SYSTEMICALLY IMPORTANT BANKS:A PERMUTATION TEST APPROACH

The framework of Systemically Important Banks (SIBs) was introduced by the financial stability board in the October of 2010 as the institutions "whose disorderly failure, because of their size, complexity and systemic interconnectedness, would cause significant disruption to the wider financial system and economic activity". The current methodology for their determination is based on balance-sheet variables and expert judgment. We propose a cross-sectional statistical procedure based on a permutation test in order to cluster SIBs separating them from the rest of the financial system. This procedure divides the sample in two subsamples choosing a quantile of suitable statistics of the considered variable, in order to reject the null hypothesis of equality in distributions. Our procedure will be applied to the European banking institutions, monitored by EBA, for which this regulator fully discloses information used in the choice of SIFIs done by the Basel committee. The analysis is done considering both single variables and through a weighted combination of them. The results obtained by the methodology we propose, taking into account properly the multivariate features of the decision process, reproduce those done by the Basel Committee for 2015 to identify the group of European Systemically Important Banks. Moreover these results, having a viable statistical methodology to select the SIBs, can open the possibility of extending the selection to a higher frequency framework, by applying the procedure to measure systemic risk, usually available daily.