Data and Code for: Does Information Break the Political Resource Curse? Experimental Evidence from Mozambique

Principal Investigator(s): View help for Principal Investigator(s) Alex Armand, Nova School of Business and Economics, Universidade Nova de Lisboa; Alexander Coutts, Nova School of Business and Economics, Universidade Nova de Lisboa; Pedro C. Vicente, Nova School of Business and Economics, Universidade Nova de Lisboa; Inês Vilela, Nova School of Business and Economics, Universidade Nova de Lisboa

Version: View help for Version V1

Name	File Type	Size	Last Modified
Data and Code for: Does Information Break the Political Resource Curse? Experimental Evidence from Mozambique		0

Project Citation:

Armand, Alex, Coutts, Alexander, Vicente, Pedro C., and Vilela, Inês. Data and Code for: Does Information Break the Political Resource Curse? Experimental Evidence from Mozambique. Nashville, TN: American Economic Association [publisher], 2020. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2020-10-21. https://doi.org/10.3886/E119785V1

Project Description

Summary: This project provides data and code for the replication of the paper "Does Information Break the Political Resource Curse? Experimental Evidence from Mozambique". The paper tests whether information can counteract this political resource curse by focusing on a large-scale field experiment following the dissemination of information about a substantial natural gas discovery in Mozambique. Outcomes related to the behavior of citizens and local leaders are measured through georeferenced conflict data, behavioral activities, lab-in-the-field experiments, and surveys. Information targeting citizens and their involvement in public deliberations increases local mobilization and decreases violence. By contrast, when information reaches only local leaders, it increases elite capture and rent-seeking.

Scope of Project

Subject Terms: Natural resources; curse; natural gas; information; deliberation; rent-seeking; Mozambique

JEL Classification:
      D72 Political Processes: Rent-seeking, Lobbying, Elections, Legislatures, and Voting Behavior
      O13 Economic Development: Agriculture; Natural Resources; Energy; Environment; Other Primary Products
      O55 Economywide Country Studies: Africa
      P16 Capitalist Systems: Political Economy

Geographic Coverage: Mozambique

Universe: Household heads, local leaders, and communities in the Cabo Delgado province (Mozambique)

Data Type(s): administrative records data; experimental data; survey data

Collection Notes:

The readme file attached to this project details the data that have been anonymized following the ethics approval of the project.

Methodology

Response Rate: A total of 2,065 heads of household were interviewed at the baseline, targeting 10 per community. Post-treatment attrition was handled through substitutions in the same household, when possible. Endline attrition rate at household-level is equal to 8.3%.

Sampling: We selected a sample of 206 communities in the province of Cabo Delgado. These were randomly drawn from the list of all 454 polling locations in the sampling frame, stratified on urban, semi-urban, and rural areas. To obtain the sampling frame, 52 polling stations with inconsistent information in either the 2009 or the 2014 general elections were excluded. Moreover, the smallest 5th percentile by voter size (corresponding to a cutoff of 207 voters) was also excluded. The polling locations were located within the following 16 districts in the province of Cabo Delgado: Ancuabe, Balama, Chiure, Macomia, Mecufi, Meluco,Metuge, Mocimboa da Praia, Montepuez, Mueda, Muidumbe, Namuno, Nangade,Palma, Pemba, and Quissanga. The Ibo district was excluded since it is an island, together with two other polling stations in another island. Eleven polling stations in Palma's posto administrativo were also excluded to avoid areas that had already experienced some violence related to relocation of communities.

To randomly allocate polling stations to different interventions, blocks of four communities were built using Mahalanobis-distance while exploiting the richness of baseline information. Within each block communities were randomly allocated with equal probability to either treatment 1, treatment 2 without the deliberation module, treatment 2 with the deliberation module, or a control group. This procedure resulted in 50, 51, 50, and 55 communities in each group, respectively. Sampling of citizens was the product of physical random walks during the baseline survey. In each house, heads of households were sampled for survey interviews and behavioral activities.

Data Source: Data sources are detailed in the readme attached to this project. Section C of the Online Appendix of the paper provides further details about the data collection and the inclusion of external data sources.

Collection Mode(s): coded on-site observation; cognitive assessment test; face-to-face interview

Scales: Different scales. See Section C of the Online Appendix of the paper for further details about the data collection and the inclusion of external data sources.

Weights: N/A

Unit(s) of Observation: Household heads, Community leaders, Communities

Geographic Unit: Community

Related Publications

Download this file

Published Versions

V1 [2020-10-21]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.