Minimum Wage Arguments

So a friend sent me a few pieces on the minimum wage, arguing from a conservative perspective that we shouldn't have one at all, or at least should keep it low.  This is in response to the President's call to raise the minimum wage to $9 an hour, which has been backed on the blogosphere by citations of studies that show (surprisingly) that moderate increases in the minimum apparently don't increase unemployment, as classical economics says it should.  After all, it's pretty basic to argue that if we artificially increase the price of labor, we will necessarily have less demand for that labor, as the "natural" price of it may be lower than the minimum wage.  But the studies linked above show that this effect doesn't seem to bear out in the real world for some reason, and therefore raising the minimum wouldn't really have a downside for the poor.

My friend thus sent me this by a writer named Don Boudreau:

Commenting on my recent open letter to Barack Obama – in which I asked, if government-mandated higher prices for imports discourage the purchase of imports, why do government-mandated higher prices for low-skilled labor not discourage the hiring of low-skilled labor? – John Burger writes that “there is ample evidence that increases in tariffs affect imports of said items,” but less-conclusive evidence that increases in the legislated minimum-wage affect employment rates.
Question: how many studies have been devoted to testing empirically the proposition that higher protective tariffs reduce imports? I’m not asking for studies that explore the question of how much an x-percent hike in protective tariffs lowers imports. Instead, I’m asking for studies that admit as reasonable the proposition that higher protective tariffs might in fact not reduce imports at all (or only by amounts too minuscule to detect). Put differently, I’m asking for pointers to studies whose authors begin by saying something like “Contrary to popular, textbook presumptions that rises in the prices of imports reduce the quantities demanded of imports, let’s look at the actual evidence to test this presumption scientifically. Perhaps it’s untrue.”
I know of no such studies, although perhaps they’re out there.
The evidence, to which Mr. Burger refers, on tariffs’ depressing effects on imports, seems to me to be established mostly from history and from common sense rather than from the sort of narrowly focused and tightly empirically controlled tests that are the stuff of minimum-wage studies. (History suggests to me that the minimum-wage hurts unskilled workers: look at the trend over the past 70 years in teenage unemployment, and especially that of the unemployment of black teenagers in America.)
The fact is that no one seriously doubts – and no one has it in his or her material interest to question – the proposition that higher protective tariffs reduce the quantities of imports demanded by domestic buyers. This straightforward proposition about the effects of government-mandated higher prices on the quantities demanded of imports is simply too obviously true to be the object of much controlled empirical testing. (If this proposition weren’t generally true, much – perhaps all – of the corpus of neoclassical economics would have to be discarded.)
Moreover, if someone did do a test and found that a higher protective tariff imposed by Uncle Sam on, say, fast-frozen french-fries imported from Canada over the years 2003-2007 in fact was followed by more imports of fast-frozen french-fries from Canada during this time span, I doubt that anyone would seize upon that finding as establishing a “new economics” of trade in which modest increases in protective tariffs have either no effect on imports or have even positive effects on imports. Everyone of sense would either question the study’s method or recognize that the ceteris in this particular historical instance wasn’t paribus. A handful of other studies reaching the same empirical result wouldn’t change matters.
...And yet, human labor somehow is exempt from this general attitude. Of all valuable goods and services bought and sold in markets, human labor is one of the few in which many people seriously believe that the law of demand – the proposition that, ceteris paribus, the higher is the cost of acquiring a unit of some given good or service, the fewer will be the units of that good or service sought to be acquired per period of time – does not necessarily apply.
 

I find this amazing- Boudreau seems to be saying that for economic questions to which he finds the answer obviously self-evident, we shouldn't pay attention to empirical studies!

Look, if well-designed studies showed that increasing a tariff modestly for fast-frozen french fries had no effect on imports, that would be very interesting, and maybe it would make economists question some of their bedrock assumptions about their discipline. At least, economists who aren't hacks would have to question things.

When confronted with an empirical study that runs contrary to conventional wisdom, such as the minimum wage studies, the response of economists should be to posit explanations and then test them. To intentionally ignore them seems, well, hackish. I'm ready to believe that those studies are flawed somehow, or leave out an important factor. But I'm also ready to believe that small increases in the minimum wage have only insignificant affects on employment. It's not a hard case to make theoretically: maybe it's just that other factors are much more important in determining employment, and a $2/hr increase in wages is insignificant in comparison.

 

But my friend wasn't done- next he sent me this, by Mark Perry at AEI:

1. Opponents of the minimum wage law generally support no minimum wage, i.e. a minimum wage of $0.00 per hour. To support a minimum wage of $0.00 per hour, the opponents can rely on economic theory, economic logic, scientific thinking, empirical evidence and cost-benefit analysis to support their position, which might be summarized as follows:
Increases in the minimum wage generate certain benefits (higher wages) for some workers, but generate costs (fewer entry-level jobs, fewer hours, fewer benefits, less-on-the-job training, reduced opportunities to acquire work skills, etc.) that outweigh the benefits, making unskilled workers as a group worse off on net from increases in the minimum wage. Further, a minimum wage of $0.00 per hour requires no regulatory mechanism and therefore no enforcement costs.
Bottom Line: A minimum wage of $0.00 is optimal because it generates net benefits to society that are greater than the net benefits of a mandated, artificially high minimum wage.
2. Proponents of the minimum wage law support periodic increases of the minimum wage, e.g. to $9.00 per hour, but never seem to provide any justification or analysis that would support a position that $9.00 per hour is somehow optimal for society. That is, why $9.00 per hour and not $9.25 or $8.75 per hour? Why not $8, $10 or $18 per hour? Why not $90 or $900? In other words, what is special or optimal about $9 per hour that justifies that hourly wage for unskilled workers? What theory, analysis, logic or rationale justifies $9 per hour over all other alternatives?
Bottom Line: If there is no economic theory or logic or cost-benefit analysis that justifies $9.00 per hour as an optimal wage for unskilled, entry-level workers, which seems to be the case, then a $9.00 minimum wage is exposed as being totally arbitrary and random. Unless and until Obama and other proponents of a $9 per hour minimum wage can provide some analysis to show that $9 is optimal and maximizes the net benefits to unskilled workers, then it’s a policy that really can’t be taken seriously. Further, the minimum wage requires a costly regulatory mechanism that administers and enforces the government-mandated wage, which is a cost that needs to be considered.
 

So here we have the opposite argument from the previous one. Whereas Boudreau studiously ignores empirical studies that challenge his assumptions, Perry essentially says that without empirical backup there can be no discussion of any minimum wage above $0.00.

But of course just because empirical information is difficult to come by at numbers above zero doesn't mean we should throw up our hands and give up. So economists have actually tried to put together studies about what happens when minimum wage is raised in one place and not in another, comparing results. That seems to show that at numbers in the $6-7 range in the '00s on the NJ-PA border, the effect is negligible.

Here's what we do know empirically: the American people are generally supportive of the current minimum wage according to polls, and even support by a majority, a higher wage. So whatever the negative job effects currently (if any), Americans are OK with those in favor of being guaranteed a bare livable wage if they're working. Now the president proposes returning that wage to have similar purchasing power to what it had when it was passed years ago, and to index it to inflation so we can stop having these battles (which help Democrats by the way, as raising the minimum wage is popular).

And once the minimum is raised to $9, economists like these guys can launch their empirical studies and see what effect it really has. Obviously we can't study a minimum wage that doesn't yet exist!