Raising the Minimum Wage Will Definitely Cost Jobs

As I show in a recent extensive survey of research on minimum wages and job loss in the U.S., this is simply not true. Most studies find that a minimum wage reduces employment of low-skilled workers, especially the lowest earners most directly affected by raising the minimum wage.

There are conflicting individual studies of the effects of minimum wages on employment. That there is disagreement shouldn’t be surprising. Economics is a social science, not a natural one. Studies of minimum wages and job loss are not laboratory experiments. They can’t be replicated and so can’t be expected to yield exactly the same results.

What’s surprising, though, is that summaries of the research literature make contradictory claims about what the overall body of evidence says. Distinguished economists like

Angus Deaton

and

Peter Diamond

signed the EPI letter asserting that the research shows little or no evidence of job loss, whereas others look at the research and conclude that it points to job loss. How can that be? Who is right?

Most economists have a strong stance on the minimum wage one way or the other. Perhaps this colors how they look at and interpret the evidence. Or perhaps there are so many studies of the employment effects of minimum wages that it is difficult to keep a “scorecard” of what the overall body of evidence says.

To provide an accurate reading of the research,

Peter Shirley

and I surveyed the authors of nearly all U.S. studies estimating the effects of minimum wages on employment published in the past 30 years. We asked them to report to us their best estimate of the employment effect, measured as the “elasticity,” or the percent change in employment for each 1% change in the minimum wage. Most authors responded, and in the few cases in which they did not, we pulled this estimate from their study.

The results are stark. Across all studies, 79% report that minimum wages reduced employment. In 46% of studies the negative effect was statistically significant. In contrast, only 21% of studies found small positive effects of minimum wages on employment, and in only a minuscule percentage (4%) was the evidence statistically significant. A simplistic but useful calculation shows that the odds of nearly 80% of studies finding negative employment effects if the true effect is zero is less than one in a million.

Across all the studies, the average employment elasticity is about minus-0.15, which means, for example, that a 10% increase in the minimum wage reduces employment of the low-skilled by 1.5%. Extrapolating this to a $15 minimum wage, this 107% increase in the states where the federal minimum wage of $7.25 now prevails would imply a 16% decline in low-skilled employment (broadly consistent with the recent CBO study). That sounds like a substantial job loss.

It’s true that some workers would experience higher incomes, and that, on net, incomes of low-wage workers would probably rise. But that doesn’t mean the minimum wage is the best way to help low-wage workers or low-income families, as research clearly demonstrates that a large share of income gains from a higher minimum wage flows to families with higher incomes. An alternative policy—the Earned Income Tax Credit—targets benefits to lower-income families far more effectively, is proven to reduce poverty, and creates rather than destroys jobs.

Our survey finds other important results. First, contrary to what is sometimes claimed, there is no tendency for the most recent research to provide less evidence of job loss. Second, the sharper a study’s focus on workers directly affected by the minimum wage, the stronger the evidence of job loss. For example, the average employment elasticity for those with at most a high school education is minus-0.24, implying that a 10% increase in the minimum wage reduces their employment by 2.4%. The only studies that produce more mixed evidence are studies of low-wage industries, like retail or restaurants. Notably, in these studies the job loss among those most affected by the minimum wage may be masked by employers substituting from lower-skilled to higher-skilled workers.

True, some studies don’t find evidence of job loss. But advocates for a higher minimum wage can claim support from the overall body of research evidence only if they discard most of that evidence. The consensus of economic research on the effects of minimum wages points clearly to job loss, and policy makers should consider this job loss in weighing the potential costs and benefits of a sharp increase in the minimum wage.

Mr. Neumark is a professor of economics and co-director of the Center for Population, Inequality, and Policy at the University of California, Irvine.