18FEB2013Even More Thoughts on the Minimum Wage
I am still in a mild state of shock that so many professional economists apparently doubt that demand curves slope downward. I admit upfront that I have not spent more than an hour or so looking through the latest literature reviews on the topic. Nonetheless, I remain unrepentant: I think increasing the price of unskilled labor by 24% will make employers hire fewer labor hours. The burden of proof is on the doubters to show why this isn?t so.
In addition to the compelling logic of ?demand curves slope downward,? we also have the casual empiricism of my last post, and now we?ve also got the below chart (brought to my attention by John S in the comments), taken from an AEI blog post but not sure who the original creator is:
Wow, look at that. It?s almost as if employers respond to incentives on the margin.
It?s true, there are papers that look at ?natural experiments? and somehow throw away logic and evidence such as the above chart. Let me run through some issues quickly:
Monopsony. One claim is that the Econ 101 logic breaks down because employers have market power. But hang on a second. If you want to tell me that the wages of, say, brain surgeons are below the competitive equilibrium, since there are only a few employers who can form a cartel, then OK I?ll at least give you a few moments to make your case. But you?re telling me there is a cartel of employers who are willing to hire unskilled labor?! That is literally the most non-specific factor of production on planet Earth. You need labor for everything, and by definition, unskilled labor is not suited for one occupation more than another.
I think the reason this might initially sound plausible to people, is that there aren?t a lot of teenagers working all over the place. You just see them concentrated in a few areas, like fast-food restaurants. But do you know why? Because of the ()#%$#$ minimum wage (and school attendance laws)!
You actually do see young people in various professional businesses and halls of government. They?re calledinterns. So we?ve got lots of young people finding employers willing to take them on at $0/hour, and yet apparently there is this ?indeterminate bargaining zone? where employers? quantity of labor demanded is the same between 1 cent and $7.25 (or $9). Does this range also count as a ?modest increase?? Or does even Krugman admit that getting rid of the minimum wage altogether would help reduce the 25%+ teen unemployment rate, while increasing it from $7.25 to $9 would be negligible in the other direction?
Studies look at employment growth, not unemployment rates. Apparently the standard thing to do in these studies is look at how much the absolute amount of employment or labor hours changes, rather than looking at the unemployment percentage. The idea (I gather) is that a high minimum wage can draw people into the labor market who can?t find a job, but these people wouldn?t have had a job anyway, so it?s not a strike against the system. Only if employers actually reduce the quantity demanded, can we say (some) workers are hurt. But even on its own terms, this argument fails. The most desperate, vulnerable people are the ones who will work for, say, $5/hour. At that rate, fewer middle-class college kids will enter the labor market. But bump up the wage rate to $7.25, and now a bunch of suburban white kids take a part time job at Pizza Hut to make a little extra money. Even if the total payroll and hours worked doesn?t change, it still means these kids bump out the new immigrant who barely speaks English and needs to get his foot in the door to establish a work history.
Studies correct employment growth for broader trends. My very quick reading of the literature suggested that the empirical studies in the olden days did find a strong connection between a minimum wage hike, and reduced hiring among teens. But, the newer wave of studies disputes that finding. One of the ?corrections? the new studies make, is to adjust the change in teen hiring compared to the broader labor market, which presumably isn?t affected by a minimum wage hike. Yet hang on a second. Even in the ?natural? experiments, I would imagine a state legislature that jacks up the minimum wage is also more likely to do other ?progressive? things that hurt employment growth. So things still move in the same direction, but now you?re not going to get as clear a signal; it?s hard to disentanglewhy the teenagers in California can?t get a job?is it because of the minimum wage hike, or because of their outrageously progressive income tax code?
Studies focus on fast-food employment across county or state lines. Again, I am not claiming to be an expert on this stuff, but it looked like a lot of the really ?compelling? studies looked at natural experiments where you had similar conditions except a chain of restaurants fell in one jurisdiction that raised its minimum wage, while the other restaurants in the chain fell in an adjacent jurisdiction that didn?t. Seems like a perfect laboratory test right? But hang on. If the minimum wage in one state makes it profitable for the restaurant to bite the bullet and install a bunch of labor-saving machinery (like the drink dispensers that you put the cup under and hit a button and walk away, unlike what they used to do when I was growing up where you had to hold the cup in place on the nozzle), then it would be pretty easy for that restaurant chain to use the same, new design when opening up new locations in other states with the original minimum wage. By the same token, even longitudinally looking at the same actual restaurant, once they redesign the place to be run by (say) 4 responsible teenagers and a manager, instead of (say) 9 goof-off teenagers and a manager, then even if that state later abolishes its higher-than-federal minimum wage, the damage is done; the restaurant isn?t going back to the old model.
How does this square with the Keynesian story about monetary stimulus? Finally, how the heck does this whole minimum wage digression line up with Krugman et al. constantly telling us that the problem in Europe and elsewhere, is that wages are too high relative to the price level? They tell us that if we engage in a currency war, we?ll all be better off because prices will rise, making it profitable for employers to hire once again. So, are they saying prices will need to rise by more than 24 percent, in order for the teen unemployment rate to budge?
I?m sorry, I just get the feeling that the story changes to fit the progressive policy of the day. And again, I am not burying my head in the sand and refusing to accept something obvious: On the contrary, I am saying demand curves slope downward, and I can point to all sorts of obvious evidence to back that up. Indeed, the Keynesians themselves think employers follow the same logic I?m talking about, when it comes to their proposals for monetary stimulus.
Yet somehow, the old empirical consensus on the minimum wage has been overturned by a wave of new studies of ?natural experiments,? so I?m giving reasons in this post why those studies might be missing the obvious conclusion staring us all in the face: Making teenagers 24% more expensive in the middle of a depression is not the way to help teenagers.
Even More Thoughts on the Minimum Wage