Trading Favors: Local Politics and Development

in Developing Countries 

How do citizens influence the distribution of public services? In my theory of “trading favors,” community associations provide a platform for voters to increase their bargaining power towards politicians, who monitor the aggregate votes of groups at polling stations. Community associations can help voters to coordinate before elections and pressure politicians after elections.


This project contributes to our understanding of distributive politics and democratic accountability by emphasizing the collective agency of citizens as they interact with brokers and politicians. I focus on access to water, which is an essential and often scarce resource that is prone to political manipulation.

I argue that 1) community activity and 2) leadership strength enable a group to coordinate its votes before an election and get the attention of politicians after the election. In communities with high community activity, group members participate in associational activities and are more likely to mobilize for a candidate and protest after elections. In communities with strong leadership, an opinion leader responds to community preferences, competes for the leadership position, and is repeatedly elected. Strong leaders are more likely to serve as political brokers and facilitate access to services.

I find that communities with high community activity and constant leadership have more reliable and secure water access. I find evidence for my main mechanism: organized communities are more likely to concentrate their vote, and bloc voting improves water access. The same communities continue to concentrate their votes over time, though many switch allegiance across elections.

During 18 months of fieldwork, I collected extensive qualitative and quantitative evidence from Ceará state in Northeast Brazil. I illustrate the theoretical mechanisms by drawing on 104 interviews with rural residents, local leaders, and bureaucrats, and I conduct statistical analysis of a large-scale household survey with 1,990 respondents from 120 rural communities in 10 municipalities merged with precinct-level electoral data.


The book project is the extension of my dissertation. To complete the book project, I added a conjoint experiment about my primary mechanisms and will use archival data to build on my current explanations of the origins of community organizing in Brazil.


I will return to Brazil in 2021 for additional interviews to conduct process tracing through in-depth comparative case studies. I will add the cases of India and Mexico, and I am collecting survey and electoral data in India through connections that I developed in my postdoctoral fellowship.