Author(s): Ferreira-Lopes, Alexandra ; Linhares, Pedro ; Martins, Luís Filipe ; Sequeira, Tiago Neves
Date: 2021
Persistent ID: http://hdl.handle.net/10316/98787
Origin: Estudo Geral - Universidade de Coimbra
Subject(s): Quantitative easing, Bank of Japan, monetary policy, economic growth, meta-probit
The Bank of Japan is the central bank that has been using unconventional monetary policies (UMP) for the longest time. We present an original meta-probit analysis for the period 2001-2020, using 45 studies, which focuses on the literature that has been studying the effect of those policies on Japanese output growth. We use impulse response functions in Vector Auto Regression (VAR) type models to assess if the effects on output of UMP are significant (or not), and if significant, positive or negative. Funnel asymmetry and precision effect tests do not provide convincing evidence of publication bias. Additionally, we also do not find a consensus regarding the output growth effects during the quantitative easing years. A thorough meta-probit analysis suggests that certain variables included in the sample such as industrial production (as a proxy for output) and the price level – as well as different specifications in the data used – have a greater effect on the probability of reporting statistically significant (positive) effects of quantitative easing on output growth. Additionally, other variables, like if it is an older study, if the study includes several quantitative easing (QE) programs, if it is published in a journal, if it is using monthly or daily data, if it is using one standard deviation in the confidence intervals, if the study does not use a monetary supply or a money base variable, but if the shock is applied to the monetary base, also increase the probability of reporting statistically significant (positive) effects of quantitative easing on output growth. However, one of the most important (policy) implications of our study is that until now the evidence for a significant real effect of unconventional monetary policy is weak, since concerning true effects, we obtain mixed results.
