"On Balance Sheet Spillovers from Nonbanks to Banks"(with Dean Corbae and Pablo D'Erasmo).
AFA 2024, SAET 2025*
Nonbank lenders such as private credit lenders have grown rapidly over the past decade. In tandem with this growth, bank lending to nonbank lenders has also increased sharply. We develop a dynamic structural model with imperfect competition where large dominant banks, small banks, and nonbanks compete in the corporate loan market. Banks finance their nonbank competitors who are also financed by equity. We study the forces that drive the increase in bank lending to nonbanks, and the effects on interest rates, aggregate credit, and financial stability of banks’ exposure to nonbanks. In a preliminary calibration, we find that an increase in capital requirements, consistent with Dodd-Frank, explains only 18% of the increase in nonbank corporate loan market share. The remaining 82% is explained by a regulation-induced 37 basis-point increase in dominant banks’ marginal cost of lending. The resulting increase in nonbank lending is financed by bank loans to nonbanks, which raises nonbanks’ leverage and consequently, their default risk.
"Bank Financial Transparency and the Lending Channel of Monetary Policy Transmission" (with Yingtong Xie )
Liberal Arts Macro (2025)*
We empirically assess the effects of banks' information environment on the bank lending channel of monetary policy transmission. Intuitively, higher information asymmetry makes it more costly for banks to draw on wholesale funding to make up monetary-policy-induced drops in insured deposits. Quantitatively, a one-standard-deviation increase in opacity corresponds to a 2.24 percentage-point decrease in loan growth given a 100-basis-point increase in the federal funds rate. Furthermore, we find that the effect of bank opacity on bank lending sensitivity is stronger for public banks. This result suggests that the wholesale funding market creditors monitor public banks more closely, consistent with the fact that public banks disclose more information than private banks.
In this note we document that CEOs of U.S. public companies with higher compensation inequality among directors of their boards are fired less frequently. Further analysis about mechanisms reveals that directors on politically partisan boards are paid more unequally than those of non-partisan boards. Our findings suggest that director compensation structure plays an important role in corporate governance strength.
Existing findings on financial disclosure's effect on corporate financial decisions (e.g., investment) can be theoretically rationalized by multiple mechanisms related to information frictions. For instance, an increase in investment associated with an increase in financial transparency is consistent with both a reduction in asymmetric information and a mitigation of moral hazard. We develop a dynamic corporate finance model to separately identify and quantify the effects of asymmetric information and moral hazard. The model features endogenous financial disclosure, signaling, and investor learning. Utilizing confidential panel data of private and public U.S. corporations from the U.S. Census Bureau, we bring unique empirical bearings of the quantitative results of our model.
Utilizing firm credit registry data from Japan, this paper conducts large-sample analyses on the debt structure of private firm. First, we document that bank loans and trade credits are the two dominant sources of debt financing for private firms. Second, there is a persistent and increasing tendency of debt specialization from 2000 to 2020. Furthermore, the usage of long-term debt is on a upward trend over our sample period.
``Composition of External Finance, Information, and Firm Ownership'' (with Dean Corbae and Katya Kazakova)
U.S. Census research projects that utilize confidential data on detailed balance sheet and income statement information of private and public corporations in the United States. Census Bureau project ID: wi2981.
Using borrower-lender-matched loan-level data, we show that firms with higher markups in the product market enjoy significantly lower loan spread when they borrow in the syndicated loan market. This finding is robust to all standard corporate finance controls (e.g., profitability) that have predictive power on the cost of debt-financing, as well as to a wide range of fixed effects. We further show that this effect is stronger for unsecured loans, credit lines, and among rated corporations. We develop and estimate a dynamic corporate finance model with endogenous firm-level markups and risky debt to explore the causal mechanisms behind our empirical findings.