Women Earn Less Than Men, Especially at the Top
By CATHERINE RAMPELLIn most jobs, the gap between men’s and women’s earnings narrows greatly when you adjust for factors like career path and experience. But at the top of the income scale — jobs paying more than $100,000 — the salary gap between equally qualified men and women is still vast.
That’s the takeaway from a vast trove of data being released by PayScale, a company that tracks self-reported salary data.
Millions of Web-surfers share their compensation data with PayScale.com, usually with the goal of finding out if their pay is comparable to that of other workers with similar experience and credentials. (In other words, PayScale has a big, but not randomized, sample; more caveats about their numbers are discussed here, in our write-up of their data on salaries by college.)
The company decided to analyze its data to see if it could help explain observed differences in what men and women earn, a subject that has long intrigued economists and women’s rights advocates. According to the Bureau of Labor Statistics, women earned a median weekly salary in 2008 that was about 80 percent of the pay for men.
PayScale chose 90 common careers that its researchers thought gave a useful cross-section of the labor market: 30 jobs that were primarily held by women (e.g., dental hygienists, of whom 95 percent are women in PayScale’s database); 30 jobs that were primarily held by men (e.g., electrical engineers, of whom 90 percent are men); and 30 that had roughly equal representation between men and women (e.g., pharmacists, of whom 49 percent are women and 51 percent are men). For each particular job, they had anywhere from several thousand to tens of thousands of data points.
The company then examined the data two ways.
First, it looked at the median pay of men compared with women within a given field, both when they started their jobs and at mid-career, using the raw data that workers had provided. By these calculations, they found that within almost every one of their 90 chosen careers, and across careers at every level of education, men earned higher pay than women.
Below is a look at the median pay of men compared with women in each of the 90 jobs examined, with each dot representing a different job. The diagonal line represents where the dots would lie if men and women earned exactly equal pay. Notice how many of the dots fall below that line, indicating that women earn less than men.
Then PayScale’s researchers decided to look at how non-gender-related factors affected salaries.
They adjusted the salaries based on hundreds of different factors they had data on, including location of the job, industry, size of the company and the budget a person managed, as well as his or her education and certifications, special skills and length of experience in a field. In some cases, they adjusted the salaries based on industry-specific data (e.g., how many beds are in the hospital a nurse works in).
The goal was to start with the typical characteristics of a male worker in a job, then adjust the characteristics of the typical woman in the job to match those of the average man to see if these two prototypical employees would earn identical salaries.
And for most jobs, they almost did.
Here are the median salaries for men and women in each of the 90 careers chosen by PayScale, after controlling for outside factors than can affect pay other than gender. Notice that for lower-paying jobs, the dots are much closer to that diagonal line representing equal pay.
In other words, for most careers the company studied, PayScale found that the pay gay gap is largely the result of outside factors. Within a specific job, before controlling for outside factors the typical female worker earns pay that is only 90 percent of the typical male worker’s pay; after controlling for these variables, she earns 94 percent of the typical male worker’s pay. For jobs paying below $100,000, the gap narrows further.
The implication is that in most jobs where a wage gap exists, it is probably not due to overt discrimination, with bosses deciding, Mad-Men-style, that women should receive unequal pay for equal work. Rather, in most jobs, the different career choices that men and women make — or perhaps the different career opportunities men and women have available to them — account for big differences in pay, says Al Lee, PayScale’s director of quantitative analysis.
But even though the gap narrows when you control for these factors, it is still large when you look at a subset of the careers PayScale examined: the high earners. In jobs that pay more than $100,000, women earn just 87 percent of what men receive, even after adjusting for outside factors. You can see this in the second chart above — around the $100,000 mark, many more dots start to fall beneath the equal-wage line.
So what is so magical about crossing the $100,000 salary mark that allows men to earn so much more than equally qualified women?
Surely everyone will have a pet theory. Mr. Lee, for his part, suggested that higher-paid jobs often have less concrete or quantifiable measures of productivity and duties.
After controlling for outside factors, some of the biggest gender pay gaps are in jobs like chief executive (in which, after PayScale adjusted the data, women earn 71 percent of what men earn), hospital administrator (women earn 77 percent of what equally qualified men earn) and chief operating officer (women earn 80 percent of what equally qualified men earn).
In each of these jobs, performance quality is a relatively subjective measure. Compare those jobs to positions like engineers, actuaries or electricians, where the criteria for a job well done might be relatively more concrete or measurable — and where the salaries earned by men and women are roughly equal.
In other words, theorizes Mr. Lee, jobs in which quality is easier to measure are more likely to be compensated based on merit, so equally qualified men and women are likely to receive equal pay. On the other hand, in jobs where quality measures are more subjective, meritocracy may not rule, and men may be better compensated for reasons other than their qualifications. For example, perhaps men are subconsciously viewed as more competent than women, or are more adept at negotiating for raises.
But then again there are tons of exceptions to this rule, where jobs with more nebulous duties are relatively egalitarian and jobs with more concrete descriptions are not.
There are certainly alternative explanations, including factors that PayScale did not control for. For example, Mr. Lee and his team were not able to control for hours worked each week, since they didn’t have that data. That seems like a pretty significant determinant of income to me.
I’m also curious to know what these trends would look like if PayScale had attempted to conduct the same analysis for every job in their database for which they had a decent sample size; perhaps these 90 jobs are not, after all, representative of the over all job market.
But again, everyone will have a different explanation for the gap in wages. What’s yours?
How would you explain why women at the top of the income scale earn so much less than their male counterparts, even when you control for outside factors?
No comments:
Post a Comment