What makes Monetary Policy More Powerful? A Big Data Approach

A large literature on monetary policy transmission has emphasised many potential non-linear drivers. Transmission strength has been shown to vary with the business cycle, financial frictions, and uncertainty, amongst other mechanisms. However, most studies focus on a small number of channels at a time. Our study takes a “big data†approach to non-linearity, allowing us to rank the relative importance of a wide range of non-linear transmission channels. We focus on asset price responses to central bank statements. Using a large, mixed-frequency macro-financial dataset, we show that US monetary policy surprises feature a small number of important non-linear drivers, relating especially to financial variables. Monetary transmission to long-term interest rates is weaker at times of high interest rates and high credit growth, consistent with diminishing effectiveness over a hiking cycle. While non-linearity over the interest rate cycle appears to dominate other channels, we find some evidence that monetary policy is weaker in booms. Using euro area data, we find a prominent role for sovereign risk in state-dependent transmission.