Climate Variability and Public-Sector Outcome Efficiency in Latin America: Dynamic Panel Evidence under Cross-Sectional Dependence

This study evaluates public-sector outcome efficiency in 19 Latin American countries during 2000–2022. A pooled output-oriented data envelopment analysis under variable returns to scale combines government consumption and health expenditure with human development, service access, and real income outcomes.

Dynamic two-way fixed-effects models are estimated on 342 observations from 17 countries, with Driscoll–Kraay standard errors, split-panel jackknife correction, a 999-replication country bootstrap, leave-one-country-out analysis, and a constant-returns efficiency alternative. Efficiency is highly persistent (lag coefficient 0.78108). Temperature deviations are negatively associated with efficiency (−0.00643; Driscoll–Kraay p = 0.068) and remain negative under every country exclusion; the bootstrap 95% interval is [−0.01432, −0.00095].