Empirical assessment of alternative structural methods for identifying cyclical systemic risk in Europe

The credit-to-GDP gap, as proposed by the Basel methodology, has become the reference measure for the activation of the Countercyclical Capital Buffer (CCyB) due to its simplicity and good predictive power for future systemic crises. However, it presents several shortcomings that could lead to suboptimal decisions in many countries if it were used as an automatic rule for the activation of the CCyB. We study to what extent the purely statistical nature of the Basel methodology is responsible for these undesired effects by considering potential complementary credit gap measures that incorporate economic fundamentals. Specifically, we analyse the performance of two alternative (semi-) structural models that may account for these factors. We assess the proposed measures using time series data from the 70’s for six European countries and compare them to the Basel gap. We find that the proposed models provide more accurate early warning signals of the build-up of cyclical systemic risk than the Basel gap, as well as lower upward and downward biases after rapid changes in fundamentals. Nonetheless, results evidence heterogeneity in the ability from different models and specifications across countries to forewarn about future crises. This result evidences the differences in the financial cycles and their drivers across countries, and shows the importance in macroprudential policy of considering flexible approaches that adapt to national specificities.