How to trust economists

EconomicsHome

The questions of whether we can trust economists and what economics is good for are back, thanks to a New York Times post last week making an outrageous claim:

The fact that the discipline of economics hasn’t helped us improve our predictive abilities suggests it is still far from being a science

At Bloomberg, economist Justin Wolfers notes that the post contains no actual evidence for that claim (though Wolfers himself does not offer any). My take is that economics almost certainly has improved our predictive ability, but I won’t attempt to make that case here. (I’ll also note that much research into whether experts predict things well tends to focus on the most difficult cases in the field. Related: Wolfers has documented the strong degree of consensus within the economics profession, which doesn’t always come through in the media.)

But apart from all this, it strikes me that whether we can trust economists is an impoverished debate. How to trust economists – how to interact with their claims, when to trust those claims and when not to – is more interesting, relevant, and important. Yes, economists too often take on the role of philosopher or political scientist. Yes, sometimes their biases shine through in their recommendations. Yes, economics doesn’t have all the answers. And yes, some of those answers even turn out to be wrong *cough* Great Moderation *cough*. But plenty of the time, listening to economists is a good idea!

So I thought I’d transcribe the last page of a book called The Assumptions Economists Make, a very pessimistic tour of economic models throughout history by HBS professor Jonathan Schlefer. It’s an entertaining read; Schlefer has a PhD in Political Science, and his quite critical take focuses on the inconsistencies in economic models throughout time. (If the book has a central flaw it’s that it tends to do battle with these abstractions in the abstract; their usefulness is mostly secondary to their validity.)

Nonetheless, his concluding recommendations are a must-read for anyone grappling with how to trust economists:

  • Economists should transparently describe critical assumptions. These assumptions should be realistic and pertinent to the situations that a particular model seeks to explain.
  • Economists should explain the structure of their models. The structure of a model constitutes the perspective it sheds on some crucial aspects of an economy. Thoughtful individuals should not believe, and policymakers should not use, an unexplained model.
  • However realistic its assumptions, a model stands an excellent chance of ignoring crucial aspects of an economy because, among other things, incorporating them might well make the model too complex to handle. Think what factors it might miss.
  • There are always conflicting models to explain any given aspects of an economy  In looking for practical conclusions, weigh conflicting models.
  • Macroeconomies are incredibly complex. One of the most useful things economists can do is explain publicly what they do not know.

Though many of these are focused on what economists should do, turn them around and you have a nice list of questions to ask economists about their recommendations. The fourth one is particular important. I’d add that we should what data the model successfully does and does not explain.

Taken together, this provides at least the beginnings of a toolkit for politicians, journalists, and citizens to engage with the recommendations of economists. And engagement, ultimately, will be more fruitful than either blind acceptance or dismissal.