Monday, September 15, 2014

An alternative Venn Diagram for Mark Perry

I really hope I don't have to explain to everyone what's so wrong about this post by Mark Perry on the minimum wage.

All the law of demand says is that demand curves slope down.

No one that I know that is doubtful about modest minimum wages hurting employment disagrees that demand curves slope down. If you think they are arguing demand curves slope up you need work a little harder at understanding the conversation you are inserting yourself into.


  1. Mmm... there's a bit of apples and oranges going on here.

    "Minimum wage reduces employment of UNSKILLED workers". It could have some other effects on employment of other workers, hence on employment in general.

    "Immigration of UNSKILLED workers hurts native UNSKILLED workers". It could have some other effects on wages/employment of other workers.

    Or another way. Long run, with capital accumulation. Solow model.

    Immigration: L goes up. Supply of labor increases. This raises MPK, which raises s*y (y is output per worker) above (n+d)*k. Capital increases to new steady state. This increases MPL. Labor demand increases. In the end, no effect on wages or native employment. Immigration does not hurt natives (at least in long run).

    Minimum wages: Economy is at steady state k*=K/L. Now minimum wage is imposed so that w(min)>MPL. Unemployment results in short run. L goes down (since we need MPL to go up). This means K/L goes up. K/L up means MPK down. This means s*y falls below (n+d)*k. Capital per worker decumulates until you get back to the right k*. K has to go down.This shifts labor demand down. Given the minimum wage this results in even more unemployment. L goes down again... whoah whoah!
    (except that the minimum wage sector isn't big enough relative to the whole economy for this whole unraveling to happen - just illustratin')

    So it's perfectly logical to think that immigration does not hurt native workers but minimum wages do. If I remember correctly, Borjas tried to pull this trick once ("if you believe minimum wages hurt employment you must believe immigration is bad, because the demand curve for labor slopes downward"). Was wrong then, is wrong here.

  2. And to be clear, I am not arguing that the above is the empirically relevant case - in particular we have a lot of trouble figuring out what the actual effects of minimum wages are. All I'm saying is that as a matter of logic, there's nothing contradictory here.

  3. Daniel, I'm not trolling you, I honestly don't know what you're saying in this post. It *sounds* like you are saying in the text, "Of course we all know that a higher wage reduces employment." Then in the diagram it sounds like you are ridiculing people who think, "A higher wage reduces employment."

    1. It's an issue of when people choose to think deeply about what's going on in the data and when they think "demand curves slope down" answers the question.

      So I would say we all agree demand curves slope down. When we get results that immigrants raise native wages people like Perry tend to think deeply about that, ask why Card's numbers are different from Borjas's, figure out that Borjas is looking at a partial equilibrium while Card is looking at aggregate effects, and then we hit on spillovers, general equilibrium effects, what have you to conclude that we do not have to abandon the idea that demand curves slope down to accept Card's numbers.

      But when Card does a paper on the minimum wage people act like he is denying basic economic theory. There are lots of ways to approach that result. The easiest way to talk about it and this downward sloping demand issue is just to point out to people that (1.) lots of people mention this monopsony thingy when they talk about the Card results and (2.) lo and behold demand is downward sloping in that model! Of course there's more to talk about than monopsony, but that alone deals with this idea that people are denying the law of demand.

      In other words, you really don't have to be all that intellectually curious to avoid this contradiction, but nevertheless there are a LOT of people in that purple area in the middle.

    2. btw - a higher wage doesn't necessarily reduce employment. Never reason from a price change. better to talk about downward sloping demand.

      And better to talk about demand than supply because supply has a greater tendency to do funky things.

    3. I first came accross "Never reason from a price change" in Scott Sumner posts. Is that something he made up or did you get it from somewhere else?


All anonymous comments will be deleted. Consistent pseudonyms are fine.