Working Paper: CEPR ID: DP13765
Authors: Rgís Barnichon; Geert Mesters
Abstract: Despite decades of research, the consistent estimation of structural forward looking macroeconomic equations remains a formidable empirical challenge because of pervasive endogeneity issues. Prominent cases ---the estimation of Phillips curves, of Euler equations for consumption or output, or of monetary policy rules--- have typically relied on using pre-determined variables as instruments, with mixed success. In this work, we propose a new approach that consists in using sequences of independently identified structural shocks as instrumental variables. Our approach is robust to weak instruments and is valid regardless of the shocks' variance contribution. We estimate a Phillips curve using monetary shocks as instruments and find that conventional methods (i) substantially under-estimate the slope of the Phillips curve and (ii) over-estimate the role of forward-looking inflation expectations.
Keywords: structural equations; instrumental variables; impulse responses; robust inference
JEL Codes: C14; C32; E32; E52
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| structural shocks (E32) | inflation (E31) |
| structural shocks (E32) | output gap (E23) |
| monetary shocks (E39) | Phillips curve slope (E31) |
| monetary shocks (E39) | unemployment (J64) |
| forward-looking inflation expectations (D84) | inflation (E31) |
| output gap (E23) | inflation (E31) |