High-yield bonds that eventually run into serious credit problems do not, as a rule, come to market looking like average-quality issues. Initial offering spreads tag them as potentially troubled situations from the outset, according to data collected on 1985 underwritings. Probably more surprising to those who consider the market reasonably efficient is the finding that the 1985 newissue spreads on the very riskiest issues, although exceptionally wide, were not wide enough to compensate for credit-related losses. In light of the data presented, reaching for maximum yield appears to be a questionable strategy for buyers of non-investment-grade bonds. Superior analytical skills, focus on secondary-market opportunities or the availability of historically large risk premiums or equity kickers might modify that conclusion, however.