Comparable worth: how do we know it will work?

Monthly Labor Review, Dec, 1985 by Carolyn Shaw Bell

The title of this article poses a question to which there is a very short answer. We don't. We are completely unable to predict the outcomes of an effective comparable worth policy, whether mandated by law or adopted by private decisionmakers. Our ignorance stems from the lack of data with which to build a viable economic model. The issue is, of course, too new for historical evidence or even case studies to provide much help.

The dearth of useful data is due primarily to the fact that comparable worth itself comprises several different issues. Most of these issues have, in fact, emerged from analyzing statistics gathered for other purposes. But comparable worth has frequently been proposed as a solution without clearly defining the problem, partly because of insufficient data, and partly because of insufficient analysis of existing data.

The following discussion will elaborate on these statements. It concludes that efforts to design data collecting systems or even to tabulate and amass hotel data that already exist lag behind efforts to litigate and legislate comparable worth. It is highly likely, therefore, that comparable worth as a policy will be adopted or rejected on the basis of factors other than reasoned analysis.

Defining comparable worth

The term "comparable worth" is difficult to define. Whatever it is, the concept emerged after the passage of the Civil Rights Act of 1964. Title VII of the act makes it an unlawful employment practice for any employer to discriminate on the basis of race, color, religion, sex, or national origin. Title VII specifically mentions hiring and discharge, compensation and conditions of employment, and the limiting of opportunities for employment.

Nowhere in the 1964 act, or in the legislative history preceding its passage, or in its predecessor, the Equal Pay Act of 1963, was the term comparable worth mentioned or its essence discussed in other ways. So the concept did not originate with Title VII, whether or not it can be justified by that legislation. Rather, the notion of comparable worth emerges from a specific interpretation of statistical estimates. These estimates show a significant and continuing disparity between men's wages and women's wages, and between the wages of blacks and whites. The data describe an existing condition, which the use of comparable worth seeks to remedy. It must be noted at once that most discussions move from simple descriptive statistics to comparable worth as the remedy with little attempt at analyzing the data, assessing their applicability, or rigorously defining the problem.

Examples of the difficulties in defining comparable worth and its aims abound in the press. When the issue arose during the 1984 Presidential campaign, one political writer identified the concept as "a means of raising the income of working women." More recently, however, another commentator defined comparable worth as a "practice . . . designed to increase the pay of workers in female-dominated fields such as nursing to a level of men in a field requiring comparable labor."

These two quotations share one characteristic: they both report comparable worth as a solution to a problem. But they identify different problems. One view, widely held, sees comparable worth as a remedy for low incomes and growing poverty among women. Another suggests that comparable worth is the remedy for the earnings differential between male jobs and female jobs. Proponents of both rely on statistics to describe the problem.

The case for comparable worth as a remedy for poverty among women is a very general or macroeconomic statement referring to women in aggregate. Thus, those who seek to remedy such a poverty quote data on earnings of women compared to men and, most frequently, the familiar figure that full-time year-round workers who are female earn about 60 percent as much as their male counterparts. They then explicitly or implicitly translate these earnings figures into income.

The preponderance of low incomes among women can be found in many different sets of statistics. To advocate comparable worth as a means of raising these incomes, however, often rests on the premise that discrimination against women exists in the workplace. The same assertion is required in the other line of advocacy, which sees comparable worth as a remedy not so much for poverty as for differences in wages.

This second notion of comparable worth has frequently been called "pay equity" and proposes to do away with obvious and sometime noteworthy differentials in wages between occupations. Again, statistical evidence can be quoted at length. However, unlike the estimates cited in support of comparable worth as a general remedy for poverty among women, these data refer to one market and, hence, constitute the microeconomic approach.

The use of data on interoccupational wage disparity can be illustrated by testimony before a 1984 congressional hearing that contrasted monthly salaries for city or State government workers in various job classifications--for example, a senior carpenter at $1,080 and a senior legal secretary at $665, or a senior accounting clerk at $836 in pay and a streetsweeper at $758. Jobs paying higher wages were found to be held almost exclusively by men, with women dominating the lower-wage jobs. Again, the proposed remedy (without any very careful delineation of the problem) was to implement comparable worth in determining wages.