This paper analyzes the political and economic determinants of volunteer use among local municipalities in California with the aim of exploring the two faces of co-production:  partnership and privatization (Bovaird, 2006; 2007).  Co-production as partnership aims to engage citizens to build civic culture and enhance services through incorporation of local knowledge.  Co-production as privatization tends to be undertaken more in the interest of cost-reduction through the shifting of responsibilities from paid staff (Ferris 1984).  Critics of the latter suggest that this shift in responsibilities may come at the price of diminished professionalism, or mask service reductions or load shedding undertaken for political reasons (Percy 1984; Matson 1986).

Hypotheses.  This paper models the use of volunteers in California local government, arguing that coproduction will take different forms in core services (police and fire) than for private services publically provided.  This paper argues that services perceived to be essential public goods, such as local police and fire, are more likely to use volunteers in a co-productive capacity.  Services that involve private goods publically provided (such as emergency medical treatment) or congestible private goods (such as parks and libraries) are more likely to use volunteers in a load shedding fashion.

The paper hypothesizes that the tendency in use of volunteers (partnership or privatization) will be associated with distinct patterns of local government political, fiscal, and institutional factors.  For example, “partnership” use of volunteers for service enhancement will be positively related to liberal political ideology and demographic diversity, with a U-shaped relationship between fiscal capacity and volunteerism.  Conversely, it is hypothesized that use of volunteers in the interests of privatization would be associated with a more conservative political environment, and negatively related to both socio-economic diversity and fiscal capacity.

Methods and Data.  The paper tests hypotheses about the determinants of volunteer use with time-series fiscal and service provision data for California municipalities for the years of 2001 to 2011, obtained from the California State Controller’s Office (5185 city-years of data).   Volunteers are more common in Police and Fire among 11 reported service arenas (respectively 44% and 32% of cities). While use of police volunteers increased by 20% (from 164 cities in 2001 to 196 cities in 2011), fire volunteerism declined by 7% (134 cities in 2001 to 124 cities in 2011).

Use of volunteers is predicted as a function of political ideology; demographic factors; and fiscal capacity, with controls for institutional form (charter or general law); governance (council/manager v. strong mayor); rural/urban location).  Multinomial logit models are used to predict use of volunteers in police and fire, essential public goods.  A zero-inflated negative binomial model is used to predict the count of other “non-essential” services that involve volunteers.

Relevance.  The findings will provide a fine-grained analysis of the complex determinants of volunteer employment among local governments, and contribute to understanding of the role that volunteers play in modern cities.