Estimating Macroeconomic News and Surprise Shocks

RePEc Short ID: ppl48
ORCID iD: 0000-0003-0604-105X

Kilian, Lutz; Plante, Michael D.; Richter, Alexander W.

doi:10.24149/wp2304r1

Working Paper Revision

Estimating Macroeconomic News and Surprise Shocks

Abstract: The importance of understanding the economic effects of TFP news and surprise shocks is widely recognized in the literature. This paper examines the ability of the state-of-the-art VAR approach in Kurmann and Sims (2021) to identify responses to TFP news shocks and possibly surprise shocks in theory and practice. When applied to data generated from conventional New Keynesian DSGE models with shock processes that match key TFP moments, this estimator tends to be strongly biased, both in the presence of TFP measurement error and in its absence. This bias worsens in realistically small samples, and the estimator becomes highly variable. Incorporating a direct measure of TFP news into the VAR model (and adapting the identification strategy accordingly) substantially reduces the bias and RMSE of the impulse responses, regardless of whether there is TFP measurement error.

Keywords: Structural VAR; total factor productivity (TFP); news; Measurement error; max-share;

JEL Classification: C32; C51; C61; E32;

https://doi.org/10.24149/wp2304r1