Correspondence

Earnings of Male and Female Physicians

N Engl J Med 1996; 335:524-525August 15, 1996

Article

To the Editor:

As members of a two-physician family we were very interested in Baker's article (April 11 issue)1 on differences in earnings between male and female physicians. We are curious about whether Baker has any data on two potentially confounding covariables. First, like many of our colleagues in two-income families, we have elected to work fewer hours in less lucrative positions. Having two incomes makes this possible; the lack of a full-time homemaker makes it essential. Baker's data show that men's annual income increases when they marry and have children, but these factors are associated with a decrease in income for women. Could this represent an increasing likelihood of a stay-at-home spouse for men but not for women across these categories? Second, the fixed nature of medical overhead — particularly malpractice insurance and office expenses, but to some degree even personnel costs — makes the marginal value of extra hours worked high. Men work more hours than women; does this explain part of the difference in hourly income?

Mark Hauswald, M.D.
University of New Mexico, Albuquerque, NM 87131-5246

Nancy L. Kerr, M.D.
4273 Montgomery NE, Suite 200E, Albuquerque, NM 87109

1 References

1. 1

   Baker LC. Differences in earnings between male and female physicians. N Engl J Med 1996;334:960-964
   Full Text | Web of Science | Medline

To the Editor:

I believe that Baker is mistaken in partly basing his analysis on the number of hours worked per physician. I worked in a setting in the early 1990s in which I saw substantially more patients per hour than some of my female colleagues. An outsider studying our pay would discover that these women and I earned the same amount per hour and might conclude that the situation seemed fair. However, if the investigator also looked at the number of patients seen per unit of time or at revenues generated per month, he or she would be forced to conclude that I was being underpaid for my work. Some observers might even wonder about reverse discrimination.

Compensation includes not only direct pay but also benefits, and I have observed substantial reverse discrimination regarding benefits for family,1 medical,2 and parental3 leave.

Baker's study would have been improved by an objective measure of physicians' productivity and by including benefits in their assessment of compensation.

Brian J. Bohlmann, M.D.
5900 Monona Dr., Suite 205, Madison, WI 53716-3960

3 References

1. 1

   Bohlmann BJ. Family leave policies neglect needs of single, childless doctors. American College of Physicians' Observer. April 1994:10.
   

2. 2

   Bohlmann BJ. Covering for impaired residents. Ann Intern Med 1992;117:267-267
   Web of Science | Medline

3. 3

   Bohlmann BJ. Pregnant surgical residents: Oh K?! JAMA 1991;266:2222-2223
   CrossRef | Web of Science | Medline

Author/Editor Response

Dr. Baker replies:

To the Editor: Bohlmann notes the importance of considering productivity in a study of earnings. Productivity is important, but I do not believe that my examination of hourly income was misleading, as he suggests. When I included the number of patients seen per hour as a control variable, the results were qualitatively unchanged. Furthermore, although Bohlmann's experience implies that male physicians may be more productive than female physicians and thus that a comparison of income generated per hour may undervalue male physicians' work efforts, this is not true in general. According to the 1991 Survey of Young Physicians (the base sample for my study),1 women saw an average of 1.8 patients per hour over the course of a week, whereas men saw 1.7.

Bohlmann also suggests the need to consider fringe benefits. I agree that incorporating benefits into the analysis would have been useful, but detailed data were not available. His implication that benefits are an important source of discrimination favoring women is not convincingly supported. The articles he cites fail to present evidence of a systematic sex bias in the allocation of benefits. Moreover, there are many other benefits besides family, medical, and parental leave (such as pensions, other retirement plans, profit-sharing plans, deferred-compensation plans, and vacations) that may or may not be subject to sex bias in either direction, but about which we lack systematic information.

Hauswald and Kerr speculate that the difference in the effect of family responsibilities on the earnings of men and women results from a greater prevalence of stay-at-home spouses among male physicians. I find this quite plausible but could not investigate it directly since specific data on stay-at-home spouses were unavailable. The full set of adjustment variables did include controls for marriage and children.

Hauswald and Kerr also advance the hypothesis that the marginal returns with respect to the number of hours worked are increasing and argue that the earnings of male physicians may be higher because, by working more hours, they more effectively exploit the increasing returns. This is also a sensible hypothesis, but it is difficult to evaluate since there is little information on marginal returns with respect to hours worked. To briefly investigate this issue, I reexamined my data for a nonlinear relation between income and hours worked. If there are significant increasing returns, income should increase more quickly for hours with higher levels of work effort. However, there was no evidence that this occurred.

Laurence C. Baker, Ph.D.
Stanford University School of Medicine, Stanford, CA 94305-5092

1 References

1. 1

   Cantor JC, Baker LC, Hughes RG. Preparedness for practice: young physicians' views of their professional education. JAMA 1993;270:1035-1040
   CrossRef | Web of Science | Medline