A New Approach to Identifying the Real Effects of Uncertainty Shocks

ABSTRACT

This article introduces the use of the sign restrictions methodology to identify uncertainty shocks. We apply our methodology to a class of vector autoregression models with stochastic volatility that allow volatility fluctuations to impact the conditional mean. We combine sign restrictions on the conditional mean and conditional second moment impulse responses to identify financial and macro uncertainty shocks. On U.S. data, we find stronger evidence that financial uncertainty shocks lead to a decline in real activity and an easing of the federal funds rate relative to macro uncertainty shocks. Supplementary materials for this article are available online.

KEY WORDS:

• Multivariate stochastic volatility
• Sign restrictions
• Uncertainty
• Vector autoregression
• Volatility-in-mean
• Wishart process