Impact of higher capital buffers on banks’ lending and risk-taking: evidence from the euro area experiments

We study the impact of higher bank capital buffers, namely of the Other Systemically Important Institutions (O-SII) buffer, on banks' lending and risk-taking behaviour. The O-SII buffer is a macroprudential policy aiming to increase banks' resilience. However, higher capital requirements associated with the policy may likely constrain lending. While this may be a desired effect of the policy, it could, at least in the short-term, pose costs for economic activity. Moreover, by changing the relative attractiveness of different asset classes, a higher capital requirement could also lead to risk-shifting and therefore promote the build-up (or deleverage) of banks' risk-taking. Since the end of 2015, national authorities, under the EBA framework, started to identify banks as O-SII and impose additional capital buffers. The identification of the O-SII is mainly based on a cutoff rule, ie. banks whose score is above a certain threshold are automatically designated as systemically important. This feature allows studying the effects of higher capital requirements by comparing banks whose score was close to the threshold. Relying on confidential granular supervisory data, between 2014 and 2017, we find that banks identified as O-SII reduced, in the short-term, their credit supply to households and financial sectors and shifted their lending to less risky counterparts within the non-financial corporations. In the medium-term, the impact on credit supply is defused and banks shift their lending to less risky counterparts within the financial and household sectors. Our findings suggest that the discontinuous policy change had limited effects on the overall supply of credit although we find evidence of a reduction in the credit supply at the inception of the macroprudential policy. This result supports the hypothesis that the implementation of the O-SII's framework could have a positive disciplining effect by reducing banks' risk-taking while having only a reduced adverse impact JEL Classification: E44, E51, E58, G21, G28