Natália S. Bueno

With: George Avelino, Ciro Biderman, Daniel da Mata, and   Leonardo R. L. Bueno, Getulio Vargas Foundation. 

Encouraging citizens to mobilize and participate in the design and implementation of development projects carries a lot of promise, especially in contexts where state presence is scarce and inconsistent, such as slum settlements. The potential gains of such community-driven solutions are particularly acute in contexts of crisis such as those created by the Covid-19 pandemic. Under which conditions can community-driven programs reduce the adversities induced by the spread of Covid-19? We study whether community-driven programs can increase compliance with measures to reduce risks of transmission and can alleviate the socioeconomic distress induced by the spread of Covid-19. Through a randomized controlled trial of an NGOled program with 611 families in 25 Brazilian slum settlements, we examine the impact of an NGO program on several outcomes related to the pandemic outbreak.

The NGO program is a bundle. On the one hand, the NGO subsidizes a wooden-made housing unit while providing a voluntary labor force to build them. On the other hand, residents must cooperate with other members of the community to make the constructions feasible. Beneficiaries must engage their previous networks and form new ones and rely on norms of trust and reciprocity. The expectation is that the combination of community work and reciprocity allows community members to plan, organize, and execute the program. We  hypothesize that the intervention increases beneficiaries' social skills which may be key in the cooperation to fight Covid-19. At the same time, better housing conditions could potentially diminish exposure to the virus, especially in densely populated slums. We test our theory by randomly distributing the housing units within a pool of eligible families that are at the bottom line of poverty in urban slums of the major Brazilian metropolises. Our randomized controlled trial (RCT) was conducted one year to three months before the pandemic break out (according to the NGO construction plan within each separate community). We test whether recipient families behave differently than the non-recipient families in a series of outcomes related to the pandemic outbreak. Additionally, our quantitative methodology is complemented by qualitative interviews with selected beneficiaries, community leaders, and NGO volunteers. We have also collected observational data on settlers that did not take part on the randomized program. 

We find that even a successful community-driven program fails to induce actions to reduce the pandemic's fallout. The program increases trust in Covid-19 related information from community leaders, yet it fails to increase compliance with Covid-19 contagion mitigation measures and fails to increase solidarity among community members and to bolster individuals' claim-making in periods of need. We find that this result partly stems from individuals involved in clientelistic relationships who are distrustful of community leaders and reticent to ask them for help. Furthermore, we find that our program strengthens solidarity among individuals who had previous connections through friendships and especially members of the same religious organizations. Overall, our results suggest that community-driven programs may have limited impact in creating solidarity and compliance to health measures due to structural inequities and political capture existent in the lives of these communities. Our work focuses on the effects of the Covid-19 pandemic, and it contributes to a growing and broader discussion on the role of civil society in development promotion, co-production of policies, and government accountability in emergency humanitarian crises.