Monday, November 15, 2010

Irregular phenomena and the macroeconomics research agenda

Many see the Great Recession, as it is now called, as a dual crisis: an economic crisis and a crisis of economics, and more specifically macroeconomics. We have lived over the past twenty years or so through a period of remarkable economic stability, which also got a fancy name, the Great Moderation, and which gave us the illusion that this stability was to last. The fact that substantial recessions are still possible is a rude awakening, in particular because this one is worse than usual. And economists are the prime suspects because first they did not see it coming, and second they did not know how to react to it.

This is a view that is shared by Alessandro Vercelli who, like others, claims that macroeconomics has had a research agenda that was fundamentally flawed because it only studied regular phenomena, and not irregular ones. Indeed, the real business cycle agenda was centered around model economies in general equilibrium at all times, economies designed to replicate salient features of past data.

Hindsight is always 20/20. The research agenda should have focused on including more features about interbank relations, creation of new assets, moral hazard and adverse selection. But one has to understand that it is very difficult to think ahead what could happen and it is easy to criticize after the fact, and especially without offering alternatives. In fact, the DSGE agenda is remarkably well suited to address new situations: it is based on fundamentals, and these micro-foundations allow to study policies and situations not observed in history. This is something the previous agenda largely based on reduced forms could not address without considerable hand-waving (remember the Lucas Critique?). And if you look at the papers written nowadays, macroeconomics seems to have picked up the ball very nicely.

At first, macroeconomists did not have answers ready, or rather they did not have the answers that politicians wanted to hear, namely that something needed to be done. In the face of a crisis, every politician wants to do "something" to show "action". If the economist says that that one should let nature run its course, that one has to bite the bullet and let some banks fail, in particular so as to avoid future moral hazard risk, then the politician will bypass the economist and fall back on the first one that will satisfy him, and he is Keynesian.

Macroeconomics did not suddenly turn Keynesian, politics did. And doing so it compounded the problem and then lead to a crisis of Keynesian nature where nobody trusts anybody and aggregate demand is seriously lacking because nobody in the economy dares to invest due to huge policy uncertainties. Is the central bank still independent from the politicians or not? Seeing Bernanke and Paulson go hand in hand to testify to Congress was the worst possible image of this crisis. What is up with fiscal policy? Are the public deficits going to be taken care of through major tax increases or inflation? The policy prescription seems quite simple: decide once for all, should some courage to impose your policy and be done with it.

2 comments:

rosserjb@jmu.edu said...

EL,

This is your weakest criticism of a critic of DSGE you have come up with yet. Vercelli knows more in his small toenail than you will in another 20 years and makes excellent points you address with empthy sloganeering.

Asserting that DSGE is better because it is based on "fundamentals" and "microfoundations" when some of those are profoundly flawed, such as ratex, as Minsky well knew, the person Vercelli cites most heavily, is eyeball-rolling-inducing. The new stuff you and your sycophants push is very weak compared to Minsky. Vercelli makes excellent points not answered by either you or that new literature.

Kansan said...

Welcome back Barkley, it was boring here without you.

Minsky did not make anything useful in a quantitative sense. His theory includes a lot of hand-waving and nothing that allows to measure anything, let alone resolve ambiguities. In particular, it does not give any indications as to how large policy interventions should be, how large their effect would be, and whether second order effects could become important. The DSGE approach is trying to get at all these questions, how would you call this weak?