies. The state of Washington conducted a study that concluded that the job of a licensed practical nurse required skill, effort, and responsibility equal to that of a campus police officer. The campus police officer was paid, on average, one-and-a-half times what the state paid the licensed practical nurse.80 In Ontario, Canada, jobs that were deemed comparable based on numerical scores displayed a similar disparity in pay. A chief librarian made $35,050, while a dairy herd improvement manager made $38,766. A computer operations supervisor made $20,193, while a forestry project supervisor made $26,947. A typist made $10,531, while a sailor made $14,097. It is this type of wage difference between jobs judged in some sense to be comparable that is controversial. The notion of comparable worth says that if jobs require comparable skill, effort, and responsibility, the pay must be comparable, no matter how dissimilar the job content may be. (In Canada and the European Union, comparable worth is called gender equity.) Consider the experience of Bell Canada and its unions. After lawyers spent over a decade disputing the quality and results of a pay equity study, Bell Canada agreed to a $104 million ($91.5 million American) settlement. The point job evaluation plan used in the study to determine pay equity was then used in the contemporary workplace in Bell's operations.81 Comparable-worth proponents in the United States continue to lobby for either new legislation or voluntary action on the part of employers that would include the comparable-worth standard. A lot of this political activity is occurring in state and local governments. This is not surprising, since over half of all women in the workforce are employed in the public sector. - The Mechanics= Establishing a comparable-worth plan typically involves the following four basic steps: 1. Adopt a single "gender neutral" point job evaluation plan for all jobs within a unit. If employees are unionized, separate plans have been prepared for each bargaining unit and take precedence over previous agreements. The key to a comparable-worth Page 653system is a single job evaluation plan for jobs with dissimilar content. What a "gender neutral" point job evaluation plan is remains open to debate. Advocates of all persuasions offer often conflicting advice, and there is little research to provide guidance. Some advocates try to distinguish between "gender neutral" and "traditional" point job evaluation.82 Close reading reveals that "traditional" refers to practices dating back 50 years that do not reflect contemporary point plan practices. 2. All jobs with equal job evaluation results should be paid the same. Although each factor in the job evaluation may not be equal, if the total points are equal, the wage rates must also be equal. 3. Identify the percentages of male and female employees in each job group. A job group is defined as a group of positions with similar duties and responsibilities that require similar qualifications, are filled by similar recruiting procedures, and are paid under the same pay schedule. Typically, a female-dominated job group is defined as having 60 percent or more female incumbents; a male-dominated job group has 70 percent or more male incumbents. 4. The wage-to-job evaluation point ratio should be based on the wages paid for male-dominated jobs since they are presumed to be the best estimate of nondiscriminatory wages. These steps are based on the state of Minnesota's law that mandates comparable worth for all public-sector employees (e.g., the state, cities, school districts, libraries). Canadian federal and provincial labor departments have also published detailed guidance on procedures.83 To understand the mechanics more clearly, consider Exhibit 17.10. The solid dots represent jobs held predominantly by women (i.e., 60 percent or more employees are Page 654female). The circles represent jobs held predominantly by men (i.e., 70 percent or more employees are men). The policy line (solid) for the women's jobs is below the policy line (dotted) for men's jobs. A comparable-worth policy uses the results of the single job evaluation plan and prices all jobs as if they were male-dominated jobs (dotted line). Thus, all jobs with 100 job points receive $600; all those with 200 points receive $800, and so on. = Market rates for male-dominated jobs are used to convert the job evaluation points to salaries. The point-to-salaries ratio of male-dominated jobs is then applied to female-dominated jobs. However, a mandated job evaluation, especially a single point plan that specifies a hierarchy of all jobs, seems counter to the direction in which most organizations are moving today. A partner of Hay Associates observed: We, ourselves, do not know of a single case where a large and diverse organization in the private sector concluded that a single job evaluation method, with the same compensable factors and weightings, was appropriate for its factory, office, professional, management, technical, and executive personnel in all profit center divisions and all staff departments.84 People who advocate a point job evaluation plan as a vehicle for comparable worth credit the technique with more explanatory power than it possesses. Nevertheless, the comparable worth debate lives on. - union developments= The amount of union support for comparable worth appears to be related to its effects on the union's membership. AFSCME and the Communication Workers of America (CWA) actively support comparable worth and have negotiated comparable-worth-based pay increases, lobbied for legislation, filed legal suits, and attempted to educate their members and the public about comparable worth. The public sector faces little competition for its services and is frequently better able to absorb a wage increase, because public employees are in a better position than taxpayers to pressure lawmakers. This probably accounts for the relative success of public employees' unions in bargaining comparable-worth pay adjustments. But trade-offs between higher wages and fewer jobs make unions in industries facing stiff international competition (e.g., the International Ladies' Garment Workers' Union and the United Auto Workers) reluctant to aggressively support comparable worth. The beauty of "equity adjustments," from a union's perspective, is that because they are a separate budget item, they do not appear to come at the expense of overall pay increases for all union members. Collective bargaining has produced more comparable-worth pay increases than any other approach.85

ANTITRUST ISSUES

In Chapter 8, we noted the need to avoid antitrust concerns in setting employee pay. A lawsuit (referred to as Hi-Tech) representing 64,000 former software engineers and programmers at Apple, Inc., Google, Inc., Intel Corp., and Adobe Systems alleged that these companies entered into an agreement between 2005 and 2009 that they would not poach (try to recruit away) each other's employees. Given that one of the main contributors to career earnings is advancement, either at one's present company or another company (typically with a higher salary), the companies' action to avoid a bidding war for talent (through using anticompetitive actions) is argued to have resulted in lost pay for their employees. The lawsuit alleges that the main mastermind of the no poaching agreement was the late Steve Jobs, the former head of Apple, Inc. (Indeed, according to one claim in the papers filed, when in 2007 a Google recruiter contacted an Apple employee about a job, Mr. Jobs complained and the Google recruiter was fired, purportedly within an hour.) The two sides in the lawsuit eventually agreed to settle the case by having the companies (those above plus later Intuit, Inc., Lucasfilm, Ltd., and Pixar) pay $435 million in back pay (and legal fees). A similar case, Animation Workers, resulted in settlements totaling $168.95 million with several companies, including some (e.g., Lucasfilm, Ltd. and Pixar) who were part of the Hi-Tech settlement, as well as other well-known companies (e.g., Walt Disney Company, Sony Pictures Animation, Inc., and DreamWorks Animation SKG, Inc.).52

EXECUTIVE ORDER 11246

Enforced by the Office of Federal Contracts Compliance Programs (OFCCP), Department of Labor, Executive Order 11246 (E.O. 11246) prohibits discrimination on the basis of race, color, religion, sex, or national origin. It requires covered government contractors to file affirmative action plans, which have three parts. First, utilization analysis compares the contractor's workforce to the available external workforce. Underutilization exists if a group (e.g., women) represents a significantly smaller percentage of the employer's workforce than of the external workforce. Second, goals and timetables are developed for achieving affirmative action. Third, action steps are Page 648developed for achieving these goals and timetables. As discussed below, the OFCCP conducts audits and seeks remedies where it finds insufficient compliance. Here we focus specifically on the steps in the OFCCP's compliance review process as it applies to compensation.69 It begins with a selection of contractors based, in part, on a mathematical model, called the Federal Contractor Selection System (FCSS), which is intended to predict the likelihood that a contractor is engaging in systemic (i.e., affecting a broad class of employees) discrimination. (Under a 1999 Memorandum of Understanding with the EEOC, individual complaints of compensation discrimination can be referred to the EEOC.) The OFCCP also selects contractors based on other factors (e.g., time since their previous review) and selects some contractors at random. In recent years, about 5 percent of all contractors have been selected for review. If selected, the first step is a desk audit. The OFCCP will notify the employer that it is conducting an audit and will instruct the employer to provide complete information on its Affirmative Action Program and all supporting personnel activity (such as hiring, promotion decisions) and compensation data within 30 days. This is "analyzed for possible systemic discrimination indicators (i.e., a potential affected class of 10 or more applicants/workers)."70 If such indicators are found, additional information for the desk audit will be requested. After the desk audit is completed, if the OFCCP decides the employer is in compliance, it ends the process by issuing a closure letter. If the OFCCP believes systemic discrimination may be present, it conducts an onsite review, where it will delve deeper into statistical analyses of data (including using multiple regression analysis) and also conduct interviews with management and nonmanagement employees for "anecdotal evidence" to consider along with statistical evidence.71 Based on its statistical analyses and anecdotal evidence, the OFCCP will decide whether there is evidence of systemic discrimination. If so, it will issue a Notice of Violation (NOV). However, that and other aspects of the investigative process have been revised (see below). If an NOV is issued, the OFCCP will seek to have the employer sign a conciliation agreement under which it agrees to stop and remedy practices identified as discriminatory. The employer may also be required to change its compensation levels for some employee groups to remedy disparities between similarly situated employees that the OFCCP judges to be the result of systemic discrimination. If the OFCCP cannot reach a settlement with the employer, it can refer the case to the Office of the Solicitor and disputes are addressed in a hearing in front of an administrative law judge. The OFCCP can also seek to disbar contractors from receiving future contracts from the government or to stop payments on current contracts.Page 649 In 2013, under President Obama, the OFCCP rescinded its previous (issued in 2006 under George W. Bush) standards for evaluating evidence of systemic pay discrimination, stating that these former standards "restrict[ed] OFCCP's ability to enforce the Executive Order's nondiscrimination mandate."72 The OFCCP issued new standards (Directive 307)73 giving it more "flexibility," which it argued was "critical because discrimination may be difficult to identify." Then, in 2018, under President Trump, the OFCCP stated that it had rescinded Directive 307 and replaced it with Directive 2018-05. In other words, this is another example of how the executive branch enforcement of laws changes with different presidential administrations (and their agency appointees). Directive 2018-05 is intended to provide greater clarity on how the OFCCP evaluates evidence to identify possible discrimination. In this vein, it goes into some detail in describing how multiple regression is used. Both the directive and a Frequently Asked Questions document are available from the OFCCP. 74 Although some forms of pay discrimination "can be easy to spot," such as paying men and women in the same job and with the same performance differently, as Exhibit 17.9 shows, other forms may be more "complex." An example from Exhibit 17.9 would be that "African-American sales workers are disproportionately assigned to territories with less potential," meaning "that no matter how well they perform, they can never have the same earnings opportunities as their white counterparts." The OFCCP goes on to state that "Title VII addresses all forms of compensation differences, including those that come from channeling a favored group into the better paying entry level jobs with better long-term opportunities, or where glass ceilings or other unfair promotion practices wrongly block advancement of talented workers on the basis of illegal criteria like race or gender. And even where base wages or salaries are fair, discrimination in access to overtime, or higher paying shifts, or bonuses, can add up to unequal take home pay in violation of federal civil rights law." Most companies already have their hands full managing relationships with customers, investors, and suppliers. An onsite visit by an OFCCP compliance investigator can mean that at least some HR employees will be forced to put their work in managing employee relationships on hold and instead deal with the compliance review. Outside legal counsel may be necessary, which can be costly. If a NOV is issued, the company will face further challenges. So, what can a company do to avoid running afoul of E.O. 11246 and the OFCCP? The simple answer is: Don't discriminate. The somewhat more complex answer is don't discriminate and collect and analyze data to document that you are not. In fact, self-evaluation by employers is actually required.75As with any analysis that can be seen as relevant to deciding whether an employer has engaged in or is engaging in discrimination, care must be taken to minimize legal risks. Obviously, obtaining prompt legal counsel is necessary. For the analyses to be privileged (and not open to discovery, for example, by a plaintiff's attorney in possible future litigation), they should be done under the direction of an attorney and strict communication protocols must be followed.76

EARNINGS GAPS

Exhibit 17.11 shows that, across all races combined and limiting the analysis to full-time, year round workers, women's median annual earnings compared to men's has changed from 60.2 percent to 80.5 percent from 1980 to Page 6552016. The wage gap between men and women varies according to wage level, being relatively large at very high wages and even larger at very low wage levels. Since the 1990s, the gap at the top narrowed slightly, due mostly to education gains by women. The gap at the bottom narrowed more, mostly due to increases in overall years worked and years worked with current employer by women.86 Exhibit 17.12 shows that Asian men's earnings have risen over time to now be higher (by 25 percent) than those of white men. Black to white men's earning ratios have varied between about 68 to 79 percent, and the ratio of Hispanic to white men's earnings varied from about 55 to 72 percent. Exhibit 17.13 compares earnings of Page 656Asian, black, and Hispanic women to white women. We see a pattern similar to what we saw with men, with Asian women having the highest earnings. Perhaps this is in part because a larger percentage of Asian women have a bachelor's degree. The gap between black and white women is less than the gap between black and white men, again perhaps partly because of education differences, as black women are more likely than black men to have a college degree.87 What do we know about why these gaps exist? Some of the more important sources, shown in Exhibit 17.14, include the following: - Work/occupation differences - Work-related behavior - Labor market conditions - Firm/industry differences - Union differences - Discrimination Sources of the Earnings Gaps - Considerable research has examined the factors shown in Exhibit 17.14, which are the central sources of the wage differences between men and women and the racial/ethnic groups. The issue, especially any proposed remedies, continues to generate research and debate. Our reading of the recent research is that the primary sources contributing to the gender gap differ from the primary sources for the race/ethnic gaps. It appears that differences in the work/occupation (e.g., technician vs. clerical) and differences in work-related behaviors (e.g., work-life balance challenges) are central to understanding Page 657the remaining gender wage gaps.88 In contrast, differences in qualifications, especially educational levels and work-related experience as well as differences in occupations, are important sources of the gaps for both blacks and Hispanics compared to white men.89 - Differences in Occupations and Qualifications= There is evidence that women of all ethnic groups are more likely than men to seek part-time and flexible work arrangements and that they are more likely than men to interrupt their careers due to family responsibilities.90 There is also evidence that gender differences in occupational choices and preferences continue to exist.91 According to the Center for Education Statistics, for every 100 women who earn bachelor's degrees today, 75 men do.92 However, men are more likely to enroll, graduate, and continue working in engineering and certain scientific specialties, though women's enrollment and graduation rates exceed men's in the biological sciences. Women may be somewhat less likely to negotiate for higher pay than men. However, that difference seems to have diminished over time.93 In the early 1970s, 53 percent of women workers were in administrative support (including clerical) and service occupations, compared to only 15 percent of men. At that time, less than one in five managers were women; professional women were frequently employed in traditionally female professions, such as nurse, teacher, Page 658dietitian, or librarian. Women were also underrepresented in blue-collar jobs, including higher-paying precision production and craft occupations. In 1960, almost half of the women who graduated from college became teachers, while today less than 10 percent do so. Today, the number of women in managerial jobs (economy-wide) is at parity with men. Despite progress, one notable concern (in addition to overall unexplained pay gaps) remains: lack of representation of women in executive and director level positions in organizations, sometimes referred to as the "Glass Ceiling" effect. Exhibit 17.15 shows how the representation of women declines at higher levels of organizations (in this case the largest 500 U.S. companies). Note that some countries (not the United States) mandate a minimum number of women directors (on boards). For example, Norway and France both require 40 percent and the European Union may consider similar mandates.94 Evidence of increased levels of occupational attainment does not automatically mean that the wage gap will close. A study of women in science and engineering finds that even though they have already cleared the hurdles of the occasional misguided high school guidance counselor and/or lack of peer support or role models, women scientists and engineers are almost twice as likely to leave these occupations as are males.95 For a variety of reasons, a relatively small wage gap among younger cohorts (i.e., recent college graduates) tends to increase as the cohort ages. Perhaps women continue to be more likely to drop out of the labor force at some point for family reasons, or perhaps the barriers to continued advancement become more substantial (the "Glass Ceiling" concept) at higher levels in the job hierarchy.96 One reviewer of all this research conjectures that some of the observed differences in wages paid in different occupations may be stuck due to inertia or, more poetically, Page 659"original sin."97 Wage differences among occupations were determined decades ago. While a variety of reasons account for these original wage differences, an important one is the belief about gender roles and "women's work" that prevailed at the time. These pay differences have persisted and influence today's occupational wage structures even though attitudes have changed.98 Level of schooling and work-related experience appear to be primary sources of the pay gaps among both black and Hispanic men and women. Almost half of Hispanic men have not completed high school and only 9 percent are college graduates.99 This may be due to the considerable differences in continuing immigration among Hispanics. Generally, the longer Hispanic immigrants stay in the United States, the better they do educationally and financially. However, the continuing inflow of poorly educated new immigrants holds down the average wage for the whole category. Black men's high school dropout rates match those of Hispanic men. In contrast, more than half of Asian men are college graduates or hold higher degrees. All this research and discussion about sources of the earnings gaps does not mean there are not any discriminatory pay practices; it does mean that important sources of the pay gaps reside in other productivity-related factors such as level and quality of education, work-life/career tradeoffs, and occupational choices. - Differences in industries and firms= Other factors that affect earnings differences among men and women and among race/ethnic groups are the industries and the firms in which they are employed. A study of middle-aged lawyers revealed large differences between men and women lawyers in the types of firms that employed them. Men were much more likely than women to be in private practice and to be at large firms (over 50 lawyers). They were much less likely than women to be in the lower-paying nonprofit sector. Clearly, these differences are related to pay: the most highly paid legal positions are in private-practice law firms; the larger the law firm, the greater the average rate of pay.100 There also may be different promotion opportunities among firms within the same industry.101 Differences in the firm's compensation policies within a specific industry is another factor that accounts for some of the earnings gap.102 As noted in Chapters 7 and 8, some firms within an industry adopt pay strategies that place them among the leaders in their industry; other firms adopt policies that may offer more work-life balance benefits compared to cash compensation. The unknown here is whether within an industry some firms are more likely to attract women or minorities than other firms because of these pay-mix differences and whether this has any effect on earnings gaps. Within a firm, differences in policies for different jobs may even exist. For example, many firms tie pay for secretaries to the pay for the manager to whom the secretary is assigned. The rationale is that the secretary and the manager function as a team. When the manager gets promoted, the secretary also takes on additional responsibilities and therefore also gets a raise. However, this traditional approach breaks down when layers of management are cut. When IBM went through a major restructuring a few years back, it cut pay by up to 36 percent for secretaries who had been assigned to managerial levels that no longer existed. IBM justified the cuts by saying the rates were way above the market.Page 660 We also know that the size of a firm is systematically related to differences in wages. Female employment is more heavily concentrated in small firms. Wages of men in large firms are 54 percent higher than wages of men in small firms. The gap was 37 percent for women in small versus large firms. Hispanic men are concentrated in construction and service firms. Other studies report that employees in some jobs can get a pay increase of about 20 percent simply by switching industries in the same geographic area while performing basically similar jobs.103 Nevertheless, a recent study concludes that this pay premium associated with changing jobs is enjoyed primarily by white males. Women and minorities who were MBA graduates from five universities did not obtain the same pay increases as their white male classmates when they switched jobs.104 To the extent that these differences in job setting are the result of an individual's preference or disposition, they are not evidence of discrimination. To the extent that these differences are the result of industry and firm practices that steer women and minorities into certain occupations and industries or lower-paying parts of a profession, they may reflect discrimination. - Challenges in Estimating Amount of Discrimination= We know that many factors affect pay and that discrimination can be one of them. Disagreement remains over what constitutes evidence of discrimination. Although the earnings gap is the most frequently cited example, closer inspection reveals the weaknesses in this statistic. Unfortunately, many studies of the earnings gap have little relevance to understanding discrimination in pay-setting practices within organizations. Some studies use aggregated data—for instance, treating all bachelor's degrees as the same, or defining an occupation too broadly (e.g., the U.S. Department of Labor categorizes LeBron James as well as the basketball game timekeeper in the same occupation—"sports professional"). Another problem is that mere possession of a qualification or skill does not mean it is work-related. Examples of cab drivers, secretaries, and house painters with college degrees are numerous. A standard statistical approach for determining whether discrimination explains part of the gap is to try to relate pay differences to the factors discussed above (occupation, type of work, experience, education, and the like). The procedure typically regresses some measure of earnings on those factors thought to legitimately influence earnings. If the average wage of men with a given set of values for these factors is significantly different from the average wage of women with equal factors, then the residual portion of the gap is considered discrimination. Unfortunately, in a sample limited to white males, such an approach explained only 60 to 70 percent of their earnings. So statistical studies, by themselves, are not sufficient evidence of discrimination.105 Even if legitimate factors fully explain gender, ethnic, and racial group pay differences, discrimination still could have occurred. The factors shown in Exhibit 17.14 themselves may be tainted by discrimination. For example, construction laborers in California are mainly Hispanic males. A slowing of the housing market will disproportionately affect them. So, measurable factors may underestimate the effects of past discrimination. Statistical analysis needs to be treated as part of a pattern of evidence and needs to reflect the wage behaviors of specific firms. One challenge is that any worker who has a history of low wages/salaries will often find it is difficult to break out of that pattern because employers regularly ask applicants what they earned in current/previous jobs (their "salary history"). So, low earnings (including unjustifiably low earnings) tend to be perpetuated. Consequently, some jurisdictions (e.g., New York City, Delaware, Oregon, Massachusetts) have prohibited organizations from asking applicants to share their salary history, at least until after the applicant has been hired and/or been already offered a specific salary.106 - Gaps are global= The gender wage gap is fairly universal. However, in a number of countries, the size of the gap is smaller than in the United States.107 One analysis concludes that the difference can be found in narrower pay structures in European countries. In many countries, as the global guide in Chapter 16 suggests, rates negotiated by federations of employers, unions, and government agencies rather than individual companies and employees mean a narrower range of pay rates for each job, as well as smaller differences between jobs.108 Earlier chapters have emphasized the wide range of pay rates in the U.S. market for any job. More centralized wage decision making permits a pay gap to be closed by political/institutional fiat. Multinational companies operating in different nations face wide differences in how wages are influenced by varying social policies and regulations.

THE EQUAL PAY ACT

The Equal Pay Act (EPA) of 1963 (which is part of the FLSA) forbids wage discrimination on the basis of gender if employees perform equal work in the same establishment. Jobs are considered equal if they require equal skill, effort, and responsibility and are performed under similar working conditions. Differences in pay between men and women doing equal work are legal if these differences are based on any one of four criteria, called an affirmative defense: - Seniority. - Merit or quality of performance. - Quality or quantity of production. - Some factor other than sex. These terms for comparison and permitted defenses seem deceptively simple. Yet numerous court cases have been required to clarify the act's provisions, particularly its definition of "equal." Definition of equal - The Supreme Court first established guidelines to define equal work in the Schultz v. Wheaton Glass case back in 1970. Wheaton Glass Company maintained two job classifications for selector-packers in its production department: male and female. The female job class carried a pay rate 10 percent below that of the male job class. The company claimed that the male job class included additional tasks such as shoveling broken glass, opening warehouse doors, and doing heavy lifting that justified the pay differential. The plaintiff claimed that the extra tasks were infrequently performed and not all men did them. Further, these extra tasks performed by some of the men were regularly performed by employees in another classification ("snap-up boys"), and these employees were paid only 2 cents an hour more than the women. Did the additional tasks sometimes performed by some members of one job class render the jobs unequal? The Court decided they did not. It ruled that the equal work standard required only that jobs be substantially equal, not identical. Additionally, in several cases where the duties employees actually performed were different from those in the job descriptions, the courts held that the actual work performed must be used to decide whether jobs are substantially equal. Definitions of Skill, Effort, Responsibility, Working Conditions - The Department of Labor provides these definitions of the four factors. 1. Skill: Experience, training, education, and ability as measured by the performance requirements of a particular job. 2. Effort: Mental or physical—the degree of effort (not type of effort) actually expended in the performance of a job. 3. Responsibility: The degree of accountability required in the performance of a job. 4. Working conditions: The physical surroundings and hazards of a job, including dimensions such as inside versus outside work, heat, cold, and poor ventilation. Guidelines to clarify these definitions have evolved through court decisions. For an employer to support a claim of unequal work, the following conditions must be met: 1. The effort/skill/responsibility must be substantially greater in one of the jobs compared. 2. The tasks involving the extra effort/skill/responsibility must consume a significant amount of time for all employees whose additional wages are in question. 3. The extra effort/skill/responsibility must have a value commensurate with the questioned pay differential (as determined by the employer's own evaluation). Time of day (e.g., working a night shift) does not constitute dissimilar working conditions. However, if a differential for working at night is paid, it must be separated from the base wage for the job. Factors Other than Sex - Of the four affirmative defenses for unequal pay for equal work, "a factor other than sex" has prompted the most court cases. Factors other than sex include shift Page 644differentials; temporary assignments; bona fide training programs; differences based on ability, training, or experience; and other reasons of "business necessity." Factors other than sex have been interpreted as a broad exception that may include business reasons advanced by the employer. A practice will not automatically be prohibited simply because wage differences between men and women result. However, an employer is required to justify the business relatedness of the practice.56 Usually a specific practice is not singled out; rather, the argument focuses on a "pattern of practices." That is what a group of female brokers at Merrill Lynch charged in their class action suit. They were concerned with how accounts from departing brokers, walk-ins, and referrals were being distributed. They felt that the top men brokers were given the most promising leads, while everyone else, including the 15 percent of brokers who were women, got the "crumbs." The women contended that Merrill Lynch discriminated against women in wages, promotions, account distributions, maternity leaves, and other areas. A negotiated settlement promised to establish a more open method for sharing leads and not to penalize brokers for time off in determining bonuses and production quotas. Because such cases tend to be settled out of court, no legal clarification of a "factor other than sex" has ever been provided. It does seem that pay differences for equal work can be justified for demonstrably business-related reasons. But what is and is not demonstrably business-related has yet to be cataloged. "Reverse" discrimination - Many people dislike the term "reverse" discrimination, saying that it is still discrimination, even if the group penalized is white males. Several court cases deal with discrimination against men when pay for women is adjusted. The University of Nebraska created a model to calculate salaries based on estimated values for a faculty member's education, field of specialization, years of direct experience, years of related experience, and merit. Based on these qualifications, the university granted raises to 33 women whose salaries were less than the amount computed by the model. However, the university gave no such increases to 92 males whose salaries were also below the amount the model set for them based on their qualifications. The court found this system a violation of the Equal Pay Act. It held that, in effect, the university was using a new system to determine a salary schedule, based on specific criteria. To refuse to pay employees of one sex the minimum required by these criteria was illegal. Viewed collectively, the courts have provided reasonably clear directions to interpret the Equal Pay Act. The design of pay systems must incorporate a policy of equal pay for substantially equal work. The determination of substantially equal work must be based on the actual work performed (the job content) and must reflect the skill, effort, responsibility, and working conditions involved. It is legal to pay men and women who perform substantially equal work differently if the pay system is designed to recognize differences in performance, seniority, quality, and quantity of results, or certain factors other than sex in a nondiscriminatory manner. Further, if a new pay system is designed, it must be equally applied to all employees. But what does this tell us about discrimination on jobs that are not substantially equal—dissimilar jobs? Fifty-eight percent of all working women are not in jobs substantially equal to jobs of men, so they are not covered by the Equal Pay Act. Title VII of the Civil Rights Act extends protection to them.

Living wage

Although living wage provisions are not part of the FLSA, we cover the topic here because of its similarity to FLSA minimum wage provisions. Rather than (solely) push for changes in the FLSA, an alternative approach in recent years has been to push for a "living wage" at local levels that provides a minimum wage tailored to living costs in an area.36 These laws have narrower coverage than minimum wage laws, as they cover only city (or state) employees and/or employers that do business with the city (i.e., contractors and subcontractors). Sometimes they cover only base wages, but more frequently they require health insurance, vacations, sick pay, job security, and provide incentives to unionize. Maryland was the first to adopt a statewide living wage ordinance, effective in 2009, covering "certain contractors and subcontractors working on State funded service contracts."37 The minimum ranges from $10.36 to $13.79, depending on location, compared to a $9.25 state minimum wage. More than 140 ordinances have been put into effect in the United States by cities, counties, universities, and other public entities. California has the most; it is followed by Michigan, New York, and Wisconsin.38 Los Angeles's law covers "contractors/subcontractors who have agreements with the city." The law mandates $12.73/hour (including health care benefits). The required wage rate is higher for certain occupations, including airport workers ($18.99/hour, including health care benefits).39 An early study of the Los Angeles law's effects found that 7,735 of the covered employees got an average wage increase of 20 percent.40 Another 149 noncovered employees got increases in order to maintain pay differentials. Employers adjusted to the law by making only very minor adjustments in employment—an estimated 112 jobs, or 1 percent of covered jobs, were lost. Benefits were cut for less than 5 percent of affected jobs, including cuts in health benefits, merit pay, bonuses, and employer-provided meals. Training for new hires stayed the same, but nonaffected firms were increasing their training. Firms benefited via reduced turnover and absenteeism. New hires tended to be better qualified, with higher levels of education and training than those hired before the law was passed. The new hires also included a higher proportion of males: 56 percent, compared to 45 percent of hires before the living wage. The study also found that 70 percent of the benefit of the law went to low-income families. Living wage laws are increasingly popular. Coalitions of union members and church groups often support them. Because they are so narrowly tailored, there is some speculation that their real intention is to reduce any cost savings a municipality might receive from outsourcing. Reduced outsourcing means more government jobs, which generally translates into more union members.41

EMPLOYEE OR INDEPENDENT CONTRACTOR?

As we saw in Chapter 12, U.S. employers are legally obligated to pay Social Security, unemployment compensation, and workers compensation taxes on wages and salaries on behalf of their employees. As we saw in Chapter 13, the average total compensation per employee in private industry was $33.72 per hour, with $23.47 of that being in the form of wages and salaries and the remaining $10.25 being for benefits. Of the $10.25, $2.62 was for the legally required benefits just mentioned.42 However, in the case of a worker who is an independent contractor rather than an employee, the employer is not obligated to pay the legally required benefits. In addition, independent contractors would also typically not receive other benefits. An independent contractor would not be eligible for overtime (time and a half). Thus, whether a worker is classified as an employee or an independent contractor can have substantial cost implications for an employer, which of course increases with the number of workers involved. Less obvious perhaps is that law-abiding employers are put at a disadvantage because complying with the law results in higher costs than those borne by employers that misclassify (unless those not in compliance are discovered). Finally, of course, workers suffer if they are denied access to benefits and legal protections that they should receive. As with the FLSA exceptions discussed a few pages earlier, the decision of whether to classify a worker as an employee or independent contractor requires careful attention to compliance issues. Both tax law—enforced by the Internal Revenue Service (IRS)—and the Employee Retirement Income Security Act (ERISA)—enforced by the Department of Labor—are relevant. To get an idea of the potential revenue losses to government of misclassification of workers, consider that Ohio's attorney general estimates that Ohio alone at one point has 92,500 misclassified workers, estimated to cost the state up to $35 million in lost unemployment insurance taxes, up to $103 million in lost workers' compensation taxes, and up to $223 million in income tax revenue. (Note: some estimates suggest that misclassified independent contractors do not report 30 percent of their income.)43 The most widely used classification criteria are provided by the IRS and shown in Exhibit 17.7. Two general criteria have to do with behavioral and financial control. The more control a firm is able to exercise, the more likely it is that the IRS will see the worker as an employee rather than an independent contractor. The IRS also considers the type of relationship, including its permanence. The Supreme Court, in Nationwide v. Darden, has applied similar criteria in deciding whether a worker is an employee under ERISA.44 Among the most often misclassified are truck drivers, construction workers, home health aides, and high-tech engineers.45 Microsoft hired workers as independent contractors. It had these workers sign agreements acknowledging their independent contractor status. However, after an audit by the IRS concluded that these workers were actually employees, Microsoft agreed to begin paying legally required taxes (see above). Microsoft had used the workers on projects, often working on teams with regular employees, doing similar work, working similar hours, and being supervised by the same managers. Microsoft also required them to work onsite and they were given office equipment and supplies.46Page 639 Next, two separate suits (Vizcaino v. Microsoft and Hughes v. Microsoft) were filed against Microsoft to compel it to retroactively provide other benefits (e.g., a discounted stock purchase program) that it provided to its (other) employees. Some of the workers had been at Microsoft for several years with a few being there as long as 10 years. Microsoft eventually settled these suits for $97 million, which after attorneys' fees will be divided between 8,000 and 12,000 people employed at Microsoft for at least 9 months during a several-year period.47 Microsoft implemented new rules, including limiting independent contractor assignments to 12 months with at least 100 days between assignments. More recently, FedEx was ordered to pay $27 million to 203 drivers in California who were ruled (Estrada v. FedEx Ground Package System, Inc.) to have been wrongly classified as independent contractors.48 And, that may not be the end of the compliance issue for FedEx. It announced that the IRS, upon tentatively deciding that FedEx has misclassified workers as contractors, was considering $319 million in tax and penalties for one year and that it was in the process of looking at other years also. FedEx continues to deal with multiple lawsuits on this issue.49

GOVERNMENT AS PART OF THE EMPLOYMENT RELATIONSHIP

People differ in their view of what role government should play in the contemporary workplace. Some call for organizations and the government to act in concert to carry out a public policy that protects the interests of employees.3 Others believe that the best opportunities for employees are created by the constant change and reconfiguring that is inherent in market-based economies; the economy ought to be allowed to adapt and transform, undistorted by government actions.4 All countries throughout the world must address these issues. However, as we will see, different countries take different approaches.5 Governments' usual interests in compensation decisions are whether procedures for determining pay are fair (e.g., pay discrimination), safety nets for the unemployed and/or those unable to work (e.g., unemployment compensation, workers compensation), and worker protection (e.g., overtime pay, minimum wage, child labor restrictions). In addition to being a party to all employment relationships, government units are also employers and purchasers. Consequently, government decisions also affect conditions in the labor market. The U.S. federal government employs 2.79 million people; state and local governments employ 19.54 million. Overall, government employment is 22.33 million, representing 15 percent of the total U.S. nonfarm employment of 148.66 million.6 In addition to being a big employer, and thus competing with private sector organizations for employees, government also indirectly Page 623affects labor demand in the private sector through its spending and purchases (military aircraft, computer systems, paper clips) and tax policy. In addition to government fiscal policy (i.e., total spending and budget, tax policy, taxes), the federal government influences overall economic growth/demand and business activity through its monetary policy (level of interest rates and money supply). Increased business activity translates into increased demand for labor and upward pressure on wages. In addition to being an employer, government affects labor supply through legislation. Laws aimed at protecting specific groups also tend to restrict those groups' participation in the labor market. Compulsory schooling laws restrict the supply of children available to sell hamburgers or to assemble soccer balls. Licensing requirements for certain occupations (plumbers, cosmetologists, attorneys, physicians, psychologists) restrict the number of people who can legally offer a service.7 Immigration policy and how rigorously it is enforced is an increasingly important factor in labor supply.8 This chapter will examine the laws and regulations that most directly affect compensation in the United States. Exhibit 17.2 provides an overview of the regulatory framework, especially as it applies to wages and salaries and other forms of direct pay. Please see Chapters 12 and 13 for information on benefits-related regulations (e.g., the Employee Retirement Income Security Act, ERISA).

FAIR LABOR STANDARDS ACT OF 1938

The Fair Labor Standards Act of 1938 (FLSA) covers all employees (with some exceptions, discussed later) of companies engaged in interstate commerce or in the production of goods for interstate commerce. In spite of its age, this law remains a cornerstone of pay regulation in the United States. The FLSA's major provisions are: 1. Minimum wage 2. Hours of work (including overtime) 3. Child labor An additional provision requires that records be kept of employees, their hours worked, and their pay. As noted earlier, in recent years, U.S. employers have paid out billions of dollars (see Exhibit 17.1) as a result of FLSA lawsuits and enforcement activity by the DOL Wage and Hour Division. Exhibit 17.3 provides a breakdown of the claim types covered in such settlements. Minimum wage - Minimum-wage legislation is intended to provide an income floor for workers in society's least productive jobs. When first enacted in 1938, the minimum wage was 25 cents an hour. It has been raised periodically; in 2009, it was raised to $7.25 and has remained there.Page 627 Exhibit 17.4 shows the purchasing power of the federal minimum wage over time, adjusted for inflation. The decline in real purchasing power could be used to argue for indexing the minimum wage to changes in the consumer price index. - Estimates from the U.S. Bureau of Labor Statistics indicate that approximately 2.2 million U.S. workers (down from 3.83 million in 2011) are paid at or below the minimum wage. The majority (1.4 million) of those earning minimum wage or less are in service occupations, mostly food service, where tips supplement hourly wages for many workers. The proportion of hourly paid workers earning minimum wage Page 628or less has trended downward since 1979 when data first began to be collected systematically. In 1979, 13.4 percent of hourly paid workers (7.7 percent of men and 20.2 percent of women) earned at or below minimum. (Note that roughly one-half of employees in the United States are paid an hourly rate.) More recently, the figures are 5.2 percent of hourly workers (3.9 percent of men and 6.4 percent of women). As a percentage of all civilian wage and salary workers, those earning at or below minimum wage has declined from 7.9 percent in 1979 to under 2 percent more recently.10 An important reason for the decline in those directly affected is that the federal minimum wage stayed unchanged at $5.15 from 1997 to 2007 (and, more recently, has stayed unchanged since 2009 from $7.25). Changes to the federal minimum wage have direct effects (on workers having a current wage between any state minimum wage and the federal minimum wage). There are also indirect, spillover effects because as legislation forces pay rates at the lowest end of the scale to move up, pay rates above the minimum often increase in order to maintain differentials. This shift in pay structure does not affect all industries equally. The lowest rates paid in the software, chemical, oil, and pharmaceutical industries are already well above minimum; any legislative change has little direct impact on them. In contrast, retailing and hospitality firms tend to pay at or near minimum wage to many clerks, sales persons, and cleaning people.11 Forty-five states plus the District of Columbia have their own minimum wages to cover jobs omitted from federal legislation. If state and federal laws cover the same job, the higher rate prevails. Over half (29) of the states have minimums higher than the federal rate, with the highest being District of Columbia ($12.50), Washington ($11.50), and California ($11.00) and Massachusetts ($11.00).12 A number of cities (and counties) also have minimum wage ordinances, including San Francisco ($15.00) and Seattle ($14 to $15.45, depending on employer size).13 Why would anyone be opposed to a mandated minimum wage or making increases to it? The concern is that the resulting higher labor costs (as noted above, not only covered employees, but also other employees to the degree pay differentials are maintained) for affected firms may lead them to decrease their demand for workers and/or their hours worked. (See our discussion of supply and demand curves in Chapter 7.) In other words, a higher minimum wage, which is intended to help low wage workers, runs the risk of reducing employment opportunities for these very workers it is intended to help. So, whether a minimum wage "works" or not depends on whether the gains through higher wages are greater than the losses of jobs and/or hours.14 Another consideration in evaluating minimum wage law effectiveness is whether wage gains go primarily to workers from low income families rather than going to workers from families with higher incomes.15 Employers certainly have a stake in minimum wage public policy and thus may seek to influence it over time. In the shorter run, employers must be in compliance and will need to consider how changes to the minimum wage will affect their labor costs, to what degree they can pass the higher costs on to customers, and to what degree they will need to take some other action to control or offset higher labor costs. Exhibit 17.5 provides an example of how an employer can run afoul of minimum wage, hours worked, and overtime provisions of the FLSA, as well as the consequences. Overtime and hours of work - The overtime provision of the FLSA requires payment at one-and-a-half times the standard for working more than 40 hours per week. However, some of the largest wage and hour monetary settlements result from private class action suits brought by plaintiffs. Paying for all hours worked and/or for overtime are often central issues. For example, Bank of America paid $73 million to settle a nationwide class action lawsuit alleging that it had a company-wide policy that required nonexempt employees to perform off-the-clock work.16 Walmart has settled or lost a number of lawsuits, including one for as much as $640 million (for allegedly not paying for all hours worked and for expecting workers to work through their breaks) and another for as much as $86 million (for not paying all wages due upon termination of employment). In another case, which it may appeal, a court ordered it to pay $188 million (for again expecting workers to skip or cut short rest and/or meal breaks). One objective of the FLSA is to share available work by making the hiring of additional workers a less costly option than the scheduling of overtime for current employees. However, the conditions that inspired the legislation have changed since the law was passed. Contemporary employers face (1) an increasingly skilled workforce with higher training costs per employee and (2) higher benefits costs, the bulk of which are fixed per employee. These factors have lowered the break-even point at which it pays employers to schedule longer hours and pay the overtime premium, rather than hire, train, and pay benefits for more employees. Again, state laws sometimes go beyond the FLSA. California, for example, requires time and a half pay for working more than 8 hours in a day and double time for working more than 12 hours in a day. It also requires premium pay for working a seventh day during a week. - Exemptions= The Wage and Hour Division of the Department of Labor, which is charged with enforcement of the FLSA, provides strict criteria that must be met in order for jobs to be exempt from minimum-wage and overtime provisions. These are summarized in Exhibit 17.6. Before his term ended, President Obama issued an executive order that instructed the Secretary of Labor to "update and modernize" the overtime exemption rules of the FLSA.17 The expectation was that the pay cutoff of $455/week shown in Exhibit 17.6 would be significantly increased. Specifically, the Wage and Hour Division proposed increasing this salary "test" to $970 per week, which would greatly increase the number of U.S. employees who are nonexempt (i.e., covered by FLSA overtime provisions). (The Society for Human Resource Management, for example, has recommended increasing the salary test to $615/week.)18 Note that In March 2019, the Wage and Hour Division, now under President Trump, proposed a new salary test of $679/week ($35,308 annually), which seems likely to take effect in January 2020. If so, each $455 in Exhibit 17.6 would change to $679. For an update, go to the Wage and Hour Division website in the next e-compensation box. ome employers try to get around the overtime requirement by classifying employees as executives, even though the work of these "executives" differs only slightly from that of their co-workers. However, in the eyes of the Department of Labor, the job title is not relevant. Rather, it is the actual nature of the work that matters. Merrill Lynch reached a $37 million settlement with financial analysts in California regarding overtime pay. A Merrill Lynch financial analyst argued that because his salary was entirely from commissions, he did not meet the "salary basis" test for the administrative exemption. The plaintiff also successfully argued that he did not exercise sufficient discretion and independent judgment. Instead, the financial analyst's work was considered "production," Merrill Lynch's practices were standard in the financial industry, and the impact of this ruling rippled through other financial service companies and brought changes to Merrill Lynch's pay system for analysts.19 Subsequently, Citigroup's Smith Barney brokerage unit settled an FLSA overtime lawsuit for $98 million, with UBS Financial Services and Morgan Stanley both also making substantial payments to settle similar suits. JPMorgan Chase & Co. reached a $42 million settlement with a class of 3,800 loan processors.20 The U.S. Department of Labor ordered Walmart to pay $4.8 million in back wages and damages to 4,500 vision-center managers and asset-protection coordinators who worked at Walmart over a four-year period.21 Insurance claims adjustors settled overtime lawsuits against Farmers Insurance for as much as $210 million and against State Farm Insurance for $135 million. Both of these lawsuits were brought under California law, under which it was more difficult than under federal law (i.e., the FLSA) to meet the administrative employee exemption. Indeed, similar lawsuits brought under FLSA have not succeeded. Another challenge in compliance is that "in an evolving, always-on workplace where employees routinely put in extra hours and shoot off e-mails late at night from mobile devices, when the workday begins and ends has become an issue for employers."22 For example, writers at ABC News asked that they be paid overtime for using their smartphones for work purposes after business hours. As a result, ABC News asked writers to sign an agreement waiving rights to overtime for such activity. Writers who declined to sign had their smartphones taken away.23 One survey reports that just over half of large firms have restricted the use of communications devices outside of the office and one-third have restricted telecommuting.24 In response to these types of lawsuits, some companies have reclassified some employees. For example, IBM voluntarily reclassified 7,000 technical and support employees after settling a class-action overtime lawsuit for $65 million. The Page 634employees, who had been earning an average of $77,000 per year, had their base salaries cut by 15 percent to account for the potential overtime that the company would need to pay them going forward.25 The impact of FLSA and other laws depends importantly on the degree to which they are enforced. A Government Accountability Office (GAO) report found that the Labor Department's Wage and Hour Division "mishandled" 9 of 10 cases brought by GAO undercover agents posing as workers who had experienced FLSA violations.26 As one example, an agent posing as a dishwasher called four times to complain that he had not been paid overtime for almost five months. His calls were not returned until 4 months later and he was then told it would take another 8 to 10 months to begin an investigation. GAO also investigated existing files. In another case, an undercover agent posing as an employer who had violated the law appeared to escape any penalty by simply saying that business was bad, so he could not afford to pay anything. The Department of Labor investigator was quoted as saying, "OK, so you're not in a position where you can pay?" and when the undercover agent said that was correct, the investigator seemed to give up and said that he would let the worker know that he could pursue the case on his own (i.e., need to hire an attorney). To increase enforcement of the FLSA, the Department of Labor since added wage-and-hour investigators.27 In Japan, unpaid overtime is (also) a major issue. The Japanese Trade Union Confederation reports that two-thirds of men work more than 20 hours of unpaid overtime each month.28 Only in 2008 did Toyota begin to pay factory workers for participating in quality control programs that were held outside of normal work hours. Some large companies have introduced "no overtime" days on which employees are to leave at 5:30 p.m. However, the concern is that many employees just take the work home, which is referred to as furoshiki, or "cloaked overtime."29 "Death by overwork" (karoshi) has resulted in lawsuits against companies in Japan. - what time is covered= As we have seen, employers are sometimes guilty of not paying employees for hours worked (i.e., of having them "work off the clock"). In addition to the FLSA, other laws come into play. Occupational Safety and Health Administration legislation specifies the number of breaks that must be provided in an eight-hour workday. The Portal-to-Portal Act provides that time spent on activities before beginning the "principal activity" is generally not compensable. The original issue that inspired the act was the time that miners were forced to spend traveling to and from the actual underground site where the mining was occurring. The meat processing industry has been the source of several cases defining time spent at work. Time spent sharpening knives and cutting tools is compensable time, as is the time spent donning protective gear and walking in this "integral" gear to the production area.30 The law is also relevant to "on-call employees" who must be available to respond outside the usual workday. Firefighters and emergency personnel are traditional examples. Today, telecommunications and software services personnel who must respond quickly to problems outside their regularly scheduled workday are newer categories of employees eligible for "beeper pay." (See also our ABC News example Page 635earlier.) In general, if employees can use this "on-call" time for their own purposes, there is no legal requirement to pay employees for such time, even if they are required to carry a beeper or must let their employer know where they can be reached. However, if they are required to stay on the employer's premises while on call, then they must be compensated for that time. Sometimes a flat rate is paid for the added inconvenience of being on call. These payments must be included when computing overtime pay.31 - what income is covered= FLSA specifies one and a half times pay for overtime, but one and a half of what? As more employees became eligible for bonuses, there was an argument over whether bonus, gain-sharing, and stock option payments needed to be included for calculating overtime pay. A 1999 advisory from the Wage and Hour Division said they did. But the extra bookkeeping and calculations provided enough of a burden that employers simply did not offer these forms of pay to nonexempt employees. The Worker Economic Opportunity Act, a 2000 amendment to FLSA, allows stock options and bonuses to be exempt from inclusion in overtime pay calculations. Gifts or special-occasion bonuses have never needed to be included, because they are at the employer's discretion rather than a pay form promised to employees if certain conditions are met. - compensatory time off= The changing nature of the workplace and of pay systems has led to calls to reform FLSA to allow for more flexible scheduling and easier administration of variable pay plans. Federal legislation has been proposed (but not yet passed) that would give employees and employers the option of trading overtime pay for time off. Rather than being paid overtime after 8 hours for a 10-hour workday, an employee would have the option of taking 2 or more hours off at another time. Or, a 50-hour workweek could be banked against a future 30-hour workweek.32 The employee would get more scheduling flexibility to attend to personal matters, and the employer would save money. This kind of change has a lot of appeal for employees who are also raising children and/or caring for elderly parents. One poll reported that 81 percent of women would prefer compensatory time off in lieu of overtime wages.33 child labor - Generally, persons under 18 cannot work in hazardous jobs such as meat packing and logging; persons under 16 cannot be employed in jobs involving interstate commerce except for nonhazardous work for a parent or guardian. Additional exceptions and limitations also exist.34 The union movement in the United States has taken a leading role in publicizing the extent of the use of child labor outside the United States to produce goods destined for U.S. consumers. Government guidelines help importers monitor the employment practices of subcontractors producing goods for the U.S. market. A recent International Labour Organization report finds that globally, child labor is declining, particularly in Latin America. Brazil and Mexico, where half the children Page 636in Latin America live, have made the greatest strides, which the study attributed to increased political will, awareness, poverty reduction, and education. The steepest declines were among children 14 and younger, and among hazardous occupations. The highest rates of child labor are in sub-Saharan Africa, where high population growth, grinding poverty, and the HIV/AIDS epidemic have left a lot of families in need of the income that children can provide.35

PAY DISCRIMINATION: WHAT IS IT?

Before we look at specific federal pay discrimination laws, which are summarized in Exhibit 17.8, let us address the more general question of how to legally define discrimination. The law recognizes two types of discrimination: access discrimination and valuation discrimination. The charges of discrimination and reverse discrimination that most often make the news involve access discrimination: the denial of particular jobs, promotions, or training opportunities to qualified women or minorities. The University of Michigan, for example, was accused of access discrimination for using differential standards among different racial groups to determine who is "qualified" for admission. Being a member of a minority group counted for 20 points, whereas the quality of the admission essay counted for 3 points. (Being an athlete also counted for 20 points.) In 2003, the Supreme Court ruled that while schools can take race into account for admission, this 20-point differential was illegal because it was applied in a mechanicalPage 642 way.53 However, the admission process for Michigan's law school was upheld because it was narrowly tailored and more flexible. Minority candidates for the law school were interviewed and their entire record was examined, in contrast to the routine addition of 20 points that the undergraduate school used. (The court did not address the issue of the preferred treatment for athletes or children of alumni or big donors.)54 Subsequently, however, the State of Michigan passed a constitutional amendment prohibiting race-conscious admission policies at the state's public universities. A second legally recognized interpretation of discrimination is valuation discrimination, which looks at the pay women and minorities receive for the jobs they perform. This is the more salient definition for our purposes. The Equal Pay Act makes it clear that it is discriminatory to pay women less than males when they are performing equal work (i.e., working side by side, in the same plant, doing the same work, producing the same results). This definition of pay discrimination hinges on the standard of equal pay for equal work. Many believe that this definition of valuation discrimination does not go far enough. They believe that valuation discrimination can also occur when men and women hold entirely different jobs. For example, office and clerical jobs are typically staffed by women, and craft jobs (electricians, welders) are typically staffed by men. Is it illegal to pay employees in one job group less than employees in the other if the two job groups contain work that is not equal in content or results but is, in some sense, of comparable worth to the employer?55 In this case, the proposed definition of pay discrimination hinges on the standard of equal pay for work of comparable worth (also called pay equity or gender pay equity). Existing federal laws in the United States do not support this standard. However, several states have enacted laws that require a comparable-worth standard for state and local government employees. Significant pay equity adjustments have also been made in Canada (see our later discussion). For an understanding of the legal foundations for both the equal work and the comparable worth standard, let us turn to the legislation and key court cases.

PREVAILING WAGE LAWS

Prevailing wage laws set pay for work done to produce goods and services contracted by the federal government. A government-defined prevailing wage is the minimum wage that must be paid for work done on covered government projects or purchases. Consider, for example, "The Big Dig," Boston's $15 billion taxpayer-financed project to put its freeways underground.50 A construction project of such magnitude attracts workers from a very wide area and distorts the labor market. Prevailing-wage laws prevent contractors from using their size to drive down wages. The law was passed in response to conditions on projects such as the construction of the Hoover Dam during the Depression. Workers who collapsed from the July heat in Nevada or were killed in accidents were quickly replaced from a pool of unemployed men who were already camping near the job site. To comply with the law, contractors must determine the "going rate" for construction labor in an area. As a practical matter, the "union rate" for labor becomes the going rate. That rate then becomes the mandated minimum wage on the government-financed project. One effect is to distort market wages and drive up the cost of government-financed projects. For example, a few years back, the market wage for plumbers in Kentucky was $18.15 an hour, according to the Bureau of Labor Statistics. Yet a wage survey for Owsley County, Kentucky, required that plumbers on public projects receive $23.75 an hour, more than 30 percent above the government's own market wage.51 A number of laws contain prevailing-wage provisions. They vary on the government expenditures they target for coverage. The main prevailing-wage laws include the Davis-Bacon Act, the Walsh-Healey Public Contracts Act, the Service Contract Act, and the National Foundation for the Arts and Humanities Act. A spate of new laws extends prevailing-wage coverage to new immigrants to the United States and to noncitizens who are working in the United States under special provisions. For example, the Nursing Relief for Disadvantaged Areas Act of 1999 allows qualified hospitals to employ temporary foreign workers as registered nurses for up to three years under a Page 640special visa program. The prevailing wage for registered nurses must be paid to these foreign workers. Similar acts target legal immigrants and farm workers. Much of the legislation discussed so far was originally passed in the 1930s and 1940s in response to social issues of that time. While this legislation has continued to be extended up to the present, the Equal Rights movement in the 1960s pushed different social problems to the forefront. The Equal Pay Act and the Civil Rights Act were passed. Because of their substantial impact on human resource management and compensation, they are discussed at length below.

COMPLIANCE: A PROACTIVE APPROACH

Compliance with laws and regulations can be a constraint and/or an opportunity for a compensation manager. The regulatory environment certainly constrains the decisions that can be made. Once laws are passed and regulations published, employers must comply. But a proactive compensation manager can influence the nature of regulations and their interpretation. Astute professionals must be aware of legislative and judicial currents to protect both employers' and employees' interests and to ensure that compensation practices conform to judicial interpretation. How can a compensation manager best undertake these efforts? First, join professional associations to stay informed on emerging issues and to act in concert to inform and influence public and legislative opinion. Second, constantly review compensation practices and their results. Be sure to consult with legal counsel in doing so, as attorney-client privilege and protection of work product are important issues to understand prior to conducting analyses on an organization's compliance. The fair treatment of all employees is the goal of a good pay system, and that is the same goal of legislation. When interpretations of what is fair treatment differ, informed public discussion is required. Such discussion cannot occur without the input of informed managers.

A 1939 pay policy handbook for a major U.S. corporation outlines the following justification for paying different wages to men and women working on the same jobs:1 The . . . wage curve . . . is not the same for women as for men because of the more transient character of the former, the relative shortness of their activity in industry, the differences in environment required, the extra services that must be provided, overtime limitations, extra help needed for the occasional heavy work, and the general sociological factors not requiring discussion herein. Basically then we have another wage curve . . . for women below and not parallel with the men's curve. The presumption that people should be paid different wages based on "general sociological factors" was still evident in the United States in the 1960s, in newspaper help-wanted ads that specified "perky gal Fridays" and in whites-only local unions. The 1960s civil rights movement and subsequent legislation were intended to end such practices.Page 620 Are you thinking you have stumbled into a history class by mistake? Not so. These historical practices and subsequent legislation still affect pay decisions. However, legislation does not always achieve what it intends nor intend what it achieves. Consequently, compliance and fairness are continuing compensation objectives. In democratic societies, the legislative process begins when a problem is identified (not all citizens are receiving fair treatment in the workplace) and corrective legislation is proposed (the Civil Rights Act). If enough support develops, often as a result of compromises and trade-offs, the proposed legislation becomes law. Employers, along with other stakeholders, attempt to influence the form any legislation will take. Once passed, laws are enforced by agencies through rulings, regulations, inspections, and investigations. Companies respond to legislation by auditing and perhaps altering their practices, perhaps defending their practices before courts and agencies, and perhaps lobbying for still further legislative change. The laws and regulations issued by governmental agencies created to enforce the laws are a significant influence on compensation decisions throughout the world. In the United States, there are three branches of federal government and each plays a role in the legal and regulatory framework in which employers work toward compliance objectives. The legislative branch (Congress) passes laws (or statutes). The executive branch, headed by the President, enforces laws through agencies and its other bodies (e.g., the Department of Labor), and the judicial branch interprets laws and considers their constitutionality. Over time, the legislative branch may change existing laws or pass new ones. The way that the judicial branch interprets laws can also change. The great interest in Supreme Court justice appointments and the difficulty sometimes encountered in gaining their confirmation is based on the belief that who the justices are will matter. Finally, enforcement priorities and intensity can vary from one presidential administration to the next. Compliance efforts by employers must take that fact into account. Of course, the regulatory environment is also a function of state and local laws, which often cover employers not covered by federal laws and/or include requirements that go beyond federal laws. For example, Title VII of the (federal) Civil Rights Act, which prohibits employment (including pay) discrimination on the basis of race, color, religion, sex, or national origin, covers employers with 15 or more employees. But, under the Wisconsin Fair Employment Act, all employers are covered and discrimination on the basis of some characteristics (e.g., sexual orientation) not included in Title VII is prohibited. As another example, in our discussion of minimum wage laws later in this chapter, we will see that some states have minimums greater than the federal minimum. We will also see that some cities have living wage laws. Finally, of course, as we saw in Chapter 16, laws differ by country.2 Our objective in this chapter is to help you become more familiar with the legal and regulatory framework of compensation. Importantly, however, you will not be an attorney after reading this chapter. Compliance will require legal advice. To help motivate you to speak promptly to an attorney should you encounter legal compliance risks, consider Exhibit 17.1, which reports payments by employers to come into compliance with regulatory actions brought by two U.S. Page 622government agencies, the Department of Labor's Equal Employment Opportunity Commission (EEOC) and its Wage and Hour Division (WHD). Compliance issues fall into two corresponding areas: employment discrimination (especially Title VII of the Civil Rights Act) and wage and hour (especially the Fair Labor Standards Act, FLSA). Over the 5-year period 2013-2017, the EEOC recovered more than $1.72 billion from employers to resolve employment discrimination issues (some, but not all, having to do with compensation issues), with most of that relating to Title VII. During that same period, the WHD recovered over $1.27 billion in back wages payments (i.e., wages for work employees previously performed, but for which they were not fully paid) from employers. Employees may also bring private lawsuits (i.e., without involving government agencies) against employers.

These are most costly when they are class action suits that include many similarly situated employees combining to bring a single joint lawsuit. In the area of employment discrimination, Exhibit 17.1 shows that the 10 largest private plaintiff class-action lawsuit settlements alone cost over $226 million per year, or about $1.4 billion over the 5-year period. On the wage and hour side, the 10 largest class action lawsuit settlements cost over $430 million per year, or more than $2.14 billion over the 5-year period. These, we think you will agree, are big numbers, and it would be great if you could help your future employer comply with the law so as not to be part of paying out these very large amounts of money to settle lawsuits.