Policy Uncertainty and Foreign Exchange Rates: The DCC-GARCH Model of the US / Japanese Foreign Exchange Rate

Abstract

Since the breakdown of the Bretton Woods system in the 1970s, the US / Japan foreign exchange rate has been largely influenced by policy changes in the United States and Japan. This study applies the multivariate dynamic conditional correlation (DCC) – generalized autoregressive conditional heteroscedasticity (GARCH) models to analyze the time-varying effects of policy uncertainty, measured by the economic policy uncertainty (EPU) index of Baker et al. (2013, 2016), on the US / Japan foreign exchange rate. Using the EPU index as a proxy variable, it shows that the dynamic conditional correlations between policy uncertainty and the exchange rate are not time-invariant, but even sign-changing in the sample period. The analysis also empirically examined what drives the evolution of the time-varying correlations. The driving force of the correlations is, however, mostly attributed to unknown random factors.