Skip to main content

TANF Responded Unevenly to Increase in Need During Downturn (with state-by-state fact sheets)

Findings Suggest Needed Improvements When Program Reauthorized

Nationally, the Temporary Assistance for Needy Families (TANF) program, which provides basic assistance to low-income families with little or no income, has only been modestly responsive to the economic downturn.  Using data collected directly from the states, [1] we estimate that between December 2007 and December 2009, TANF caseloads increased by just 13 percent, while Supplemental Nutrition Assistance Program (SNAP, formerly known as food stamps) caseloads grew by 45 percent and the number of unemployed persons doubled.  (See Figure 1 and Appendix Table B-1.)  Moreover, in 22 states, TANF caseloads responded very little or not at all to the recession:  16 states had caseload increases of less than 10 percent and six states had caseload declines. 

This paper examines TANF’s responsiveness during the first two years of the economic downturn both nationally and at the state level; accompanying fact sheets for each state and the District of Columbia document trends in the TANF caseload since October 2006. [2]  This paper is the first in a series that will examine how the TANF caseload is responding to the rise in joblessness and the accompanying economic hardship that has defined the Great Recession, which officially began in December 2007 and ended in June 2009.  While the recession may have technically ended, hard economic times continue and a long and slow recovery is anticipated.  Future analyses will report on how the TANF caseload changes as the economic recovery progresses.   

The Context:  Sustained High Unemployment Creating Widespread Hardship

Image
1-25-11tanf-f1.jpg
The recent recession has been characterized not only by high unemployment, but also by the speed at which the ranks of the unemployed swelled and the long periods of time for which many people have remained unemployed.  During the period of our analysis, from the start of the recession in December 2007 through December 2009, the unemployment rate more than doubled, increasing from 4.9 percent to 10 percent.  (It peaked at 10.1 percent in October 2009.)  Over this two-year window, the number of persons who were unemployed also more than doubled, increasing from 7.4 million to 15.0 million.  (See Appendix Table B-2.)  And in December 2009, 6 million of the 15 million unemployed workers had been unemployed for 27 weeks or longer. [3]  The unemployment rate has since declined slightly but remains very high (9.4 percent as of December 2010) and is predicted to remain high for the foreseeable future. [4] 

The recession followed a disappointing economic recovery that started at the end of 2001 and culminated in 2007.  That recovery marked the first time on record that poverty and the incomes of typical working-age households worsened despite six years of economic growth.  The share of Americans in poverty climbed from 11.7 percent in 2001 to 12.5 percent in 2007, while the median income for working-age households (those headed by someone younger than 65) declined from $59,640 in 2001 to $56,575 in 2007 (in inflation-adjusted 2009 dollars). [5] 

The situation worsened from 2007 to 2009:  the number of people in poverty increased by 6.3 million to 43.6 million, the poverty rate rose to 14.3 percent (its highest level since 1994), and the median income of working-age households declined by 4.6 percent or $2,700, after adjusting for inflation (falling to its lowest point since 1996). 

The percentage of people in deep poverty — that is, with incomes below half the poverty line — rose from 5.2 percent in 2007 to 6.3 percent in 2009, which is the highest level on record with data going back to 1975.  (Half of the poverty line corresponds to an annual income of $9,155 for a family of three.)  The 19 million people in deep poverty made up 43.7 percent of all poor people, which was the highest share ever recorded.

Recent economic news suggests poverty will be even higher in 2010 and 2011 than in 2009. Poverty is likely to remain high even longer than unemployment does:  in the last three recessions, poverty did not begin to decline until a year after the annual unemployment rate started to fall.

After unemployment insurance, SNAP historically has been the most responsive federal program during economic downturns, providing additional assistance to families and individuals who see their economic circumstances change during periods of high and rising unemployment.  This economic downturn has been no exception:  national SNAP (Food Stamp Program) enrollment is at an all-time high (see Figure 2).   

Image
1-25-11tanf-f2.jpg
In December 2009, 17.9 million households with 39 million individuals were participating in SNAP.  Nationally, caseloads rose by 5.6 million households (45 percent) since the official start of the recession. [6]  Caseloads increased in every state, with all states (including DC) experiencing all-time highs since the recession began. [7]  (See Appendix Table B-1.)

Some of the states hit hardest by the recession saw the largest food stamp caseload increases between December 2007 and December 2009.  For example, in Nevada, where the number of unemployed increased by 157 percent, the number of food stamp cases increased by 89 percent during this period.  In Florida, where the number of unemployed increased by 149 percent, the number of food stamp cases increased by 83 percent. 

Food stamp caseloads can grow for two reasons:  because more households are becoming eligible for the program and enrolling, or because a larger share of already-eligible households is signing up for the program.  During most of the early to mid-2000s, both were occurring.  With a weaker economy, the number of eligible households climbed from 14 million in 2000 to 18 million by 2005.  Meanwhile, the participation rate among eligible households increased from 50 percent in 2000 to 63 percent in 2006, as states and the U.S. Department of Agriculture (USDA) worked to increase enrollment by eligible households.  The rapid caseload growth in more recent years appears to primarily reflect the fact that more households are becoming eligible because of the recession.  Between 2006 and 2008, USDA estimated a modest increase in the participation rate (from 62.7 percent to 63.7 percent), which suggests that the growth in participation rates may have leveled off. [8]    

Moreover, while the recession is the principal reason for food stamp caseload increases over the 2007 to 2009 period, other factors could have contributed to caseload growth.  In particular, effective April 2009, the economic recovery act temporarily lifted the 3-month time-limit that unemployed childless adults otherwise face.  In addition, during 2007 to 2009 period, several states adopted a state option to expand food stamp eligibility modestly (known as expanded categorical eligibility) and states may also have engaged in targeted outreach (for example, to provide application assistance to seniors or working families) or improved their enrollment practices (by, for example, adopting an online application or call center). 

TANF Caseloads Have Risen Substantially in Some States But Not in Others  

In marked contrast to food stamps, TANF has been only modestly responsive to rising unemployment and greater economic hardship.  The total number of families with children receiving cash assistance increased by just 13 percent between December 2007 and December 2009, despite a doubling of the number of unemployed persons.  However, the national data mask significant state-by-state variation (see Figure 3):

  • Sixteen states had caseload increases of at least 20 percent; in eight of them, caseloads increased by more than 30 percent. 
  • Thirteen states had caseload increases of between 10 and 20 percent. 
  • In the remaining 22 states, TANF caseloads responded very little or not at all:  16 states had caseload increases of less than 10 percent and six states had caseload declines.

As Figure 4 shows, six of the eight states with the biggest caseload increases were in the West and Southwest.  The states whose TANF caseloads declined or increased very little were concentrated in the central and northeastern regions of the country.

A comparison of TANF caseloads over a different time period could reach different conclusions than our analysis.  Most states experienced caseload declines after the 2006 TANF changes in the Deficit Reduction Act, but the points at which caseloads subsequently began to rise vary significantly among states.  A few states’ caseloads were already rising in the 15 months prior to the start of the recession (October 2006 to December 2007); other states’ caseloads only began rising during the recession, and some states’ caseloads did not begin to increase until after December 2009.  For some states, caseloads were near peak or flattening out by December 2009, while in others, the growth was just beginning.  

Changes in state TANF caseloads prior to the recession are also an important factor to consider when assessing TANF’s responsiveness to the recession.  In the 15 months prior to the recession, Colorado’s caseload declined by 28 percent, more than any other state.  While Colorado also had the largest caseload increase from December 2007 to December 2009 (48 percent), its December 2009 caseload was only modestly (6 percent) above its October 2006 caseload because of the dramatic pre-recession decrease.  This is a contrast to Nevada, whose caseload increased by 20 percent between October 2006 and December 2007 and another 30 percent during the recession; the pattern here is one of continual climb.  (See Appendix Table B-4.) 

Variations in Unemployment Don’t Fully Explain Different TANF Caseload Responses

Image
1-25-11tanf-f3.jpg
The economic downturn has been severe but has not hit all states equally.  For example, the peak unemployment rate was 14.5 percent in Michigan but just 4.4 percent in North Dakota.  Given this wide variation, we would expect changes in TANF caseloads during the recession to vary as well.  However, there is less of a relationship between a state’s unemployment situation and changes in its TANF caseload than one might expect — states with similarly high or low unemployment rates registered very different changes in their TANF caseloads.  For example, in Michigan, with a peak unemployment rate of 14.5 percent, the TANF caseload increased by just 6 percent, whereas in Nevada, with a peak unemployment rate of 13.0 percent, the TANF caseload increased by 30 percent.  Oklahoma and Montana both had peak unemployment rates of 6.7 percent, but Montana’s TANF caseload increased by 21 percent while Oklahoma’s increased by 10 percent.

Likewise, states with similar changes in TANF caseloads registered very different changes in unemployment.  The eight states with the largest caseload increases had peak unemployment rates that ranged from 6.8 percent to 13 percent, and the six states with caseload declines had peak unemployment rates that ranged from 4.4 percent to 12.7 percent.  Three of the six states with caseload declines saw the number of unemployed more than double during the recession. 

Rhode Island experienced the nation’s largest caseload decline (29 percent), even though its peak unemployment rate of 12.7 percent was the third highest in the nation.  Rhode Island’s TANF caseload decline can be explained, at least in part, by significant policy changes the state implemented near the start of the recession:  between July 2008 and July 2009, Rhode Island both shortened its TANF time limit and began cutting off benefits to the entire family (rather than just the parents) when the time limit was reached.  In the 15 months prior to the start of the recession, Rhode Island’s caseload declined by 11 percent; this was followed by the additional 29 percent decline after the recession hit. 

Image
1-25-11tanf-f4.jpg

TANF Not Designed for Hard Economic Times

Since Congress created TANF in 1996 to replace AFDC, researchers and policymakers have speculated about how it might respond to an extended period of joblessness.  Some have indicated that the real success of welfare reform could only be determined after witnessing the program’s response to a significant economic downturn.[9] 

AFDC protected states from the full economic costs of serving more families when the economy weakened, since the federal government shared the costs of increased caseloads with the states.  TANF operates very differently.  States generally do not get more federal funding when caseloads increase in hard economic times (the TANF Emergency Fund is an exception, as explained below),   and because TANF is a block grant, decisions on whether or how to reallocate funds to address greater economic hardship rest solely with the state.  Moreover, three key features of TANF create a disincentive to serve more families during periods of greater need: 

  • The block grant structure means that if a state uses more TANF funding for cash assistance, it will have less for other TANF-funded programs.  Because states can allocate TANF funds to many different purposes, they have a significant financial incentive to keep their cash assistance caseloads low.  This incentive is even greater when the economy is weak because this is when state budget pressures are greatest.  When caseloads declined during the early years of welfare reform, states shifted TANF funds to many other worthwhile purposes such as child care, homelessness prevention, and child welfare services.  Once TANF funds have been reallocated to other purposes, it becomes very difficult to reclaim them to provide cash assistance to more families, no matter how great the need. 
  • TANF’s primary performance measure, the work participation rate, rewards states for reducing their TANF caseloads, even if the economy is weak. [10]   States receive a “caseload reduction credit” toward their work participation rate requirement if their TANF caseload declines below its 2005 level.  The credit has helped many states meet their work participation rates, but it gives states an incentive to keep TANF caseloads down, as states that allow their caseloads to rise in response to increased need generally would be penalized by a reduced credit. [11]  In addition, because states must achieve specified work participation rates, the number of families they must engage in work-related activities to achieve the rate increases as caseloads increase; this further discourages states from allowing their caseloads to grow when need increases. 
  • The activities through which TANF recipients can meet their work requirement are narrowly focused on employment, even if jobs are not available.   States primarily rely on unsubsidized employment of TANF recipients to meet their work participation rates.  It is more difficult for states to get credit toward the work rate for recipients’ participation in other activities; for example, there are limits on states’ ability to count participation in job search, post-secondary education, and activities to remove barriers to work (such as substance abuse treatment) toward their work participation rates.  Hence, when the economy weakens and fewer jobs are available, it becomes more difficult for states to meet their prescribed work participation rates unless they keep caseloads down, even as the number of families in need is swelling.

In addition, the work participation rate structure encourages states to concentrate their efforts on the most employable families — those that are already working or who have demonstrated their ability to participate in a work program almost full time — rather than families with the greatest barriers to employment, even though the latter would likely benefit the most from assistance.  Because the work participation requirements do not change when the economy is weak, the incentive to focus services on the most employable becomes stronger when few jobs are available.    

The TANF Emergency Fund:  Extra Help During a Time of Unprecedented Need

The TANF Emergency Fund, created as a part of the 2009 Recovery Act, provided much-needed assistance to help states respond to increased need.  Without it, TANF caseloads likely would have been even less responsive to the economic downturn.  The fund made $5 billion available to states to provide additional help to needy families in three categories:  basic assistance, one-time non-recurring assistance, and subsidized jobs.  For 2009 and 2010, a state could draw down up to 50 percent of one year’s TANF block grant allocation to cover 80 percent of the state’s increased costs of providing aid in those three categories. 

By the time the fund expired on September 30, 2010, 49 states (all except Wyoming), the District of Columbia, Puerto Rico, the Virgin Islands, and 25 Indian tribes had received assistance from the fund.  A total of $1.6 billion from the fund was used to cover the increased cost of providing basic assistance to families.  The largest share of the fund, $2.1 billion, was spent for increases in one-time non-recurring benefits, and $1.3 billion was spent to provide subsidized jobs to low-income parents and youth.  The emergency assistance and subsidized jobs provided through the TANF Emergency Fund undoubtedly helped some families to stay off the regular TANF rolls.       

Some states also were eligible to receive assistance from the permanent TANF Contingency Fund, which contained $1.4 billion at the start of the recession — all of which states had claimed by December 2009. [12]  To qualify for funds from the Contingency Fund, states had to meet an economic distress trigger as well as a maintenance-of-effort (MOE) requirement that is more stringent than the standard TANF MOE requirement.[13]  Because of the special MOE requirement and other complicated requirements, only 19 states — and only six of the 24 states whose peak unemployment rates exceeded the national average — received funds from the Contingency Fund.  In addition, while the TANF Emergency Fund required states to increase their expenditures in the three specified categories to receive additional funding, the regular Contingency Fund has no such requirement. 

Policy Changes to Increase TANF’s Responsiveness During Hard Economic Times

The recession has exposed serious weaknesses in TANF’s ability to respond to significant changes in the economy.  TANF is often the safety net of last resort for parents who cannot find work and do not qualify for unemployment insurance.  Thus, TANF has an important role to play in helping families with children weather the crises that result when jobs are not available.  TANF reauthorization provides an opportunity to make the program more responsive in such situations.  Potential changes that policymakers should consider include:     

  • Redesign the Contingency Fund.   The Contingency Fund has not been effective in helping many states respond to the economic downturn.  Because of its complex design and requirement that states significantly increase their TANF spending, fewer than half of the states qualified for money from the Contingency Fund, even though nearly all states met the economic distress triggers.  Moreover, the fund’s current design does not require that states use the resources to fund programs or services that explicitly respond to families’ special needs during an economic downturn.  They must, however, be used for purposes allowed under TANF.  It is important that TANF include a Contingency Fund that states can draw on during future downturns and that helps strengthen the responsiveness of TANF in difficult economic times. 

    The redesigned Contingency Fund should incorporate several features that helped make the TANF Emergency Fund so effective.  States should be able to get extra federal help during difficult economic periods if they are devoting more resources to help families meet their basic needs through an increase in expenditures for basic assistance, subsidized employment programs, or other supports for families facing hard times (such as one-time emergency assistance to avert homelessness).  In addition, the economic distress trigger should be redesigned to account for long recessions and to make it possible for states to qualify for a specified period after their economy begins to recover.  The current period is a case in point — the economy is technically in recovery, but the unemployment rate remains stuck at 9.4 percent, and indications are that poverty is continuing to rise. 
  • Systematically track, by state, measures that assess the effectiveness of TANF as a safety net for deeply poor children.  Since TANF was created in 1996, its effectiveness as a safety net for very poor families with children has declined dramatically.  To begin to reverse this trend, HHS could begin tracking and reporting on multiple measures of the program’s effectiveness as a safety net, including responsiveness during economic downturns.  These measures should focus on assessing the extent to which a state’s TANF program serves children in deep poverty and the progress that a state is making in decreasing the share of deeply poor children that fail to receive such aid.    
  • Replace the work participation rate and caseload reduction credit with a new performance measure that focuses on employment outcomes or other measures of family well-being, with adjustments during hard economic times.   The work participation rate has proven to be an ineffective performance measure for TANF.  A better approach would be to identify a new performance measure that rewards states for positive outcomes such as increasing employment or earnings or improving recipients’ education and skills.  Such a measure would maintain the program’s emphasis on work but would not reward states for simply removing families from the TANF caseload.  Also, the performance measure should adjust to reflect the availability of jobs; states cannot be expected to meet the same employment-related outcomes when unemployment is high as when it is low.    
  • Make modest changes in the types of work activities that states can count in order to refocus TANF employment activities on improving outcomes for families.   The current work participation rate structure drives state TANF programs to use very narrowly defined pathways to work for program recipients.  To the extent that the work participation measure is retained, either for some or all families, now seems the ideal time to consider expanding what can count toward the work rate so as to remove disincentives that often lead states to prevent recipients from pursuing the most promising path to employment.  While immediate job placement is likely to continue to be the preferred strategy for most recipients, parents who have the interest and ability to pursue an educational or training path that will lead to higher wages and less dependence on government assistance over the long term should be able to do so.

    This is especially important when the economy is weak.  Improving one’s education or skill level is a productive activity when few job opportunities are available.  Similarly, families experiencing personal and family crises should be encouraged to participate in activities that will improve their circumstances over the long term. 
  • Provide additional funds to help states maintain and build on the successful subsidized employment programs they created through the TANF Emergency Fund.  States were enormously successful in creating subsidized employment programs during this recession.  In many cases, states built these programs from scratch and implemented them in a short period of time.  These programs allowed states to provide new job opportunities to individuals with the least favorable employment prospects.  They also allowed TANF agencies to forge new partnerships with workforce development agencies and private-sector employers.  Because Congress did not extend the TANF Emergency Fund beyond September 30, 2010, a number of states have now ended their programs (or will shortly) since they do not have the funds to sustain them; many of these states would rekindle (or continue) their programs if new dedicated funding were available.  These programs have an important role to play in enabling TANF agencies to help TANF recipients find jobs.  While these programs may be more important during economic downturns, they would strengthen TANF work programs even when the economy is stronger.   

Appendix A

Methodology:  TANF Caseload Data Used for This Analysis

This paper looks to national and state trends — over time and across states — in caseloads for cash assistance to families with children, often referred to as state welfare caseloads or state TANF caseloads.  The analysis uses caseload data collected directly from the states rather than the official data reported by the Department of Health and Human Services (HHS) in order to more consistently reflect the number of families with children receiving cash welfare in each state over time.

The Deficit Reduction Act of 2005 and its implementing rules made it more difficult for states to meet the TANF work participation rates.  Since states that fail to meet these rates face a potential fiscal penalty, many states responded by changing the configuration of funding in their TANF programs.  Two of these changes in particular affect the caseload data that states report to HHS. [14]

  • Solely state-funded programs:  Over half the states serve some groups of families outside of the TANF or MOE funding structure. [15]  Often these are families whose needs cannot be met through the narrowly defined set of “countable” TANF work activities — for example, families needing extended time in activities designed to help them address personal and family challenges that limit their employability, or families participating in post-secondary education for longer than a year.  Many states also serve two-parent families in solely state-funded programs since these families are subject to a higher (and difficult to achieve) work participation rate.  Failure to account for these programs could lead one to underestimate the TANF program’s responsiveness to the recession if these families are removed from the state’s TANF caseload near the start of the recession and to overestimate the program’s responsiveness if they are added back in during the recession (as some states have done due to state budget shortfalls).   
  • Worker supplement programs:   Over one-third of states provide some type of supplemental “assistance” payment to low-income working families outside of their regular TANF or welfare caseload in order to provide additional work support and better achieve state policy goals. [16]  Most states provide these as transitional benefits to families that work their way off of TANF.  A few states also have served a broader group of low-income working families that receive other benefits such as SNAP.  Because these supplemental payments are supported by TANF or MOE funds, the recipients are included in caseload data reported to HHS, but they are not eligible for cash assistance under the state’s regular eligibility rules.  Depending in part on when a state began providing these payments to working families, including these cases as a part of the TANF caseload could overestimate a state’s TANF responsiveness to the economic downturn because these programs include families that are not really considered part of the state’s basic cash assistance caseload. 

The methodology we employed treats TANF caseload numbers differently from HHS with respect to these two populations.  Families served in solely state-funded programs are not included in the HHS data but are included in our data; working families that receive monthly cash assistance supplements are included in the HHS data but not in our data.  These two methodological decisions ensure that we are comparing a consistent cash assistance caseload over time and across states.  While the HHS data accurately reflect the number of families that states have chosen to serve with their TANF block grant and MOE funds over time, they do not always reflect the state’s caseload of families with children receiving cash assistance.  Moreover, because state choices have changed over time, the HHS data are not well suited to monitoring changes in the number of individuals or families serviced over time even within a state. 

We collected data from state websites and contacted state officials for data when the website data were not available, incomplete, or unclear.  Because states post their caseload data in different ways — with some separately identifying separate state programs, solely state-funded programs, or worker supplement programs and others not separately breaking out data — we selected data that correspond to the definition of caseload that we are using in this analysis.  Some states also report their TANF-funded child-only or kinship care assistance caseload separately; we included these data as part of state caseloads to ensure that the caseload data are comparable across states.

Because the caseload data are obtained from the states, for some states they may differ somewhat from the data submitted to HHS even if a state does not have a solely state-funded program or a worker supplement program.  (For some states, the data exactly match the HHS data.)  This is because state caseload data are often more immediate and represent all benefits issued in a month, while the HHS data that are submitted some months later include adjustments.  Welfare caseloads are very dynamic, with families entering, exiting, and having brief closures with subsequent reopenings.  While there may be some small differences between the state caseload data and the data reported to HHS that go beyond the adjustments for solely state-funded or worker supplement programs, our consistent use of the state caseload data over a period of time means that the data used for each state are comparable over the period of time and that the trends within a state should be consistent.

Appendix B - Data Tables

APPENDIX TABLE B1:
Changes in TANF and Food Stamp Caseloads
DECEMBER 2007 TO DECEMBER 2009
  TANF Food Stamps
Dec-07 Dec-09 % change Dec-07 Dec-09 % change
Alabama 18,830 21,330 13.3 226,401 335,354 48.1
Alaska 2,989 3,082 3.1 20,969 28,379 35.3
Arizona 37,887 39,858 5.2 245,225 429,370 75.1
Arkansas 7,684 7,957 3.6 158,080 195,332 23.6
California 466,853 561,909 20.4 871,358 1,327,580 52.4
Colorado 9,226 13,681 48.3 106,650 169,691 59.1
Connecticut 18,736 19,595 4.6 117,550 170,340 44.9
Delaware 4,628 5,833 26.0 31,330 48,356 54.3
District of Columbia 14,853 16,453 10.8 47,404 63,253 33.4
Florida 48,608 61,097 25.7 707,984 1,296,324 83.1
Georgia 22,719 21,444 -5.6 401,851 650,006 61.8
Hawaii 7,676 8,984 17.0 47,344 67,278 42.1
Idaho 1,537 1,770 15.2 38,568 71,636 85.7
Illinois 26,621 29,582 11.1 589,819 760,630 29.0
Indiana 40,985 36,989 -9.7 260,901 338,287 29.7
Iowa 16,459 17,781 8.0 112,647 152,274 35.2
Kansas 12,837 13,599 5.9 84,038 116,377 38.5
Kentucky 29,323 30,243 3.1 278,716 345,797 24.1
Louisiana 11,178 11,740 5.0 276,247 346,880 25.6
Maine 13,169 14,302 8.6 83,059 110,917 33.5
Maryland 22,338 28,486 27.5 160,154 252,148 57.4
Massachusetts 45,915 50,822 10.7 254,061 392,999 54.7
Michigan 74,666 79,203 6.1 575,711 816,056 41.7
Minnesota 28,851 32,349 12.1 137,943 201,924 46.4
Mississippi 11,641 12,598 8.2 201,694 245,662 21.8
Missouri 42,951 42,589 -0.8 307,570 401,304 30.5
Montana 3,189 3,859 21.0 34,978 48,600 38.9
Nebraska 9,849 10,309 4.7 51,658 68,114 31.9
Nevada 8,732 11,377 30.3 63,429 119,919 89.1
New Hampshire 4,584 6,505 41.9 30,066 47,201 57.0
New Jersey 38,615 39,233 1.6 206,285 287,923 39.6
New Mexico 14,060 20,103 43.0 92,080 142,174 54.4
New York 153,949 161,179 4.7 979,989 1,414,173 44.3
North Carolina 25,634 27,562 7.5 409,244 587,490 43.6
North Dakota 2,423 1,991 -17.8 21,540 26,709 24.0
Ohio 80,212 102,489 27.8 515,483 731,783 42.0
Oklahoma 9,238 10,179 10.2 174,799 243,143 39.1
Oregon 19,126 26,599 39.1 233,413 359,243 53.9
Pennsylvania 81,420 85,534 5.1 548,518 718,299 31.0
Rhode Island 10,929 7,780 -28.8 38,898 66,664 71.4
South Carolina 15,736 20,854 32.5 248,868 348,116 39.9
South Dakota 2,918 3,250 11.4 25,660 38,458 49.9
Tennessee 55,161 62,760 13.8 398,134 561,024 40.9
Texas 54,858 49,764 -9.3 963,856 1,310,466 36.0
Utah 5,390 7,397 37.2 51,399 87,858 70.9
Vermont 4,884 5,759 17.9 26,828 41,643 55.2
Virginia 29,680 37,545 26.5 240,425 351,911 46.4
Washington 49,908 65,421 31.1 298,755 450,032 50.6
West Virginia 9,689 10,924 12.7 122,736 162,769 32.6
Wisconsin 18,327 20,777 13.4 170,508 305,216 79.0
Wyoming 278 355 27.7 9,496 13,743 44.7
US Total 1,747,949 1,982,781 13.4% 12,300,319 17,866,825 45.3%
APPENDIX TABLE B2:
Changes in the Number of Unemployed
DECEMBER 2007 TO DECEMBER 2009
  Number of Unemployed Unemployment Rate
Dec-07 Dec-09 % change Dec-07 Dec-09 Peak Rate
Dec07- Dec09
Alabama 84,572 224,175 165.1 3.9 10.9 10.9
Alaska 21,975 31,008 41.1 6.2 8.6 8.6
Arizona 131,047 287,041 119.0 4.3 9.2 9.5
Arkansas 68,689 104,513 52.2 5.0 7.6 7.6
California 1,049,107 2,233,878 112.9 5.8 12.3 12.3
Colorado 116,631 191,887 64.5 4.3 7.3 8.3
Connecticut 91,071 165,861 82.1 4.9 8.8 8.8
Delaware 16,860 37,614 123.1 3.8 8.8 8.8
District of Columbia 18,321 39,526 115.7 5.5 11.9 11.9
Florida 433,134 1,078,998 149.1 4.7 11.7 11.7
Georgia 245,224 482,245 96.7 5.1 10.3 10.3
Hawaii 19,057 43,310 127.3 3.0 6.8 7.0
Idaho 26,791 68,263 154.8 3.5 9.1 9.1
Illinois 369,584 725,293 96.2 5.5 11.0 11.0
Indiana 148,134 300,834 103.1 4.6 9.7 10.6
Iowa 65,302 109,797 68.1 3.9 6.5 6.5
Kansas 59,121 99,155 67.7 4.0 6.5 7.2
Kentucky 112,846 219,656 94.7 5.5 10.6 10.8
Louisiana 76,729 151,513 97.5 3.8 7.3 7.3
Maine 33,303 56,998 71.1 4.7 8.1 8.2
Maryland 107,337 217,291 102.4 3.5 7.4 7.4
Massachusetts 151,638 322,527 112.7 4.4 9.3 9.3
Michigan 357,575 699,663 95.7 7.1 14.5 14.5
Minnesota 136,620 219,267 60.5 4.7 7.4 8.4
Mississippi 79,371 135,891 71.2 6.1 10.5 10.5
Missouri 163,087 288,047 76.6 5.3 9.6 9.7
Montana 19,575 33,149 69.3 3.9 6.7 6.7
Nebraska 28,927 44,754 54.7 2.9 4.6 4.8
Nevada 69,243 178,261 157.4 5.2 13.0 13.0
New Hampshire 25,365 51,132 101.6 3.4 6.9 6.9
New Jersey 202,826 452,499 123.1 4.5 10.0 10.0
New Mexico 34,489 79,024 129.1 3.6 8.2 8.2
New York 448,055 857,578 91.4 4.7 8.9 8.9
North Carolina 223,502 494,208 121.1 4.9 10.9 11.0
North Dakota 11,018 15,625 41.8 3.0 4.3 4.4
Ohio 337,602 637,513 88.8 5.6 10.8 10.8
Oklahoma 62,245 121,400 95.0 3.6 6.8 6.9
Oregon 100,917 205,681 103.8 5.2 10.6 11.6
Pennsylvania 288,632 559,643 93.9 4.5 8.8 8.8
Rhode Island 34,132 72,750 113.1 6.0 12.7 12.7
South Carolina 117,917 268,850 128.0 5.6 12.4 12.4
South Dakota 12,233 20,943 71.2 2.8 4.7 5.0
Tennessee 167,299 319,962 91.3 5.5 10.7 10.9
Texas 508,268 987,161 94.2 4.4 8.2 8.2
Utah 43,008 88,684 106.2 3.1 6.6 6.8
Vermont 14,347 23,945 66.9 4.0 6.7 7.3
Virginia 132,124 280,791 112.5 3.2 6.8 6.9
Washington 157,744 324,477 105.7 4.6 9.2 9.2
West Virginia 33,068 70,576 113.4 4.0 9.0 9.0
Wisconsin 138,684 258,156 86.1 4.5 8.5 8.9
Wyoming 8,125 22,085 171.8 2.8 7.6 7.6
US Total 7,402,471 15,003,098 102.7% 5.0% 10.0% 10.1%
APPENDIX TABLE B3:  
Changes in TANF and Food Stamp Caseloads and the Number of Unemployed
DECEMBER 2007 TO DECEMBER 2009
  TANF Cases Food Stamp Cases No. of Unemployed
% change Rank % change Rank % change Rank
Alabama 13.3 22 48.1 20 165.1 2
Alaska 3.1 44 35.3 37 41.1 51
Arizona 5.2 36 75.1 5 119.0 12
Arkansas 3.6 42 23.6 50 52.2 49
California 20.4 16 52.4 17 112.9 16
Colorado 48.3 1 59.1 9 64.5 46
Connecticut 4.6 41 44.9 23 82.1 37
Delaware 26.0 13 54.3 15 123.1 10
District of Columbia 10.8 27 33.4 40 115.7 13
Florida 25.7 14 83.1 3 149.1 5
Georgia -5.6 47 61.8 8 96.7 26
Hawaii 17.0 18 42.1 27 127.3 8
Idaho 15.2 19 85.7 2 154.8 4
Illinois 11.1 26 29.0 46 96.2 27
Indiana -9.7 49 29.7 45 103.1 22
Iowa 8.0 32 35.2 38 68.1 43
Kansas 5.9 35 38.5 35 67.7 44
Kentucky 3.1 43 24.1 48 94.7 30
Louisiana 5.0 38 25.6 47 97.5 25
Maine 8.6 30 33.5 39 71.1 41
Maryland 27.5 11 57.4 10 102.4 23
Massachusetts 10.7 28 54.7 13 112.7 17
Michigan 6.1 34 41.7 29 95.7 28
Minnesota 12.1 24 46.4 21 60.5 47
Mississippi 8.2 31 21.8 51 71.2 39
Missouri -0.8 46 30.5 44 76.6 38
Montana 21.0 15 38.9 34 69.3 42
Nebraska 4.7 40 31.9 42 54.7 48
Nevada 30.3 8 89.1 1 157.4 3
New Hampshire 41.9 3 57.0 11 101.6 24
New Jersey 1.6 45 39.6 32 123.1 9
New Mexico 43.0 2 54.4 14 129.1 6
New York 4.7 39 44.3 25 91.4 33
North Carolina 7.5 33 43.6 26 121.1 11
North Dakota -17.8 50 24.0 49 41.8 50
Ohio 27.8 9 42.0 28 88.8 35
Oklahoma 10.2 29 39.1 33 95.0 29
Oregon 39.1 4 53.9 16 103.8 21
Pennsylvania 5.1 37 31.0 43 93.9 32
Rhode Island -28.8 51 71.4 6 113.1 15
South Carolina 32.5 6 39.9 31 128.0 7
South Dakota 11.4 25 49.9 19 71.2 40
Tennessee 13.8 20 40.9 30 91.3 34
Texas -9.3 48 36.0 36 94.2 31
Utah 37.2 5 70.9 7 106.2 19
Vermont 17.9 17 55.2 12 66.9 45
Virginia 26.5 12 46.4 22 112.5 18
Washington 31.1 7 50.6 18 105.7 20
West Virginia 12.7 23 32.6 41 113.4 14
Wisconsin 13.4 21 79.0 4 86.1 36
Wyoming 27.7 10 44.7 24 171.8 1
US Total 13.4%   45.3%   102.7%  
APPENDIX TABLE B4:  
Changes in TANF Caseloads Before and After the Start of the Recession
  Prior to Recession
Oct-06 - Dec07
Two Years After Start
Dec-07 - Dec-09
Over Entire Period
Oct-06 - Dec-09
% change Rank % change Rank % change Rank
Alabama -5.4 23 13.3 22 7.2 20
Alaska -9.0 32 3.1 44 -6.2 39
Arizona -2.9 13 5.2 36 2.1 28
Arkansas -8.1 28 3.6 42 -4.9 35
California 0.1 8 20.4 16 20.4 9
Colorado -28.3 51 48.3 1 6.4 21
Connecticut -4.6 19 4.6 41 -0.3 30
Delaware -18.5 46 26.0 13 2.8 26
District of Columbia -5.3 21 10.8 27 4.9 23
Florida -2.0 11 25.7 14 23.2 7
Georgia -16.2 42 -5.6 47 -20.9 48
Hawaii -18.8 47 17.0 18 -5.0 36
Idaho -12.5 36 15.2 19 0.7 29
Illinois -21.8 50 11.1 26 -13.2 46
Indiana -3.5 17 -9.7 49 -12.9 44
Iowa -4.8 20 8.0 32 2.9 25
Kansas -19.7 48 5.9 35 -15.0 47
Kentucky -8.4 29 3.1 43 -5.5 38
Louisiana -6.2 25 5.0 38 -1.5 31
Maine -2.8 12 8.6 30 5.5 22
Maryland 1.2 4 27.5 11 29.0 4
Massachusetts 0.2 7 10.7 28 10.9 17
Michigan -15.3 39 6.1 34 -10.1 42
Minnesota -3.0 14 12.1 24 8.7 19
Mississippi -5.4 22 8.2 31 2.4 27
Missouri -3.4 16 -0.8 46 -4.2 34
Montana -4.0 18 21.0 15 16.2 12
Nebraska -17.0 44 4.7 40 -13.1 45
Nevada 19.9 1 30.3 8 56.3 1
New Hampshire -20.8 49 41.9 3 12.3 15
New Jersey -6.6 26 1.6 45 -5.1 37
New Mexico -0.3 9 43.0 2 42.5 3
New York -6.0 24 4.7 39 -1.6 32
North Carolina -13.0 37 7.5 33 -6.4 40
North Dakota -9.0 31 -17.8 50 -25.2 50
Ohio -0.5 10 27.8 9 27.2 5
Oklahoma -11.5 35 10.2 29 -2.5 33
Oregon 9.6 2 39.1 4 52.4 2
Pennsylvania -16.1 41 5.1 37 -11.8 43
Rhode Island -10.7 34 -28.8 51 -36.4 51
South Carolina -16.9 43 32.5 6 10.2 18
South Dakota 0.4 6 11.4 25 11.8 16
Tennessee -18.0 45 13.8 20 -6.7 41
Texas -15.5 40 -9.3 48 -23.4 49
Utah -14.8 38 37.2 5 17.0 11
Vermont 3.0 3 17.9 17 21.4 8
Virginia -9.9 33 26.5 12 13.9 14
Washington -3.2 15 31.1 7 26.9 6
West Virginia -8.6 30 12.7 23 3.0 24
Wisconsin 1.1 5 13.4 21 14.6 13
Wyoming -7.0 27 27.7 10 18.7 10
US Total -6.3%   13.4%   6.3%  

End Notes

[1] We use caseload data collected directly from states for this analysis rather than caseload data reported by the states to the Department of Health and Human Services (HHS) in order to more consistently reflect the number of families with children receiving cash welfare in each state over time, and across states.  In response to TANF changes in the Deficit Reduction Act of 2005, many states changed their funding configurations (and did so at various times in the past several years) and the caseload data that states report to HHS may not consistently represent the number of families relying on cash assistance in the state.  See fuller discussion of the reasons why states made these changes and of the data choices we made in Appendix A.

[2]   See Appendix B tables for state-specific data. Additional state-specific fact sheets and the underlying monthly data used in this analysis can be found at https://www.cbpp.org/research/index.cfm?fa=topic&id=42.

[3] http://data.bls.gov/cgi-bin/surveymost

[4] The Employment Situation – January 2010, Bureau of Labor Statistics, http://www.bls.gov/news.release/pdf/empsit.pdf.

[5] Robert Greenstein, Sharon Parrott, and Arloc Sherman, “Poverty and Share of Americans Without Health Insurance Were Higher in 2007 — And Median Income for Working-Age Households Was Lower — Than at the Bottom of Last Recession,” Center on Budget and Policy Priorities, August 26, 2008, https://www.cbpp.org/cms/?fa=view&id=621.

[6] The food stamp data are for December 2009, which is the most recent month for which we have comparable TANF caseload data for every state. In the ensuing eight months, food stamp caseloads continued to grow and stood in August 2010 at 19.7 million households containing 42.4 million individuals (nearly one in seven Americans and one in four children).

[7] This calculation excludes months that a state operated a disaster food stamp program, usually in the aftermath of a hurricane or other natural disaster.

[8] “Trends in Supplemental Nutritional Assistance Program Participation Rates: 2001 to 2008,” USDA, Table 2, (Households), at  http://www.fns.usda.gov/ora/menu/Published/SNAP/FILES/Participation/Trends2001-2008.pdf .

[9] LaDonna A. Pavetti, “What Will States Do When Jobs Are Not Plentiful:  Policy and Implementation Challenges,” and David N. Figlio and James P. Ziliak, “Welfare Reform, the Business Cycle, and the Decline in AFDC Caseloads,” in Sheldon H. Danziger, ed., Economic Conditions and Welfare Reform, W.E. Upjohn Institute for Employment Research. 1999.

[10] States are required to meet two work participation rates — 50 percent for all work-eligible families and 90 percent for two-parent families — and each of these rates is reduced by the caseload reduction credit to set the target work rate that a state must meet.  If states fail to meet their target rates, they are subject to financial penalties. 

[11] ARRA included a temporary provision that allows a state an option to use its caseload reduction credit based on circumstances from 2007 or 2008 in lieu of the most recent fiscal year in setting its target work rate for 2009, 2010, and 2011.  This provision temporarily lessens some of the pressure on states to keep caseloads from increasing despite rising need. 

[12] Additional funds initially were added to the Contingency Fund for FY 2011 but subsequently rescinded effective December 8, 2010.  The outcome of this was that 21 states received a total of $334 million in Contingency Funds for October through December and no additional Contingency Funds are available for the remainder of 2011.

[13] States must meet a special 100 percent MOE requirement to receive help from the Contingency Fund, and the extent to which a state exceeds this 100 percent level determines whether the state can keep or must repay the Contingency Funds it receives.  Moreover, unlike a state’s regular MOE requirement, the Contingency Fund MOE does not include state spending on child care or separate state programs.

[14] For a fuller discussion of these state changes and the impact on caseloads, see LaDonna Pavetti, Linda Rosenberg, and Michelle K. Derr, “Understanding Temporary Assistance for Needy Families Caseloads After Passage of the Deficit Reduction Act of 2005, Final Report,” Mathematica Policy Research, September 21, 2009, http://www.mathematica-mpr.com/publications/PDFs/family_support/TANF_caseloads.pdf .    See also, “Temporary Assistance for Needy Families: Implications of Recent Legislative and Economic Changes for State Programs and Work Participation Rates,” GAO-10-525, May 2010.

[15] For additional information on solely state-funded programs, see Liz Schott and Sharon Parrott, “Designing Solely State-Funded Programs: Implementation Guide for One “Win-Win” Solution for Families and States,” Center on Budget and Policy Priorities, revised January, 2009, https://www.cbpp.org/sites/default/files/atoms/files/12-7-06tanf.pdf.

[16] For more information about worker supplement programs, see Liz Schott, “Using TANF or MOE Funds to Provide Supplemental Assistance to Low-Income Working Families,” Center on Budget and Policy Priorities, revised September 2008, https://www.cbpp.org/sites/default/files/atoms/files/5-24-07tanf.pdf.