Natural Disasters, Economic Growth, and Carbon Emissions in Pakistan: PLS-SEM Mediation Analysis

Abstract

This study quantifies how natural disasters affect carbon emissions in Pakistan, both directly and indirectly through economic growth, and identifies policy levers that can jointly enhance resilience and decarbonization. Pakistan is highly exposed to floods and storms and has experienced rising CO₂ alongside expanding energy use. Prior studies typically treated the disaster–growth nexus and the growth–emissions nexus separately, leaving the growth-mediated channel from disasters to emissions underexplored. Using annual data for 1980–2024, the analysis estimates a structural equation model based on partial least squares (PLS-SEM). A reflective latent “disasters” construct (floods, storms, droughts, earthquakes) explains real GDP; CO₂ emissions are modeled as a function of GDP, disasters, and controls (energy consumption, urbanization, population growth). Bootstrapped (5,000 resamples) path coefficients, indirect effects, and predictive diagnostics (R², Q²) are reported, with robustness checks (alternative scalings, per-capita CO₂, energy proxies, leave-one-out indicators, sub-periods). Disasters significantly reduced GDP; GDP and energy increased CO₂. The indirect effect (Disasters → GDP → CO₂) was negative and significant, while the direct Disasters → CO₂ path was small and only marginally positive, consistent with carbon-intensive reconstruction. The total effect of disasters on CO₂ was negative in annual data, reflecting activity suppression rather than structural decarbonization. Policy priorities include risk-informed infrastructure and land use, green reconstruction standards, and energy-system decoupling through efficiency, grid modernization, and renewable integration, supported by green finance and pricing reforms.