Is Collaboration a Good Investment? Modeling the Impact of Government Support for Nonprofit Collaborative Watershed Management Councils.

This paper couples data concerning 2500 grants given to local watershed management councils by an Oregon state agency, the Oregon Watershed Enhancement Board (OWEB), with 20 years of monthly water quality data sampled at 141 sites around the state. I examine whether public input, in the form of grant funding, to nonprofit watershed councils impacts water quality, and further compare the impacts of different grant types, such as restoration projects or stakeholder outreach. These data present some of the first evidence regarding the effect of collaborative management outputs on environmental outcomes (Carr et al. 2012; Koontz and Thomas 2006). In modeling these effects, this paper also makes a methodological contribution by demonstrating how spatio-temporal ecological and epidemiological modeling techniques can be used to test policy theory and analyze policy impacts using extant data. Specifically, I use integrated nested Laplace approximation (INLA) (Rue et al. 2009) and stochastic partial differential equations (SPDE) (Lindgren et al. 2011) to fit a hierarchical Bayesian model that accounts for spatial and temporal dependency. I find that increased public support (in the form of OWEB grant funds) for nonprofit watershed council education and outreach actions are strongly linked to water quality improvements. Support for council administrative purposes and scientific or technical activities are also associated with improved water quality over longer timer periods. The impacts of restoration projects are positive on average but more uncertain.