Evaluating tail risks for the US economic policy uncertainty

Abstract

The goal of this paper is to employ a relatively new methodological approach to extract quantile-based economic policy uncertainty risk forecasts using the Quantile Autoregressive Distributed Lag Mixed-Frequency Data Sampling (QADL-MIDAS) regression model recommended by Ghysels and Iania (2018). This type of modelling delivers better quantile forecasts at various forecasting horizons. The forecasting results not only imply that the risk measure of economic policy uncertainty measure is linked to the future evolution of the index itself, but also it help constructing explicitly EPU risk measures, which are used to identify what drives such risk policy measures, especially across certain sub-sample periods associated with major global events, such as the collapse of the Lehman Brothers, the Trump’s election, and the trade-war tensions between the US and China. The findings offer a new empirical perspective to the existing economic policy uncertainty literature, documenting that special world events carry a strong informational content as being a primary key to understand the dynamics of the economic policy tails.