Money-Output Revisited: Time-Varying Granger Causality Evidence from Forty-Three Countries

We revisit the old issue of the causal link between money and output using time-varying Granger causality. Despite the proliferation of studies on this issue to which the evidence is heavily focused on the United States, a consensus on this topic has remained elusive. Using more recent data, we then re-examine this issue not just for the United States, but also for forty-two other countries. We find the existence of a robust and uni-directional Granger causality from money to output for thirteen countries with identified causal episodes that vary across these countries. We also detected from our implementation of the time-varying Granger causality test the period corresponding to the first half of the 1980s for the United States, a period which generated intense controversy and debate in this literature. However, we find this episode to exhibit feedback effects (bi-directional) between money and output. Given the significant changes that have transpired over time in the way that countries conduct and operate their monetary policy, our evidence reinforces a present and growing trend on the re-emergence of the important role of money in the economy.