The instability of leading indicators in forecasting Austrian inflation: lessons from the COVID-19 pandemic and the energy crisis

This analysis tests 25 macroeconomic indicators for their ability to predict Austrian HICP inflation and evaluates three methods of combining these indicators into a composite forecast. The key findings are: First, for the evaluation period from early 2007 to the fourth quarter of 2023, competitors’ import prices, oil prices and domestic output prices for consumer goods are found to be the best individual leading indicators across various forecasting horizons (one, four and eight quarters ahead). Second, indicator performance varies over time. The forecasting performance of the output gap, for instance, declined considerably during the COVID-19 pandemic and the energy crisis, while that of other indicators like oil prices and competitors’ import prices improved. Third, for the period before 2020, composite indicators produced better forecasts than individual indicators over the entire forecasting horizon. This no longer holds when we include the pandemic and the energy crisis in the evaluation period. Then, two of the top three individual indicators, namely competitors’ import prices and domestic output prices for consumer goods, outperform combined indicators over the medium- and longer-term horizon (four and eight quarters ahead). Fourth, both individual and composite indicators outperformed autoregressive forecasts, especially in medium- and long-term predictions.