The prediction that consumption-income ratios should decline as income rises in cross-sectional data is a feature of Friedman's (1957) permanent income hypothesis and other consumption-smoothing models. The theory thus provides a link between longitudinal income data and cross-sectional expenditure data: given measured income variability and a functional relationship between consumption and permanent income, we predict cross-sectional expenditure patterns and compare those predictions to actual values.