Nick Rowe writes,

What should the government do, if there is an increased desire to save, and the central bank is unable or unwilling to cut real interest rates enough to offset it? The answer is that the government should cut the growth rate of government spending, to shift the IS curve left, which raises the natural rate of interest (i.e. prevents it falling), because the IS curve slopes up when we have the growth rate of transitory income on the axis…

Fiscal policy in Old and New Keynesian models is even more different than John Cochrane thinks it is. And understanding why is hard.

Read the whole thing. It is graduate-school macro, which is harder than undergraduate macro.

I will just toss in a wrinkle, which is that in these sorts of models, sometimes the way to cut the growth rate of something is to increase its level immediately. So “cutting the growth rate of government spending” can mean immediately jumping to a high rate of government spending, from which you then have a slower growth rate.

There are various dichotomies in macro. The classical dichotomy separates nominal from real. The money supply only affects nominal variables (the price level) while leaving real variables (real wages, real GDP) unaffected. The crude Keynesian dichotomy is between prices and quantities. Instead of intersecting supply and demand curves, you have quantities affecting quantities. Spending creates jobs, and jobs create spending. Prices adjust to wages, and wages adjust to prices, but neither adjusts to employment or output.

Another dichotomy is between undergraduate and graduate macro. In undergraduate macro, fiscal policy works the way old Keynesian intuition says it works–by injecting spending into the economy. In graduate macro, fiscal policy works in these weird ways of twisting the path of desired consumption.

In communicating with other academics, a Keynesian economist will manipulate these dynamic optimization equations and derive a result that says that “fiscal policy works.” But when the economist communicates with undergraduates and other ordinary mortals, they hear a completely different story–basically Old Keynesian.

For the record, I think both stories have serious problems. The problem with the Old Keynesian story is that it rules out by assumption the operation of the adjustment mechanisms that we ordinarily take for granted, in which prices respond to excess supply and demand, and resources shift in response to profits and losses. However, I will still respect you in the morning if you say you are an Old Keynesian who believes that those adjustment mechanisms operate only very slowly in the real world. You can get from there to my PSST view by simply adding the proposition that throwing more M or G at the economy does nothing to speed up this adjustment process.

I have less respect for New Keynesians. (1) Although in a charitable mood I can try to pass a Turing test as a defender of the need for “microfoundations,” in my heart of hearts I believe that these representative-agent, dynamic optimization models are a useless exercise in mathematical masturbation. (2) In policy discussions, New Keynesians seem to gloss over how their models work, wave their hands, and deliver that Old-time Keynesian religion, hoping that nobody will call them out on it. Unfortunately for them, John Cochrane and Nick Rowe are around to try to keep them honest.

You should also read Brad DeLong and Paul Krugman on this topic. My sense is that they regard New Keynesians as useful idiots. But I think that they actually prefer something closer to Old Keynesianism. I would if I were them.