The limitations of the official federal poverty measure are widely recognized (Blank, 2008; Iceland, 2005; Citro & Michael, 1995). The Supplemental Poverty Measure (SPM), introduced by the U.S. Census Bureau and Bureau of Labor Statistics in 2010, addresses many of the shortcomings of the official poverty measure. The SPM incorporates a more modern definition of the family unit, as well as poverty thresholds that are based on more current household expenditures for basic needs and adjusted for geographic differences in the cost of living. Moreover, the SPM includes near-cash benefits such as food stamps and the Earned Income Tax Credit in household resources, and subtracts non-discretionary expenses such as medical bills and child care from household income, thus better reflecting the resources actually available to households to pay for basic needs (Short, 2011).

Implementing the Supplemental Poverty Measure in multiple major national datasets enables comparison of annual SPM poverty estimates across different data sources and examination of different aspects of SPM poverty. This paper provides poverty estimates using the Supplemental Poverty Measure in the Panel Study of Income Dynamics (PSID), for the biennial years 1998 through 2010 (PSID, 2014). The PSID is a nationally representative, long-term panel study that contains comprehensive data on income, health, wealth, child development, and other topics, making it a broad research resource. As a long-term longitudinal dataset, the PSID particularly facilitates examination of the persistence of poverty.

This paper describes the methods used to construct the SPM in the PSID, and includes annual SPM poverty estimates for the years 1998, 2000, 2002, 2004, 2006, 2008, 2010 in the PSID, with comparison to SPM estimates for the same years derived from Current Population Survey data by the Census Bureau (Short, 2011) and independent researchers (Wimer, et al., 2013). Poverty estimates are calculated for the overall population and for three specific age groups: children, working-age adults, and seniors.  This paper also describes the persistence of poverty measured using the Supplemental Poverty Measure from 1998 through 2010, an exercise made possible by the longitudinal structure of the PSID.  The proportion of households poor for 1, 2, 3, 4, 5, 6, and 7 of the biennial survey years from 1998 to 2010 are examined using the SPM, with comparison to the proportion poor for the same numbers of years using the official poverty measure.