Spatial diagnosis for precision policy: a tri-model framework to target poverty interventions and measure spillovers in Tanzania

Persistent spatial disparities in poverty across Africa necessitate a decisive shift beyond uniform policy approaches. This study develops and applies an integrated geospatial framework for the precision targeting of poverty interventions and the measurement of their cross-border spillovers within Tanzania. The novel methodological contribution lies in the synthesis of three spatial econometric techniques—spatial autoregressive, geographically weighted regression and spatial Durbin models—enabling the simultaneous quantification of global spillovers, local heterogeneity and context-specific direct and indirect effects. Applied to high-resolution data, the framework reveals significant geographical variations in the drivers of poverty that are obscured by national-level analysis. Key findings demonstrate that the economic returns on education investments exhibit profound spatial heterogeneity, with premium returns in urban centers and constrained returns in rural agricultural regions. The analysis further identifies significant positive spillover effects, confirming that interventions generate measurable economic benefits beyond administrative boundaries, and pinpoints priority zones where high poverty incidence converges with acute climate vulnerability. This integrated tri-model approach provides policymakers with a practical tool to optimize regional resource allocation, target climate-poverty hotspots and improve the efficiency of public investments. The framework offers a replicable, scalable diagnostic for enhancing the spatial precision of poverty reduction strategies across Tanzania and comparable sub-Saharan African contexts.