Replication Data for "The Dual Benefit Principle: When Collective Institutions Deliver Both Private and Social Returns"

This repository contains complete replication data and code for "The Dual Benefit Principle: When Collective Institutions Deliver Both Private and Social Returns" (submitted to Journal of Political Economy). CONTENTS: 1. SCHEME-LEVEL DATA (79 schemes) - Complete data for 79 community-based health insurance (CBHI) schemes - Coverage: India (32 schemes), Nepal (18), Philippines (13), Uganda (16) - Time period: 2008-2024 - Variables: treatment assignment, retention rates, baseline characteristics, outcomes - Format: CSV and Excel files with complete codebooks 2. NEPAL HOUSEHOLD MICRODATA (4,176 records) - Individual household enrollment data from Banke district, Nepal - Validates scheme-level findings with household-level patterns - Personal identifiers removed for privacy protection - Time period: 2011-2015 3. ANALYSIS CODE - Complete R scripts to generate all manuscript tables - Master script (RUN_ALL_ANALYSES.R) runs entire analysis pipeline - Appendix generation scripts (JavaScript for formatted Word documents) - All code fully documented and ready to run 4. DOCUMENTATION - Comprehensive codebooks for all variables - Data provenance documentation (complete source tracking) - Dataset relationships and linking documentation - Extraction logs with timestamps - Validation results KEY FINDINGS:- Participatory schemes: 74.6% retention (25 schemes)- Imposed schemes: 58.0% retention (54 schemes)- Difference: 16.6 percentage points (p < 0.001)- Effect size: Cohen's d = 1.63 (very large)- Participatory failure rate: 0% vs Imposed: 29.6% REPLICATION:All analyses can be replicated by running scripts in 02_Analysis/ folder. Complete instructions provided in repository README. DATA QUALITY:- Zero missing values on key variables- All ranges validated (retention 0-100%, trust 0-1)- Source provenance documented for every variable- Complete audit trail from raw data to final results