Before taking a week off for summer holidays, quick note on two recently accepted papers at the Review of Finance. While my tenure at the RF ended more than 18 months ago, I am still handling papers originally submitted and invited for resubmission
during this period. And it is always a pleasure to see papers finally make it to the acceptance stage, significantly improved and with important conclusions. And as the coincidence has it, both papers concern banks’ risk-taking behaviour
in the securitization market.
In Arbitraging the Basel Securitization Framework: Evidence from German ABS Investment, Mattias Efing shows
how German banks use regulatory arbitrage for additional risk-taking. Specifically, he uses data on ABS investments on the balance sheets of German financial institutions between 2007 to 2012 and shows that banks go for higher-yielding (but therefore
also higher risk) securities within the same credit rating (and thus capital risk weight) class. This effect is especially strong for more capital-constrained banks (not quite betting the bank behaviour, but along the same lines).
Revenge of the Steamroller: ABCP as a Window on Risk Choices, Carlos Arteta, Mark Carey, Ricardo Correa and Jason Kotter explore banks’ incentives to upload on tail risk through
bank-sponsored asset-backed-commercial papers. These securities offer relatively low but safe returns during normal times, but large negative returns during distress times due to maturity mis-match and reliance on wholesale funding. The authors find an interesting
interaction between governance and government-provided financial safety net (also known as bail-out expectations). On the one hand, banks with managers that had better aligned incentives with shareholder returns were on average less likely to become sponsors.
However, for such banks, which also operated with very high implicit government guarantees, this relationship reverses and such banks are more likely to sponsor such vehicles. Another important reminder that assessing the governance of banks is incomplete
without considering the role of government (guarantees).