What are the effects of a negative AS shock & how should the central bank respond?
THREAD, partly based on a just-published paper of mine (AEJ Macro April 2020)
https://www.aeaweb.org/articles?id=10.1257/mac.20180207
(1/7)

In macro we often think of AD shocks as pushing inflation & output in the same direction but AS shocks as pushing them in opposite directions. For example, a negative AS shock is thought to trigger a recession + inflationary pressures. This is in general WRONG
(2/7)

Why? This confuses impulse vs. propagation. The impulse is the shock itself, the propagation is the general eq’m effect. The shock may directly affect AS, but the feedback to AD may be even stronger than the shock itself
(3/7)

This typically occurs in models with imperfect (unemployment) insurance and time-varying precautionary savings. A <0 AS shock that persistently raises unemployment risk triggers a drop in consumption demand, hence in AD, that may cause BOTH inflation and output to fall
(4/7)

How should the CB respond to such shocks? Well, in a very similar way as with a negative AD shock: lean against deflation, and doing so will also limit the damage on output & employment. The strong AD feedback of the AS shock implies that there is less of a policy tradeoff
(5/7)

In fact, in my paper I find that under imperfect insurance policy rates should be cut after negative AS shocks affecting productivity or production costs (the opposite is true under perfect insurance). Here is the figure for a <0 productivity shock
(6/7)

A recent paper by @VeronicaGuerri7, @guido_lorenzoni, @IvanWerning & @ludwigstraub looks at a different feedback from AS to AD, and with a proper “covid shock”
( https://economics.mit.edu/files/19351 ). They also look at fiscal policy.
(7/7)