Estimating after-tax income using survey and administrative data

Abstract

For many years, the United States has not had a comprehensive, recurring series on after-tax income containing detailed demographic and socioeconomic information. In response to this need, statisticians at the Census Bureau have developed a model that simulates the payment of four different types of taxes: Federal individual income taxes, state income taxes, social security payroll taxes, and property taxes on owned properties. The simulations are based on a unique combination of data from household surveys and administrative data. The estimates can be compared with independent data to assess the quality of the simulations. Estimates of after-tax household income are presented for 1974 and 1980, and changes in the distribution of income during this period are examined. The analysis focuses on changes in tax incidence and burden for each of the different types of taxes. Extensions and applications of the tax model can be used to produce new data series, refine existing series, and address important tax policy issues. Some of the new extensions include the development of estimates of discretionary income and intercensal estimates of households and income.