Wednesday, May 20, 2015

Paul Romer Confused On Capital Theory

I have noted Paul Romer's confusions before. For example, consider the following passage:

"In the conventional specification, total capital K is implicitly defined as being proportional to the sum of all different types of capital. This definition implies that all capital goods are perfect substitutes. One additional dollar of capital in the form of a truck has the same effect on the marginal productivity of mainframe computers as an additional dollar's worth of computers. Equation (1) expresses output as an additively separable function of all the different types of capital goods so that one additional dollar of trucks has no effect on the marginal productivity of computers." -- Paul Romer (1990).

Does Romer think that the so-called factor price curve for all techniques must be an affine function? That price Wicksell effects are always zero? Or maybe he just is trying to buffalo his reader with an ill-thought out use of mathematical symbols.

On his twitter feed, he expresses a disinterest in knowing what he is talking about:

"Sorry, but the capital controversies were a waste of time. No relevance then or now." -- Paul Romer, 16 May 2015, 1:09 PM.

I suppose one might possibly be able to defend this view:

"Economists usually stick to science. Robert Solow (1956) was engaged in science when he developed his mathematical theory of growth. But they can get drawn into academic politics. Joan Robinson (1956) was engaged in academic politics when she waged her campaign against capital and the aggregate production function." -- Paul M. Romer (2015).

One might say Solow was looking to empiricalism when he developed his non-rigorous, loose theory of growth. And, I suppose one could say that some political views were involved in Joan Robinson's insistence that Keynes' theory applies to all runs, both the short run and the long run. And in her attempt to combat the development of pre-Keynesian theories after Keynes, even if such developments were the product of those who called themselves Keynesians in some other context.

But to make such an argument, one would have to have read at least some of Joan Robinson's work from the era. It is clear that Romer has not:

"When I learned mathematical economics, a different equilibrium prevailed. ...when economic theorists used math to explore abstractions, it was a point of pride to do so with clarity, precision, and rigor. Then too, a faction like Robinson’s that risked losing a battle might resort to mathiness as a last-ditch defense, but doing so carried a risk. Reputations suffered.

If we have already reached the lemons market equilibrium where only mathiness is on offer, future generations of economists will suffer... Where would we be now if Robert Solow’s math had been swamped by Joan Robinson’s mathiness?" -- Paul M. Romer (2015).

When, during the Cambridge Capital Controversy, did Robinson try to buffalo readers with pretend rigorous manipulations of imprecisely defined mathematical symbols. How about never? Is never good for you?

Update (21 May 2015): Reactions to Romer from Peter Dorman, Justin Fox, Joshua Gans, Noah Smith, Lars Syll, and Matias Vernengo

Update (24 July 2015): Marc Lavoie and Mario Seccareccia also comment on Romer's confusion.

References
  • Romer, Paul M. (1990) Endogenous Technological Change, Journal of Political Economy V. 98, N. 5 (Oct): S71-S102.
  • Romer, Paul M. (2015). Mathiness in the Theory of Economic Growth, American Economic Review, V. 105, N. 5: pp. 89-93.

5 comments:

ivansml said...

"Does Romer think that the so-called factor price curve for all techniques must be an affine function?"

Why would he think any such thing? He's stating an assumption he's using in his specific model. You may not like it, (Romer himself discusses some of its limitations in the next paragraph), but to interpret the statement as some kind of general claim on all possible technologies is just plain wrong.

Robert Vienneau said...

Does Romer (1990) assume the existence of heterogeneous physical capital goods? I find it hard to read his paper as anything but incoherent. I do not think my reading as idiosyncratic, although others (e.g., Ian Steedman) focus on other failures of clarity in the definition of variables.

Romer is muddled on other subjects. It is not the case that only in science is it possible for people to truly listen to one another and be persuaded to change their minds. Maybe Romer can get McCloskey to explain Habermas to him. (I only know of Habermas from secondary literature.)

Nevertheless, I agree with Romer on some points. For example, I agree that Milton Friedman wrote his paper on methodology so as to dismiss the theory of monopolistic competition.

But his apparent beliefs about Joan Robinson are just ludicrous.

Anonymous said...

Nobody cares what you think, in case you didn't know that already.

Blissex said...

«Nobody cares what you think»

I care.

R Viennau's thoughts are interesting to me. Perhaps because my background in political economy studies (quite some time ago) was much the "european" one that he has been discovering (Pareto, Walras, Keynes, Robinson, Sylos-Labini, Tinbergen, Schumpeter, Sraffa, ...) so I am amused to follow the evolution of this thoughts. I am also working in engineering currently, so that's another point of affinity driving interest.

I stopped looking into the technical details long ago, when I figured the big tricks and sophistries of the neoclassical approach, and for me it is also interesting to me to read his detail investigations of the more reliable parts of the published literature and learn new things about that.

Stefano Zambelli said...

Dear Vinneau,
I have been following your comments and work on the capital controversy for many years.

You have cited my work from 2004 with thew title "The 40% Neoclassical Aggregate Theory of Production". For this I am very thankful.

Now I have, finally, published an article which is a much developed follow up of that work. Using empirical data I show that "The Aggregate Production Function is NOT Neoclassical". That is, the it is in fact the "The 0% Neoclassical Aggregate Theory of Production".

The article is now available on line through the Cambridge Journal of Economics Homepage. I would very much appreciate to know what you think about it.

If I had your personal internet address I would have written to you directly.

Thank you again.

Stefano Zambelli