Mike Konczal is alarmed to find the following:
I’m not going to control for all the important things that are necessary to do this right – education, age, gender, race, etc. Just looking at the raw numbers: The 2009 average salary across the averages for all districts is $48,267. The Census Bureau’s Median Household Income by State – Single-Year Estimates has the medium household income at $51,237. A random household is going to take home a larger salary than a teacher in Wisconsin.
Does the median household in Wisconsin have only one wage-earner? That is entirely possible. But it does seem as though a more useful comparison would be between the average salary for a teacher and the median worker.
1. Salary information skews high, so the average (mean) is higher than the median. The average teacher’s salary is lower than the median household in Wisconsin. Since the average household income in Wisconsin is going to be even higher than the median, this is a very low-ball estimate. (Can we get average household income by state?)
The more you control for things, the sharper this distinction will get. According to the Census, only around 22.4% of people in Wisconsin have a college degree. Since college education is expected for teachers, we aren’t even comparing the same education bucket and they are already behind.
Mike is very confident that this distinction will get sharper. But should we assume that all college-educated workers are interchangeable, and that they ought to earn 1.5x or 2.1x what a non-college-educated worker earns?
2. If you take being “middle-class” as having a median wage, then being a teacher is not a middle-class job as it stands. A household with a teacher will need to put additional labor into the marketplace in order to have a median amount of wages.
To know if that’s true or not, we’d need to look at the median amount of wages for individuals. Fortunately, there is a wonderful tool called The Measure of America that can help us with that, which a friend of mine helped design for the Social Science Research Council.
As of 2009, median earnings for an individual in Wisconsin were $29,167. If two individuals earning the median amount of wages were to form a household, they would have a combined household income of $58,334. If two teachers were to form a household, they would have a combined household income of $96,534.
But what about the median for college-educated workers? The National Center for Education Statistics tells us that the median annual earnings for full-time, full-year wage and salary male workers in the United States as of 2008 was $55,000. It was $45,000 for female workers. Multiply the male number by two adults and you get $110,000, which is indeed higher than the two-teacher household. If it is a man and a woman forming a household together, that number would be $100,000. Of course, these are median annual earnings for full-time, full-year wage and salary workers in the United States, not Wisconsin.
And Wisconsin has a somewhat lower cost of living than other regions in the U.S. Wisconsin has a regional CPI of 168.4 vs. a U.S. city average of 173.8. That is not a tremendous difference, but consider that college-educated workers are not evenly distributed across U.S. cities. Rather, they are heavily concentrated in high-cost metropolitan areas, for a variety of reasons. A salary of $96,534 in Racine or Janesville, Wisconsin might be a handsome income in contrast to $110,000 or $100,000 in Queens, New York.
None of this is dispositive. But it does seem important to keep in mind. I don’t get the sense that college-educated workers in the private sector at the median in Wisconsin are earning wages that are much higher than teachers. As far as I can tell, Mike’s data source is not correcting for the duration of the school year, but it is entirely possible that I’ve missed something. (It’s worth noting that at least some teachers like having unstructured time during the summer months, and one can imagine a universe in which one might choose more leisure at the expense of a somewhat lower hourly wage.) I don’t think Wisconsin’s Department of Public Instruction intended for its data to be used for a comparison with private sector college educated workers. Keefe does correct for hours.