New Approaches for Measuring Poverty
Friday, Jan. 3, 2020 8:00 AM - 10:00 AM (PST)
- Chair: David S. Johnson, University of Michigan
Potential Improvements to the Census Bureau’s Supplemental Poverty Measure for 2021
AbstractFollowing decades of research on poverty measurement and guidance from a National Academies of Sciences Panel and the Interagency Technical Working Group on Developing a Supplemental Poverty Measure (SPM), in 2011 the Census Bureau released the first annual SPM report. Since this first publication, no major changes have been made to the SPM, but research has been ongoing at the BLS and Census Bureau on potential improvements and validation of prior assumptions. The Office of Management and Budget (OMB) has established a new interagency technical working group, which has set 2021 as a target for making methodological improvements to the measure.
This paper examines the impact of various proposed improvements to the SPM on poverty rates overall and for major demographic groups. The proposals under consideration include several changes to the methodology for establishing the SPM poverty thresholds including, but not limited to: changing the range of expenditures which serve as the basis for the thresholds, expanding the estimation sample upon which the thresholds are based, imputing noncash transfers into thresholds, and applying alternative geographic adjustments using Regional Price Parities produced by the Bureau of Economic Analysis and/or an adjustment to reflect amenities. The final change to the SPM under consideration in this paper would change the methodology for assigning a value to nutritional assistance programs in the calculation of available resources. Using data from the Survey of Income and Program Participation (SIPP) as well as administrative data, this proposal explores the possibility of imputing school breakfast values (in addition to school lunch) and allowing WIC values to vary by state. Poverty and related statistics will be presented which highlight the impact of the planned improvements to the SPM.
Poverty in the United States Using the Comprehensive Income Dataset
AbstractThis paper provides new estimates of poverty in the United States using a groundbreaking set of linked survey and administrative data. The administrative data cover earnings, asset, and retirement income from IRS tax records, as well as transfer income for a myriad of safety net programs including Social Security, SSI, SNAP, housing assistance, and veterans’ benefits. We link these data to the Current Population Survey (the source of official poverty and inequality statistics) and the Survey of Income and Program Participation (the most comprehensive survey of income sources in the U.S.). Linking the administrative data to the surveys is vital given that surveys miss a large and rising share of income sources. Using these linked data, we find in the CPS that 55% fewer individuals are in poverty after incorporating taxes and in-kind transfers and correcting for measurement error using the administrative data. The corrections for measurement error are more important in percentage terms than the conceptual changes to income. Accounting for service flows from home and car ownership in the SIPP removes an additional 12% of individuals from poverty. We observe a demographic shift in the composition of the poor, with our adjustments leading to significantly more single individuals and fewer families with children in poverty. These estimated reductions in poverty are partly due to surveys underestimating the value of the anti-poverty effects of government programs like SNAP, the EITC, housing assistance, and SSI. We also find that incorporating taxes and in-kind transfers leads to a weaker correlation between income poverty and material deprivation, while accounting for assets and the administrative data leads to a stronger correlation between income poverty and material deprivation.
David S. Johnson,
University of Michigan
University of Wisconsin-Madison
- I3 - Welfare, Well-Being, and Poverty