« Back to Results

Credit Markets

Paper Session

Friday, Jan. 7, 2022 10:00 AM - 12:00 PM (EST)

Hosted By: American Economic Association
  • Chair: Hassan Afrouzi, Columbia University

Leverage Risk and Investment: The Case of Gold Clauses in the 1930s

Mete Kilic
University of Southern California


We study the impact of the 1933 abrogation of gold clauses on the slow recovery of corporate
investment following the Great Depression. Legal challenges to the constitutionality of the
abrogation of gold clauses exposed many firms to the possibility of a 69% increase in required
payments to bondholders. We show that public firms with higher exposure to this risk
reduced their investment in 1933 and 1934. For these firms, investment recovers quickly
following the Supreme Court’s 1935 decision to uphold abrogating gold clauses. In the cross-section
of firms, decreases in investment over 1933 and 1934 coincide with an increase in
equity payouts. Our estimates imply that the risk of higher financial leverage accounts for
about one-third of the decline in aggregate investment by public firms over 1933 and 1934.
This channel complements existing explanations of the slow recovery based on bank credit
supply which public firms did not rely on.

Leverage Limits in Good and Bad Times

Juanita Gonzalez Uribe
London School of Economics
Cynthia Balloch
London School of Economics


How do leverage limits affect lending? We examine a regulatory change to the business development company (BDC) lending sector, which allowed lenders to double their regulatory leverage constraint. Exploiting the staggered timing of approvals, we show that this allowed firms to slowly adjust loan portfolios and increase leverage, but suddenly increase the unrealized losses reported on their loans. These patterns around approvals suggest that the slackness of regulatory constraints has important effects on lenders’ incentives to accurately assess fair value. In addition, we explore how the pandemic differentially affected BDCs that were close to or far from their leverage limits. Our results shed light on an important lending sector for small businesses, which has grown dramatically since the Great Recession.

Bank Capital, Lending Distortions, and Misallocation

Miguel Faria-e-Castro
Federal Reserve Bank of St. Louis
Pascal Paul
Federal Reserve Bank of San Francisco
Juan M. Sanchez
Federal Reserve Bank of St. Louis


Using loan-level supervisory data for the United States, we show that low-capitalized banks systematically distort their risk assessment of firms and their capital ratios. To avoid potential losses, low-capitalized banks extend relatively more credit to underreported borrowers. These effects are driven by larger loans, are stronger for low-productivity firms, and translate into real outcomes at the firm-level. We develop a simple model of relationship lending where we show that lenders may have an incentive to evergreen loans by offering betters terms to less productive and more indebted firms. We incorporate this mechanism into a dynamic model of industry dynamics to show that relationship lending and lender capitalization can affect misallocation, the size distribution, and the entry rate of firms.

Cross-Subsidization of Bad Credits in a Lending Crisis

Nikolaos Artavanis
Virginia Tech
Brian Jonghwan Lee
Columbia University
Stavros Panageas
University of California-Los Angeles
Margarita Tsoutsoura
Cornell University


Using a unique dataset we study the corporate-loan pricing decisions of a major Greek bank in the years surrounding the recent Greek financial crisis. The unique aspect of the dataset is that besides the loan interest rates, firm characteristics, loan characteristics, etc. we can also observe the breakeven rate that is provided to the account managers by the bank’s loan pricing department. This allows a straightforward measurement of the extent to which the loan’s breakeven rate is passed through to the customer. We document positive and significant markups to the low breakeven-rate borrowers (“good borrowers”) and zero or even negative markups to the high breakeven-rate borrowers (“bad borrowers”). Pass-through is sluggish in the time-series and asymmetric across good and bad borrowers. The discrepancy between interest rates and breakeven costs does not predict changes in the borrower’s future financial conditions, suggesting that interest rates do not embed additional information that is not available to the bank’s pricing department.
JEL Classifications
  • G2 - Financial Institutions and Services