Part 2: Maximum obtuseness

Evidence does not support a minimum wage increase.

Derek Olson

Last week, I discussed the fundamental problem with raising the minimum wage: It increases unemployment. In December, our duplicitous President Barack Obama said, “There’s no solid evidence that the minimum wage costs jobs.” But after his recent “If you like your health care plan, you can keep it” line, many people recognized that the president is a liar.

David Neumark and William Wascher, economists from the University of California at Irvine and the Federal Reserve Board, respectively, conducted a survey of more than 100 studies on the minimum wage. They report a sizeable majority — 85 percent — of credible studies support the theoretical prediction of the disemployment effect, or that raising the minimum wage increases unemployment.

Additionally, they offer methodological critiques of those studies that did not find such an effect. It’s worth noting this alone ought to be quite convincing. You wouldn’t go to 10 doctors, and after eight of them diagnosis cancer, trust the one or two who say there’s no evidence you’re sick.           

It’s important to understand that data in the social sciences does not match the precision of data in the hard sciences. Economic data is more easily subject to error and incomplete information. Even the best empirical work cannot revolutionize basic economic theory. When a multitude of economic factors are in play, high-powered statistics are necessary to comb through imperfect data and isolate the effects of a single policy. We should not be surprised if a small portion of studies fail to uncover those effects.

Sophisticated defenders of the minimum wage almost invariably jump to the work of David Card and Alan Krueger, two economists who studied employment at fast food restaurants in New Jersey and Pennsylvania after New Jersey raised its minimum wage but Pennsylvania did not.

Card and Krueger found no decrease in employment due to the law change; however, there is a potentially catastrophic flaw with this and other similar studies. As is common, the legislation did not take effect until a year later. Rather than waiting until the last possible day, businesses could have adjusted their hiring and attrition leading up to the new law. Theoretically, when the law took effect, there would be zero layoffs. The disemployment effect could be entirely invisible yet completely present.

In addition to his work on the minimum wage, Card has also studied the effect of low-skilled immigration on native wage rates. He found that large increases in the supply of low-skilled labor did not have a large effect on wage rates, which suggests a near-horizontal curve. However, his study of the minimum wage suggests a near-vertical demand curve. With opposite results, one of these conclusions must be wrong.

In economics, empirical results demand theoretical explanation. It takes wildly implausible economic theory to believe the minimum wage has no adverse effects. It requires believing that the law of demand is somehow suspended for low-skilled labor markets. In the debate over the disemployment effect, the burden of proof should be on those attempting to contest the law of demand in favor of an absurdly abstract theory.

Fortunately, the weight of the evidence demonstrates that Economics 101 prevails. Wage floors destroy jobs for precisely those individuals lawmakers designed it to help.

In my last part of this series next week, I will explore alternatives to minimum wage.