After more than a generation since the Equal Pay Act of 1963 and the Civil Rights Act of 1964 together barred employment and wage discrimination, the gap between men's and women's average earnings is still wide. In 1999, women's median weekly earnings for full-time workers were 76.5 percent of men's-a gender wage gap of 23.5 cents for every dollar earned by the median man.
Many believe that the wage gap is a good measure of the extent of gender wage discrimination, which occurs when men and women are not paid equal wages for substantially equal work. Irasema Garza, director of the Women's Bureau of the U.S. Department of Labor, recently testified to the widespread nature of this view in policy-making circles. Before Congress last June, she outlined steps being taken by the current Administration to eliminate the gender wage gap. Ironically, the gap has increased since 1993, when the Administration took office. After falling steadily between 1979 and 1993, it rose in four of the six years from 1993 to 1999, ending the period a little more than one-half of a cent higher.
This uncomfortable trend, however, has little to do with a failure to fight wage discrimination. The weight of evidence suggests that little of the wage gap is related to wage discrimination at all. Instead, wage discrimination accounts for, at most, about one-fourth of the gap, with the remainder due to differences between men and women in important determinants of earnings such as the number of hours worked, experience, training and occupation. Moreover, even this one-fourth of the gap may have less to do with wage discrimination than with the accumulated effects of shorter hours and interrupted careers on women's earnings and promotion prospects. To see this, let's break the wage gap numbers down in greater detail.
Breaking Down the Numbers
The first step in understanding the composition of the gender wage gap is to see if the correct measure of wages is being used. Because the average woman works fewer hours per week than the average man, defining the gap in terms of weekly earnings, as the Department of Labor usually does, inflates the wage gap artificially.1 Shifting the focus to hourly wages alone eliminates almost one-third of the gap: In 1999, women's median hourly earnings were 83.8 percent of men's, leaving a 16.2 cent gap in hourly earnings.
Defining the gender wage gap in terms of hourly earnings not only makes more sense statistically, but also illuminates the labor market gains made by women. As the accompanying chart shows, during the last two decades the gender gap in hourly earnings has fallen faster than the gap in weekly earnings. This has occurred as more women entered the labor force, including much larger proportions of women with children.2 Because the average woman with children works fewer hours per week, this trend has tended to increase the difference between the two measures of the gender wage gap.
Still, the gender wage gap of 16.2 cents that remains after correcting for the number of hours worked per week is rather substantial. The next question to examine is how much of the gap is due to human capital variables--such as education and experience--and other variables--such as industry, occupation and union status--that make wages differ between any two groups of workers. A 1997 study by Francine Blau and Lawrence Kahn is representative of the research done on this question. This study was relied upon in a recent analysis of the gender wage gap by the President's Council of Economic Advisers (CEA), which was subsequently cited by Director Garza in her statement to Congress.3
The Blau and Kahn study attributes 62 percent of the gap in hourly wages to such differences-one-third to differences in human capital variables, and 29 percent to differences in industry, occupation and union status. After applying these numbers to the 16.2 cent gap in hourly earnings, 6.2 cents of the gender wage gap remains unexplained. …