Disentangling supply-side and demand-side effects of uncertainty shocks on U.S. financial markets : identification using prices of gold and oil

This paper investigates the effects of uncertainty shocks on selected U.S. financial asset prices by decomposing a traditional uncertainty shock into its supply-side and demand-side components. Following the approach by Piffer and Podstawski (2018), we identify uncertainty shocks using the price of gold and enhance this strategy by introducing the price of oil as a second variable. By examining daily price changes during significant events that trigger uncertainty, we provide evidence suggesting that despite an increase in gold prices, supply-side uncertainty shocks (e.g. armed conflicts or natural disasters) tend to result in higher oil prices, while demand-side uncertainty shocks (e.g. political and economic events) lead to declining oil prices. By exploiting this information with help of sign restrictions, we create two proxy variables and estimate Bayesian Vector Autoregression (BVAR) models to identify supply-side and demand-side uncertainty shocks. Our findings indicate that while gold prices alone can identify uncertainty shocks for most variables, the inclusion of oil prices reveals an additional dimension. The effects of these shocks differ in their impact on inflation expectations and may thus be a potential source of price puzzles if only the price of gold is considered.