The Wall Street Journal, April 10, 2010, Saturday

The Wall Street Journal

April 10, 2010, Saturday

The Wall Street Journal

Not All Differences in Earnings Are Created Equal

By CARL BIALIK

Later this month, American women will start earning money for the first time this year.

That seemingly nonsensical statement stems from this fact: For the average full-time female worker to earn the same amount of money as the average man, she would need to work nearly four additional months—until the third week of April—because of the 23-percentage-point gap in pay between the sexes.

That is why several advocacy groups have designated April 20 as Equal Pay Day. Lawmakers in the U.S. and the U.K. have taken up the cause, and President Barack Obama, who cited the pay gap on the campaign trail in 2008, repeated the statistic while commemorating International Women's Day last month. The Senate is considering a bill meant to address underlying gender discrimination some believe explains the pay gap, and the U.K. Parliament passed such a bill this week.

But do women really earn that much less than men? It depends on how you interpret the numbers.
Journal Community

Vote: What is the best explanation for the gap between men's and women's income?

People often assume that the gender gap measures how much a woman is paid for doing the same job as a man. Instead, the figure is based on a broader look at employment. In 2008, the typical American woman working at least 35 hours a week, year-round, earned 77.1% of what the typical American man did, according to the latest figures from the Census Bureau. That gap has changed little since 2001, ranging from 75.5% to 77.8%.

But restrict the comparison to those working 40 hours a week or more, and the gap narrows, with women earning 87% of what men did.

Also, the gap covers all workers, regardless of job title. But men, on average, work in higher-paying fields.

Cornell University economists Francine D. Blau and Lawrence M. Kahn found that after adjusting for factors such as education, experience, occupation and industry, the remaining, "unexplained" gender gap in 1998 was nine percentage points. Women also are likely to interrupt their careers, often to start a family, and such breaks can derail promotions and raises.

"When you first see the numbers, you would say there is a glass ceiling," says Harvard University economist Claudia Goldin. "And yet when you scrutinize the data, you find lots of evidence of people making rational choices."

But advocates for closing the pay gap say these differences in industries, job titles and career tracks might themselves be evidence of discrimination, rather than of free will in the job market. If women are underrepresented in higher-paying fields, that could be due to hostile environments in those fields rather than women's preference to work elsewhere. "Employers are doing a terrible job of hiring women in large enough numbers into nontraditional occupations," says National Organization for Women President Terry O'Neill.

Also, not all gender differences explain men's higher pay. U.S. working women earn less than men even though they are more educated on average than working men, points out Jeffrey Hayes, senior research associate for the Institute for Women's Policy Research, a nonprofit research group.

The importance of how the pay gap is calculated was highlighted in a recent battle in the U.K. The Government Equalities Office last year said that women earn 23% less than men, based on the median hourly wage of all employees. But the U.K. Statistics Authority, which audits government numbers, pointed out that the gap looks much smaller when focusing on full-time workers, among whom men make 13% more than women; or part-timers, among whom women make 3% more than men. The gap is bigger overall than in either of the subsets of workers because more men than women work full-time, and full-time workers earn more than part-timers per hour.

But the contrast between full- and part-time income gap shows how advocates on either side of the issue can find a number to back their case. "Everybody wants one statistic to use to measure the gap between men's and women's pay, and there isn't one statistic that really serves satisfactorily," says Richard Alldritt, head of assessment for the U.K. Statistics Authority, "which is quite an unusual situation for a concept that is really quite a simple concept."

A survey of global businesses released last month sheds light on the shortage of information about the gender gap. Just 28% of 600 major global employers whose human-resources managers responded to a survey conducted by the World Economic Forum attempt to track differences in pay at their companies by gender. "You can't solve what you can't measure," says Saadia Zahidi, a co-author of a report on the survey for the World Economic Forum, organizers of the annual Davos forum.

For planning purposes, the National Committee on Pay Equity, a coalition of women's and civil-rights organizations, schedules Equal Pay Day far in advance—before the latest Census numbers are available. Committee chair Michele Leber says the most-recent data would point to holding Equal Pay Day on April 19, not April 20, a Tuesday. "A number of years ago, we decided to select a Tuesday in April—Tuesday because that is how far into the [next] work week a woman has to work to earn as much as men do," Ms. Leber says.

Nonnumerical considerations also come into play. "We try to avoid major holidays and other conflicts," Ms. Leber says. "And we have been trying to schedule it when Congress is in session."