Measuring Inequality And Poverty When Income Data Are Missing: Can Household Assets Serve As Practical Proxies?

Abstract

This thesis investigates whether administrative records of housing and vehicles can serve as reliable proxies for measuring economic inequality and poverty in data-scarce environments. Using data from the United States (PSID/SCF) and Indonesia (IFLS), the study evaluates these proxies against a comprehensive suite of indices, including the Gini, Atkinson, and FGT measures. The results demonstrate that asset-based distributions, particularly vehicle ownership, carry a genuine distributional signal that closely tracks income inequality and reacts consistently to macroeconomic shifts like changes in unionization and unemployment. While the findings show that these proxies effectively capture broad social stratification, they also highlight a structural limitation in capturing the extreme upper tail where wealth shifts from physical assets to financial instruments. Ultimately, the study concludes that administrative asset registries can offer a robust, low-cost, and independent monitoring mechanism for identifying distributional patterns when traditional household surveys are unavailable or unreliable.