42 billion of the speculative-grade bonds that defaulted over the 13-year period of 1982 through 1994.
As we see below, the observed primary-market, initial-issue yield premium for speculative-grade bonds relative to Aaa-rated bonds, 366 basis points, lay within the 95 percent confidence interval when both types of correlation were considered; however, if both types of correlation were ignored, it fell outside the 95 percent confidence interval.
Altman (1989, 1991, 1992) suggests speculative-grade bond risk premiums of 300 to 500 basis points over Treasuries, whereas Hickman (1958) found that, from 1900 to 1943, the annual yield spread between speculative-grade bonds and high-grade bonds was approximately 300 basis points.
Given the results displayed in Tables 3a and 3b, we are 95 percent certain that the true population mean annual loss rate is less than 422 basis points using equal-weights or 626 basis points using market-weights for speculative-grade bonds based on the 13 years covered in this study.
0394 Notes: (a)=par value of defaulting speculative-grade bonds (b)=total par value of speculative-grade bond market excluding defaulted issues (c)=traditional frequency measure is aggregate par value of defaulting bonds in year t divided by aggregate par value of outstanding bonds in year t; we normalize [F.
We collected data on speculative-grade bonds back to 1976; however, because of the small size of the market, we elected not to include data prior to 1982.
Market weights are determined by the cumulative par value of all speculative-grade bonds outstanding.
The differences between investment-grade and speculative-grade bonds are even more striking when the data are disaggregated by maturity.
Holding maturity constant, the option values of investment-grade and speculative-grade bonds are closer.
Assuming the Kalotay, Williams, and Fabozzi model is accurate, the call dates of the speculative-grade bonds are not sufficiently early to offset the effects of their shorter maturities and higher call prices.