The New Keynesian model was built to explain a world of moderate fluctuations in GDP. It features temporary price rigidities, and the macroeconomic aggregates in the model are consistent with the optimizing behavior of individual consumers and producers. For certain types of questions – how should policymakers behave to stabilize an economy with mild fluctuations induced by price rigidities – it is the best model to use. Hence it’s popularity during the “Great Moderation” from 1984-2007 when there were no large shocks to the economy…
The IS-LM model, on the other hand, was built in the aftermath of the Great Depression to examine precisely the kinds of questions we faced throughout the Great Recession, issues such as a liquidity trap, the paradox of thrift, and how policymakers should react in such an environment. Why is it surprising that a model built to explain a particular set of questions does better than a model built to explain other things?
Read the whole essay. My guess is that even those who are inclined to be sympathetic to Thoma on this point, myself included, will have a hard time accepting the idea that one should switch models depending on circumstances. Yet I would argue that the evidence is that economists have done exactly that over the past 50 years.
Oy. Thoma’s article is disheartening.
His opening analogy regarding driving and digging a hole is terrible. Maps are not models. There actually *is* a useful unifying model which abstracts away unimportant details for those two activities. It’s called Newtonian physics. A map of “underground pipes, electrical wires, etc” explains how a hole is dug as much as a map of the human body explains how surgery is performed.
“how should policymakers behave to stabilize an economy with mild fluctuations induced by price rigidities – it is the best model to use. Hence it’s popularity during the “Great Moderation” from 1984-2007 when there were no large shocks to the economy.”
This attitude, that economic shocks are exogenous, is the most obvious flaw in mainstream macroeconomics, and is why it’s seen as a joke by everyone outside of econ departments. They just think the economics gods shake our snow globe and cause recessions? Or do recessions like 2007 start because of problems building up underneath the surface of the moderate periods?
Here’s a prediction: future scholars will look at today’s mainstream macroeconomic models with as much respect as we look upon the “four humors” model of the body.
Given my priors I find your criticism compelling. But could Thoma, et al, respond that the models aren’t useful so much for telling us what is truly going on under the surface, but what policies are helpful given a particular set of circumstances?
On the other hand (and I speak as an absolute novice) I think your point about the “four humours” is likely true. I just have a hard time believing that economists understand the economy all that well, and not only by virtue of the fact that it’s hard to do controlled or natural experiments on an economy. Another issue, it seems to me, is just how new, as phenomena, modern economies are. How could we have a big enough sample at this point in time (or even a proper notion of what should delimit a sample) to have a decent grasp on a plurality of factors that could produce a particular state of affairs in an economy? Considering this in the light of technological and social changes further muddies the water.
Going back to the map imagery: perhaps it’s not that economics isn’t useful, but that economists are roughly in the position of Age of Exploration mapmakers. You need them, but (through no particular fault of their own) you can’t trust them.
Macro economists need to admit that there are different schools of economics with different assumptions and approaches. There are paleo-Keynesians, like Krugman, New Keynesians, neo-classicals, monetarists, socialists and Austrians. I highly recommend Roger Garrison’s book “Time and Money.” Garrison compares all the schools of macro.
Having learned neo-classical and new keynesian in school, then Austrian afterwards, it’s clear to me that Austrians have the inside track.
As you mentioned, economists cannot conduct controlled experiments like the natural sciences. Equilibrium analysis was originally an attempt to do mental/logical controlled experiments. Econometrics was supposed to be the controlled experiment for economics, but we find that we often don’t have the data we need; there are too many variables with high levels of covariance; misspecification is still a problem; Dr. Kling has written a lot on the problems with macro and econometrics.
We need to apply what we know for certain in micro to macro. Micro is the most certain side of economics. Macro should not contradict it. For example, monetary and fiscal policies in most macro tend to violate the micro principles of opportunity cost and diminishing marginal returns.
Most importantly, mainstream macro ignores the capital structure of the economy. Austrian econ shows how government and Fed policies impact the capital structure and the effect on the economy.
How do you know which model to use ex ante? Other than that, I am all for using different models.
As a poilcy adviser I have trained myself to read and study the most variegate and alternative models. I have realized that models are (almost) always internally consistent; they differ because their assumtion differ. Now different assumptions describe different worlds,, better: different state of the world. Successful advisers pick up the model that better describes the ‘current’ state of the world and derive its policy implications.
Two failed models are our only choices? How sad. It advertises the bankruptcy of mainstream macro.