Theories of nonprofit density have assumed a variety of dispositions toward the state, including opposition, suspicion, indifference, and mutual dependence. In this article, we conduct the first large-scale simultaneous empirical test of the two most prominent nonprofit theories: government failure theory and interdependence theory. The former characterizes nonprofit activities as substitute or oppositional to state programs, accounting for the limitations and failures of government-provided services and more reflective of the heterogeneity of demand for services. The latter emphasizes the more complementary and collaborative nature of nonprofit activities, focusing on the overlapping agendas of nonprofits and the state and the mutual dependency that arises from partnership. The theories are difficult to test empirically because both predict the same relationship between state capacity and the size of the nonprofit sector, albeit for theoretically distinct reasons. A true joint test requires the separation of government support from private support for nonprofits. Using a newly constructed panel dataset in which we separate out nonprofit revenue sources normally agglomerated in the Internal Revenue Service 990 data, we examine the empirical merits of both theories to answer the question of whether human service nonprofit organizations thrive when government fails or when government collaborates. Our findings suggest that government funding has a more favorable effect on nonprofit density than private donations. The findings raise several policy and management implications that need evaluation.