Are commodity prices good predictors of inflation? The African perspective

This paper examines the predictive ability of commodity prices in forecasting inflation in two largest African economies. We consider an augmented Phillips curve model with symmetric and asymmetric commodity prices to forecast headline and core inflation using different methodological approaches. We found evidence of persistency, endogeneity and conditional heteroscedasticity in the predictors; thus, we apply the Westerlund and Narayan (2015) estimator which allow for these features in the predictive model. Consequently, we extend our model to account for asymmetries in commodity price changes. Our results show that commodity price indexes are good predictors for both headline and core inflation in both Nigeria and South Africa. Therefore, amidst all the competing multi-predictor framework, commodity prices outperform all other variants both in the in-sample and out-of-sample forecasts.