Informal Civil Society, Electoral Politics and Local Public Goods Provision: Evidence from Mexico

How do informal civil society organizations, such as community assemblies and committees affect resource distribution across sub-national units in decentralized systems? Previous literature has outlined different mechanisms how these civil society institutions affect distributive politics: they could serve a brokerage function in exchanging funds for concentrated community vote, they might help to mobilize community support for developing projects or they might serve as mobilizing tools for small-scale protests demanding discretionary resources. This paper specifies political incentives underlying these distribution strategies, and tests these mechanisms rigorously against each other with village-level data from rural Mexico. I argue that when sub-municipal localities and their leaders have the capacity to mobilize their citizens, they prefer contentious action over vote concentration, as the latter is inherently risky for marginalized communities. I also argue that mobilizing community contributions attracts resources only under very low levels of state capacity, where municipal and state level politicians lack the ability to implement projects themselves. I provide evidence for this theory using an original village-level data set and identification strategy based on instrumental variable design.