Policy Uncertainty in Japan

Abstract

We develop new economic policy uncertainty (EPU) indices for Japan from January 1987 onwards building on the approach of Baker, Bloom and Davis (2016). Each index reflects the frequency of newspaper articles that contain certain terms pertaining to the economy, policy matters and uncertainty. Our overall EPU index co-varies positively with implied volatilities for Japanese equities, exchange rates and interest rates and with a survey-based measure of political uncertainty. The EPU index rises around contested national elections and major leadership transitions in Japan, during the Asian Financial Crisis and in reaction to the Lehman Brothers failure, U.S. debt downgrade in 2011, Brexit referendum, and Japan's recent decision to defer a consumption tax hike. Our uncertainty indices for fiscal, monetary, trade and exchange rate policy co-vary positively but also display distinct dynamics. VAR models imply that upward EPU innovations foreshadow deteriorations in Japan's macroeconomic performance, as reflected by impulse response functions for investment, employment and output. Our study adds to evidence that credible policy plans and strong policy frameworks can favorably influence macroeconomic performance by, in part, reducing policy uncertainty. JEL Classification Numbers: D80, E20, E52, E62, F13

Cross-Category, Trans-Pacific Spillovers of Policy Uncertainty and Financial Market Volatility

“…To measure monetary, fiscal and trade policy uncertainty in the US and Japan, we use the newspaperbased EPU indices of Baker et al (2016) and Arbatli et al (2017), respectively. In order to construct these indices, both research teams scan local-language newspaper archives for articles that meet predefined keyword criteria.…”

“…In some cases, there are also non-negligible technical barriers. Arbatli et al's (2017) exchange rate policy uncertainty index, for instance, shows many isolated jumps, making it difficult to reconcile the index with the DY framework's assumptions -see the next section and JPN_FX in Appendix Figure A1. 7 The VIX index as the "most commonly used proxy for overall economic uncertainty" (Baker et al 2016(Baker et al , p. 1620 is not computed before 1990.…”

“…During the post-crisis years, various (political) events that created further economic uncertainty unfolded in both the US and Japan. Examples include the continued utilisation of untested policy instruments by the two countries' central banks, a number of partisan clashes over fiscal policy in the US, and the introduction of "Abenomics" in Japan -an ambitious reform package by PM Shinzo Abe that entailed fundamental changes to both monetary and fiscal policy (see, e.g., Baker et al 2016, Davis 2017or Arbatli et al 2017. Likely as a result, system-wide connectedness fluctuates around a relatively high level of about 58% between 2010 and 2016.…”

“…As shown in Panel e), for instance, the FX market is strongly affected by US TPU between 2013 and 2016 (up to -17.7). At the time, talks about the Trans-Pacific Partnership (TPP) agreement were well underway and, even before TPP was finally signed in early 2016, several political events had already cast doubt on the treaty's future ratification -see, e.g., Arbatli et al (2017). Similarly, between March and December 1995, the JYVIX's net spillover index in Panel f) falls by more than twenty points (from +6.9 to -15.9) before settling well into negative territory.…”

“…Moreover, since IV indices are widely regarded as proxies for overall economic uncertainty (see, e.g., Bloom 2009, Nodari 2014or Baker et al 2016, by including these measures we are also able to control for fluctuations in non-policy related uncertainty when investigating within-country or trans-Pacific EPU spillovers. 3 In short, we estimate a system of nine variables: Three newspaper-based measures of monetary (MPU), fiscal (FPU) and trade policy uncertainty (TPU) for the US and Japan, respectively (see Baker et al 2016 andArbatli et al 2017), two indices of option-implied stock market volatility, again one for each country, and one measure of foreign-exchange (FX) market volatility. Afterwards, we compute bilateral, multilateral, and system-wide measures of spillover intensity to characterise the variables' relationships.…”