Krugman's Optimal Fiscal Policy Illustrates Everything That Is Wrong With Modern Macroeconomics

Monday December 29, 2008

Paul Krugman on the need to model the current situation:

One thing that’s been bothering me about the discussion over fiscal stimulus is the virtual absence of fully worked-out models, with all their t’s dotted and eyes crossed, or something. Not that a rigorous model is always better than a rough-and-ready but more realistic approach, but I like to have both on hand. So I’ve tried a very rough sketch of a full, intertemporal maximization yada yada analysis of the fiscal policy issue...

I don't disagree that having a model is helpful. And I highly recommend that anyone interested in macroeconomics read Krugman's paper; a PDF is available here.

Having studied at Rochester and being primarily a micro guy, I must admit that I am not an expert on the New Keynesian Economics literature. But I can, however, recognize the limitations of an economic model.

Krugman constructs a model where "consumers maximize an intertemporal utility function" that is strictly increasing in consumption, quantity of money held, and government spending. Seems reasonable enough. Government spending is paid for by lump-sum taxes, which is enough to throw the model into doubt, but I have seen worse. But there is one other variable in there I find puzzling:

I’ve introduced a shift factor s on consumer demand, as a crude way to allow for the kind of negative demand shock we’re experiencing right now.

With that variable, Krugman can show that:

But what if the required nominal rate is negative? In that case monetary policy can’t get you there: once the interest rate hits zero, people will just hoard any additional cash – we’re in the liquidity trap.

And thus monetary policy is ineffective.

But what caused this s to shock downward in the first place? The model does not address this. Why would, say, a helicopter drop of money with an expiry date, placebo (like getting everyone to wear a button), or other non-standard monetary policy not increase s? We have a variable that we cannot see, cannot measure, do not fully understand, but is driving the entire result of the model.

None of this is unique to Krugman or New Keynesianism. In almost any macroeconomic model that is both mathematically tractable and gives reasonable sounding results, unreasonable assumptions and questionable concepts lie underneath the hood.