Title: Uncertainty and economic activity: Identification through cross-country correlations
Authors: Ambrogio Cesa-Bianchi - Bank of England (United Kingdom) [presenting]
Hashem Pesaran - University of Southern California (United States)
Alessandro Rebucci - Johns Hopkins (United States)
Abstract: A multi-country approach is proposed for the analysis of the interaction between uncertainty and economic activity both within and across economies. Two global factors, labeled real and financial, are considered and, assuming different patterns of cross-country, correlations of country-specific innovations to output growth and stock market volatility are identified. It is assumed that only the real factor is required to model cross-country correlations of growth innovations, but both factors are needed to model cross-country correlations of volatility innovations. These assumptions are shown to be in accordance with stylized facts of the data and the estimated innovations. The identified factors are then used in a factor-augmented VAR to quantify the impact of common and country-specific shocks. It is found that most of the unconditional correlation between volatility and growth can be accounted for by the two common factors. While unconditionally volatility behaves countercyclically, once we condition on the factors the growth-volatility correlation becomes essentially negligible. The share of volatility variance explained by the real factor and by country-specific growth shocks is shown to be relatively small. Similarly, shocks to the financial factor explain only a small fraction of the country-specific growth variance. Finally, country-specific shocks are important for domestic volatility and growth, but have limited spillover effects to other countries.