How the Urban Institute

How the Urban Institute Assessed Impacts of Policy Improvements

Child Poverty Measurement

Child poverty impacts were measured using the Supplemental Poverty Measure (SPM), an alternative poverty measure developed by the U.S. Census Bureau and the Bureau of Labor Statistics based on a broader range of income sources and costs than the official poverty measure. While the official poverty measure counts only earnings and cash benefits (such as social security and unemployment benefits), the SPM also counts in-kind benefits such as food, housing and energy assistance, and tax credits like the Earned Income Tax Credit (EITC) and the Child Tax Credit. It also deducts expenses such as taxes and child care, commuting and health care costs, and it takes into account variations in housing costs in different parts of the country.49 As a result the SPM is a more comprehensive measure of poverty that accounts for the impact of the government anti-poverty programs and policies CDF wanted to measure (see Appendix 2 for details).

The only policy change not fully captured by the SPM is increased availability of child care subsidies. In contrast to housing subsidies and SNAP benefits, the SPM does not count the value of child care ­subsidies, but instead deducts families’ out-of-pocket costs. Child poverty impacts measured by the SPM will therefore only be based on changes to families’ out-of-pocket child care costs and will appear lower, relative to costs, than the impacts of housing subsidies and SNAP benefits.

Simulation Model

The Urban Institute modeled the impact of the nine policy changes using TRIM3, a validated microsimulation model of the tax and benefits programs affecting U.S. households. TRIM3, which models the U.S. non-institutionalized population based on U.S. Census and federal program and tax data, is a well-respected tool used for over 40 years to assess the operation of the U.S. safety net and to estimate the potential impacts of changes to safety net programs and policies.50

The analysis was based on data representing the U.S. in 2010, the most recent year available at the time the project began. While the economy has improved since 2010, the number of children below 100 percent of the official poverty threshold has decreased only 10 percent from 16.2 million in 2010 to 14.7 million in 2013.51 All policies in place in 2010 were assumed to be in effect in the simulation, including the ARRA SNAP benefit increase and the improvements to the EITC and the CTC. The only provision that was excluded from the baseline was the temporary Making Work Pay tax credit, which was in place only in 2009 and 2010. 52

Child SPM Poverty in 2010

Prior to the policy changes there were 10.9 million poor children in 2010 according to the Urban Institute’s SPM calculations, resulting in a child poverty rate of 14.6 percent. This child SPM estimate is lower than the Census’ SPM estimate of 18.2 percent because TRIM3 corrects for under-reporting of certain survey-reported resources, including receipt of SNAP, subsidized housing, Supplemental Security Income and TANF, and because TRIM3 uses a different methodology to impute taxes paid. Characteristics of poor children based on Urban Institute’s model are presented in Appendix 2.

For more details about the Urban Institute’s methods please refer to the Urban Institute’s technical report.53