Alice: A New Inflation Monitoring Tool

The ability to anticipate future inflation developments and the understanding of its driving factors are of the greatest value. Inflation forecasting is an important part of the broader economic analysis, albeit it is well known that inflation is inherently hard to forecast. While most commonly used approaches aim at quantitative inflation forecasts, this study focuses on (composite) leading indicators and qualitative inflation signals, which is a complementary tool to gauge future developments in inflation. This method is also largely statistical but it aims to predict turning points in the inflation cycle rather than directly forecasting the level of inflation. The aim is to provide early-warning signals with respect to the direction of the movements in inflation. We make several important contributions to the existing literature. Firstly, this is the first study to construct a composite leading indicator for both the headline as well as core inflation cycle. Secondly, the sample period goes well beyond the late 1990s and includes the Great Recession as well as the euro area debt crisis. This allows taking advantage of actual euro area data in addition to “synthetic” euro area data that in part go back to the 1960s. The third contribution is a careful analysis of over 150 potential leading series covering different parts of the economy in order to select component series for the leading indicators. Finally, we provide a pseudo real-time evaluation of the performance of the two constructed indicators. The current paper takes a non-model based approach to construct the composite leading indicator of the euro area inflation cycle. We apply the deviation cycle definition with respect to the euro area inflation rate. The random walk filter by Christiano and Fitzgerald (2003) is employed in the current paper to obtain the cyclical components of the reference series. We remove from the series the frequencies that are higher than 12 months and lower than 120 months. This choice is in line with the OECD system of composite leading indicators for the business cycle that is based on the double HP with 12-month and 120-month lower and upper limits for frequency bands. The two constructed Area-wide Leading Inflation CyclE (ALICE) indicators for the euro area headline and core inflation consist of nine and, respectively, seven leading series, with a lead time between 3 and 25 months. The leading series have a broad economic coverage, ranging from external factors, prices and costs measures, economic activity variables, “soft” survey data, financial variables, and market-based inflation expectations. The headline and core ALICE identify ex post major cyclical movements in inflation quite well, especially since 1999. A pseudo real-time analysis confirms these findings and shows that the ALICE for both headline and core inflation perform well, i.e. they indicate turning points of the reference cycle in advance, and do not suffer from major revisions over time. Thus, these indicators appear to have a potential to be useful for the real-time monitoring, analysis and forecasting of inflation developments in the euro area.