Association for Public Policy Analysis & Management

What makes an organization innovative? This is an enduring question in literature with a variety of models explaining innovation adoption in public organizations. This study contributes to this research by introducing substandard performance as a determinant of innovation adoption, taking the example of the adoption of CompStat systems in U.S. police departments. CompStat is a significant innovation in policing that was first operationalized by the New York Police Department in mid-1990s, and is consistently gaining popularity among police departments in the U.S. and abroad. I employ survival analysis of 342 small to mid-size U.S. police departments over a 14-year period. Event history and Cox hazards modelling shows that poor pre-adoption performance for violent crime is significantly related to CompStat adoption; and the weaker a department’s pre-adoption performance, the earlier it adopts CompStat. Property crime, on the other hand, is not found to have a significant impact on the adoption of CompStat. These results emphasize the importance of past performance on salient dimensions on innovation adoption.