The research identifies non‑monetary correlates of household consumption to predict poverty incidence using survey data. Separate welfare functions for urban and rural areas are applied to provincial and district levels, enabling poverty estimation through light monitoring tools such as the CWIQ. The results show that 33 percent of the population was poor in 2001-02, with rural areas facing higher incidence, depth, and severity than urban counterparts. The findings provide policymakers with a consistent framework for tracking poverty and designing targeted interventions.
