International Inflation's Predictive Ability

Abstract

In this paper we build forecasts for Chilean year-on-year inflation using simple time-series models augmented with different measures of international inflation. Broadly speaking, we construct two families of international inflation factors. The first family is built using year-on-year inflation of 18 Latin American (LA) countries (excluding Chile). The second family is built using year-on-year inflation of 30 OECD countries (excluding Chile). We show sound in-sample and pseudo out-of-sample evidence indicating that these international factors do help forecast Chilean inflation at several horizons. Incorporating the international factors reduce the Root Mean Squared Prediction Error of pure univariate SARIMA models statistically speaking. We also show that the predictive pass-through from international to local inflation has increased in the recent years. As a robustness check we construct another international inflation factor as an average of the inflation of fifteen countries from which Chile gets a high percentage of its imports. With the aid of this factor the models outperform our univariate benchmarks but also underperform the results obtained with the broader factors built with LA or OECD countries, suggesting that imported inflation is not the only channel explaining our findings.