MOST MINIMUM WAGE WORKERS ARE NOT WORKING POOR. Most working poor earn more than minimum wage. That's something I learned working through a paper I had to comment on at the Western Economic Association conference in San Diego. (Yes, there were responsibilities that intruded on the sightseeing.) There's a summary of what we know about these non-overlapping subsets of the work force at Betsy's Page that I recommend you read through (with the cross-references. Disregard the polemics.)

At about the time I was packing to leave, there was a bit of posting about the economics of the minimum wage at Marginal Revolution and Cafe Hayek, much of which focused on ideologies when it wasn't focusing on econometric niceties. Cafe Hayek's Don Boudreaux points to a column he wrote for Tech Central Station that suggests the econometric niceties might be, well, not so nice.

First, a higher proportion of empirical research in the social sciences is subject to legitimate -- oftentimes irresolvable -- dispute. Second, as a consequence, in the social sciences theoretical considerations inevitably play a larger role in navigating around these disputes and in forming judgments about desirable public policies.

And so it is with the minimum wage. Almost any empirical study of this government mandate can be challenged for ignoring this variable, for mis-identifying that variable, for focusing on an inappropriate time period, or for countless other possible errors. Therefore, following Tyler Cowen, we are justified in being extraordinarily skeptical of empirical findings that are inconsistent with widely accepted theoretical foundations.

Minimum-wage studies that find no negative effects on the employment prospects of low-skilled workers are whoppingly inconsistent with basic economics -- a fact that means that they are probably inaccurate.

On the other hand, there might be room for further research. (There's always room for further research. There is no shortage of interesting questions in economics. Tractable in a way that survives scrutiny is another matter.) Take that "inconsistent with widely accepted theoretical foundations." That could mean "the four or five technically proficient theorems that satisfied referees at American Economic Review and Econometrica." What's that John McGee line about "never heard of Williamson and Scherer and simply blunders in" with respect to strategic behavior? Perhaps agents are acting under different constraints or facing incentives that haven't been as well thought through yet. (High occupancy toll lanes exist despite any number of theorems about the optimality of a single price for all users of a congestible facility.) That "no negative effects" merits further review as well. The effect of a minimum wage on hours worked might be "negative but not statistically significant," a phrase that often substitutes for thought. (On the other hand, we may have more precise estimating methods as part of the quest for theoretically pure data that yields the right results.) But does that phrase suggest that fewer people are working, yet total compensation to minimum wage workers has increased? Room for further research, indeed.