I don’t like Freakonomics.
The occasion for this outburst of ire is a post by Tyler Cowen in the New York Times Upshot titled “The Lack of Major Wars May Be Hurting Economic Growth.” The article is a pretty fine example of everything I dislike about Freakonomics. The game looks something like this:
- The economist (all aspiring freakonomists are economists) picks a field in which she is not an expert. The field is one in which she can feel confident in having superior statistical training relative to the median practitioner of said field. Any social science is a pretty good choice. So are the humanities. The aspiring freakonomist will never choose a field like computer science, physics, or biology; it is too likely that the practitioners in these fields have the requisite training to expose the freakonomist as a buffoon. In the case of Tyler Cowen’s article, he chose history.
- Find an obvious correlation. This is pretty easy; there are lots to be had. In this case the correlation is between incidence of war and GDP growth in the developed world.
- Spin up a plausible story that explains the correlation. Tyler’s story is of course not the typical Keynesian story about warfare spending spurring economic growth. Instead he claims that war creates high stakes for government policy makers, causing them to make better economic decisions. He cites both R&D investment and economic liberalization policies.
- Act confidently that the story you have spun up is the story. After all you are an economist; you have the kind of statistical training necessary to spot these things. Bask in your superior insight. Freakonomics is a cousin of mansplaining. Noah Smith suggests Freakonomics may be a form of of derp.
In the case of the article I linked it’s pretty easy to poke holes in the story. Cowen restricts his analysis to a very small swathe of history, specifically the period from WWII to the end of the Cold War. While it is true that this period exhibited a correlation between GDP growth and incidence of war, there were-needless to say-other things happening. And while it is also true that the post Cold-War era has witnessed both a decline in the rate of economic growth and violent conflict, there are other plausible explanations for the relatively low growth rates of the past twenty or thirty years. Tyler Cowen has even written a book on them. Further, his analysis does not consider the entirety of pre-industrial human history, in which there was both a lot of violent conflict and very little productivity growth.
I am an aspiring economist; I wouldn’t be engaged in that project if I didn’t believe that economics is a powerful tool with which to understand myriad important social problems. However I think it ‘s also important to employ these tools with a sense of humility and with an appreciation of their limits. Normally Tyler Cowen is someone I look up to in that regard; in that sense his Upshot piece is a disappointing outlier.