No abstract

Evidence from recent randomized …eld experiments on community-based monitoring reveals substantial heterogenous treatment e¤ects. Using data from a randomized experiment in primary health in Uganda, we test whether social heterogeneity can explain why some communities managed to push for better health service delivery while others did not. The results suggest that income inequality, and particularly ethnic fractionalization, adversely impact collective action for improved service provision. (JEL: H41, I19, O15) Acknowledgement: We thank Alberto Alesina and Ekaterina Zhuravskaya for sharing their data on ethnolinguistic fractionalization, Kjetil Bjorvatn, participants at the 2009 EEA conference in Barcelona, and speci…cally Erik Ø. Sørensen for valuable comments.

Evidence from recent randomized …eld experiments on community-based monitoring reveals substantial heterogenous treatment e¤ects. Using data from a randomized experiment in primary health in Uganda, we test whether social heterogeneity can explain why some communities managed to push for better health service delivery while others did not. The results suggest that income inequality, and particularly ethnic fractionalization, adversely impact collective action for improved service provision. (JEL: H41, I19, O15 ) Acknowledgement: We thank Alberto Alesina and Ekaterina Zhuravskaya for sharing their data on ethnolinguistic fractionalization, Kjetil Bjorvatn, participants at the 2009 EEA conference in Barcelona, and speci…cally Erik Ø. Sørensen for valuable comments.