To ask or not to ask? Bank capital requirements and loan collateralization*

We exploit the 2011 EBA Capital exercise, a quasi-natural experiment that required a number of banks to increase their regulatory capital. This experiment makes secured lending for the affected banks more attractive vis-a-vis unsecured lending, because secured loans require less regulatory capital. Using loan-level data covering the universe of bank loans in Portugal, we identify how banks require collateral on new loans when facing higher capital requirements: relative to the control group, treated banks require loans to be collateralized more often after the shock. We find the affected banks partially shield their relationship borrowers. The increased collateralization also has economically relevant real effects. Treated banks reallocate funds towards sectors with greater asset tangibility. Firms and sectors borrowing to a greater degree from treated banks exhibit lower growth and tilt their investments towards tangible assets. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

By utilizing the 2011 EBA Capital exercise, it was found that affected banks tend to require collateral on new loans and partially shield relationship borrowers. The increased collateralization has economically relevant real effects, leading to fund reallocation towards sectors with greater asset tangibility and a tilt towards tangible asset investments by firms and sectors borrowing from treated banks.