The Assumptions in Growth Theory

Dietz Vollrath has a new post that goes a long way toward clarifying the battle lines in the fight over the foundations of growth theory. If you haven’t read it, go read it now, then come back. Any attempt I might make to summarize it here will only add noise. The trilemma he lays out is crystal clear.

He acknowledges conversations with Pietro Peretto, who apparently deserves some of the credit for the distilled clarity of the post.  This shows that as usual, we are deep into meta-theory. The whole point of his post is that is never enough credit/compensation to go around when people like Dietz and Pietro produce a valuable nonrival good like this post.


I think that Dietz is correct that the most interesting part of the Boldrin-Levine analysis is the discussion of open source software. As he says, the right way to make sense of their claim is that they are treating ideas as pure public goods that earn no compensation. (I’m going to post something separately on how to model why people supply some new ideas as voluntary public goods. BL do not provide a model for this.) So it is correct to say that BL drop #2 from the trilemma that Dietz lays out.

But it is worth recalling that in the “theory” section that has all the mathiness, BL also make the astonishing claim that in a competitive equilibrium they can drop #3 (that rival inputs are paid their marginal product) if the competitive equilibrium has capacity constraints. This isn’t neo-Marshallian theory. It is paleo-Marshallian. From economists who claim to be members of the one true church of dynamic general equilibrium theory, it is a real head-scratcher.


After twenty years of playing whac-a-mole with the die-hard price-taking neo-Marshallians, I can tell you exactly where the argument will go next. When the choices are stated as clearly as Dietz does, the neo-Marshallians will claim that there is no such thing as a nonrival good. Nick Rowe has already started laying down a smoke screen of obfuscation on this point.

His argument illustrates what we can expect:

I will use words only because with some math people will see right through what I am saying. And because I’m a professor, I can rely on unaided introspection as a guide in any discussion about the production and distribution of ideas.

So here is my verbal argument. A nonrival idea like the post that Dietz has written has to be combined with the human capital of the reader before it can produce any value. So there is no such thing as an input that is nonrival idea. Despite what you might think, the post by Dietz actually does not exist.

This is a familiar point. In agriculture there is no such thing as an input called land because land has to be used with labor. Can you get any food without labor? Preposterous!

So obviously there is no such thing as a blog post, no such thing as a nonrival idea, and no such thing as land. So Marx was right and we have to go back to the labor theory of value.

Oops.