Minimum Wage Arguments

I recently came across this excellent piece on the minimum wage. Actually, it is more generally a discussion of the so-called bargaining power of big businesses:

Indeed, despite being cited by the CEA report, the previous Furman research, coauthored by Peter Orszag, stands in contrast to the basic monopsony theory. In the previous piece, they argue that lack of competition is giving some firms economic rents, and this is boosting pay at those firms by so much that it significantly contributes to inequality. So, are big firms bargaining down wages for their workers or are they generating massive rents that are boosting the pay of their workers? The reconciliation between these two theories is incomplete at best.

Second, if big firms are bargaining down wages then why do labor economists consistently find a large firm wage premium? To take one example from many, one recent study on retailers found that after controlling for individual and store characteristics, firms with at least 1,000 employees pay 9% to 11% more than those employing 10 or fewer.

Third, if firms’ bargaining power over their employees is growing, then why are they increasingly contracting out for work? Lawrence Katz and Alan Krueger argue that from 2005 to 2015, the share of workers hired out through contract companies grew from 0.6% to 3.1%. A company with labor market power wouldn’t want to contract out work to another company. They’d want to hire workers directly to take advantage of that power.

Fourth, the CEA report points to the minimum wage literature as evidence of monopsony power. Leaving aside the debate over whether the minimum wage reduces employment (I say yes, the report says no) the literature clearly shows that the minimum wage increases prices. As Daniel Aaronson and Eric French have pointed out, the monopsony model implies that the minimum wage should increase employment and output, thereby decreasing prices. That prices rise is inconsistent with the monopsony model.

It reminded me of something I hate about the minimum wage debate: it is so one-dimensional. Even the "sophisticated" version of it, made by economists who favor the minimum wage, goes something like, "Large firms have monopsony (single-buyer) power and therefore can bid down wages below what is economically efficient." If this is correct, then a minimum wage might be justified. (Still, only under certain conditions.) So they create some theoretical model of employers having monopsony power, or they empirically measure something that sounds like market power (industry share, company size, capital share of gdp) and insist that it suggests monopsony power. Combine this with the mixed empirical record of minimum wages on low-skilled unemployment, and you have a plausible "market power" story. This argument seems to be obsessed only with the question of whether increasing the minimum wage does or does not have a measurable effect on unemployment, as if that were the only relevant point. Opponents of the minimum wage say, "You are putting people out of work." Proponents of the minimum wage say, "No, look at these studies. (Ahem, ignore those other studies over there for the moment.) This pared-down list of studies shows no effect on employment. Therefore there's no reason not to." We can't just point out a few pieces of evidence that are consistent with the "monopsony" story. This story makes a lot of predictions, and the Adam Ozimek piece above points out that many of these predictions are failures. Much, much more in that same vein, Don Boudreaux (who has been absolutely dogged in this fight) quotes an excellent comment by an anonymous economist. Basically, the point here is, "Market power? Monopsony you say? Great. Go explain all this other stuff, too."

I want to take this "Prove to me that there's a disemployment effect" line and turn it on its head. I've made this argument before in a previous post. It's not on us minimum wage skeptics to "prove" that minimum wages have all the harmful effects predicted by economic theory. The onus is on the minimum wage proponents to demonstrate empirically that raising the minimum wage has benefits. That's what you are supposed to do for any kind of treatment, be it medicine or policy: prove that it works. Prove that it's safe. See the figure from my previous post. I pulled that table from Minimum Wages by David Neumark and William L. Wascher. Sure, you get a mixed empirical record on the unemployment question. (Anyone who says there is "no evidence" of an unemployment effect is distorting the empirical record pretty badly. There is evidence of sizable effects, evidence of small effects, evidence of no effect, and evidence of small positive effects. Some take this mixed record to mean the unemployment effect doesn't exist at all, which arbitrarily (selectively?) edits out most of the literature.) But it's not like you get empirical validation of the benefits suggested by minimum wage proponents. You don't see a reduction in poverty, or a more equitable income distribution (whatever that might mean). I think it's not sufficient for minimum wage proponents to argue that the negative side-effects don't materialize. If they are so dedicated to empirical validation, they should be proving that the benefits actually do materialize.