Working Paper: NBER ID: w21738
Authors: Pooyan Amir Ahmadi; Harald Uhlig
Abstract: We propose a novel identification strategy of imposing sign restrictions directly on the impulse responses of a large set of variables in a Bayesian factor-augmented vector autoregression. We conceptualize and formalize conditions under which every additional sign restriction imposed can be qualified as either relevant or irrelevant for structural identification up to a limiting case of point identification. Deriving exact conditions we establish that, (i) in a two dimensional factor model only two out of potentially infinite sign restrictions are relevant and (ii) in contrast, in cases of higher dimension every additional sign restriction can be relevant improving structural identification. The latter result can render our approach a blessing in high dimensions. In an empirical application for the US economy we identify monetary policy shocks imposing conventional wisdom and find modest real effects avoiding various unreasonable responses specifically present and pronounced combining standard recursive identification with FAVARs.
Keywords: Bayesian FAVAR; Dynamic Factor Models; Identification; Sign Restriction; Gibbs Sampling; Monetary Policy Shocks
JEL Codes: C22; E52
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Contractionary monetary policy shock (E49) | federal funds rate increases (E52) |
| Contractionary monetary policy shock (E49) | prices decrease (E31) |
| Contractionary monetary policy shock (E49) | nonborrowed reserves decrease (E51) |
| Contractionary monetary policy shock (E49) | M1 decrease (E51) |
| Contractionary monetary policy shock (E49) | industrial production decreases (L16) |
| Contractionary monetary policy shock (E49) | real GDP effect less pronounced (F69) |
| Contractionary monetary policy shock (E49) | employment decreases (J63) |
| Contractionary monetary policy shock (E49) | total unemployment shows initial puzzling drop (J64) |
| Contractionary monetary policy shock (E49) | yield spreads initially fall (E43) |