Credit to GDP gap as an indicator for upcoming financial crisis

According to the Basel III Framework, the gap between Credit to GDP Ratio and its long-run trend is the single best indicator for setting the Countercyclical Capital Buffer (CCB). The aim of setting the CCB is to increase the Capital Adequacy Requirement (CAR), in order to increase the resilience of the banking system in the case of upcoming financial difficulties. For calculating the long run trend of the Credit to GDP Ratio, the Basel Committee suggests to use the Hodrick-Prescott (HP) filter. In order to detrend the Credit to GDP Ratio, the HP filter only relies on the Credit to GDP Ratio itself and does not take into account other variables, which may be relevant to the risks to financial stability. Economic theory immediately suggests the Real Gross Domestic Product and the Real Estate Price Index as these relevant variables. During periods of negative Real Gross Domestic Product and Real Estate Price Index gaps, a high Credit to GDP Gap is less dangerous than is indicated by the HP filter. The reverse is true when gaps of these two variables are positive, that is a high Credit to GDP Gap is more dangerous for the financial system and the economy than is indicated by the HP filter. The present paper provides a theoretical and empirical justification of using Real GDP and Real Estate Price Index gaps in the process of detrending Credit to GDP ratio. Since the HP filter cannot work with different variables simultaneously, the paper introduces the Kalman filter as a solution. Comparing credit to GDP gaps calculated using different filters, the paper shows two cases when the Kalman filter outperformed the HP filter in Georgia between the years 2000 and 2016. The first case is the financial crisis of 2007-2008, during which the HP filter could only signal that a crisis was occurring after the fact, while the Kalman filter could work as an early warning indicator, informing about an upcoming crisis in the beginning of 2006. The second case is the first half of 2016, when the HP filter suggested to set the CCB while there was no financial crisis, which was correctly indicated by the Kalman filter.