Consider this a Robin Hanson-style post.

Basic textbook economics says that workers’ wages are determined by marginal productivity. (Here’s a quick explanation from Mankiw intended for a popular audience. Deep in this column you can see Krugman saying the same thing.) Now, one of the major issues in labor economics is to what degree this basic model holds. We know there are all sorts of distortions and strange things going on in the real world. I won’t get into the debate at great length in this post, but, just to get started, I’d recommend you Google and read debates about, e.g,, A) sticky wages, B) whether the output recovery without employment recovery following the great recession showed that many workers had zero marginal product, C) how immigration restrictions are massively inefficient and keep third world wages incredibly low. (I’m not taking a stand here on A or B, though I have endorse C in print and continue to do so.) And of course there’s Frank’s famous paper here and the hundreds of reactions to it.

Many of the responses to my previous post (which purposefully kept the economics simple in order to ask a moral question) have been rather obviously ideological. I discussed a case where I stipulated that a particular worker’s marginal product was extremely low, lower than whatever someone might reasonably consider a living wage, in order to ask whether employers still owe that worker a living wage, even at a net loss to themselves. Many libertarians took that to mean I decisively refuted the Walmart living wage argument, even though I warned them I had not, while many left-wingers had various fits and tantrums.

If you’re on the Left, you might be tempted to believe something like the following: “With the exception of a few disabled people, perhaps, all adult workers are productive enough that their potential employers can hire them at a living wage [however you might define that] and still make a profit.” If you find yourself thinking something like that, ask, did you come to this conclusion after studying labor economics closely, or did you come to it because it would be convenient for your ideology? Unless you define the “living wage” as very low, it would be pretty surprising if this were true. So, for instance, this website defines a living wage for a single adult living alone in the District of Columbia as $13.68/hr, a single adult with one child living as $26.37/hr, and one adult with 2 children as $32.97/hr. Do you think many low-income workers in DC are producing $32.97/hr?* Even in DC, that doesn’t seem very plausible. Is it even plausible to think that all potential employees will produce $13.68/hr? Many low-income workers are dysfunctional, badly behaved, or just not very skilled. Sorry.

If you’re a libertarian or a conservative, on the other hand, you might be tempted to believe that whenever someone gets a low wage, this is because she has low productivity. This is probably often the case. But you don’t have to accept Marxian whack-a-doodle-nomics–you can look at real social science–to see arguments that some workers are getting paid low wages because they have a bad bargaining position or because there is a market or government failure. (C above is just such an example. Sweatshops would be far less sweaty with open borders.) It’s awfully convenient, ideologically, if it turns out that all poorly paid people are just doing low productivity work. But, if you’re going to argue that, you’d better be prepared to answer a string of objections from labor economists who claim that econ 101 marginal productivity theory frequently doesn’t hold in the real world. Libertarians will have an easier time ideologically with this than conservatives, though, since many of them already hold that corporate rent seeking causes market distortions that bias wages down.

Relatedly, Will Wilkinson responds here, in the Economist, making a further case for social insurance over the living wage. Sam Freeman says Rawls rejected the minimum wage:

He [Rawls] thought we ought to get rid of a minimum wage and let the labor market just go as low as it would and let employers just pay two, three dollars an hour if they could and let the government come in and supplement that.

…So he supported – he would support wage subsidies…

So, for all the unfortunate angry mood affiliation and team signaling that’s taking place here, Rawls appears to be on my side on this.

UPDATE: Skoble has a winning comment: “Of course, in DC, many high-income workers don’t produce $32.97/hr either.”