The paper studies the properties of the methodology for estimating the credit-to-GDP gap currently used by Banca d'Italia for making its decisions on the Countercyclical capital buffer (CCyB). To this end, the methodology is applied to a panel of 26 countries between 1971 and 2018. The study aims to assess in particular whether the positive properties of the methodology already shown for Italy in a previous research, are also confirmed for other countries.
The results show that the methodology used by Banca d'Italia mitigates some of the weaknesses of the methodology proposed by the Basel Committee on Banking Supervision. The use of different methods is allowed by the regulation.
In particular, the proposed methodology: 1) obtains more accurate estimates of the credit-to-GDP gap; 2) is a better predictor of financial crises; 3) increases the probability of the CCyB being at its maximum one year before the beginning of a crisis.