How does a regulator's reputation affect the public observability of its regulatory errors? I present a verbal model in the policy domain of drug safety that suggests that media coverage of the regulator's errors is a function of the regulator's predominant basis of reputation. Media coverage will be lowest when the regulator has a reputation for scientific expertise in preapproval drug evaluation (or when it “shadows” decisions made by regulators that have reputation for expertise) and highest when it has a reputation as a guarantor of public safety in the media. Empirical tests of the model in the United States, the United Kingdom, Germany, Canada, South Africa, Australia, New Zealand, Ireland, Israel, and Switzerland between 1975 and 2004 supports the model's prediction and therefore, undermine Carpenter's assumption that regulators cannot recover reputation losses resulting from the approval of a truly dangerous drug.