To adjust or not to adjust? Spatial price variation and the measurement of poverty*

Abstract

Publicly provided goods and services influence household consumption levels. Consumption estimates based solely on private expenses are therefore biased. Given that we usually are unable to account for consumption from public provision, I argue that we often should ignore spatial price differences in inter-household comparisons. The key reason is that prices and levels of public provision are likely to be positively correlated across space. I provide a simple framework to evaluate this, and I apply it to the measurement of poverty in India. The analysis reveals that a uniform poverty line might be preferable to the currently used price-adjusted poverty lines.