Update 1/11/07:First-rate article on this subject by Timothy Egan in the New York Times. It’s easy to see how he won the National Book Award.

***

There’s been a lot of talk recently about the 110th Congress raising the federal minimum wage. One thing you hear a lot is that raising the minimum wage leads to higher unemployment. (Here’s one risible example; and here it is filleted by David Goldstein.)

But is this canard actually true? I’m skeptical—here’s why.

First off, in deference the economists out there, I should mention that there has been any number of academic papers arguing this question. But one thing that I’ve never seen—although I don’t follow the debate closely—is an argument with recourse to a fairly obvious natural experiment: What’s happening in each of the 50 states?

In this corner, we have 21 states who share the federal minimum wage of $5.15, the lowest allowable wage. Among these states, some have the same wage as the feds, some no state minimum wage, and one, Kansas, has a lower wage. But federal law trumps.

In the other corner, there are 30 states (counting DC), that have a minimum wage higher than $5.15. There’s not much uniformity among these states’ wages and the details vary considerably. Still, we have something like a rough comparison between the stingy states and the generous states.

So what’s the difference in the unemployment rates of stingy and generous states? Exceedingly little.

    • Of the 21 states with the lowest allowable wage, 9 of them (or 43 percent) have an unemployment rate higher than the national average of 5.1.
    • On the other hand, of the 30 states with higher minimum wages, just 10 (or 33 percent) have an unemployment rate above the national average.

    For a more accurate picture, I calculated the aggregate unemployment rate for the stingy states versus the generous states (basically, using each state’s labor force as a weighting factor for its unemployment rate). The difference is negligible.

    The stingy states have a combined unemployment rate of 5.16 percent. Meanwhile, the generous states do almost imperceptibly better, with a combined unemployment rate of 5.07 percent.

    Now, there may still be any number of reasons to believe one way or the other based on econometric models or academic theories. But the results on the ground suggest that the minimum wage does not make much of a difference to the unemployment rate. If anything, the state by state experience hints at the opposite conclusion. So maybe we can start retiring this old saw?

    Note: All unemployment figures in this post refer to 2005, the last year for which we have complete data.

    Update: Something about this post was bugging me on the bus ride home last night. I finally realized what it was: in late 2005 Hurricane Katrina pretty severely screwed up the economies of Louisiana and Mississippi, both of which are “stingy” states that provide only the lowest allowable wage.

    So this morning I recalculated things without those two. Here’s what I found:

    • 7 of the 19 stingy states (37 percent) had an unemployment rate above the national average (as compared to 33 percent of the generous states).
    • The aggregate unemployment rate of stingy states falls to 4.98 percent (as compared to 5.07 percent for the generous states.

    My reading of the numbers? Still too close to call—and that’s basically my point. Recent state by state data does not lend support to the oft-repeated claim that a higher minimum wage is tantamount to higher unemployment.