Do Lenders Extract Rents When Financing Bankrupt Firms?

Firms filing for Chapter 11 bankruptcy often simultaneously arrange ‘debtor-in-possession’ (DIP) financing. It has been shown that DIP financing is central to the preservation of the going-concern value of Chapter 11 firms and their survival. However, little is known about how these loans are structured and whether they are competitively priced. Our paper is the first to examine these important questions by assembling one of the most comprehensive datasets of DIP loans using US bankruptcy court filings such as DIP motions and master credit agreements retrieved from the Public Access to Court Electronic Records (PACER) system.

Our study sample is comprised of 545 DIP-loan facilities by 359 large Chapter 11 firms over the period of 2002-2019. We first show that all of our sample DIP loans are fully repaid (both principal and interest). In terms of the payment default risk alone, DIP loans appear to be no worse than investment-grade loans. We further show that these short-term loans, accounting for a small fraction (on average 15%) of a Chapter 11 firm’s book assets, effectively insulate these loans’ payment default risk from their borrowers’ underlying asset risk. Moreover, these loans contain a large number of protective covenants and milestones tied to liquidity, performance, capital expenditures, production, asset sales, and even corporate governance. These protective covenants and milestones provide DIP lenders with the ability to react quickly if and when their borrowers’ asset quality starts to deteriorate further.

We next show that the average spread on our sample of 545 DIP-loan facilities is 628 basis points (bps) over the London Interbank Offered Rate (LIBOR). Strikingly, this spread is almost five times the average spread on investment-grade loans matched on year, industry, and size. Moreover, the 628 bps spread is almost double the average spread in a sample of matched highly risky leveraged loans issued by non-bankrupt firms. These excess DIP-loan spreads remain when using spreads on distressed loans issued by the same bankrupt firms within three years of their Chapter 11 filings as the risk benchmark.

To investigate whether high spreads reflect economic rents extracted by super-priority lenders, we employ regression analysis to assess whether lender protections and firm characteristics help explain the spreads and find that there is no association between lender protection and DIP-loan spreads. We next examine the loan placement process. Performing textual analysis of motions filed by Chapter 11 firms, we find that about 70% of the debtors in our sample have approached prospective lenders. However, these debtors receive interest from more than one lender (typically the prepetition lender) in only a third of the cases, and interested lenders proceed to a formal round of bidding in only one tenth of the cases. We find no evidence that approaching prospective outside lenders results in lower spreads, but loan spreads are 142 bps lower if the number of interested lenders is above the sample median. We conclude that some search and bidding do help lower the high DIP-loan spreads.

We next explore whether lender type and prior lending relationship help explain DIP-loan spreads. We note that prepetition secured lenders supply close to 80% of the DIP loans in our sample. In 54% of the DIP-loan facilities provided by prepetition secured lenders, DIP lenders have a lending relationship with the borrower. Examining spread differences between DIP loans provided by relationship prepetition lenders and those by non-relationship prepetition lenders, we find that a strong lending relationship yields only a small amount of savings for borrowers. The evidence suggests that relationship lenders exploit their informational advantage and charge a spread that is not much lower than an otherwise similar loan by non-relationship lenders who likely face higher costs for information production. Finally, we show that in more than 60% of the cases, unsecured junior creditors file objections to DIP-loan terms. Contested DIP loans carry spreads that are on average about 85 bps higher than non-contested DIP loans. Moreover, we find that unsecured creditors’ recovery is 16.4 percentage points lower in contested cases than in uncontested cases. Nonetheless, courts routinely approve DIP loans.

We explore a number of other possible explanations for the high spread. One potential explanation is that it captures large costs associated with monitoring and renegotiation. We note that direct costs of loan origination and monitoring are usually covered by fees, not by spreads.

We also show that DIP-loan fees are substantially higher than fees for both investment-grade loans and leveraged loans. Another potential explanation is related to the lending market structure. We note that the DIP-loan market is quite concentrated, with the majority of loans provided by a small group of lenders. We find that lending institution heterogeneity has strong statistical power in explaining the loan spreads, suggesting the importance of lender-specific skills and their relatively unique access to information on loan spreads. A third potential explanation is the agency problem likely arising from incentive alignment between DIP-loan providers and management—Chapter 11 firms’ management may grant DIP-loan providers opportunities to extract rents to preserve their own human capital.

The full article is available here.

B. Espen Eckbo holds the Tuck Centennial Chair of Finance at the Tuck School of Business at Dartmouth.

Kai Li holds the Canada Research Chair in Corporate Governance and W. Maurice Young Professor of Finance at Sauder School of Business at the University of British Columbia.

Wei Wang is a Professor of Finance at the Smith School of Business at Queen’s University, British Columbia.