Frequency-dependent real-time effects of uncertainty in the United States: evidence from daily data

ABSTRACT

In this paper, we analyse the impact of uncertainty shocks at the daily-frequency on key macroeconomic variables for the United States. In doing so, we use a vector autoregressive (VAR) model, including the inflation rate, a real-time measure of economic activity and a measure of monetary policy as endogenous variables and decompose uncertainty effects into short, medium and long-term based on a discrete-time Fourier transformation. Aggregate results (prior to decomposition) show that an increase in economic uncertainty has a significant expansionary impact on monetary policy. However, when we decompose uncertainty into its short-, medium- and long-run components, we find that economic activity is affected negatively in a statistically significant manner to shocks in low-frequency uncertainty, while, statistically significant monetary expansion is observed under shocks to relatively high frequencies of uncertainty.

KEYWORDS:

• Uncertainty
• frequency-dependence
• daily data

JEL CLASSIFICATION:

• C32
• E31
• E32
• E52

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The data is available for download from: https://www.philadelphiafed.org/research-and-data/real-time-center/business-conditions-index.

² The data is downloadable from: http://www.thebillionpricesproject.com/.

³ We also replaced the SSR with the Expected Time to Zero (ETZ), and the Effective Monetary Stimulus (EMS), as alternative measures of monetary policy developed also by Krippner (2013). However, our main results were qualitatively similar, and complete details are available upon request from the authors.

⁴ The SSR is available for download from: https://www.rbnz.govt.nz/research-and-publications/research-programme/additional-research/measures-of-the-stance-of-united-states-monetary-policy.

⁵ Complete details of the unit root tests are available upon request from the authors.

⁶ Alternative ordering, whereby the overall EPU or the decomposed versions of the same were ordered last as in Colombo (2013), did not qualitatively affect our results (of course barring the first period), complete details of which are available upon request from the authors.