Skip to main content

Declines in Unemployment Benefits and Government Employment Shaped Poverty Trends in 2011, Preliminary Data Suggest

The Census Bureau will release official 2011 poverty figures on September 12.  If the figures show that poverty rose, as some analysts predict, two key factors will almost certainly be a reduction in unemployment insurance benefits in 2011 and a decline in public-sector jobs, particularly among state and local government workers.

In Early Data, Two Factors Explain a Rise in Poverty

A preliminary analysis of Census data for the first 11 months of 2011 (which are the latest data now available) indicates that the average share of people with monthly cash income below the official poverty line rose in 2011 by about half a percentage point.  The combination of a decline in unemployment insurance (UI) benefits and a loss of public-sector earnings accounted fully for this increase in monthly poverty, according to the analysis.  (The weak performance of the broader economy in 2011 accounts for the lack of improvement in the poverty ratebetween 2010 and 2011.)

These preliminary estimates are likely to differ somewhat from the official poverty figures that will be released on September 12, which will come from a different and larger Census survey and will be based on annual rather than monthly data.  But the finding that declines in UI benefits and public-sector employment combined to exert substantial upward pressure on poverty rates is solid. 

The preliminary monthly data are from a CBPP analysis of the Census Bureau’s Survey of Income and Program Participation (SIPP).[1]   The SIPP data show that:

  • Income from UI benefits kept a monthly average of 0.8 percent of the U.S. population above the federal poverty line in the first 11 months of 2011.  This was down from the same months of 2010, when UI benefits kept an average of 1.3 percent of Americans out of poverty.  Thus, the decline in UI added half a percentage point to the monthly poverty rate in 2011, relative to what the rate otherwise would have been.
  • Declining income from public-sector employment similarly added 0.2 percentage points to the poverty rate during the first 11 months of 2011 compared with the same months of 2010, affecting government workers and their family members.
  • UI and public-sector jobs together kept an average of 25.0 million people above the monthly poverty line in the first 11 months of 2011.  This was about 1.9 million fewer than in the same period in 2010 (26.9 million).

The following figures help to explain this effect.  The SIPP data, which follow the same individuals from one year to the next, indicate that roughly 900,000 people, representing about 0.3 percent of the U.S. population, appear to have dropped below the poverty line in 2011 after someone in the family lost a substantial amount of UI income.  (These 900,000 people are members of families that lost more than $1,000 of UI benefits between 2010 and 2011 and were kept above the poverty line by their UI income in 2010, but fell into poverty in 2011.)  Another 666,000 people lost significant family earnings from state and local government work in 2011 and became poor as a result.

Together, these two factors appear to account for all of the half-a-percentage-point increase in the monthly poverty rate that the SIPP data show to have occurred in the first 11 months of 2011.

As explained below, the declines in UI benefits and government employment reflect a combination of forces including the expiration of temporary anti-recessionary initiatives, the ongoing weakness of the economy, and other factors.

Contributions of Reduced UI and Government Job Losses to Poverty are Consistent with Other Data

The finding that the decline in UI and public-sector job losses increased poverty from 2010 to 2011 is broadly consistent with Labor Department data confirming sizable declines in UI payments and reductions in public-sector payrolls in 2011.

  • Unemployment benefits, from both federal and regular state UI programs, reached 7.7 million workers in an average week in 2011, down from 9.8 million in 2010.  The Labor Department has reported that the inflation-adjusted value of unemployment benefits declined by 25 percent, or $36 billion, between 2010 and 2011.[2]   (About $10 billion of the decline was due to the expiration in late 2010 of a temporary provision of the 2009 Recovery Act that had paid an extra $25 a week to UI beneficiaries.  In addition, large numbers of workers continued to exhaust their long-term UI benefits before finding a job, although the number of new cases entering the UI system declined as well.)
  • Government jobs declined by 386,000, or 1.7 percent, from 2010 to 2011, according to full-year Labor Department figures.  This includes a loss of 110,000 state and local education jobs, 157,000 other state and local government jobs, and 28,000 postal service jobs (as well as a number of temporary federal positions related to the decennial census).  By contrast, the number of private-sector jobs grew by 1.7 percent.[3]

The relationship of UI and government jobs to poverty is clear in poverty data for 2010.  According to the official, full-year Current Population Survey (CPS) data from the Census Bureau, UI benefits kept 3.2 million Americans above the poverty threshold in 2010.[4]    The earnings of workers whose primary job in 2010 was for local, state, or federal government kept another 20.5 million workers and family members above the poverty line, according to a Center analysis of CPS data.[5]   Since both UI and public-sector jobs kept large numbers of Americans out of poverty in 2010, it is not surprising that declines in these sources of income would keep the poverty rate elevated. 

Final 2011 Poverty Rates May Not Match These Preliminary Findings

Estimates in the analysis of SIPP data are likely to differ somewhat from the final, official poverty figures for 2011 for at least three reasons.  First, the official figures will come from a different survey — the Annual Social and Economic Supplement of the CPS.  The SIPP and the CPS each have strengths and weaknesses.  The CPS is based on a larger (and thus slightly more reliable) sample, and CPS respondents appear to report earnings more accurately.  SIPP is smaller but tends to measure most government benefits more accurately.

Second, the official CPS figures are based on a family’s annual income compared with an annual poverty line, while the SIPP analysis is based on a family’s monthly income compared with a monthly poverty line.  The SIPP analysis is therefore more likely than the CPS figures to be affected by changes in brief spells of joblessness and other short-term economic setbacks or advances.[6]

Third, the analysis based on SIPP data covers only the first 11 months of each year.

Whether or not the official, full-year poverty rate rose in 2011, the data from SIPP and the Department of Labor make clear that reductions in unemployment benefits and public-sector layoffs put upward pressure on poverty.  This pressure may have been large enough by itself to increase the poverty rate, despite the modest improvements in some other leading determinants of poverty, such as private-sector employment.

[2] CBPP compilation of data for UI regular state programs (at and the Emergency Unemployment Compensation program and the Extended Benefits program (at, adjusted for inflation.

[3]   Bureau of Labor Statistics data from the Current Employment Statistics survey at

[4] U.S. Census Bureau, “Income, Poverty, and Health Insurance Coverage: 2010,” slide presentation, September 13, 2011.

[5] In total, UI and earnings from government jobs kept 23.7 million people above the poverty line in 2010, according to the CPS data.  This estimate is similar but not identical to the 26.9 million number under SIPP, with a difference of this size not being surprising given the differences between the surveys and the difference between annual and monthly poverty status.

[6] Data from previous years show that monthly poverty rates tend to be slightly higher than annual poverty rates, reflecting the fact that some spells of monthly poverty do not last a full year.  In 2010, an average of 16.5 percent of the population was poor on a monthly basis in the first 11 months of 2010, according to SIPP data, while CPS data based on annual income showed that 15.1 percent of the population was poor in 2010.  Both sets of figures increased from 2009 to 2010; the average monthly poverty rate in the SIPP rose by 0.5 percentage points and the annual CPS poverty rate rose 0.8 points.

End Notes

[1] U.S. Census Bureau, Demographics Survey Division, Survey of Income and Program Participation branch, revised June 6, 2012,