Monday, January 24, 2011

Your Inner Glass Ceiling?

Do you know what I find incredibly weird? That so many people don't appear to be able to understand that something may have many causes, not just one cause. If I asked you what makes a chocolate-strawberry cake, none of you would insist that it's the baking powder or the eggs or the flour or even the chocolate and strawberries alone. The gateau has many components.

Just like that the reason why women earn less, on average, than men cannot be made into ONE reason. And the various explanations are not necessarily mutually exclusive alternatives but true at the same time! Sorta like a chocolate-strawberry cake. But when we move from the cake to the gender gap in earnings people somehow start focusing on one single reason. What that reason is depends on the person's prior political stance, pretty much. Conservatives, for instance, argue that women choose to earn less, either because they just do or because evolution has made them that way.

The reason I'm writing about this, again, is a study summary I came across. It had to do with testing the possibility that women are more risk-averse than men, less willing to take work with rewards which depend on competition. Here's the actual study synopsis, not the write-up I want to talk about later:

We know that women, often working at the same kind of job as men, frequently are not paid as much as men,” said John List ( , professor of economics at UChicago and an author of the paper, “Do Competitive Work Places Deter Female Workers? A Large-Scale Natural Field Experiment on Gender Differences in Job-Entry Decisions ,” published by the National Bureau of Economic Research.

“Some of the explanations for the differences contend they are caused by discrimination or by women leaving the workforce to have children and then returning,” he said. “Other people have suggested that men are more attracted to competition than women, and that accounts for the differences .”

To test whether differences between men’s and women’s interest in competition actually affects their job choices, List and a research team created two advertisements on Internet job boards.

They posted jobs for administrative assistants, the most common job in the United States. One ad, which was gender-neutral, described the job responsibilities as preparing reports based on news stories and fulfilling typical office tasks. The second ad, for a sports news assistant, was similar, except that the job would entail writing reports about sports stories.

The advertisements were placed on job boards in 16 of the nation’s largest cities between January and April 2010. The team then presented respondents with additional information to describe different forms of compensation. Some applicants were told the job paid $15 an hour. Others were told the pay was based on individual competition, with a base salary of $13.50, and a $3 bonus depending on how he or she
did in comparison to other workers. Another package offered a $12 hourly base pay with a $6bonus if the employee outperformed other workers. Still others were told the job had a competition-based wage, but that comparisons would be based on the productivity of people working in teams.

Of the 6,779 people who responded to the ads , 2,702 applied once they knew the wage structure. Those included 1,566 women and 1,136 men. (About 20 of the applicants were actually hired.)

“When the salary potential was most dependent on competition, men were 94 percent more likely to apply than women,” List said.

The study found that although women were much less likely to pursue jobs where individual competition was a factor, the deterring effect on women could be overcome by having workers compete in teams, rather than individually.

Women were more deterred by jobs in which competition was a factor in determining pay if the local wages in their city were high. For instance, women were less inclined to pursue jobs with competitive wage situations in San Francisco, Washington, D.C., and Boston, where the median wages for other local administrative assistant jobs were about $13, close to the base pay for the jobs the researchers offered. In cities with lower local wages, the dissuasive effect on women of
competition-based pay diminished.

For example in Houston, where the local wage was $10, women actually showed slightly more interest in applying to jobs with a competitive pay structure
than men.

List said socialization of women and men may play a factor in the gender differences in the way men and women respond to pay incentives based on competition. Boys receive more encouragement growing up to be competitive, particularly in sports, while girls frequently are encouraged to be more cooperative, he said.

Note that the study is about applying for jobs, not about who gets picked for a particular job out of those who apply. This means that what it studies cannot really be viewed as an alternative total theory for the gender gap in earnings. And because it's only about applications the study tells us nothing what happens after a person has been hired for the job.

Note, also, that the kind of risky pay the study discusses is exactly the way restaurant servers are paid, for example. Yet women are very common in that particular profession.

The results about Houston where women were more interested in the competitive jobs than men suggests that the underlying reasons have at least something to do with local circumstances. In short, women can be more competitive than men under certain conditions, and one of those conditions may be the average local wage rate for a comparable job.

Those are some of my quick thoughts after reading the synopsis (I have several more complicated thoughts having to do with what the comparable wage rate might be for women and men, in general, and how that would feed into the study results). I couldn't find the actual study to see what the sports-related job's function was. My guess is that it was some kind of a control in the study.

Now for the summary about the inner glass ceiling: First, it omitted the bit about women in Houston actually being more competitive than men and replaced it with this:
Researchers also found that women who worked in cities with typically higher wages were less inclined to put themselves in a competitive pay scheme. Female residents of cities like San Francisco, Washington, D.C., and Boston, where median wages were closer to what researchers offered, were less inclined to apply than those in cities like Houston, where the median wage was around $10 at the time, indicating that the potential earning power had to be considerably higher than the local average for a woman to enter the competitive arena.
Do you see how the flavor has changed?
Second, it then goes on:
One interpretation suggests that women themselves are solely responsible for creating their own glass ceiling, but List offers the more likely cause that the social norms behind women’s apparent aversion to more competitive positions are established well before they enter the workforce.
Boys receive more encouragement growing up to be competitive, particularly in sports, while girls frequently are encouraged to be more cooperative, List explained.
There's that glass ceiling! Remember the strawberry-chocolate cake? We are now talking about baking powder alone. Sure, women can help themselves in the job markets by asking for raises more often and by demanding a higher initial salary if possible. And sure, realizing that one might be avoiding certain types of behavior because of one's upbringing doesn't mean that one shouldn't try to change those behaviors.

But to suggest as one interpretation that the glass ceiling is all inside women's own heads plays into the game where we try to pick one ingredient in the cake as the decisive one. It also discounts the possibility that women, on average, know how that competition is judged and that the rules might not be as objective as we wish.