Third-party certification in new issues of corporate tax-exempt bonds: standby letter of credit and bond rating interaction

Financial Management (Financial Management Association), Spring, 1996 by Roger D. Stover

III. Empirical Results: Testing the Relative Certifying Effects

Figure 1 presents results for the total sample. Panel A of Table 2 condenses the path results from Figure 1 to illustrate the total effect of each variable on the bond's interest cost. In Figure 1, the indirect effect of the letter of credit on the bond interest cost transmitted through the bond rating variable is statistically significant, which provides evidence concerning how the standby letter of credit affects the bond's interest cost. In contrast, no significant direct relationship exists between the standby letter of credit and bond interest cost. Thus, simply obtaining a standby letter of credit does improve the interest cost to the borrower but principally because of the reaction of the rating agencies. A comparison of the nonstandardized coefficients in Panel A of Table 2 shows that the total effect of employing a standby letter of credit is a 17-basis-point reduction in the new issue interest cost.

[TABULAR DATA FOR TABLE 1 OMITTED]

While the level of interest rates has the expected (significantly) positive direct effect on bond interest cost, issuers of corporate tax-exempt bonds also reduce the maturity of issues as interest rates rise. Given the typically upward-sloping yield curve for tax-exempt bonds, this reduction in maturity tends to offset the direct effect of the interest rate level. However, while both standardized effects are significant, the direct effect is substantially greater. Again, observing the nonstandardized coefficients in Panel A of Table 2, the direct effect of the level of interest rates on bond interest cost of 118 basis points was only slightly mitigated by the borrower's reduced maturity; the total effect of INT was 102 basis points.

While standby letters of credit do influence the interest cost on corporate tax-exempt debt issues in Figure 1, the certification hypothesis suggests an additional effect based on the quality of the bank issuing the letter of credit. Based on the institutional evidence in Section I, such a quality measure should reflect not only the creditworthiness of the bank issuing the letter of credit but also that bank's reputation in the market for this form of credit enhancement. The credit quality measure used is Moody's rating for the bank's senior unsecured or subordinated debt (BANKRAT). Because of the relative prevalence of foreign banks issuing standby letters of credit, I contacted Moody's Investors Service Ratings Information System to assure that the ratings were consistent for both domestic and foreign banks. The same ranking system used for RAT is used for BANKRAT.

In addition to this credit quality measure, the initial public offering literature suggests that the market reacts to the reputation of investment bankers and auditors (see Carter and Manaster, 1990). Therefore, to provide a more complete assessment of the effect of bank quality on bond interest cost, proxies for bank reputation in this debt market must also be employed. Based on the study of banks issuing standby letters of credit by Koppenhaver and Stover (1991), the bank's asset size and profitability can be used to proxy for its reputation. The source of both is the listing of the world's top five hundred commercial banks by Euromoney magazine in the year in which the credit-enhanced bond was issued. Asset size (ASSET) is measured in terms of total U.S. dollars. Profitability is proxied in terms of the ranking by shareholder equity (RANK). The RANK variable ranges from one for the highest ranking down to five hundred for the lowest quality bank.