Date of Graduation
Davis College of Agriculture, Natural Resources and Design
Agricultural & Extension Education
Dale K. Colyer.
While the traditional methods of measuring income inequality reveals interesting and important features of labor markets, estimates of income inequality do not provide complete summary statistics of the distribution of well-being. Interpreting the Gini coefficient of family earnings as a measure of disparity in welfare implicitly assumes that households with the same level of before tax income are equally well-off. Utility is derived from the consumption of goods and services and there is ample empirical evidence which indicate that the distribution of total expenditure is likely to be different from the distribution of income. As Friedman indicated households with low income levels are disproportionately represented by with those temporary reductions in current income and will typically have high ratios of consumption to income. Households with high income levels are over represented by those with transitory increases in income and will exhibit low ratios of consumption to income. The implication is that, all other things equal, one would expect less dispersion in the distribution of total expenditure relative to the income distribution. The problems of using family income as a measure of well-being go beyond the differences in the expenditure and income distributions. Treating heterogeneous households symmetrically, as is common in income inequality studies, indirectly assumes that two households with the same level of income but different sizes are equally well-off. If household characteristics influences household expenditure patterns, consumption needs and welfare, such effects are likely to have an important effect on both the level and trend of inequality in the distribution of welfare. Consumption rather than income may be a better measure of the actual economic welfare of a household than its current income.;The main objectives of this study are to: (1) Measure the impact of demographic characteristics on the distribution of individual expenditures on the consumption of goods and services. (2) Examine the inequality in the distribution of household consumption expenditures using the Gini coefficient.;The data are drawn from the 1980 through 1994 interview surveys conducted and gathered by the U.S. Labor Statistics. An econometric method using the translog model is specified for seven commodity groups. We estimated six equations using Seemingly Unrelated Regression procedure.;The parameters estimated to measure the impacts of demographic characteristics on the consumption of goods and services indicated that the allocation of budget shares for different consumer goods is affected by the composition and characteristics of households.;The decomposition of the Gini provides specific information concerning the concentration of consumption expenditures by budget components, and information about how the marginal changes in particular expenditures affect overall inequality. Unlike in income distribution, the Gini estimates in expenditure distribution appear to be closer. It is possible to conclude that expenditure on goods and services does depend not only on current income, but also on income earned through the years. This substantiates the permanent income hypothesis theory articulated by Friedman (1957). Being able to measure these impacts can be useful for policy makers interested in the effect that certain programs may have on the spending patterns of households. Results obtained in this study substantiate the importance of evaluating the differential impacts of proposed and enacted policies on subgroups of the population and differences in inequality which can result when expenditures for budget components change. Without adequate evaluation, policies and programs intended to decrease inequality may lead to the opposite outcome in the distribution of material well-being across household units in the population.
Berhanu, Samuel, "Econometric analysis of household expenditures" (1999). Graduate Theses, Dissertations, and Problem Reports. 3124.