Comparable worth is a policy implemented by a small number of state, local, and national governments to remedy wage disparities between men and women due to occupational segregation, a situation in which women are more likely to be employed in low-paying occupations and men are more likely to be employed in high-paying occupations. Comparable worth requires “equal pay” for “equal worth,” where worth or value is not determined by the market but via a job-rating scheme that assigns point values to different job attributes, such as skill level, responsibility, and working conditions, and aggregates them to a measure of job worth or value. Jobs with equal value would be paid the same wage; jobs with higher value would be paid a higher wage; and jobs with lower value would be paid a lower wage. In addition to raising wages in female-dominated occupations, comparable worth might also lower wages in male-dominated occupations.
Proponents of comparable-worth legislation view the negative relationship between wages and the proportion of females in a given occupation as an undervaluation by society of “women’s work” relative to “men’s work.” For these proponents, such undervaluation of work performed primarily by females is thought to arise from two sources. The first source, crowding, refers to the large number of women forced into certain occupations as a result of discrimination. The large supply of women to these occupations depresses wages. The second is society’s devaluation of work performed by women simply because it is performed by women. Evidence presented for this thesis is that earnings are statistically negatively related to the proportion of women in an occupation, even after controlling for observable and measurable job characteristics. Comparable worth is proposed as a way to raise the wages of women in female-dominated occupations to their “true” higher value.
Some opponents of comparable-worth legislation note that salaries set according to comparable worth do not consider the personal characteristics of workers or the unobservable or unmeasured job attributes that affect pay, such as flexibility and overall family-friendliness, and so this disparity in occupational choice may not be due to discrimination. Women choose to enter these lower-paying occupations because these unmeasured job attributes compensate for the lower pay. In the jargon of economists, women are willing to pay a compensating-wage differential (that is, accept a preferred working environment for less monetary pay) in order to better combine work and child-rearing responsibilities. These opponents point out that since this is not discrimination, no remedy is needed.
Opponents who accept discrimination as a factor that leads to occupational segregation note that interfering with market wages has negative unintended consequences, even though increasing women’s wages to combat discrimination may be a laudable goal. One of these potential unintended consequences is a reduction in the employment of women in female-dominated occupations. Employers faced with paying higher wages in these occupations will substitute other inputs for these workers and decrease production overall in response to the increased cost of labor. Another consequence of interfering with market wages is the creation of unemployment in these female-dominated occupations due to the surplus of women desiring to work at the now-higher wages. Yet another negative consequence is the depression of the wages of men and women in the male-dominated occupations caused by an increase in the supply of workers to these occupations—a result of women who are unable to find employment in the now-higher-paying female-dominated occupations moving into the male-dominated occupations. Finally, if wages are artificially lowered as part of a comparable-worth policy that both increases some wages and decreases others, this will lead to the decreased quality of the workers in the male-dominated occupations.
Finally, even if the negative effects on employment and wages are deemed acceptable, it is difficult to put the concept of comparability into practice, given the subjective nature of the job evaluation process and the need to allow market forces and the efforts of workers to play a role. Evidence has shown that different job evaluation studies assign different points or weights to various skills, leading to different job comparisons and wage adjustments depending on the particular evaluation used. In addition, different localities are subject to different market forces. A private national employer with employees all across the country will need to take these local market forces into account in order to remain profitable. If a firm is not allowed under comparable worth to address such market forces with wage differences, an employer might choose to adjust production in such a way as to employ workers in the female-dominated occupations only in labor markets where wages are low and employ workers in male-dominated occupations only in labor markets where wages are high. To encourage individual employee productivity, merit bonuses and raises for extra effort may be necessary, while to discourage turnover, seniority may also need to be rewarded. Finally, overtime may be necessary for some workers but not others. These important factors are ignored when computing wages strictly according to the “equal value” principle of comparable-worth policy.
History of Comparable Worth
Early U.S. legislation attempting to address the inequality of pay between men and women included the U.S. Equal Pay Act of 1963 and Title VII of the U.S. Civil Rights Act of 1964. Similar legislation was put in place in the 1970s in other industrialized countries, such as the United Kingdom, Canada, Spain, and France. The focus of this legislation was the provision of “equal pay for equal work” performed by both men and women. However, critics claimed that because of occupational segregation, men and women were rarely performing equal work. As a response to this perceived inequality, comparable-worth policy was first proposed in the United States in the late 1970s to instead promote “equal pay for equal value.” In the United States, comparable worth was primarily adopted by state and local governments rather than private employers, as public employers are not subject to profit-maximizing constraints. However, when other industrialized countries followed suit in the early 1980s, some required comparable worth in the private sector as well. Some people believe that unions have the potential to bring comparable worth to the private sector in the United States as a result of the increasing feminization of labor unions.
Job Evaluation Schemes in Determining Equal Worth
Job-rating schemes have long been used by employers with internal labor markets, including public employers and large private firms that participate in collective bargaining with unions over wages. These schemes assign points to various characteristics of a job, such as skill level, level of responsibility, working conditions, and physical and mental effort required. The points are then summed (and possibly weighted) to assign a particular total value to a particular job. The point total is used to compare the particular job to other jobs in determining “equal worth.” Equal points mean equal worth and equal pay; higher points mean higher worth and higher pay; and lower points mean lower worth and lower pay.
A problem with this technique of assigning points, however, is that it is quite subjective. Characteristics such as physical strength that are more often associated with male-dominated jobs may be rated more highly than characteristics such as communication skills that are more often associated with female-dominated jobs, leading to lower values for female jobs. However, over the years, several gender-neutral evaluation methods have been developed and used to implement comparable worth.
Pay Based on Job Evaluation Schemes: Determining Comparable-Worth Wages
Once each job is assigned a total point value, total points are plotted against wages, and a line is fit to these data using regression analysis, an accepted statistical technique. Once the equation of the line representing the relationship between wages and value is determined in this manner, it is often revealed that most female jobs are below the line, while most male jobs are above the line, where the standard for categorizing an occupation as female dominated is 70 percent or more female workers and the standard for categorizing an occupation as male dominated is 70 percent or more male workers. This suggests that as long as no important job attributes that affect wages have been omitted from the calculation of the total point value, female-dominated jobs are paid less simply because they are performed by females.
The most transparent method of increasing female wages is to directly match female-dominated jobs to male-dominated jobs that are deemed equivalent because they fall within the same range of overall job point values. The wage in the female-dominated occupation is then assigned the wage paid in the matching male occupation. If more than one male-dominated occupation matches a particular female occupation, a rule must be followed in determining which wage will be applied. In practice, however, this can lead to the lowest wage being applied. It can also lead to female-dominated occupations with higher point values being assigned lower wages than female-dominated occupations with lower point values, given idiosyncrasies in the markets for workers in male-dominated occupations.
There are, however, several other ways to determine female wages under comparable worth that are not subject to these concerns. One method is to estimate a wage equation for male-dominated occupations and use that estimated equation to compute a wage for each female-dominated occupation. Another method is to estimate an earnings equation over all occupations, controlling for total job points and the proportion of females in an occupation, and then add the estimated reduction in wage due to the proportion of females to existing wages in each of the female-dominated occupations. Finally, another way is to estimate an earnings equation using data on wages and job values for all occupations and then use that equation to adjust all workers’ wages, both male and female. In this scenario, some individuals, primarily females, would experience an increase in wages, while others, primarily males, would experience a decrease in wages. This is less politically feasible, however, than any of the other methods.
Economic Theory and Comparable-Worth Occupational Segregation and Crowding
Economic theory predicts a negative effect of comparable worth on women’s employment. This is because increased wages in female-dominated occupations would lead employers to make fewer of these jobs available. Employers cut back on production when faced with higher input costs, requiring less of all inputs. In addition, employers substitute other inputs for the now-more-expensive female workers. Theory also predicts an increase in unemployment in female-dominated occupations due to an increase in the number of women looking for work at the now-higher wages at the same time the number of available jobs is declining. Economic theory also predicts that the supply of females to male-dominated occupations where wages are not directly adjusted via comparable-worth policy will increase due to the decline in employment in the female-dominated occupations covered by the policy. This will depress wages in the male-dominated occupations and will possibly even lower wages for the women below those they would have received without comparable worth in the female-dominated occupations.
Finally, if comparable worth not only increases wages in female occupations, as is often proposed, but also artificially decreases them in male occupations, economic theory predicts there will be a shortage of workers in male-dominated occupations, as fewer individuals will be willing to work at these lower wages. As a result of this shortage of workers in male-dominated occupations, the quality of workers in these occupations could decline, as firms might need to accept lower-quality workers in order to meet staffing requirements. Thus, comparable-worth policy affects not only the employment and unemployment of workers in female-dominated occupations but also the quality of workers hired and hence the quality of the final products or services produced by workers in the male-dominated occupations.
Empirical Evidence on Comparable Worth
Effectiveness of Comparable-worth Policies with Respect to Women’s Wages and Employment
Empirical evidence suggests that the benefits of comparable worth in the form of a lowered wage gap between male- and female-dominated occupations are smaller than proponents of comparable-worth legislation expected and that the decline in employment is also much smaller than opponents expected. One potential reason for these small estimated effects is that comparable worth is limited primarily to the public sector. This sector makes up only a small portion of employment in many, if not most, female-dominated occupations. Another reason suggested by empirical research is that male-female pay differences within an occupation are predominantly interfirm and interindustry, not intrafirm, and thus not subject to comparable-worth policy that affects pay differences only within a particular firm or establishment.
Effectiveness of Comparable-worth Policies in Satisfying the Prevailing-Wage Principle
Wages paid by public employers are usually set with the prevailing-wage principle in mind. This principle states that public employees are to receive wages equal to those they would be paid if they were working in an equivalent job in the private sector. If the ratio of public to private wages is equal to 1, then the prevailing wage principle is satisfied. While perfect correspondence between wages in the public and private sector does not occur in the absence of comparable worth, the estimated ratio is close to 1. Empirical results indicate, however, that when comparable-worth policy requires wage increases for female-dominated occupations but disallows decreases for male-dominated occupations (as is often the case in practice), public sector wages are moved away from this prevailing-wage standard. However, if decreases in the wages paid in male-dominated occupations are allowed, although in practice this is usually not politically feasible, then one would be moved toward the prevailing-wage standard. Hence, both the competing principles of prevailing wage and comparable worth must be considered when designing comparable-worth policy.
A Different Application of Comparable Worth
A situation in which the issue of equal pay for equal worth arises outside the context of gender is in the case of university professors. Efforts by unions to organize academic faculty often note the disparity in pay between faculty with similar training (a PhD) and years of experience (a particular rank) but different fields of research and/or teaching. Such disparities arise due to differences in the supply and demand in the markets for different types of professors. For example, economics or finance professors often earn relatively high salaries compared with sociology professors. This may be due to a higher demand for economics classes than for sociology classes, a smaller supply of economics professors due to more lucrative alternative opportunities in the nonacademic market, or a combination of both. However, unions and comparable-worth advocates would argue that the work performed by both types of professors is of equal worth, as both types of professors engage in teaching, research, and service activities, and so both economics and sociology professors should be paid equally. Opponents, however, would point to the negative effects of a decline in the employment of sociology professors, an increase in the unemployment of sociology professors, and the potential shortage of and corresponding decline in the quality of economics professors that would result if the wages of economics professors were set artificially low as a result of comparable-worth policy.
- Ames, L. 1995. “Fixing Women’s Wages: The Effectiveness of Comparable Worth Policies.” Industrial and Labor Relations Review 48:709-725.
- Ehrenberg, R. G. and Smith, R. S. 2003. Modern Labor Economics: Theory and Public Policy. Boston, MA: Addison-Wesley.
- Eyraud, F. 1993. “Equal Pay and the Value of Work in Industrialized Countries.” International Labour Review 132:33-48.
- Hoffman, S. D. and Averett, S. 2005. Women and the Economy: Family, Work, and Pay. Boston, MA: Addison-Wesley.
- Hundley, G. 1993. “The Effects of Comparable Worth in the Public Sector on Public/Private Occupational Relative Wages.” Journal of Human Resources 28:318-342.
- Johnson, G. and Solon, G. 1986. “Estimates of the Direct Effects of Comparable Worth Policy.” American Economic Review 76:1117-1125.
- Orazem, P. F. and Mattila, J. P. 1998. “Male-female Supply to State Government Jobs and Comparable Worth.” Journal of Labor Economics 16:95-121.
- Sorenson, E. 1990. “The Crowding Hypothesis and Comparable Worth.” Journal of Human Resources 25:55-89.