For decades, the conventional wisdom among economists was that raising the minimum wage increases unemployment. On its face, the argument seemed reasonable enough: no employer will hire a worker at $6/hour if the worker only brings in $5.35/hour in revenue, so raising the minimum wage might cause some companies to shed jobs. A spate of empirical studies seemed to back up the theory, helping to kill any upward momentum on the minimum wage.
But recent research (summarized in this Slate article by Steven Landsburg) suggests that much of the previous research was wrong, and that the minimum wage has almost no effect on unemployment levels in the real world. Put simply…
the power of the minimum wage to kill jobs has been greatly overestimated. Nowadays, most labor economists will tell you that that minimum wages have at most a tiny impact on employment.
It’s worth noting that the author of this article actually opposes increasing the minimum wage, on the grounds that it places too much of the burden of higher pay on employers, and not enough on society at large. But other economists (including my favorite blogger on economic topics, Brad DeLong), say that even that contention is flawed: the burden of minimum wage increases fall on the customers of companies that employ minimum wage labor, which includes most of us. So raising the minimum wage is like a small tax on consumers that’s funnelled directly to the people at the bottom of the income ladder.
So all this leads me to believe that increasing the minimum wage would be an unalloyed good, rather than a mixed blessing.
Find this article interesting? Please consider making a gift to support our work.
One more plug for the Slate article: it provides a fascinating look at how subtle and unconscious biases can have an enormous impact on academic research.
Apparently, the bulk of research published on the topic confirmed the standard theory, i.e., that increases in minimum wage cause unemployment. But the more studies there were, the more troubled economists became: the bigger studies proved to be no more statistically significant than the smaller ones. This is definitely not what one would expect. If a statistical correlation is valid, studies with large sample sizes will show the relationships much more clearly than studies with small sample sizes.
So researchers looking at the body of published results concluded that there was a pervasive bias in play—empirical studies that confirmed economic theory were written up and published, but those that didn’t confirm the theory never saw the light of day. The result was a corpus of academic research that came to the wrong conclusion, simply because the correct results didn’t seem interesting enough to be worth publishing.
Apparently, when enough great minds think alike, they can all be wrong together.