Paul Romer on math masquerading as science
from Lars Syll
I have a new paper in the Papers and Proceedings Volume of the AER that is out in print and on the AER website …
The point of the paper is that if we want economics to be a science, we have to recognize that it is not ok for macroeconomists to hole up in separate camps, one that supports its version of the geocentric model of the solar system and another that supports the heliocentric model …
The usual way to protect a scientific discussion from the factionalism of academic politics is to exclude people who opt out of the norms of science. The challenge lies in knowing how to identify them.
From my paper:
“The style that I am calling mathiness lets academic politics masquerade as science. Like mathematical theory, mathiness uses a mixture of words and symbols, but instead of making tight links, it leaves ample room for slippage between statements in natural versus formal language and between statements with theoretical as opposed to empirical content.”
Persistent disagreement is a sign that some of the participants in a discussion are not committed to the norms of science. Mathiness is a symptom of this deeper problem, but one that is particularly damaging because it can generate a broad backlash against the genuine mathematical theory that it mimics. If the participants in a discussion are committed to science, mathematical theory can encourage a unique clarity and precision in both reasoning and communication. It would be a serious setback for our discipline if economists lose their commitment to careful mathematical reasoning …
The goal in starting this discussion is to ensure that economics is a science that makes progress toward truth. A necessary condition for making this kind of progress is a capacity for reaching consensus that is grounded in logic and evidence. Given how deeply entrenched positions seem to have become in macroeconomics, this discussion could be unpleasant. If animosity surfaces, it will be tempting to postpone this discussion. We should resist this temptation.
I know many of the people whose work I’m criticizing. I genuinely like them. It will be costly for many of us if disagreement spills over into animosity. But if it does, we can be confident that the bad feelings will pass and we should stay focused on the long run …
Science is the most important human accomplishment. An investment in science can offer a higher social rate of return than any other a person can make. It would be tragic if economists did not stay current on the periodic maintenance needed to protect our shared norms of science from infection by the norms of politics.
One of those economists Romer knows and — rightfully — criticizes in his paper is Robert Lucas.
Lucas is as we all know a very “mathy” person, and Romer is not he first to notice that “mathiness” lets academic politics masquerade as science …
Added 20:00 GMT: Joshua Gans has a post up on Romer’s article well worth reading, not the least because it highlights the nodal Romer-Lucas difference behind the “mathiness” issue.
In modern endogenous growth theory knowledge (ideas) is presented as the locomotive of growth. But as Allyn Young, Piero Sraffa and others had shown already in the 1920s, knowledge is also something that has to do with increasing returns to scale and therefore not really compatible with neoclassical economics with its emphasis on constant returns to scale.
Increasing returns generated by non-rivalry between ideas is simply not compatible with pure competition and the simplistic invisible hand dogma. That is probably also the reason why so many neoclassical economists — like Robert Lucas — have been so reluctant to embrace the theory wholeheartedly.
Neoclassical economics has tried to save itself by more or less substituting human capital for knowledge/ideas. But knowledge or ideas should not be confused with human capital.
In one way one might say that increasing returns is the darkness of the neoclassical heart. And this is something most mainstream neoclassical economists don’t really want to talk about. They prefer to look the other way and pretend that increasing returns are possible to seamlessly incorporate into the received paradigm. Romer’s view of human capital as a good example of non-“mathiness” not-withstanding, yours truly is of the view that talking about “human capital” — or as Lucas puts it,”knowledge ’embodied’ in individual people in the short run” — rather than knowledge/ideas, is only preferred because it makes this more easily digested.
Added 20:55 GMT: Romer has an even newer post up, further illustrating Lucasian obfuscations.
Added May 17: Brad DeLong has a comment up on Romer’s article, arguing that Lucas et consortes don’t approve of imperfect competition models because they are “intellectually dangerous” since they might open up for government intervention and “interventionist planning.” I agree with Brad, but as I’ve argued above, what these guys fear even more, is taking aboard increasing returns, since that would not only mean that policy preferences would have to change, but actually would bring havoc to one of the very fundaments of mainstream neoclassicism — marginal productivity theory.