Thursday, June 28, 2018

Thought for the day

From Paul Krugman [emphasis added]:
What the freshwater school did was to take the actual experience of business cycles and say, “We don’t see how to formalize this experience in terms of maximizing equilibrium models; therefore it doesn’t exist.” It only looks as if recessions result from inadequate demand and that monetary or fiscal expansion can create jobs; our models tell us that can’t happen, so it’s all an optical illusion.


Anyway, this isn’t about me (well, it sort of is, but never mind.) The important point shouldn’t be “don’t formalize”; it should be that formalism is there to open your mind, not close it, and if the real world seems to be telling you something inconsistent with your model, the problem lies in the model, not the world.


  1. Mark:

    That's an interesting thought, but I think Krugman's only halfway there. The problem is that it is often through models that we know what "the world" is. Consider "the economy," which we understand through various measures such as unemployment, inflation, GDP, trade balance, etc., all of which are measured in some model-based way.

    To put it another way, if the simple distinction implied between "the model" and "the world" really existed, with "the world" preferable to "the model" whenever the two disagreed, then why have models at all?

    I'm not trying to pick at Krugman's particular point regarding business cycles---a topic I know nothing about---but rather to make a more general point. I think we should take our models seriously, push them hard, and fix them when we fail. Comparison to the real world is essential. But it's a two-way street. A good model does not just "open your mind"; it can also change how you measure and understand the world.

    1. Andrew,

      That's a large family-sized can of peas you just opened and it's too late to address the points in detail, but here are a few quick thoughts:

      In this context, I think the distinction between models and theories gets a little vague.

      If you're looking for more familiar territory, evolutionary psychology (specifically the naïve variety) is somewhat analogous. When faced with evidence that contradicts their theories (like men pursuing sex without procreation), the proponents tried to ignore or explain away the inconvenient examples.

      Yes, there are big problems with the model versus real world framing. Krugman is writing in a short format for a lay audience and he is cutting some corners. A better framing (and one I've suggested before) is looking at models, first principles/common sense, data, and anecdotal experience. Normally, we would expect these four to correspond. When three of these point in one direction and the fourth points in a completely different direction, we should start by questioning the one, then take a hard look at the remaining three.