Working Paper: NBER ID: w15459
Authors: John H. Cochrane
Abstract: Bennett McCallum (2009), applying Evans and Honkapohja's (2001) results, argues that "learnability" can save New-Keynesian models from their indeterminacies. He claims the unique bounded equilibrium is learnable, and the explosive equilibria are not. However, he assumes that agents can directly observe the monetary policy shock. Reversing this assumption, I find the opposite result: the bounded equilibrium is not learnable and the unbounded equilibria are learnable. More generally, I argue that a threat by the Fed to move to an "unlearnable" equilibrium for all but one value of inflation is a poor foundation for choosing the bounded equilibrium of a New-Keynesian model.
Keywords: No keywords provided
JEL Codes: E0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| bounded equilibrium not learnable (C62) | agents cannot observe monetary policy shock (E19) |
| agents cannot observe monetary policy shock (E19) | agents cannot accurately estimate parameters for learning (C51) |
| unbounded equilibria are learnable (C73) | agents can infer information from observable variables (C20) |
| bounded equilibrium (t = ut + et) being unlearnable (C62) | inability of agents to observe (et) directly (D89) |
| unbounded equilibria allow for learning through observable dynamics (C73) | agents can learn (L85) |