The relationships between types of wealth components and consumption expenditures were investigated as a means of better understanding retired elderly household well-being. Specifically, the concept of mental account was used to identify the characteristics of different types of wealth components, and four mental accounts were identified: flow of investment, current asset A, current asset B, and future income. Based on the traditional life-cycle hypothesis, the behavioral life-cycle hypothesis, and neo-classical demand theory, the consumption functions for the total and 17 subcategorical consumption expenditures were formulated. These consumption functions were formulated to study linkages between household portfolio behavior during the working years and household consumption behavior during retirement. A tobit linear regression model was utilized to estimate parameters in consumption functions. The data were drawn from the Consumer Expenditure Survey, Interview Survey, 1990. The flow of investment mental account includes Social Security benefits, pension benefits, and transfer payments from public programs. The current asset A includes balances in checking and savings accounts, and the current assets B includes balances in stocks and bonds. The future income includes market values of home equity and real estate. The findings supported that the total and subcategorical consumption expenditures are the most sensitive to changes in flow of investment and the least sensitive to changes in future income. Further, among retired elderly households, the four mental accounts differ in influence on subcategorical consumption expenditures. The flow of investment was positively related to food at home, food away from home, utilities, household operation, clothing, transportation, entertainment, personal care, and cash contributions. Current asset A was positively related to health care, reading and education, and alcoholic beverages, and negatively related to food at home. Current asset B was positively related to clothing, transportation, entertainment, reading and education, and alcoholic beverages. Future income was positively related to food at home, utilities, household operation, and personal care. The research findings may help public policy makers understand or predict consumption expenditures as wealth components change in retired elderly households. Further, the economic well-being of retired elderly households should be discussed in terms of different types of wealth components rather than in terms of total wealth. / Graduation date: 1995