Monetary policy shocks identified using the entire yield curve: an alternative approach

ABSTRACT

Conventional monetary policy (MP) shocks are typically identified as unexpected changes in the short end of the yield curve. During the zero lower-bound period, central banks attempt to provide stimulus by operating on the long end of the yield curve with new tools, such as forward guidance and large-scale asset purchase. This implies that the identification of MP shocks should be done throughout the entire yield curve. This study proposes an alternative approach to analysing the effects of MP shocks, using factor-augmented vector autoregression (FAVAR) of monthly data. We therefore used information from large datasets and made changes over the entire yield curve as shocks to the VAR, which were identified using external high-frequency instruments. Furthermore, empirical analyses using economic and financial variables showed that contractionary yield curve changes decrease the inflation rate and increase risk aversion and uncertainty, which is consistent with conventional wisdom.

KEYWORDS:

• Monetary policy effects
• risk aversion
• uncertainty
• external instruments
• factor-augmented vector autoregression

JEL CLASSIFICATION:

• E44
• E52
• E58

Acknowledgement

We would like to thank Mark Taylor (editor) and the anonymous referee for the helpful comments and constructive suggestions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ Mertens and Ravn (2013) considered only the case of $k=s=2$.

² We are grateful to Eric Swanson for sharing the decomposed factor data.

³ The Fed started announcing target rate changes from February 1994.