The wisdom of Nick Rowe
By Scott Sumner
Nick Rowe recently retired from teaching, and the Economist has provided a well-deserved tribute. During the past decade he’s the blogger that had the greatest influence on my views of macroeconomics. I’ve absorbed so many of his ideas that I’ve lost track of where my ideas end and his begin.
There’s probably no one in the blogosphere whose views are closer to mine on macro issues, and yet the field is so complex that even Nick and I frame fundamental issues slightly differently. Consider this paragraph:
Monetarists think the medium of exchange is distinctive for a variety of reasons. With any other good or asset, when people want more they must buy it. If they want more money, however, they can simply refrain from buying other things, a drop in spending characteristic of a recession. Similarly, if any other asset or good is in hot demand, its price will rise until the demand is quenched. But because everything is priced in money, it has no price of its own. It can rise in value only if the price of everything else falls, a deflationary pressure also characteristic of recessions.
The first half of the paragraph discusses money as a medium of exchange and the second half focuses on money as a medium of account. I don’t want to overdo that distinction; the medium of exchange and the medium of account are usually identical. But it does represent a subtly different way of framing the problem.
It seems to me that Nick’s model of a business cycle has two key components, sticky prices and bouts of excess demand for the medium of exchange. That model allows one to go directly from monetary shocks to fluctuations in real output.
I view that model as being superior to 99% of the other macro models out there, but nonetheless prefer a slightly different way of framing the problem. My approach is a sort of two-step process. I start with flexible prices, and then first try to model NGDP cycles. I argue that changes in the money market (medium of account) cause fluctuations in NGDP because the nominal price of money is fixed and hence the only way for the value of money to adjust is for NGDP to change. But it’s not yet a business cycle model, just an NGDP cycle model.
In stage two I add sticky wages. Now fluctuations in NGDP lead to similar changes in RGDP, but only in the short run (when wages have not yet fully adjusted.)
If we try to compare approaches, it’s surprisingly difficult to tease out the essential differences. And not just between Nick and I, but also between our views and standard New Keynesian models. Or Milton Friedman’s monetarism. Or Leland Yeager’s vision. Or Ralph Hawtrey’s vision. Or Irving Fisher’s. In some cases we are using different language to talk about the same underlying concept (monetary instability), or differentiating between things that might be different in principle but almost never are in practice, or we are disagreeing about the relative importance of wage and price stickiness, even while recognizing that each one matters.
A philosopher once said, “That which has no practical implications, has no philosophical implications.” One reason I believe our two approaches are quite close is that I can’t see any real world examples of where our two views have different practical implications. A second reason is that our critiques of non-monetarist models are often quite similar. (The don’t pay enough attention to money, they misinterpret interest rates, etc.)
PS. I was discouraged to read this:
Indeed, Mr Rowe attributes part of his success as a teacher to his shortcomings as a mathematician. He quotes Joan Robinson, another clear expositor of macroeconomics: “I never learned maths, so I had to think.” Because the answers did not leap out at him from the equations, he had to dwell on the economic behaviour underneath the algebra.
Shortcomings? Compared to me, Nick is a Field Prize candidate.