A comparative assessment of alternative ex ante measures of inflation uncertainty

The existence of unconventional monetary and fiscal policy arrangements in industrialized economies has been raising concerns about the future evolution of inflation rates ever since the onset of the financial and sovereign debt crisis in 2008. However, the question of how inflation uncertainty should be quantified is an open issue. We assess the informative content of alternative ex ante quantifications of inflation uncertainty by predicting ex post squared inflation forecast errors in an out-of-sample forecasting contest. We find that the average across distinct models' levels of ex ante uncertainty offers a greater predictive content than other uncertainty measures based on the cross-sectional variance of point forecasts, GARCH or stochastic volatility models.