EPM investigates why starting salaries for women emergency physicians appear to be nearly $13,000 lower than male counterparts.
Forty years after Gloria Steinem hit the scene, could it still be possible that there are gender differences in starting salaries amongst some of our most educated and elite professionals? The answer is yes, according to a provocative study in last month’s Health Affairs, which put starting salaries for male emergency physicians $12,653 higher than that of females.
Dr. Lo Sasso and his team looked at survey data from more than 8000 New York state physicians graduating from their residency programs and asked them questions about starting salaries. They found that there was a $35,000 gap between women and men. Clearly there are a lot of things which clearly influence salary, such as specialty, location, type of practice, anticipated hours, and loan repayment obligation. When comparable data from 1999 went through these factor adjustments, they found most gender differences disappeared. However, when Lo Sasso adjusted their 2008 numbers he found a residual and unexplainable $16,819 gap. While emergency medicine fared slightly better in the study, the numbers still beg explanation. There are several plausible theories for what is causing this disparity, from persistent discrimination to negotiation biases to confounding variables.
Legislative changes have made blatant gender discrimination in employment illegal, but perhaps subtle hiring discrimination still flies under the radar. Two provocative studies support this. Cecelia Rouse (The American Economic Review 2000) examined historical hiring practices of elite, big city orchestras in relationship to their adoption of “blinded” first-round auditions in which the identity (and gender) of the musician was hidden behind a screen. She attributed the increase in the number of new women hires in major orchestras from about 10% to 35% over the past 30 years to these blinded auditions. The second study by K Schilt (the B.E. Journal of Economic Analysis and Policy 2008) approaches this question in a rather novel way by actually making the individual their own matched control by examining their salary line pre and post sex change. She found that women transitioning to men were paid slightly higher while men transitioning to women lost up to 1/3 of their salary.
Perhaps it is not the initial salary put on the table but the effectiveness of the negotiation process that causes the disparity in question. It takes two to tango, and although contract malleability on the employer end is likely dependent upon job market and practice type, contract negotiation by the potential employee may be influenced by gender. Anecdotally, a few years ago a large contract management group took over a local ER and was hounding local EPs for coverage. Although I was not interested, I was a bit flattered by the generous moonlighting rate the aggressive recruiter offered me. This wore off pretty quickly, however, when I found out they had offered a male colleague far more. As I started to fume, my colleague said, “C’mon. You just needed to ask. They’ll pay anything. They’re so desperate.” This begs the question, are women as likely than men to “ask” for the extra goodies? Babcock (Harvard Business Review 2003) looked at the starting salaries of MBA grads from Carnegie Mellon and found that men negotiated their salaries 57% of the time while only 7% of the women did. As those who negotiated their salaries started out with approximately 7% higher base salary, the potential long-term financial implications of successful upfront negotiation are clear, even if all future raises are equal standard percentages.
So, is the answer just to encourage women to be tougher negotiators? Ironically, a disturbing little study suggests that this could backfire, putting women in a Catch 22. HR Bowles (Organizational Behavior and Human Decision Processes 2007) performed a set of experiments that concluded that women were less likely to initiate negotiations if a man was evaluating them and were more likely to be penalized than a man if they did attempt to negotiate. The limitation of this study is that all material was scripted (men and women recited identical lines), which leaves the door open to explore different types of negotiating styles to determine if women may be more successful with some alternate approach.
Gender differences in estimated salary projections
Is it possible that this $16,000 is simply the result of men innocently over-inflating their anticipated salary line? There is some precedence for this type of phenomena as nicely reported in Kolata’s 2007 New York Times article titled, “The Myth, the Math, the Sex.” Kolata discussed the High School Prom Theorem that examines logically implausible studies that purport significant gender-based differences. Using the example of a study in which men reported almost double the number of sexual partners as women, she concluded that the reported differences were very likely rooted in a slight, actual gender difference but that these differences were then grossly inflated by conscious and subconscious social memes, leading to over-reporting by men and possibly under-reporting by women. In support of this theory, Lo Sasso’s study did not actually double check what the residents were going to be paid. So perhaps there were a few chest puffing men and overly modest women.
Confounding missing variables?
Perhaps women choose jobs for perks other than a hefty paycheck. Considerations such as collegial environment, anticipated workload, degree of non-clinical obligations and long-term schedule flexibility may play an important role. I think this is what we would all like to believe, and hopefully, we are right. However, we still need to study the data. Just as patients can be misdiagnosed with a panic attack when they really have SVT, we should be a little wary of writing off these very real salary differences as simply “lifestyle choices” without further investigation.
April 12, 2011 is Equal Pay Day. Think about it.