That was the topic of a conference organized by the Center for Free Enterprise at West Virginia University that took place on Saturday, April 25. The conference was divided in two sessions: one where theoretical aspects of NGDP were discussed and another that took a more empirical approach to the matter. I presented in the second session on how to spot if NGDP targeting is, in fact, too loose.
Besides the presence of Scott Sumner, probably the best known proponent of NGDP targeting, other presenters included Thomas Hogan, Alexander W. Salter, Ryan Murphy, Joshua Hendrickson, Robert Lester, and Vipin Veetil. While all papers endorsed NGDP targeting in one way or another, at least as a superior norm to other principles such as price stability or the Taylor rule (Hendrickson and Lester), the papers also focused on either if this holds under a free banking system (Salter) or if the rule could put the economy in an inferior equilibrium (Hogan). Even though I found all presentations interesting, I’ll briefly comment on just two of them.
Salter and Young build on a free banking model by Selgin to explore if under free banking NGDP would be stabilized. They find that NGDP is stable in the case of a demand shock, but that if there is no price stickiness, then in the case of a supply shock part of the shock is absorbed through price stability. This means that free banking can automatically switch between NGDP targeting and price stability, and this is a very hard mechanism to replicate by a central bank. I find this outcome a little less surprising than the authors for two reasons. First, what is, in theory, stabilized is NGDP per capita (or per unit of factor of production). This means that if the supply shock is due to more factors of production, then total NGDP would increase. Second, a “menu cost” effect. If there is a positive supply shock, then if special conditions hold, there are less final prices that would need to be adjusted if the price level is allowed to fall than if total intermediate price should adjust if the price level of final goods is to remain stable. While I don’t think there is an explicit treatment of this, it seems that the outcome in a free banking system would be a combination of NGDP stable and price stability depending on the relative “menu costs” of final and intermediate prices. Interestingly, Selgin’s does not account, at least explicitly, for these two assumptions (he does discuss them somewhere else). It seems that there is a possibility to reconcile Salter and Young’s model with these two principles.
Veepil and Wagner’s paper was a (friendly) criticisms of NGDP Targeting. Their argument is that NGDP is a spontaneous outcome of multiple individuals interacting (think of a complex agent based model), but that the targeting of NGDP cannot guarantee economic efficiency at the microeconomic level. Namely, one cannot impose “equilibrium” by targeting NGDP, NGDP needs to be the outcome of market coordination. By targeting NGDP, noise is added to the market (ie “cantillon effects”) and therefore NGDP is chasing a “mirage.” I agree with this diagnosis, but the negative effects pointed out by this paper may not be as serious as the alternative. The alternative is not a free banking system (that monetary reform is not feasible in the short-run, medium-run, and probably neither in the long-run). The alternative is a different policy by the central bank. Now, NGDP Targeting is focused on maintaining monetary equilibrium. Let’s say, for the sake of argument, that this is the right policy. What we have, then, is a central bank that aims at the right target but misses. The error of missing the target is still expected to be less than the error of not aiming at the right target in the first place.
It is noteworthy that a conference on this particular topic took place and that all papers presented where interesting and bringing something new to the topic, rather than just rephrasing what we already now about NGDP targeting. Let’s hope that this type of sound money discussion continues in the future.