Economist, heal thyself
By Roger Martin
The opinions expressed are his own.
Reuters invited leading economists to reply to Mark Thoma’s Op-Ed on the “great divide” in economics and will be publishing the responses. Below is Roger Martin’s reply. Here are responses from Ashwin Parameswaran, James Hamilton, Dean Baker, Lawrence Summers, and a recap of Paul Krugman’s.
I have a great deal of sympathy with Mark Thoma’s thesis of the great divide between academic and practitioner economists, though I have a somewhat different perspective on the reasons why the great divide exists. I am less convinced that the divide is spurred by a desire on the part of academic economists to look down their noses at practitioner economists. I think it has more to do with the strictures that they have slowly but surely placed on themselves.
Social scientists – economists included – have long bristled at the degree to which the hard scientists stare down their noses at them because they are not ‘rigorous’: their models aren’t specified precisely enough and their data aren’t pure enough to satisfy the physicists, chemists, biologists, etc. Over time, some social scientists – led by the economists – battled back by becoming more quantitatively rigorous, by making the methodologies they used more similar to those of the hard scientists, with more thoroughly specified models, more high-flying math, more precise data and bigger sample sizes. That helped the economists in particular stand up to the scientists and say: my math is as big as your math; my data is as sound as your data. This made them happy and helped them look down their noses at ‘softer’ social scientists – like psychologists and historians and anthropologists. And as they felt that rush of satisfaction, they were inclined to continue to up the ante some more to move closer to the hard scientists and distance themselves more from the soft social scientists.
But it created a problem. If things are complicated in the hard sciences, they are even more complicated in the social sciences. People (not to be confused with human bodies) are really complicated, especially when they interact with other people. And lots of things that they do are really difficult to model using the techniques approved by the union of hard scientists. So economists have increasingly decided to model only what they can model using the required techniques. If they don’t use the requisite techniques, they get lumped back with the softies – and given the emancipation they have felt by doing rigorous analysis, they would never go back.
Practitioner economists, on the other hand, live in and face the problems of that horribly complicated world filled with ambiguous data, small sample sizes – sometimes of one – and confounded causal relationships. They need to make sense of their world and do the best they can to model the difficult-to-model world that confronts them. They want to know: how fast will the US economy recover from 2008-9? The highly quantitative macroeconomists predict the course of recovery using the exact same rigorous, quantitative models that, as late as the second quarter of 2008, failed completely to predict the economic crash that happened a mere quarter later. These are rigorous quantitative models that the practitioners look at and say: rigorous but irrelevant.
The academic economists don’t refuse to engage in these problems because they think the problems are unworthy or that the practitioners are beneath them. Rather, many of them are like Snowy, our Bichon Frise, in the yard of our house back in Wellesley, Massachusetts. Because like many Bichon Frises, Snowy was not a deep and careful thinker, we installed an invisible dog fence (with a collar that zapped her if she crossed the line) to make sure she didn’t run out into the street and get mowed down by a car. Snowy got zapped by her collar a couple of times and that was it: she was never going to cross the barrier again. And that even meant not crossing it to the side when she saw her favorite play partner, Oliver – another foofy little dog breed, the name of which I can’t remember – in the neighbor’s front lawn.
They don’t hate or mock their proverbial Oliver. In fact, they look longingly and lovingly at Oliver. But they will be damned if they cross that line and get zapped by their collar. The main difference between those academic economists and Snowy is that we put the zapping collar on Snowy and the economists put on their own collars.