Studies of economic inequality and mobility depend on valid and reliable measures of economic resources. Income data is available from many sources, including census surveys, tax records, and other administrative data. Each data source has strengths and weaknesses. The Current Population Survey (CPS) and American Community Survey (ACS) have demographic information on population characteristics, but some income components (such as government benefits, self-employment income, interest, dividends, rents, and royalties) are underreported, especially for individuals in the lower and upper tails of the distribution. Tax data are more accurate than self-reports to surveys, but lack demographic detail and information on non-taxable income such as food stamps and the earned income tax credit. Administrative data from government agencies have information not available in Census surveys or Internal Revenue Service (IRS) records, but are typically only available for program participants and hold little information beyond what is needed to administer the program. None of these data sources independently provides a comprehensive measure of income.
Bruce Meyer and James Sullivan will construct a Comprehensive Income Dataset (CID) that combines survey, tax and administrative program data. This dataset will provide a better measure of income for families and households, with rich demographic and socioeconomic measures. The tax data will include individuals missed by the other datasets, more accurate earnings information, and income from other sources, such as realized capital gains, self-employment, interest and dividends. The administrative data will provide information on income not reported to the IRS such as Temporary Assistance to Needy Families (TANF), the Supplemental Nutritional Assistance Program (SNAP), and Supplemental Security Income (SSI). Linking these datasets will produce a more accurate income measure for researchers and policymakers. When joined with consumption data from the Bureau of Labor Statistics, the CID will foster analyses of the differences between the distribution of family income and the distribution of consumption.