Yeah, we're imperialist!
Kevin Munger has an interesting piece on economics imperialism — the tendency of economists to study topics that seemingly belong to other disciplines. He asks economists to push back. Fine, I’ll bite. I work in an economics department but took my PhD in political science, so I've seen both sides of the fence. And I do work in genetics, so I practice imperialism (though maybe piracy is a better word).
Bluntly, I may not need to defend the current system. Kevin does too good a job of it himself! He calls our papers “admittedly excellent”, “impeccably crafted” and “well-cited”. Sounds good! Then he says
“Economists cannot develop entire literatures in mass shootings and mammograms, and the other disciplines lack the quantitative skill... to push back against or refine their findings.”
So to sum up, we’re writing good papers in other people’s disciplines, and the people in those disciplines lack the skills to challenge them. And this is our problem? Not theirs?
Kevin makes two substantive arguments against economic imperialism.1 One is:
“Given a finite number of cites from within Economics, the motor of Economics imperialism is the existence of other disciplines with untapped citation potential.”
The model of science here seems to be that there’s a finite pie of citations, and we need to divide it up fairly. Economists are getting listened to, and that means we’re not! Well, if sociologists cite the latest cleverly-designed, carefully analysed economics paper instead of something from within their own discipline, how is this anything but good work coming to the fore? Social sciences are there to learn about the world, not as podia to pontificate from.
A more serious point is this:
“The culture I’m criticizing is, in effect, redistributing Economics production so that there is a single point for each research question and thousands of research questions instead of twenty. That is, we will never be able to make this graph with a row with dots representing each of the causally well-identified school shootings.”
In other words, we are spreading ourselves too thinly. The graph that backs this claims up is from a meta-study of variation in treatment effects. Essentially, we need to run many RCTs on a single topic in order to understand the different dimensions of variability. Fine, I agree, we need many RCTs on a single topic. And yes, the top-five culture of producing one exhaustively backed-up paper instead of five might lead to a misallocation of effort. (Though is it really worse than academic cultures which require you to produce 10 half-baked multiple regressions every year?)
But it’s hard to see what this has to do with economic imperialism. If all the economists doing RCTs on school shootings or bed nets stopped and did RCTs on unemployment benefits instead, we’d learn more about unemployment benefits and less about school shootings. I assume everyone agrees society needs to study both. So, why shouldn’t economists study both? Especially since it’s they who for 20 years have been pushing the RCT methodology/credibility revolution/identification police, or whatever you want to call it, on to the rest of the social science. You need a story where somehow if economists exit the school-shooting-RCT market, the slack is taken up by criminologists with equal quantitative skills — which seems unlikely if those guys “lack the quantitative skills… to push back”.
To be clear, it’d be great if — no, is great when — sociologists, criminologists et al. do rigorous work on their core topics. I am sure those papers get widely cited. Until that happens more often, I hope that all the highly-cited economics papers continue to exert market pressure for them to move in that direction.
The history of economic imperialism is overwhelmingly positive for social science. Would anyone really not want Gary Becker to have written about the family, or crime? Even those who disagree with his approach can surely accept it was a valuable challenge. Economics keeps throwing out these challenges. Other disciplines should rise to them.
[Update: Kevin Munger responds. He calls me “incurious and self-satisfied” but you should read the whole piece….]
If you liked this piece, then I would love you to do three things:
Subscribe to this newsletter. It’s free and spam-free.
Share Wyclif’s Dust on social media, or forward it via email. By telling your friends and/or followers, you’ll be doing me a huge favour.
Read about the book I’m writing. It’s called Wyclif’s Dust, too. You can download a sample chapter.
He also makes an argument that economics is methodologically conservative. I think this is debatable — we’re happily taking on machine learning and big data — but anyway, it’s orthogonal to the question of imperialism. The papers that invade other fields and are hard for the natives to critique are clearly more methodologically sophisticated, not less, than the native product.
One comment. Rachel Croson once gave a keynote on Economic Imperialism. Her advice was: work with good natives (don't think she used that word), go to their conference. Publish in their journals. She did it in OR, but I think this is where most of economics fails. The rush to publish in top-5, and to a lesser degree in top fields leaves little scope for communicating with other disciplines. We have both published econ-style papers in other disciplines. We know this has a cost. I was told explicitly, upfront and ongoing, that these publications do not count towards promotion. I suspect it's worse in leading US departments. Most people are not interested in broad communication, especially as those come at the expense of publications within economics (I know anecdotally of one case where a top-5 editor explicitly rejected a paper because the authors published aspects of the research in a top non-econ journal). This not only stunts scientific progress, but leads to some cases where economists publish research that is methodologically rigorous, but completely misses the point, as they are not experts in the domain.