We first study a stylized model of self-fulfilling panic among agents with differing fragilities to strategic risk and show that depending on the severity of coordination problems, the panic trigger threshold can depend only on one type's fragility. We then present a model of systemic panic among financial institutions with heterogeneous fragilities to financial spillovers. Concerns about potential spillovers generate strategic interaction, triggering a pre-emption game in which one tries to exit the market before others to avoid spillovers. Although financial contagion originates in weaker institutions, systemic risk can critically depend on financial health of stronger in the contagion chain. In this case, bolstering the strong, rather than the weak, more effectively enhances systemic stability.

We examine the relationship between banks’ corporate social responsibility (CSR) scores, measured at the peak of the boom, and their lending behaviors during the Great Recession. High-CSR banks, expected to better support local borrowers during critical periods, instead reduced small business lending more sharply than their low-CSR counterparts. This paradox arises from CSR scores’ emphasis on visible but less material attributes, such as employee benefits, which are easier to measure during booms but deplete financial slack necessary for sustained lending under stress. Our findings highlight a misalignment between CSR metrics and material social impacts, underscoring the need for more reliable performance measures to implement stakeholderism effectively.

We test whether lenders’ screening incentives weaken when faced with the possibility of loan sales. We adopt a new measure of lending standards, the processing time for mortgage applications at the loan level, and use the collapse of the nonagency mortgage-backed securities issuance market as a natural experiment. Secondary market liquidity for nonconforming loans decreased significantly at the end of 2007, but the market for securitizing conforming loans did not experience the same breakdown. Following this event, lenders spent significantly more time screening applications for loans larger than the conforming loan limits than for those below the limits. The processing-time gap widened more for banks with lower capital, greater involvement in the originate-to-distribute model, and larger assets.

We consider a macroprudential approach to analyze the optimal lending policy for the central bank, focusing on spillover effects that policy exerts on money markets. Lending against high-quality collateral protects central banks against losses, but can adversely affect liquidity creation in markets since high-quality collateral gets locked up with the central bank rather than circulating in markets. Lending against low-quality collateral creates counterparty risk but can improve liquidity in markets. We illustrate the optimal policy incorporating these trade-offs. Contrary to what is generally accepted, lending against high-quality collateral can have negative effects, whereas it may be optimal to lend against low-quality collateral.

We analyze the effects and interactions of monetary policy tools that differ in terms of their timing and their targeting. In a model with heterogeneous agents, more productive agents endogenously expose themselves to higher interim liquidity risk by borrowing and investing more. Two inefficiencies impair the transmission of monetary policy: an investment- and a hoarding inefficiency. Heterogeneous agents respond disparately to ex-ante, conventional and ex-post, unconventional monetary policy. However, we show that the two policies are equivalent due to the endogeneity of hoarding. In contrast, targeted interventions such as discount-window lending can alleviate both inefficiencies at the same time.