An Enduring Partnership: Incorporating Administrative Data Into Sample Design for the Survey of Consumer Finances

• Activity Number: 162
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Barry Johnson*+
• Companies: Statistics of Income, IRS
• Keywords: Administrative Records ; Dual Frame Sample ; Household Survey ; Weights ; Imputation

Abstract

The triennial Survey of Consumer Finances (SCF) is distinguished as an outstanding source of information on income and balance sheets, demographics, and economic expectations of U.S. households. It includes representative coverage across the income and wealth spectra, enabling users to study assets that are widely distributed in the economy as well as those that are much more concentrated. This is made possible by a dual frame sample design that includes both a traditional area probability sample and a list sample that is used to oversample high-wealth households. Since its institution, the list sample has been developed using administrative data accessed through a partnership between the Federal Reserve Board of Governors (FRB) and the Statistics of Income Division (SOI) of the Internal Revenue Service. This partnership recognizes the essential contribution of the SCF in helping policymakers understand the impact of current and proposed tax policies on American households.

Section 1 provides background on the SCF, a brief history of SOI's support of the survey, and a discussion of the importance of these data to tax policy research. Section 2 describes the evolution of the list sample while Section 3 discusses benefits to SOI and FRB of partnering on the SCF. Some direct comparisons of SOI and SCF estimates are presented in Section 4 in order to demonstrate the importance of benchmarking results for improving data quality, and Section 5 provides concluding remarks. The goal of this paper is to document the benefits of partnering on projects of National interest as a model for advancing a wide range of institutional goals.