The Wall Street Journal seems to be mistaken about economists' pay

Yesterday the Wall Street Journal ran a story on p. B1, "Wage Winners and Losers", that contained an amazing chart (link may require a subscription to the Journal). The chart showed the "average annual wages" for twenty-two different professions. The highest paid profession in the chart, higher than physicians, higher than airline pilots, and much higher than lawyers and judges, architects, natural scientists, and computer programmers was . . . "economics teachers". The chart shows us earning a bit more than halfway between $125,000 and $150,000 per year, or $137,500+.

Can this be right?

Apparently not. The source for the numbers in the chart, the Bureau of Labor Statistics, takes the annual salary of workers and divides by the reported number of hours worked per year to compute median hourly earnings. Since many--I believe, most--college professors have summers off the clock, the number of hours worked for economics professors is estimated by the BLS at 1558. The BLS computes the median hourly earnings for economists at $66.45.

The Journal, however, computes the annual salaries in the chart "by multiplying average hourly wages by average number of hours worked per week for workers in those occupations, assuming a standard 52-week year." Meaning, apparently, that the median hourly earnings for economists was multiplied by 52 times 40 = 2080. (That yields an annual salary of $138,261, a bit more than $137,500, and consistent with the chart.) Clearly, this is incorrect: the Journal should have computed the annual salary for economics professors as $66.45 times 1558 = $103,529.

But even this corrected figure seems too high. Does the median economics professor actually earn six figures (fringe benefits and consulting or other outside income are not included)? One of my colleagues, Doug Pearce, suggested that perhaps highly-paid finance professors were being grouped with economics teachers. So I tried to find out how the BLS defines "economics teachers". A quick look didn't turn up a definition, but did reveal a list of the occupational categories BLS uses. In that list, under "Teachers, College and University", we find, in order: Psychology Teachers, Economics Teachers, History Teachers, Political Science Teachers, Sociology Teachers, and Social Science Teachers, Not Elsewhere Classified. The latter category indicates that economics teachers are grouped with the social sciences; I think that makes it unlikely that finance teachers are grouped with us, but I don't know for sure. (One of the other categories is "Business, Commerce, and Marketing Teachers". I would think finance professors are categorized there.)

If "economics teachers" are just economics teachers in the BLS data, what explains the $103K annual salary? I don't know. But there are two other puzzling results in the BLS data. For almost all the categories of college teachers, median hourly earnings are less than mean hourly earnings, what I would expect. But for economics teachers the reverse is true: the median is $66.45 but the mean is $63.98. Even more surprising is the time series of hourly earnings for "economics teachers, postsecondary". Between 1997 and 1999, our hourly wage falls by 20%. (If that link doesn't work, try this one.) And bizarrely, between 1999 and 2003 our hourly wage rises by more than 50%!

I admit that I've had no experience with these data and maybe I'm making stupid mistakes in using them. But for now I'll assert that it doesn't look like they are at all reasonable in describing what economics professors actually earn.

Your tax dollars at work.