off the charts

The 2009 Recovery Act — Even Better in Preventing Poverty Than We Thought

We previously described the 2009 American Recovery and Reinvestment Act (ARRA) as one of the “most effective pieces of anti-poverty legislation in decades,” saying its temporary expansion of the safety net kept 4.5 million people out of poverty in 2009.  Actually, the impact was greater than we thought. We now know that, when one follows recommendations from the National Academy of Sciences (NAS) on how to measure poverty more comprehensively, seven of ARRA’s provisions kept more than 6 million Americans above the NAS poverty line – that is, out of poverty – in 2009.
These ARRA effects (shown in the graph in red) included two new benefits — the Making Work Pay tax credit and a one-time payment of $250 to certain recipients of retirement and disability benefits — as well as expansions of existing benefits, such as the Child Tax Credit, Earned Income Tax Credit, unemployment insurance, and SNAP (formerly known as food  stamps). Unlike the official poverty figures, which count family income from cash, the NAS-based measures count family income from both cash and cash-like sources and tax credits.  They also subtract child care and other necessary expenses from income and use a slightly modernized poverty line.  (The newest refinement of the NAS-based measures is called the Supplemental Poverty Measure, which the Census Bureau will unveil later this month.) NAS-based poverty measures generate poverty rates that are modestly above the official 14.3 percent poverty rate (in this case, 15.7 percent in 2009 under the NAS version that most closely matches the original NAS recommendations) but are designed to reveal more accurately the composition of who is poor and who is not.  A major reason that our revised ARRA estimate is so much larger appears to be that the NAS measure — with its deductions for child care and other necessary expenses and its slightly higher poverty line — focuses more than the non-NAS measure we originally used on identifying the needs of low-income working families, a group that received substantial help from ARRA. Later this month, the Census Bureau will release 2010 data under a variety of experimental poverty measures that are based on the NAS recommendations, including the new Supplemental Poverty Measure.  Like the official poverty figures released last month, these figures are likely to show that poverty got worse in 2010.  But they also will enable analysts to identify the effects on poverty of assistance that ARRA provided to families and individuals in the form of tax credits, nutrition assistance, and other programs. In addition, in a paper for an upcoming academic conference, I note that, compared with the official poverty measure, the NAS measures show much less of an increase in poverty between 2008 and 2009.  That’s both because the safety net responded automatically — as it should — to the collapsing economy, absorbing growing numbers of increasingly impoverished applicants and participants, and because of the ARRA provisions discussed above.