Credit-to-GDP gap calculation using multivariate HP filter

Periods of excessive credit growth can imply emergence of systemic financial stress which may result in financial crisis causing severe losses in the real economy. The base indicators of overheatedness in the credit markets are the expansion of the credit-to-GDP ratio and its deviation from its long-term trend, the credit-to-GDP gap. When calculating the latter, the major methodological challenge is to develop a model capable of executing the most reliable trend-cycle decomposition. This study presents a multivariate Hodrick-Prescott approach for the decomposition process, which defines the cycle with the inclusion of explanatory variables chosen by considering both statistical and economic selection criteria, successfully solving the problems raised by previous Hungarian research. The model also plays a role in the Hungarian macroprudential policy as in the future it will serve a basis for the calculation of the country specific, additional credit-to-GDP gap: one of the main quantitative factors influencing decisions regarding the countercyclical capital buffer (CCyB).