Economics Gets Sucked Into Dark Corners
Olivier Blanchard, the International Monetary Fund's chief economist and a leading inventor of modern macroeconomic theory, is known as one of the field's gentler critics. In 2008 he wrote a famously ill-timed paper declaring that "The state of macro is good." The biggest financial crisis since the Great Depression followed soon after. In that paper, Blanchard presented New Keynesian models -- which he had a big hand in developing -- as the dominant approach; at the same time, two prominent macroeconomists at the Federal Reserve Bank of Minneapolis were releasing a paper declaring those models "not yet useful for policy analysis!"
So it's little surprise that Blanchard's latest critique of macro is a bit of a softball. Blanchard generally praises the "hundred flowers" that have bloomed in the field since the triumphs he praised in 2008 blew up. He is also happy that macro models have started to include finance (a bit belatedly, perhaps). And he believes that standard macro models can still provide policy makers with a lot of insight -- as long as we stay away from the "dark corners" where things like financial collapses cause the models to collapse as well:
If macroeconomic policy and financial regulation are set in such a way as to maintain a healthy distance from dark corners, then our models that portray normal times may still be largely appropriate. Another class of economic models, aimed at measuring systemic risk, can be used to give warning signals that we are getting too close to dark corners, and that steps must be taken to reduce risk and increase distance. Trying to create a model that integrates normal times and systemic risks may be beyond the profession's conceptual and technical reach at this stage...The main policy lesson is a simple one: Stay away from dark corners.
There's at least one theoretical problem with this idea: How do we know which standard model to use in normal times? For example, take the Smets-Wouters model, which is the main modern model used by central banks. How do we know the Smets-Wouters model is a good one? Well, people who like this model would say that it matches certain features of past business cycles -- the variance of gross domestic products, the correlation of GDP and investment, and the like.
But what if some of those past business cycles were caused by the "dark corners"? How can we use Smets-Wouters to describe normal times -- but not abnormal times -- when its parameters are fit to data that includes both normal and abnormal times?
There is also at least one practical problem with Blanchard's suggestion. Econ models -- unlike models in, say, physics -- don't come with any guide to when to use them. To take a well-known example from physics, it's easy to know when to use quantum mechanics -- you use it when things are very small or very cold. If all you want to do is put a man on the moon, you don't need QM.
Macro models aren't like this. Suppose you have a model of a "dark corner" -- the Diamond-Dybvig model of bank runs. When should a policy maker pay attention to this model? It isn't clear. Blanchard hopes that macroeconomists will develop good quantitative indicators of systemic risk, but so far that is just a hope.
But all of this discussion is -- quite literally -- academic. In reality, what macroeconomists say in their papers has very little effect on what policy makers actually do.
Take fiscal stimulus, for example. Do you think U.S. representatives and senators, or the president, paid any attention to the academic literature on stimulus? Surely some advisers, such as Larry Summers and Christina Romer, knew the models. But in the end, the size and type of stimulus was determined by politics, not by the best academic estimates of the parameters of the most popular dynamic stochastic general equilibrium model.
One might hope that monetary policy would be a bit different. The Fed is staffed with academically trained economists who know how to sling Smets-Wouters or draw on Diamond-Dybvig. But actually, you see the Fed relying a lot on older models that academic macro gave up on long ago. You also see the Fed relying on pure judgment, especially when assessing the likelihood of one of Blanchard's "dark corners."
Is this a problem with policy makers or with academia? Are policy makers ignoring the useful advice of trained experts? Or have the experts focused so much on exploring ideas -- on letting a hundred flowers bloom -- that they have neglected to think hard about how those ideas might be practically applied?
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