Related BLS programs | Related articles
December 1994, Vol. 117, No. 12
Imputing income in the Consumer Expenditure Survey
Geoffrey D. Paulin and David L. Ferraro
In the Consumer Expenditure (CE) Surveythe only U.S. government survey that releases expenditures of consumers in the United States to demographic characteristicsincome is the characteristic most often used in research and analysis. Indeed, nearly 90 percent of respondents to a recent survey of CE Survey data users reported using the data by income class.1
Income data are used in two ways: as a way to classify and as a demographic characteristics. Consumer units2 are classified by income quintile and level of income before taxes in tables published routinely from the CE Survey. Many users of the data analyze the relationship between income as a demographic characteristic and expenditures. For example, John Sablehouse used the data to assess the impact of taxing consumption versus income,3 and William S. Reece estimated the relationship of income of different types of charitable contributions.4 Other researchers have used income data from the survey to compare expenditure patterns across various groups5 and to estimate linear Engel curves.6
Household surveys, such as the CE Survey, are subject to nonresponse and underreporting. Some respondents provide information on some, but not all, sources of income or on some, but not all, members of the family. The respondents does not always know detailed information on family income, and sometimes the respondent refuses to provide information. In all such cases, the families are classified as nonrespondents, at least with regard to the particular member or source not reported. Underreporters are families who answer the questions, but reduce the real amount. E. Raphael Branch (p. 47, this issue) evaluates the underreporting of income and expenditures.
This excerpt is from an article published in the December 1994 issue of the Monthly Labor Review. The full text of the article is available in Adobe Acrobat's Portable Document Format (PDF). See How to view a PDF file for more information.
Read abstract Download full text in PDF (543K)
1 Consumer Expenditure Data User Survey Results, Report 832 (Bureau of Labor Statistics), March 1993, p. 2.
2 A consumer unit is the basic unit of analysis in the CE Survey. It is defined as single person living alone or sharing a household with others, but who is financially independent; members of a household related by blood, marriage, adoption, or some other legal arrangement; to or more persons living together who share responsibility for at least tow of three major types of expenses - food, housing, and other expenses. The terms family and families are substituted throughout for convenience.
3 John Sabelhouse, "What Is the Distribution Burden of Taxing Consumption?" National Tax Journal, September 1993, pp. 331-44
4 William S. Reece, "Charitable Contributions: new Evidence on Household behavior," American Economic Review, March 1979, pp. 142-51
5 Older workers and nonworkers (Thomas Moehrle, "Expenditure Patterns of the Elderly: Workers and Nonworkers," Monthly Labor Review, May 1990, pp. 34-41); single- and dual-earner families (Rose M. Rubin, Bobye J. Riney, and David J. Molina, "Expenditure Pattern Differences between One-Earner and Dual-Earner Households: 1972-73 and 1984," Journal of Consumer Research, June1990, pp. 43-52); and ethnic groups (F.N. Schwenk, "Income and Consumer Expenditures of Household headed by Hispanic and Elderly Women," Family Economics Review, 1994, pp. 2-8).
6 Barbara A. Sawtelle, "Income Elasticities of Household Expenditures: A U.S. Cross-Section Perspective," Applied Economics, May 1993, pp. 635-44.
Within Monthly Labor Review Online:
Welcome | Current Issue | Index | Subscribe | Archives
Exit Monthly Labor Review Online:
BLS Home | Publications & Research Papers