Development and Human Rights

This study addresses the methodological challenge of measuring poverty and vulnerability at fine territorial scales, where traditional household surveys lose statistical representativeness.To overcome the limitations of direct estimators-which often exhibit high variability in municipalities with small sample sizes-the ALIVIO methodology is implemented within the Small Area Estimation (SAE) framework.This approach adopts a unit-level model that borrows strength from auxiliary sources and spatial structures to stabilize municipal estimates.The results reveal a marked heterogeneity in the incidence of vulnerability across the country (where 1 represents a 100% probability).At the upper end of the spectrum, the states of Chiapas and Oaxaca concentrate the most critical levels.For instance, in Chiapas, municipalities such as Altamirano show an incidence of 0.90, reflecting near-total vulnerability validated by the model's spatial structure.In contrast, the lowest levels are located in northern Mexico and metropolitan areas; municipalities such as Mexicali (0.109) or Playas de Rosarito (0.095) in Baja California present significantly lower probabilities.A key finding of the ALIVIO application is the shrinkage or smoothing effect, which reduces erratic figures in small municipalities by pulling them toward regional structural patterns without introducing systematic biases.This provides authorities with a reliable tool for public planning, enabling the identification of poverty clusters that require urgent intervention.In conclusion, the methodology demonstrates that the integration of Bayesian models and administrative data allows for an accurate, calibrated, and statistically coherent mapping of poverty, which is essential for the design of targeted social protection policies.