Assessing the scoreboard of the EU macroeconomic imbalances procedure: (machine) learning from decisions

This paper uses machine learning methods to identify the macroeconomic variables that are most relevant for the classification of countries along the categories of the EU Macroeconomic Imbalances Procedure (MIP). The random forest algorithm considers the 14 headline indicators of the MIP scoreboard and the set of past decisions taken by the European Commission when classifying countries along the MIP categories. The algorithm identifies the unemployment rate, the current account balance, the private sector debt and the net international investment position as key variables in the classification process. We explain how high vs low values for these variables contribute to classifying countries inside or outside each MIP category.