Felix Geiger

A new working paper of mine, forthcoming in JPE: Macroeconomics, makes a simple point with data. The point is old and well understood, but easy to forget. When output is supply-constrained, or close to it, with a steep aggregate supply curve, a central bank can engineer rapid disinflation at little cost to output. Raising rates aggressively does nothing more than withdraw the excess aggregate demand. Many missed this during COVID and the recent energy shocks. A central bank that accommodates when output is supply-constrained is pushing on a string. The same withdrawal of excess demand can come through fiscal policy, and in some circumstances, that is the better tool. This was Keynes’ central insight in his 1940 essay, “How to Pay for the War.” My read is that many central banks drew the wrong lesson from the late 2000s and early 2010s. Expansionary monetary and fiscal policy did not raise inflation then, but only because output was demand-constrained. Accommodation buys output cheaply when demand is the constraint. It buys inflation when supply is. There is an equivalent reading through the FTPL. For space, we set it aside in the paper, but we are really making a statement about the real value of government debt. When output is supply-constrained, an excess of nominal claims chasing a fixed bundle of goods shows up as inflation: the price level rises until the real value of outstanding nominal liabilities equals the present value of the primary surpluses that back them. Disinflation then requires raising that present value, either through higher future surpluses or by retiring claims today, which is exactly what an absorption of excess demand does. Keynes’ compulsory saving scheme and the central bank’s rate hike become two routes to the same revaluation. An ungated copied can be found here: sas.upenn.edu/~jesusfv/State…