How reliable are household expenditures as a proxy for permanent income? Implications for the income-nutrition relationship

Abstract

Measurement error in short-run expenditures from household surveys may attenuate estimated effects of permanent income on economic outcomes. Repeated observations on households during the year are used to calculate reliability ratios and estimate errors in variables regressions of the impact of income on calorie intakes. In contrast to influential studies finding no effect of income, the results suggest significant nutritional responses to income in poor countries.