PURPOSE: Theorists and public administration practitioners have long advocated for increased use of public participation in administrative decision making. Citizen participation can take many forms, but the basic definition requires that a citizen interacts with and provides feedback to government officials at some part of the policy process or implementation. While public participation is normatively desirable, to date, the public management literature does not empirically validate a link between public participation and citizens’ satisfaction with public services. We establish a base positive relationship between the two variables, indicating that as citizen participation increases, satisfaction with services also increases. Then, we provide nuance to the finding by estimating the joint effect of citizens’ trust for their local government on the participation-satisfaction relationship. Finally, we test a possible pathway (the perceived severity of the recent economic crisis) to explain the findings.

DATA: We use data from the Latin American Public Opinion Project (LAPOP) to test three hypotheses about participation and satisfaction of municipal services. LAPOP surveys about 1,500 citizens in each of 26 different Latin American countries every two years. These data provide in-depth, individual level data on democratic values and behaviors in the Americas.

METHODS: Our research questions are threefold. First, can we establish an empirical relationship (which has previously only been theorized) between citizen participation and satisfaction with services? LAPOP surveys citizens about their satisfaction with municipal services, their participation in local government meetings, and demands made upon local government officials. Then, we estimate whether this relationship is consistent at all levels of government, or whether trust (a theoretically important construct in participation literature) in local government will moderate that relationship. Finally, we explore the role of the economy on these relationships. The cross sectional nature of the data limit our analyses to OLS models, but the richness of the data allow us to test for moderating and mediating effects.