Ebook Does the Secondary Loan Market Reduce Borrowing Costs?
Syndicated loans represent one of the most important debt financing vehicles in the US economy, funding more than half of aggregate corporate external capital demand. The rapid development of the primary syndicated loan market is matched by an even faster-growing secondary loan trading market, increasing from $8 billion in 1991 to $342 billion in 2007 a compound annual growth rate of 26.5 percent. Despite the importance of the loan resale market, the economic implications of the development of this market remain open to debate.
Making use of the secondary loan market database jointly provided by the Loan Syndication Trading Association (hereafter LSTA) and the Reuters Loan Pricing Corporation (hereafter LPC), we show that this market is primarily a clearinghouse for ex ante relatively riskier loans, and that the ex ante probability of loan resale dramatically impacts the primary market loan spread. This impact on the primary market is above and beyond the effect of the ex ante risks and other characteristics of borrowers, raising the spread by roughly 70 basis points for non-investment grade loans, a finding we speculate is due to concerns about monitoring of the resold loans. We also find that the lifting of loan covenants restricting resale lowers the loan spread by as much as 59 basis points for such loans – the option to access external funds brings lower costs. The average net impact of simultaneously lifting resale constraints and raising the probability of resale across the full sample is to lower spreads by 12-14 basis points (or $140-$164 thousand savings on the annual interest expense for the typical (median) size loan). While this beneficial net impact for issuers on loan spreads is absent for firms issuing investment grade loans as well as for firms with little risk of default, it is strong for those issuing non-investment grade loans (32 basis points), non-rated loans (25 basis points) and at risk of default (by measure of Altman’s (1968) Z, 27 basis points), prima facie evidence of the benefit of the secondary loan market.
We understand these results in the context of Gorton and Pennacchi’s (1995) motivation of the empirical regularity of loan resales in the face of clear moral hazard for purchasers of this debt. Their framework relies on implicit contractual agreements between loan purchasers and sellers to manage the moral hazard issues (impaired monitoring with resale), and assumes high internal costs of funds for the seller to make the exchange profitable (hence motivating the notion that access to external funds is valuable). Gorton and Pennacchi (1995) argue persuasively that the secondary market solves one problem, a shortfall of inexpensive funding for loans, and introduces another, a shortfall in monitoring with resale. Parlour and Plantin (2008), a closely related paper, note that a liquid secondary market can lower the cost of the loan by removing the liquidity premium but incentives to monitor before resale of the loan are impacted by the possibility of resale. Similar to Gorton and Pennacchi (1995), their model indicates that loans will be less expensive if the option to resell is available.
In relation to these models, the option to resell the loan, ceteris paribus, is an unambiguously good thing for the bank originating the loan, if internal funding is relatively costly for the loan syndicate. This should induce a syndicate to lower the spread it charges if a borrower will allow its loan to be sold. However, the higher the probability of resale, ceteris paribus, the more likely that bank monitoring will be negatively impacted and the greater the loan spread required to entice other lenders to join the loan syndicate. In this way, loan resale probability proxies for reduced monitoring and increased moral hazard that can follow resale. We employ loan covenants restricting resale to proxy for the constraints on the option to resell. The option to resell, not necessarily the intention to resell or even the likelihood of resale, is captured by this variable. We note that restrictions on the option to resell a loan have little impact on the likelihood that a loan will be resold. The probability that a loan will be resold is actually weakly negatively correlated with the option to resell, a phenomenon similar to loans secured by assets tending to have higher default probability than unsecured loans, in spite of their secured status – loans most in need of monitoring will be more likely to have constraints placed on resale. While the data are not available to observe changes in monitoring with loan resales or to establish that buyers of debt on the secondary market have access to cheaper funding sources than the sellers, we provide strong indirect support for Gorton and Pennacchi (1995).
Our result provides nuance to theories of loan securitization which predict that banks choose to sell better (safe) loans where the value of monitoring is negligible and thus the associated moral hazard and adverse selection problem is minimized (e.g., Greenbaum and Thakor (1987), Pennacchi (1988), Berger and Udell (1993)) as well as to the literature outlining the certification hypothesis for the initial loan distribution in the primary market (Dennis and Mullineaux (2000), Sufi (2007) and Panyagometh and Roberts (2010)). We document that safe (investment-grade) loans are resold, but we also document that the typical loan resold is an ex ante relatively riskier loan (consistent with the findings of Gorton and Pennacchi (1995)), and that a higher price is paid on these loans with increased ex ante probability of resale.
Download
PDF Ebook Does the Secondary Loan Market Reduce Borrowing Costs?
Posted in :
