I define the simplified version of the Sumner Critique as follows: If the Central Bank is targeting Nominal GDP precisely then all other macroeconomic effects become classical in nature.

This is obviously an extreme position as it requires not just omnipotence but omniscience and for lack of a better term omnibenevolence on the part of the Central Bank.

A useful way to interject the Sumner Critique into all of our reasoning, however, is simply to ask what the Central Bank Multiplier (CBM) is.

So, do your analysis in whatever framework you prefer and then when you get to the end attach a CBM to all of your final estimates. Under Scott Sumner’s vision the CBM for all aggregate demand moves is zero.

Thus if you imagine that the Fiscal Cliff for example will subtract 4% from GDP then you apply a CBM of 0 and you get that the Fiscal Cliff will subtract 0% from GDP or will have no net effect.

A less extreme version might place the CBM a .5, so that your traditional analysis says that the fiscal cliff subtracts 4% but after applying the CBM you get 2%.

Now, to be sure there will not be a single the “CBM.” We don’t want to get trapped in that intellectual cul-de-sac again.

The CBM you use will not only depend on the nature of your thought experiment but the framework that you used in the first place to get your original estimate. If you included no type of monetary feedback then the CBM will be larger smaller.

The CBM concept also allows us to usefully go all the way back to IS-LM. Since we are going to apply monetary reaction at the end we don’t need it in the base model.

I also suspect, but it would require some work, that just about everything that you would want in an inter-temporal model can either be collapsed into assumptions about the way the IS curve responds to expectations or into the CBM.

I don’t think you need any other explicit inter-temporal modeling. The simple reason is that periods in time must either “push-on” each other ether through their effects on real goods and services which will show up in the IS curve or through inter-temporal prices, which will be collapsed into the CBM. In short the CBM works by manipulating intertemporal prices so as to achieve the effect the central bank wants.

So, from an outside perspective it doesn’t matter what agents think inter-temporal prices will be, the CBM will force them to be what they need to be to make the CBM work.

That does of course mean that you have to do a lot of thinking when trying to figure out what the CBM will be, but once you have done that you don’t need to carry the baggage around with you everywhere. Just do the simple statics with real expectations imbedded in IS and then apply the CBM.

About these ads