Do monetary indicators lead euro area inflation?

This paper assesses based on euro area data whether the strong in-sample long-run link between money growth and inflation that has been highlighted by a number of recent empirical studies is exploitable to predict inflation out of sample. The results from standard bivariate forecasting models suggest that no monetary (nor any other) single indicator significantly outperforms a simple benchmark forecast. The further analysis shows that it would be premature however to discard based on such evidence the usefulness of monetary (and all other) indicators. First, it is shown that based on judgemental adjustments to monetary indicators correcting for the effects of identifiable, persistent velocity shifts, it appears possible to significantly improve predictive ability in real time. Second, I find that a factor forecasting model combining monetary and economic indicators, which can be regarded as a generalized quantity theory- or two-pillar Phillips Curve-forecasting model, delivers a fairly good and stable forecasting performance.