We provide insights into determinants of the rating level of 371 issuers which defaulted in the years 1999 to 2003, and into the leader-follower relationship between Moody?s and S&P. The evidence for the rating level suggests that Moody?s assigns lower ratings than S&P for all observed periods before the default event. Furthermore, we observe two-way Granger causality, which signifies information flow between the two rating agencies. Since lagged rating changes influence the magnitude of the agencies? own rating changes it would appear that the two rating agencies apply a policy of taking a severe downgrade through several mild downgrades. Further, our analysis of rating changes shows that issuers with headquarters in the US are less sharply downgraded than non-US issuers. For rating changes by Moody?s we also find that larger issuers seem to be downgraded less severely than smaller issuers.