Liquidity regulation and bank funding costs
Summary
Focus
Bank regulators argue that prudential rules can partly pay for themselves: safer banks should face lower funding costs as creditors demand a smaller risk premium. The Liquidity Coverage Ratio (LCR), introduced under Basel III, requires banks to hold enough liquid assets to cover a 30-day stress scenario. Whether it has actually made banks' wholesale funding cheaper has been difficult to assess, as several post-crisis reforms arrived as a package and detailed data on bank funding costs are scarce. This paper studies how the LCR affected the cost and maturity of US banks' borrowing from money market funds (MMFs).
Contribution
We provide causal evidence on how the LCR affects what banks pay for funding and the maturity structure of their liabilities. We use detailed data on the short-term instruments through which banks borrow from MMFs (certificates of deposit, commercial paper, asset-backed commercial paper and repurchase agreements) from February 2011 to December 2015. This lets us compare how the same fund prices loans in the same month to banks that are subject to the LCR but had different liquidity holdings before the LCR came into effect (their effective "LCR gap").
Findings
Banks with a larger pre-regulation LCR gap saw a markedly steeper fall in funding costs after the LCR was introduced. A one standard deviation larger gap is associated with funding costs that are around 8% lower, or roughly 2.5 basis points at the sample mean of 31 basis points. The fall is more pronounced for longer-maturity instruments: about twice as large at maturities of 31 days or more as at one day. More exposed banks also borrow larger amounts and tilt their borrowing towards longer maturities. The results support the view that the LCR, by making banks safer, lowers the risk premium creditors demand, partly offsetting the costs of compliance.
Abstract
We establish a causal link between liquidity regulation and a lower cost of bank wholesale funding. For identification, we use pre-determined variation in banks' liquidity coverage ratio (LCR) in a difference-in-differences setup. Granular instrument-level data allow us to carefully control for any observable and unobservable time-varying factors at the creditor, instrument type, and macroeconomic levels. We find that banks with greater LCR exposure see a steeper decline in their wholesale funding costs. Consistent with seminal theoretical papers on bank liquidity risk, we provide novel evidence that wholesale funding costs decline by more for longer-maturity instruments and that banks shift from short to longer maturity liabilities. Our results support the argument that bank regulation can– at least partly– offset its costs to intermediaries through cheaper wholesale funding.
JEL classification: G21, G23, G28
Keywords: liquidity coverage ratio, liquidity risk, Basel III, money market funds, market discipline
