This paper presents a contract-theoretic model in which banks choose pre-arranged and ex-post funding to finance firms' liquidity needs through credit lines. In states with high liquidity needs, pre-arranged funding is key to sustaining lending and reducing the number of liquidated firms. Yet, in the presence of a pecuniary externality on firms' liquidation values, competitive banks choose insufficient pre-funding compared to a constrained social planner. Constrained efficiency can be restored using regulatory liquidity ratios. The optimal regulatory ratio depends on the frequency of high liquidity need states, the value lost after a firm liquidation, and the premium on pre-funding. 

Stress testing of banks is a tool widely used by supervisory authorities. The public disclosure of bank-level results provides valuable information to market participants, but it can entail adverse consequences for under-performing entities. We uncover a novel cost of disclosing stress test results in the form of credit line runs. Using the Spanish Credit Register and the 2011 stress test of the European Banking Authority, we find that, following the release of the results, firms drew down approximately 10 pp more available funds from lines granted by banks with a worse performance in the stress test. Moreover, before the release date, worse-performing banks reduced more the size of credit lines, and those underperformers with higher credit line balances were also more likely to cut term lending ex-post. 

This paper studies the impact of supervisory expectations regarding non-performing loans (NPLs) on banks' NPL disposal and lending behavior as well as on the real economy. Using the supervisory intervention announced by the European Central Bank in 2018q1 as a quasi-natural experiment, we show that banks disposed of old NPLs at a higher rate after the policy. Furthermore, banks that were more heavily exposed to the policy tightened their lending standards, especially for risky firms. We also find that firms borrowing from banks affected by the policy intervention experienced a decline in the growth rate of their total assets, investment, employment, and sales. Our results highlight the importance of timely recognition of NPLs as a factor that can affect credit allocation. We also shed light on the importance of strong bank fundamentals in cleansing banks' balance sheets without disrupting credit intermediation.

This paper analyzes how lending relationships affect the incentives of borrowers to default using loan-level data in Spain. We provide new evidence showing that firms first default on loans from non-main banks. This effect is stronger in small firms and the lower the bank solvency. Our results suggest that firms have lower incentives to default to their most important banks to preserve the most valuable lending relationships. Our findings also indicate that banks internalize this borrower behavior in their credit risk management because most important banks recognize lower discretionary loan impairments. The results are robust to alternative specifications and control for potential bank forbearance, loan characteristics, and a variety of time-varying fixed effects.

The rise of FinTech lenders offers an opportunity to promote financial access but may disrupt banks' banking efforts. This paper presents a banking model where an incumbent bank specializes in some niche markets. When FinTech enters, it intensifies competition for certain niches, reducing the bank's lending relationship gains. Although FinTech serves some unattended niches, the bank may abandon others, creating an ambiguous impact on financial inclusion. The overall effect depends on the initial level of financial inclusion and on FinTech's efficiency and competitiveness.