I have three main concerns:
1. The CEA post does nothing to distinguish between the short-run and long-run effects of the minimum wage (or of low-skill wage levels more generally, a topic oddly absent from most minimum wage discussions). The famous Card and Krueger paper estimates a very short-run effect. It is not at all surprising, if you think that firms are forward-looking and that most of the action revolves around the substitution of capital for labor, that the short-run effect would be zero or very small. Substantively minor short run effects are quite consistent with substantively large long-run effects. Both the short and the long run matter for policy; one might think the long run should matter more. The failure of the CEA post to mention the distinction is disappointing. I fear that the post has run afoul of one of the most important (and most popular) false folk theorems of empirical economics: "Quantities that are hard to credibly estimate must equal zero".
2. Near the end of the post we find this bit:
Overall the logic for the finding that raising the minimum wage does not result in large adverse impacts on employment is that paying workers a better wage can improve productivity and thereby reduce unit labor costs. These adjustments, along with others that firms can make, help explain why the increase in the minimum wage need not lead to a reduction in employment. Higher wages lead to lower turnover, reducing the amount employers must spend recruiting and training new employees. Paying workers more can also improve motivation, morale, focus, and health, all of which can make workers more productive. In addition, by reducing absenteeism, higher wages can increase the productivity of coworkers who depend on each other or work in teams.Perhaps I misunderstand, but I read this as saying that firms throughout the economy are leaving huge amounts of money on the table by not setting their wages correctly. Really?
3. In my view, the post puts too much weight on the Booth School survey of economists about the minimum wage. I have two concerns here. First, the questions in the Booth survey are not worded as well as they could be, especially the first one. In particular, in addition to a sort of general vagueness (e.g. what does "noticeably" mean), they do not distinguish between the short-run and the long-run effects of the policy. For this reason alone the responses should be discounted. Second, though the respondents are all indeed "experts" at something as described in the CEA post, they are not all experts on the minimum wage or even on labor economics. Many of them are experts in quite different bits of economics. As I read it, the CEA post gives the impression that the poll reflects mainly those familiar with the economics and evidence of the minimum wage, when in fact many poll respondents are theorists, macro-economists and so on and so are quite unlikely to be familiar with the relevant literature in any detail.
In closing, as always, I promise to meditate on the substantive unimportance of low-skill wages (and labor costs more generally) to the organization of production (and thus to the quantity of low-skill labor demanded) next time I fill up my own soft drink in a restaurant, or fill my own gas tank at a gas station (yes, I am old enough to remember when they had workers who did this for you), or check myself out at the drug store, or wander the long aisles of a discount retailer trying to find an employee, or tap my order into a screen at a restaurant, or scan my own receipts for reimbursement, etc.
Full disclosure / small world of economics: Betsey Stevenson of the CEA is my Ford School colleague at Michigan. Based on the positive impression I have of her as an economist via repeated interactions in academic contexts I assign responsibility for the defects of the CEA post to some combination of her co-poster and her political minders in the administration.