Wonks Abandon an Economic Dream
There has been an interesting debate in economics blog-land during the past couple of weeks. Basically, a lot of people outside econ academia are very annoyed with macroeconomics for failing to foresee the possibility of a big crisis, right up until a big one hit in 2008. A lot of these people blame the type of models that academic macroeconomists were using at the time (and are mostly still using now).
The question here is whether economists should focus most of their energy on what happens in normal times, or focus on what happens when things go crazy.
Here’s the meat of the debate. Mainstream macro models are just big systems of equations, and if you focus on only one small region, they start to look like linear equations. That makes them easy to work with. Lines can only intersect at one point, so your equilibrium -- which represents your prediction for what happens to the economy -- is unique. In other words, there’s only one thing that can happen.
But if you go far away from that equilibrium point, the curves can bend back around and meet in some faraway location. Then you have a second equilibrium -- another possible future for the economy.
So what happens then? No one knows. If models have multiple equilibria, then they predict that big sudden shifts can happen in the economy. To know how close you are to such a shift, you have to rely on data. But data in macroeconomics isn’t very good, and the model will almost certainly not be exactly correct. And when you go nonlinear, even seemingly simple models get very weird, very fast. So in other words, if you go with the full, correct versions of your models, you stop being able to make predictions about what’s going to happen to the economy.
Economists like to make predictions. They like to sound confident and to feel confident about policy. In macro, that confidence is rarely warranted by the facts, but that doesn’t negate the near-universal human desire to believe that we understand stuff. Also, linearized models -- the kind that let you convince yourself that you know what’s going to happen -- are a heck of a lot easier to work with, mathematically.
So some economists suggest that we should use the simple, linearized models in normal times -- after all, normal times are what these models are made to describe -- and then look around for things that might throw the economy off of its normal track. Subprime mortgages, overly interconnected financial companies, housing bubbles and things like that. The biggest promoter of this idea has been International Monetary Fund Chief Economist Olivier Blanchard.
Lots of people, mostly but not all of them outside academic econ, are angry at this idea. One such critic is finance blogger Frances Coppola. Another is economics columnist Wolfgang Munchau. But Roger Farmer, a respected macroeconomist at University of California-Los Angeles, and someone who has been warning about multiple equilibria for years (in vain, of course), has added his voice to the critics.
Other macroeconomists protest that they have, in fact, been thinking about multiple equilibria. Paul Krugman wrote a paper about it in 1999, using it to explain the Asian financial crisis of the late 1990s. And plenty of macroeconomists are even now hard at work exploring the weird, wacky world of nonlinearity.
So Coppola, Munchau, and the other critics may have been too quick to issue a blanket condemnation of the entire macro field. But thinking along the lines of Blanchard’s -- that models should describe only the normal, calm times -- is still very common. Most macroeconomists are still focusing mainly on how to stabilize the economy during the times when it’s already pretty stable. That approach may feel good to central bankers who want to congratulate themselves on a job well done, but the critics are right that it’s not going to help predict when danger is coming.
The problem is, multiple-equilibrium models might not help very much. Those models can describe the phenomenon of big, sudden shifts in the economy. But they also tend to be very fragile -- if you alter the model just a little, all the results can change dramatically.
But the final alternative -- moving away from formal, all-encompassing models -- means giving up on the dream of describing business cycles as one unified phenomenon. But maybe the critics are right. Maybe that dream has to go. As formal macroeconomic models have become more realistic, they’ve become nastier and less usable. Maybe their days are simply numbered.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Noah Smith at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com