Economic policy uncertainty and equity markets : evidence from univariate models

Abstract

This work attempts to study the impact of economic policy uncertainty on stock markets, while investigating long-range dependence in the mean and the volatility of equity markets returns. A recent index, developed by Baker, Bloom, and Davis (2016), is intended to measure economic policy uncertainty (EPU) through a news-based approach. For this sake, EPU indices of the United States and Europe are incorporated in autoregressive fractionally integrated moving average (ARFIMA) and fractionally integrated general autoregressive conditional heteroscedastic (FIGARCH) contexts. Monthly returns on equity market indices for France, Germany, Italy, the Netherlands, and the United Kingdom in the period from early 1987 to late 2016/ early 2017 are investigated. While findings of the ARFIMA model estimation show no evidence of long-memory in returns on European equity markets, the impact of European and U.S. uncertainty on returns proves to be negative and significant. Results related to dual long-memory show that the ARFIMA-FIGARCH model doesnt support long-memory dynamics in all return series. However, long-memory behaviour in volatility, is found to be statistically significant for few equity markets under study, particularly the United Kingdom and Germany. Results also indicate that the impact of U.S. uncertainty is negative and statistically significant for most European markets under study. An extension to a set of emerging markets (namely the Gulf Cooperation Council countries), included as a subsection, finds no significant impact of U.S. uncertainty on the GCC stock markets returns and their volatility.