Heterogeneity of rural poverty: an application of the “ordered data” model

This document analyzes the heterogeneity of rural poverty in Peru, proposing a classification that goes beyond the traditional distinction between the poor and the non-poor. Based on the 1997 National Survey of Living Standards and using a probabilistic model of ordered data, the study identifies up to four types of rural poverty defined by income level and the size of the agricultural holding. The results show that rural poverty is not a homogeneous phenomenon: factors such as education, access to basic services, housing characteristics, household size, use of agricultural technology, and participation in agricultural activities differentially affect the probability of belonging to each poverty category. The study concludes that policies to combat rural poverty must combine poverty alleviation and reduction strategies with more precise targeting mechanisms based on the productive capacities, available assets, and specific needs of each group of rural households.