Friday, April 15, 2016

Democratic debate last night, and the $15 minimum wage

I watched some of the Democratic debate last night, which seemed to have imbibed just a bit of the Republican debates' spirit (with shouting and interrupting), albeit conducted on a more substantive and less fantastical (in the sense of fantasies and falsehoods) plane.

One topic of interest to me was the minimum wage, which I've written about.  In my 1997 U of Chicago Law Review piece, The Minimum Wage, the Earned Income Tax Credit, and Optimal Subsidy Policy, I addressed the two alternatives noted in the title from a public economics tax/transfer standpoint, and also in light of the then-still-recent research by David Card and Alan Krueger that found little or no disemployment effects from modest increases to the minimum wage.

From a straight public economics standpoint, the minimum wage has an odd design.  One can think of it as follows.  Suppose we are looking at a $10/hour minimum wage.  That's equivalent to the case where there was no minimum wage, but when the market wage was less than this, imposing a tax and a transfer.  Suppose that, in the absence of the minimum wage, I would have worked an hour at McDonald's for $8.  Leaving aside administrative issues, the actual order of cash flows, etc., the min wage is equivalent to allowing that transaction to occur, but also having the government pay me a $2 transfer, financed by a $2 tax on the McDonald's franchise owner.

Two notable things about the transfer.  First, it only goes to low-hourly-wage who actually are employed, not those who would like a job but can't get one (or don't like this one enough to take it).  In that respect, it is just like the earned income tax credit (EITC).

Second, unlike the EITC, it identifies recipients based purely on the difference between the hypothetical market wage they would otherwise have gotten, and the mandated minimum hourly wage.  The EITC, by contrast, looks at annual earnings and income, as well as at household information (e.g., whether one has 0, 1, or at least 2 dependent children).  So the EITC is better targeted in terms of people who are actually poor - as distinct from, say, teenagers from affluent households.

Let's now look at the financing.  The EITC is financed out of general revenues, whereas the minimum wage is "collected" from low-wage employers.  Now, the EITC's funding design certainly sounds better, although of course there are optical advantages (really, from the structure of the spending side) to the minimum wage setup.

The "tax" on low-wage employers, from the minimum wage's funding design, presumably is not borne by those parties at equilibrium - one would expect to be passed on somehow, e.g., through effects on consumer prices and low-wage employment levels, as well as through allocative shifts in the business sector.

Standard price theory would, of course, suggest that the employment level must drop.  Low-wage workers might still collectively gain - in effect, they have been cartelized to demand at least the minimum wage - but their increased net wages as a group (if the price was not set too high) wouldn't be evenly shared, if some lost their jobs.

But here's where the Card-Krueger research kicks in.  It showed that modest minimum wage increases might not reduce employment levels.  The core reason, to put it very broadly, is that labor markets are different from, say, boring commodity markets with fungible items.  For example, labor supply may respond to perceived wage fairness.  And on the demand side, low-wage employers are making various choices, e.g., should we pay less and get worse workers who also quit a lot, or should we pay a bit more and get slightly better workers (not otherwise distinguishable to the employer) who also quit less.  Even aside from the quality trade-off, there is that between (a) paying lower wages but having to scramble more to replace and at least minimally train new workers, and (b) paying higher wages but reducing those problems.

So, from that last factor alone, one could luck into increased, rather than reduced, employment (measured by total hours) by finding the sweet spot in which there is enough employer switching from the high-turnover to the low-turnover strategy.

That makes the min wage's "funding instrument" potentially appealing in a way that standard price theory as applied to tax would miss.  But there can be no doubt that, at some point, the neoclassical view that minimum wages will reduce employment, will kick in.  After all, suppose we made the min wage $50 per hour, indexed to inflation.  That is unlikely to turn out well.

OK, back to Bernie and Hillary.  I realize one needs to score dramatic political points, when one is trailing in the late stages of a contested primary campaign, but I was nonetheless made uneasy by Bernie's flag-waving about the national $15 minimum wage.  I am not an expert on this, but I strongly suspect that, for much of the country, a minimum wage at that level is just too high.

Indeed, I don't know enough to deny that it might be too high everywhere - but at least high-price-level urban centers, such as New York City, Los Angeles, and Seattle, have a better chance of finding the sweet spot rather than proving too high.  Nonetheless, we get Bernie bludgeoning away over the $15 minimum wage, treating it as a matter of principle that Hillary is simply too craven and right-wing to accept unless she's forced to.

I'd like to think that Bernie is just grandstanding for the TV audience and in truth knows that $15 might be too high.  Better a bit of cant to help in the election process than genuine closed-mindedness to possible contrary evidence.  But I have no grounds for believing that he knows better.  He seems, rather, to think that a $15 national minimum wage is just a matter of principle, with no further analysis being needed.

If that's the right way to look at it, I was saying to myself during this part of the debate, why stop at $15?  Why not $50 or $100 per hour?

This gives me a bit of sympathy with the following Paul Krugman comment in his blog today:

"What you see ... on multiple issues, is the casual adoption, with no visible effort to check the premises, of a story line that sounds good .... In each case the story runs into big trouble if you do a bit of homework; if not completely wrong, it needs a lot of qualification.... It's about an attitude, the sense that righteousness excuses you from the need for hard thinking."

Krugman could be kinder to one of the reasons why Sanders and his supporters are inclined to think this way - money talks big in politics, certainly including to members of the Democratic establishment (and not just Hillary).  So genuine good faith commitment to a set of objectives is truly an issue here, and people have reason to be looking for evidence of it.  But one has to be able to think dispassionately about the actual effects that high-minded, well-intentioned policies might have.  Otherwise, one is selfishly treating the opportunity to feel virtuous as if it were more important than the welfare of the individuals whom one ostensibly wants to help.

1 comment:

DM Hasen said...

One probably needs, as well, to consider the interaction of current, low wages with various government benefit programs that act as wage supplements. For example, there is evidence that employers adjust wages downward in light of the availability of the EITC, food stamps, and similar programs. It seems clear that the incidence of these programs is not intended to be on employers, suggesting a "minimum" minimum wage is needed to clear some threshold that ensures that government programs have intended effects - at least as a first-order matter.