Many New Keynesians point out that there is a class of flexible inflation targeting rules that perform better (from a welfare perspective) than NGDP targeting. This is true. In a previous post I pointed out that I agreed with Woodford’s claim that NGDPLT is not in general the “optimal” policy rule. After reading some comments I realize that I need to amend that post. That’s because I don’t accept the validity of the New Keynesian models that show flexible inflation targeting is superior to NGDPLT, and should not have implied that I did. I view a completely different rule—nominal wage targeting—as being the theoretical ideal.
[My first line was poorly worded--I meant "a class of models with flexible inflation targeting rules that perform better . . .]
I just noticed that George Selgin left a comment that nicely characterizes my objection to new Keynesian monetary models:
I think that Scott and others are granting far too easily the “nominal GDP targeting isn’t optimal but it’s a good pragmatic solution” argument. Having looked closely at Woodford’s work, and at other work in the same vein, I find that somewhere there is always a loss function, not always implicit, that treats fluctuations in P as “bad,” but without supplying any compelling justification for doing so. Instead, there’s some rather transparent hand-waving that amounts to saying that this treatment itself is “not optimal but pragmatically justified” or something like that. (The same, by the way, can be said of Wicksell’s own argument for treating zero inflation as optimal–that is, for equating a policy that achieved it with one that kept interest rates at their “natural” levels. In fact his argument here goes through only if productivity growth is constant, which of course it usually isn’t.)
Of course, it’s almost certainly true that no particular NGDP (or nominal income) target is “optimal.” But it doesn’t follow that it isn’t at least as optimal as the usual alternatives. And there are now a number of studies, DSGE and otherwise, supporting this conclusion.
I agree with George 100%. I’ve often argued that the supposed “welfare costs of inflation” are better thought of as the welfare costs of volatile or excessive NGDP growth. This also has implications for the supposed “lack of models” in the market monetarist camp. I believe that I have provided models, but that other economists don’t recognize them because they don’t know what a model is. They think (wrongly) that a model is a bunch of equations. Don’t get me wrong; equations can be useful. I once published a model with equations showing the optimality of wage targeting. Another time I did the same for NGDP futures targeting. But in all truthfulness the equations in those papers merely demonstrated what was already intuitively obvious given the model’s assumptions.
Here’s what a lot of economists don’t realize. We aren’t yet at the stage where mathematical models can show which policy is best. We simply don’t know enough about the welfare costs of inflation. Obviously if you think the CPI is a good proxy for the welfare costs of inflation, then inflation is likely to play a role in your optimal policy rule. And if you think (as George and I do) that NGDP is a better proxy for the welfare costs of inflation then NGDP may play a role in your policy rule. That’s just common sense.
In the 1970s Milton Friedman argued that macroeconomics had advanced beyond Hume in one respect only—we now know how to deal with the first derivative of nominal changes. In the 1920s economists debated macro issues using words, not equations. There’s something to be said for going back to that approach.