We compare the long-run performance of IPO firms that do not receive analyst coverage (orphans) to those that do (non-orphans). During the 1996 to 1998 time period, orphaned
IPOs significantly underperform non-orphans. Further analysis reveals that the outperformance by non-orphans stems from multiple analyst coverage. Specifically, IPO firms outperform when the lead underwriter initiates overage and their recommendation is corroborated/confirmed by an accompanying (non- lead) underwriter`s report. This result is consistent with the confirmation hypothesis presented in Bradley, Jordan, and Ritter (2003), but largely inconsistent with the conflict of interest hypothesis in Michaely and Womack (1999). We also find that, in the bubble years of 1999-2000, both orphaned and non-orphaned IPOs significantly underperform their style-matched benchmarks. This suggests that research coverage was no longer selective and any certification value analyst coverage had was lost during this time of excessive over-optimism as almost all IPOs received favorable coverage.

