One of the oldest debates in contemporary social science is why women earn less than men. Conservatives tend to argue that because women anticipate taking time off to raise children, they have fewer incentives to work hard in school, and they choose careers where on-the-job training and long hours are less important. Liberals tend to focus on sex discrimination as the explanation. Obviously some mixture of those factors is at work, but academics have long been frustrated when they try to estimate which force is greater: women's choices or men's discrimination.

A new study looks at this problem in a wonderfully inventive way. In previous studies, academics have looked at variables like years of education and the effects of outside forces such as nondiscrimination policies. But gender was always the constant. What if it didn't have to be? What if you could construct an experiment in which a random sample of adults unexpectedly changes sexes before work one day? Kristen Schilt, a sociologist at the University of Chicago and Matthew Wiswall, an economist at New York University, couldn't quite pull off that study. But they have come up with the first systematic analysis of the experiences of transgender people in the labor force. And what they found suggests that raw discrimination remains potent in U.S. companies.

Schilt and Wiswall found that women who become men (known as FTMs) do significantly better than men who become women (MTFs). MTFs in the study earned, on average, 32% less after they transitioned from male to female, even after the authors controlled for factors like education levels. FTMs earned an average of 1.5% more. The study was just published in the Berkeley Electronic Press' peer-reviewed Journal of Economic Analysis & Policy.

The men and women in the study had already gone to school and made their career choices. Some of them changed jobs after they transitioned, and some stayed in the same jobs. Some were out to their employers; others started completely new lives as members of the opposite sex. Regardless, the overall pattern was very clear: newly minted women were punished, and newly minted men got a little bump-up in pay.

Still, the paper is complex, so it's useful to step back first and look at where the larger debate over the gender wage gap stands. After all, isn't that gap narrowing to the point of obscurity? Actually, no. The Russell Sage Foundation published the most authoritative work on the gender wage gap in 2006, The Declining Significance of Gender?. In the book, Francine Blau and Lawrence Kahn, both Cornell economists, show that the average full-time female worker in the U.S. earns about 79% of what the average full-time male worker makes. Women employed full-time actually tend to have slightly more education than men, but women are still more likely to work in clerical and service jobs. Blau and Kahn say women do make different choices when they decide on college majors and jobs — even highly educated women more often choose "female" occupations that pay less — but the authors also note that discrimination persists. As one example, they cite a 2000 study which found that when symphony orchestras switched to blind auditions — those in which the musicians play behind a screen — women had a significantly better chance of being hired.

The good news is that the gender wage gap has narrowed. In 1978, full-time women workers earned just 61% of what full-time men did, compared to 79% now. But what to make of the big difference in the experiences of those transgenders who have become women versus those who have become men? Schilt, one of the authors of the new article, interviewed a female-to-male transgender attorney a few years ago. As a younger attorney, the lawyer had been Susan; now he was Thomas. He told Schilt that after he transitioned from female to male, another lawyer mistakenly believed that Susan had been fired and replaced by Thomas. The other lawyer commended the firm's boss for the replacement. He said Susan had been incompetent; "the new guy," he added, was "just delightful." (Later, Ben Barres, an FTM neurobiology professor at Stanford, told The Wall Street Journal of a similar experience. An attendee at one of his lectures leaned over to a colleague and said, "Ben Barres' work is much better than his sister's.")

Such stories help explain an interesting feature of transgender life: men who want to change outward gender wait an average of 10 years longer to transition than women, according to the new article by Schilt and Wiswall. "MTFs attempt to preserve their male advantage at work for as long as possible," they write, "whereas FTMs may seek to shed their female gender identity more quickly." It should be noted that many transgender men do experience discrimination, especially if they are short and if they don't look convincingly male. Also, it's harder for MTFs to pass than FTMs: men who become women still have large hands and bigger frames. The less-convincing appearance of MTFs probably explains part of the reason they earn so much less after they transition. Still, the new paper suggests an entirely new vein of research in the field. It also suggests that if you're thinking about changing sexes, you should carefully consider the economic consequences.