Scartascini, Carlos.
2025. "NIMBYism and Participatory Bias in Latin America." AEA RCT Registry. AEARCTR-0016601.
https://doi.org/10.1257/rct.16601-1.0
Download:
BibTeX
·
RIS— for reference managers without a browser
translator. Zotero and most
others pick up this page's citation metadata directly.
We run a set of survey experiments to evaluate participation in public consultation meetings and preferences for development projects
Intervention
A survey experiment designed to study citizen responses to proposed infrastructure projects—specifically metro stations and highways—in Latin American cities. First, we examine how local opposition shapes urban development outcomes, focusing on how geographic proximity to a project influences policy preferences. Second, we review evidence on who tends to participate in public hearings, emphasizing how concentrated costs and diffuse benefits generate a participatory bias. Third, we analyze how the perceived legitimacy of decision-making actors—such as government agencies versus citizen panels—can alter levels of support. Fourth, we consider how the public or private nature of preference expression influences stated positions, particularly in contexts where social norms or reputational concerns are salient.
Experimental design
Randomization of: 1. Proximity to project site 2. Source of recommendation of project development 3. Secret vs public voting
Design details
Randomization method
Randomization was embedded in the LAPOP survey and conducted by them (computer). The experiment employs a randomized factorial design that manipulates two dimensions: (1) the respondent’s spatial proximity to the proposed infrastructure project, and (2) the institutional source of the project’s recommendation. Each dimension includes three and two levels, respectively, resulting in a total of six treatment
arms. Respondents are randomly assigned to each spatial condition with equal probability (1/3), and to each recommendation source with equal probability (1/2), yielding an even distribution across the full 3×2 design.
Randomization unit
Individual
Clusters
The final analytic sample includes 4,147 respondents: 1,006 in Argentina, 1,021 in Brazil,
1,016 in Colombia, and 1,104 in Mexico.
Observations
The final analytic sample includes 4,147 respondents: 1,006 in Argentina, 1,021 in Brazil,
1,016 in Colombia, and 1,104 in Mexico.
Arms
Each dimension includes three and two levels, respectively, resulting in a total of six treatment arms. Respondents are randomly assigned to each spatial condition with equal probability (1/3), and to each recommendation source with equal probability (1/2), yielding an even distribution across the full 3×2 design.
Primary outcomes
1. Self-reported willingness of respondent to attend a local meeting
2. Preferred project location
Cited by
No citation data available for this registration yet.