A Comparison of Syndicated Loan Pricing at Investment and Commercial Banks
Harjoto, Maretno, Mullineaux, Donald J., Yi, Ha-Chin, Financial Management
We reject the hypothesis that investment and commercial banks have identical loan-pricing policies. We find that compared to commercial banks, investment banks lend to less profitable, more leveraged firms, price riskier classes of term loans more generously, and offer relatively longer-term credits, usually with term, not commitment contracts. Investment banks typically establish higher credit spreads, although the premium declines when a commercial bank joins as syndicate co-arranger. Investment banks also price riskier classes of term loans more generously to borrowers than do commercial banks. Commercial-bank funding advantages do not appear to be a source of the pricing differences.
In recent years, commercial banking organizations have gained a substantial share of the market in securities underwriting. A number of studies, such as Gande, Puri, Saunders, and Walter (1997), show that bank holding companies' entry into this area has influenced the pricing of new issues, underwriting fees, or both. Perhaps as a competitive response to "turf invasion" by commercial banks, investment banks entered the commercial lending business, placing their primary emphasis on syndicated loans. From 1996-2003, investment banks led over 1,300 commercial loan syndications, raising almost $768 billion in funding. However, there has been little research on this development. The primary objective of our study is to determine whether syndicated loans originated by investment banks are priced differently from those arranged by commercial banks.
In this article, we estimate models for credit spreads on syndicated loans originated by investment and commercial banks. We first examine whether spreads are relatively higher or lower at investment banks, assuming that the same model generates the outcomes. Our measure of spread captures fees as well as loan rates. We find that investment banks charge higher spreads than do commercial banks, ceteris paribus, on both term and revolving loans (revolvers). We also find that loan pricing is affected when investment and commercial banks act as co-arrangers in a loan syndication. We then relax the assumption that the same process across institution type determines spreads and find evidence that loan pricing processes differ at investment and commercial banks. Investment banks appear to charge less on the margin for assuming credit risk, for instance.
In Section I, we consider why loan pricing could differ for syndicated loans originated by investment banks rather than by commercial banking organizations. Section II describes the data and presents the results of various model estimations. Section III presents some robustness analysis and Section IV concludes.
I. Why Loan Pricing Might Differ
There are several important distinctions between investment and commercial banks that may have implications for loan pricing. Investment and commercial banks differ in at least five ways: (a) sources of funds, (b) regulation, (c) the relevance of customer relationships, (d) prospects for economies of scope, and (e) the relevance of accounting rules. We briefly consider the roles of each from a loan-pricing perspective.
A. Sources of Funds
For commercial banks, deposits are relatively inexpensive, reliably stable sources of funds. Core deposits have low rates primarily because they are, for the most part, insured and hence risk-free. However, the default-free status of deposits brings both explicit and implicit costs. The explicit costs are deposit insurance premiums and the implicit cost is the regulatory burden imposed on banks to preserve the viability of the FDIC. These associated costs of deposits offset at least some, if not all, of the benefits of their low interest expense.
Unlike commercial banks, investment banks cannot accept deposits. However, some investment banks have gained access to insured deposits by acquiring or chartering banks or bank-like organizations. …