Abstract
This paper employs a previously unexplored panel dataset to test whether government grants crowd out private donations to charitable organizations, controlling for changes in the organizations' fundraising behavior. The data covers all registered charitable organizations in Sweden between 1989 and 2003. We have a total of 361 organizations where the largest group is health related. The panel data allows us to control for unobserved organizational heterogeneity and time fixed effects. Furthermore, we use a 2SLS specification to control for possible endogeneity in government grants and fundraising expenditures. Complete crowding out can be strongly rejected. In the 2SLS regression, the estimated crowd-out is small and highly significant in the full sample, on average 5.0%. In the disaggregated sample, we cannot reject zero crowding out for any type of organization in the 2SLS regressions. Furthermore, we find strong evidence that organizations are net revenue maximizing, indicating that fundraising activities are efficient.