Pasar al contenido principal

What to Look for in Wednesday’s Poverty Data — And What the Official Data Won’t Tell Us

On September 12, the Census Bureau will release official poverty figures for 2011, as well as additional data related to the impact of various safety net programs in keeping people out of poverty last year.

What to Look For:

  • Poverty could rise again; if it does, it will have risen significantly in 8 of the last 11 years.  Such an increase in 2011 would continue a trend that began with the recession of 2001, continued through the last economic expansion (an unusual development since poverty usually declines during economic recoveries), and accelerated after 2007 due to the Great Recession.  The unusually weak recovery from 2001 to 2007 marked the first sustained economic expansion on record in which growth was so weak and its monetary benefits sufficiently unevenly shared that the official poverty rate was higher by the end of the recovery in 2007 (12.5 percent) than when the recovery began in 2001 (11.7 percent).
  • Declines in Unemployment Insurance (UI) benefits and government jobs appear to account for much, if not all, of the upward pressure on the official poverty rate in 2011.  An analysis of preliminary, monthly Census data for January through November 2011 finds that losses of income resulting from declines in unemployment benefits and public-sector job loss pushed the monthly poverty rate upward by more than half a percentage point, compared to the same months of 2010.  Rising poverty in response to the loss of UI income reflects several factors, including the expiration of a temporary $25 per-week increase in UI benefits that the 2009 Recovery Act had put in place, as well as the exhaustion of UI benefits for many long-term jobless workers whose benefits ran out before they could find work.  The loss of 386,000 local, state, and federal government jobs in 2011 also contributed to the expected rise in poverty. (By contrast, private-sector employment grew in 2011.) These Census data and findings are described in a CBPP analysis issued September 7.  The Census data issued September 12 will include data on the number of people kept out of poverty by UI in 2011 and shed further light on this issue.
  • The effect of SNAP (food stamp) benefits and the Earned Income Tax Credit on poverty.  As part of the data that it issues on September 12, Census is expected to provide data on the impact of SNAP and the EITC on poverty if those benefits are counted as income in measuring poverty, as most analysts across the political spectrum believe they should be.  This information will be relevant to current debates about whether to cut such programs substantially.  (For example, the budget the House passed last spring would cut SNAP by more than $133 billion over ten years.)

What the Official Poverty Figures Will Not Show:

  • A meaningful comparison back to the 1960s.  If the official poverty rate is found to have increased in 2011, it may well have reached its highest level since the 1960s, but this will not be a meaningful comparison.  That is because the official poverty definition is based on families’ pre-tax, cash income.  It ignores all non-cash benefits, such as food stamps, low-income housing assistance and the EITC.

    This leads to severe distortion when official poverty figures for the 1960s and 1970s are compared to today’s figures.  Since those decades, cash welfare assistance — the principal means-tested benefit for many years, and one that is counted in the official poverty figures — has declined substantially while non-cash benefits for food and housing and refundable tax credits (like the EITC) — which are not counted in the official measure — have expanded substantially.  The result is an apples-to-oranges comparison that obscures more than it illuminates. 
  • The full effect of the safety net.  For several years, Census has reported a variety of alternative poverty measures (culminating in its “Supplemental Poverty Measure”) that reflect a fuller range of benefits, including EITC, SNAP, housing assistance, school lunches, and others, as well as the effect of income and payroll taxes, work expenses, a modestly modernized poverty line, and other changes in the measurement of poverty that leading scholars have called for.  The measures are based on the recommendations of the National Academy of Sciences (NAS) and are preferred by many experts.  From 2007 to 2010, as the official poverty rate rose sharply, these NAS-based measures were much flatter — thanks chiefly to expansions in working-family tax credits and SNAP that the official poverty rate ignores.[1]   Under these more sophisticated and comprehensive alternative poverty measures, the increases in poverty since 2007 have been strikingly modest given that the economy has gone through the most serious downturn since the 1930s.  The Census Bureau expects to release NAS-based measures for 2011 later this year, but those measures will not be available on September 12.  (Note: even these alternative measures of poverty will not permit comparisons back to the 1960s or 1970s, because no NAS-based or similar poverty estimates are available for those years.)

End Notes

[1] U.S. Census Bureau, “Official and National Academy of Sciences-based Poverty Rates, 1999-2010,”