BOND quality ratings, that is, the ratings assigned to bond issues by organizations such as Moody's Investors Service, have constituted a source of general interest to economists and practical men of affairs for some time. Of particular importance, however, have been the following three questions: (1) how are ratings determined? (2) do ratings accurately discriminate among issues on the basis of default risk; for example, do default rates vary inversely with quality ratings? (3) to what extent, if any, do ratings have an independent influence on bond yields; that is, do ratings-as opposed to the underlying economic factors that influence ratings-have an impact on the market prices of bonds? Over the past two decades, the first two questions have received considerable attention, with the result that we now know a good deal about the determinants of ratings and the relationship between ratings and bond default experience.' In contrast to the progress that has been made in these areas, however, virtually no systematic empirical evidence has been assembled concerning the impact that ratings may have on yields. No doubt this void primarily reflects an inherent difficulty in devising "meaningful" tests of the relationship between ratings and yields.2 There is, after all, a high degree of correla-