The nonresidential solar energy market has grown rapidly in the United States. Companies considering the adoption of solar energy have an important choice: whether to purchase the system (direct ownership, or DO) or opt for an agreement with a service provider that offers third-party ownership (TPO). Although this decision is financially very important, empirical research comparing the two models from an operational perspective has been scarce. These issues are particularly relevant considering that federal and state incentives are commonly used to stimulate adoption. This paper presents an empirical analysis of the role of TPO in the operational performance of nonresidential solar energy installations in California. Using a panel data estimation framework, and accounting for factors that influence TPO adoption, the analysis shows that TPO systems performed, on average, approximately 4% better than DO systems) in terms of their production yield. This performance improvement is consistent with better system design under TPO, but not with better solar panel technology selection by TPO providers.