Evaluating Tax Data Generated Using the Survey of Consumer Finances and TAXSIM

• Activity Number: 181
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): John Sabelhaus and Kevin Moore*+ and Robert Argento
• Companies: Federal Reserve Board and Federal Reserve Board and Federal Reserve Board
• Keywords: Survey of Consumer Finances ; TAXSIM ; tax data

Abstract

This paper details the method used to generate federal and state tax rates and liabilities using the Survey of Consumer Finances (SCF) and NBER TAXSIM. The detailed information in the SCF on income, assets and liabilities allows us to create the variables necessary for the TAXSIM calculator. One critical issue in computing tax rates and liabilities is that the unit of observation in the SCF is the household, not the tax unit. We take two approaches to dealing with this issue; the first is to assume the household and tax unit are identical, while the second attempts to split households into tax units. The paper compares the tax data generated by the two methodologies on the number of filers by filing status, the number of itemizers, tax rates and tax liabilities to estimates published by the Internal Revenue Service (IRS). We find that assuming that the household and tax unit are identical generates too few single and head of household filers relative to the methodology that splits households into tax units. The results for tax liabilities and rates are more ambiguous, as neither methodology appears to dominate in matching the distribution of IRS tax rates and liabilities.