Madam Chairman and members of the Equal Employment Opportunity Commission, thank you for inviting me to testify before you today. It is a great pleasure to be able to come and clarify the misconceptions involving average and adjusted wage gaps, and I ask your patience while I explain the data and how they are used.

Last Thursday, April 8, was dubbed Equal Pay Day by the National Committee on Pay Equity. The National Committee on Pay Equity joined the National Organization for Women and the AFL-CIO to try to persuade the nation that women are paid only 74 cents on a man’s dollar. Their organizational literature proposed stunts such as selling hamburgers for $1 to men but for 75 cents to women; selling cookies with one quarter removed; distributing dollar bills with holes in them to reflect the gaps in women’s pay; and organizing a New Year’s party on April 8 to recognize that women have begun a new year after catching up to men’s earnings from 1998. Such claims draw media attention, but do not accurately describe women’s compensation in the American workplace.

If these groups are to be believed, then American women are still second-class citizens, as they were before they had the right to vote. But, before declaring another crisis, it is worth looking at how these numbers were put together and some of the reasons behind the differences. This is what my coauthor, Christine Stolba, and I do in Women’s Figures: An Illustrated Guide to the Economic Progress of Women in America, which I would like to submit for the record.

During the nineteenth century, employers usually operated on the assumption that women in the labor force earned wages that were merely supplemental to household income. This assumption was reflected in women’s average earnings, which, according to most historians, were approximately one-third of men’s in 1820, rising to approximately 54 percent of men’s by the end of the nineteenth century. Women’s average wages continued to rise relative to men’s wages during the twentieth century, reaching 74 percent of men’s in 1998.

The 74 cent figure is derived by comparing the average median wage of all full-time working men and women. So older workers are compared to younger, social workers to police officers, and, since full-time means any number of hours above 35 a week (and sometimes fewer), those working 60-hour weeks are compared with those working 35-hour weeks. These estimates fail to consider key factors in determining wages, including education, age, part- or full-time status, experience, and, perhaps most importantly, consecutive years in the workforce. This average wage gap, as it is known, says nothing about whether individuals with the same qualifications who are in the same jobs are discriminated against.

When discrimination occurs, and, as this Commission knows all too well, it does occur, we have laws to deal with it. We need to focus on individuals rather than averages, and apply the Civil Rights Act and the Equal Pay Act to eradicate cases of discrimination as they occur.

How much less do equally-qualified women make? Surprisingly, given all the misused statistics to the contrary, they make about the same. Economists have long known that the adjusted wage gap between men and women—the difference in wages adjusted for occupation, age, experience, education, and time in the workforce—is far smaller than the average wage gap. Even just adjusting for age, as can be seen in Figure 1, removes a lot of the gap, because we can see that in 1993 women aged 16 to 29 made 92 percent of what men made.

The wage gap shrinks dramatically when multiple factors are considered. Women with similar levels of education and experience earn as much as their male counterparts. Using data from the National Longitudinal Survey of Youth, CBO director June O’Neill found that, among people ages twenty-seven to thirty-three who have never had a child, women’s earnings are close to 98 percent of men’s. Dr. O’Neill notes that "when earnings comparisons are restricted to men and women more similar in their experience and life situations, the measured earnings differentials are typically quite small."

What about the remaining gap, often referred to as the unexplained statistical residual? Economists Solomon Polachek and Claudia Goldin suggest that different expectations of future employment, or human capital investment, may explain the residual. In other words, since 80 percent of women have children, they may plan their careers accordingly, often seeking employment in fields where job flexibility is high and where job skills will deteriorate at a slower rate. This allows them to move in and out of the workforce with greater ease, or to shift from full-time to part-time work, if they so choose. This flexibility has a cost: lower wages in these fields.

Tenure and experience are two of the most important factors in explaining the wage gap. According to the U.S. Bureau of the Census, women on average spend a far higher percentage of their working years out of the workforce than men. As demonstrated by economists such as Francine Blau, Andrea Beller, David Macpherson and Barry Hirsch, this means that upon returning to the workplace, women will not earn as much as their male or female counterparts who have more uninterrupted experience.

There are reasonable explanations for the differences in average wages between men and women. First, in the 1960s and 1970s women received fewer undergraduate, graduate, and professional degrees than men. It was only in 1982 that women began to earn more than half of B.A. and M.A. degrees. In 1970 women earned about 5 percent of law and business degrees awarded, compared with about 40 percent today, as can be seen in Figures 2 and 3. It is these 1970 graduates who are now in their 50s, at the age of peak earnings, and there are more highly paid men than women.

Second, as Figures 4 through 7 show, many women still choose to major in specialties which pay less. Women get more degrees in public administration and communications and fewer degrees in math and engineering.

Third, as mentioned above, many women choose jobs that enable them to better combine work and family, and these pay less than those with rigid or extensive hours. Even in higher-paying professions such as medicine, many women choose to go into pediatrics, psychiatry, and family practice, all lower-paying fields than surgery, which is more demanding in terms of hours.

Many studies link increased numbers of children with decreased earnings. Professor Jane Waldfogel of Columbia University compared the gap in wages between men and women with the same education for two groups, mothers and women without children. She found that in 1991, women without children made 95 percent of men’s wages, but mothers made 75 percent of men’s wages, as can be seen in Figure 8. The difference can be explained by choices of occupations and hours worked, two variables which were not included in her study.

Naturally, there are different explanations for these data. One is that children take time away from women’s careers, both in terms of time out of the workforce to bear the children and in terms of time put into work effort afterwards.

A second explanation is that women who qualify for high-paying jobs—who major in business or math, or who go to the trouble of getting professional training, for example—quite naturally choose to work more. With a high-paying career, it is more tempting to delay having children, or have fewer of them, or none at all.

Of course, many people would say that there is a third explanation: employers discriminate against women who have children. So mothers are paid less for the same work or forced into positions of low pay. But data show that employers do not pay unmarried women less: why should the employer care if a woman is married? If employers were against marriage, they would pay married men less. But data show that married men are paid more than unmarried men.

If women were systematically discriminated against, as some assert, then some entrepreneur would be able to step forward and take advantage of this. We would see that firms hiring only mothers would make larger profits than others. In the same way, if women were truly paid only 74 cents on a man’s dollar, then a firm could fire all its men, replace them with women, and have a cost advantage over rivals. We do not observe this happening.

And this brings us to comparable worth, because the only way to get rid of the average wage gap of 74 or 75 cents on the dollar is to require equal pay for different jobs. And that is precisely what is being proposed by President Clinton in his Equal Pay Initiative, and by Senator Daschle and Representative DeLauro in their Paycheck Fairness Act.

Under comparable worth plans, a job’s worth would be so-called "objectively" measured by having officials examine working conditions and the knowledge or skill required to perform a task. These criteria not only favor traditionally female occupations over male ones (secretaries over truck drivers), but favor education and white-collar jobs over manual, blue-collar work. Neither experience nor risk, two factors which increase men’s average wages relative to those of women, are included as job-related criteria. And men’s jobs are more dangerous—ninety-two percent of workplace deaths are male.

Comparable worth has been rejected in courts all over the country as a requirement for states’ compliance with Title VII of the Civil Rights Act. But the president could impose comparable worth on American business, circumventing Congress, through regulation. His Equal Pay Initiative calls for the Department of Labor to issue guidelines for job categories deserving of equal pay. These guidelines are described as "voluntary," but there is nothing to prevent President Clinton from issuing an executive order forbidding the federal government from doing business with companies who do not adopt the standards.

Advocates of comparable worth deny that they support a centrally-planned economy, and say that all they want to do is stop discrimination against women. But a preference for more time at home with less pay and less job advancement over more time at work with more pay and advancement is a legitimate individual choice for women. Similarly, the choice of some men to retire early and forego additional earnings, a continuing trend, does not prove inequality between young and old. Neither of these phenomena is a policy crisis calling for government interference.

One of the greatest harms that feminists have inflicted on American women is to send the message that women are only fulfilled if their salaries are equal to men’s, and that a preference for more time at home is somehow flawed. Neither men’s nor women’s education and job choices prove social inequality.

The main question in the wage gap and comparable worth debate is who will bear the costs of having and raising children, women or employers? The practical consequences of forcing employers to bear these costs include less hiring—fewer jobs and more machines. In an international economy that means more jobs abroad instead of at home. Women’s wages made the biggest strides in the 1980s, a time of strong economic growth but one in which the minimum wage shrank in real terms and affirmative action enforcement was not a priority. There are also issues of fairness. Artificial increases in working women’s wages at the cost of lower salaries for men, or higher prices in stores, hurt non-working women who rely on men’s incomes. And why stop at comparable worth for men’s and women’s jobs? Why not have it for jobs between blacks and whites, or the disabled and the healthy, or tall and short people?

The average wage gap is not proof of widespread discrimination, but of women making choices about their educational and professional careers in a society where the law has granted them equality of opportunity to do so. Comparable worth promotes a dependence for women, and a reliance on government for protection. Given women’s achievements, such dependence is unnecessary. American women enjoy historically unparalleled success and freedom, and the progress they have made in the past half century will continue.

 Figure 1
 Figure 2
 Figure 3
 Figure 4
 Figure 5
 Figure 6
 Figure 7
 Figure 8

Diana Furchtgott-Roth is a resident fellow at AEI.