TANF Continues to Weaken as a Safety Net
October 27, 2015
Though conservative policymakers tout the “success” of the 1996 welfare reforms, the reality is that the Temporary Assistance for Needy Families (TANF) program, the cornerstone of the 1996 reforms, is not the success that some proclaim and should not be used as a model for other safety net programs. TANF provides a temporary safety net to few families who are poor and may have no other means to meet their family’s basic needs — and its reach continues to shrink.
Instead of providing the help that families need when they fall on hard economic times, TANF has put poor families — and especially their children — at risk of much greater hardship with the potential for long-term negative consequences. In addition, TANF does little to connect families to work to mitigate their need for assistance. Recent congressional discussions to reauthorize the program, which was last renewed in 2005 for five years and has continued since 2010 through short-term extensions, include some improvements to TANF’s work efforts, but fail to bolster it as a cash safety net. The end result is that TANF does less to lift families out of deep poverty than its predecessor Aid to Families with Dependent Children (AFDC), and it has contributed to a rise in families living in extreme poverty.
In 2014, for every 100 families in poverty, only 23 received cash benefits from TANF. This is down from the 68 families for every 100 in poverty that received cash assistance when TANF was first enacted in 1996. This ratio, which we call the TANF-to-poverty ratio (TPR), has declined nearly every year since 1996 and reached its lowest point in 2014.
While the national TPR highlights how much the cash safety net for families has weakened since the advent of TANF, it tells only part of the story as this ratio varies widely among states, ranging from a low of 4 to a high of 78. In 12 states, the ratio is less than 10, meaning that for every 100 families living in poverty, fewer than 10 receive TANF cash assistance. Many states have made policy and administrative changes that have significantly cut their TANF caseloads, even as families struggled financially during and after the Great Recession.
We use the TANF-to-poverty ratio in this paper to describe how TANF’s role in helping poor families meet their basic needs has decreased over time, with an emphasis on what happened between the start of the Great Recession and 2014, the most recent year for which we have data. We calculate the ratio by dividing the number of TANF cash assistance cases by the number of families with children in poverty from the Census Bureau’s Current Population Survey (CPS). We use two-year averages for our state-level calculations to improve the reliability of the data. (See Appendix A for further details about our methodology.) When this ratio falls, it means TANF is less responsive to need than in previous years. A decline in the TPR can happen because: (1) the number of families receiving cash assistance from TANF falls without a corresponding drop in the number of families living in poverty; or (2) poverty increases without a corresponding rise in the number of families receiving TANF benefits.
When it created TANF, Congress handed responsibility to the states for providing a safety net for families in need. But the result has been a weaker cash safety net that has increased deep poverty. The data we present in this paper show why Congress should act to remove incentives for states to weaken TANF’s role as a safety net and to hold states accountable for serving families in need. It also shows why TANF is not a model for other programs.
A Weak Safety Net Getting Weaker
In 2014, for every 100 families in poverty, just 23 families received TANF cash assistance. This is down from 68 families that received cash assistance for every 100 in poverty in 1996, when TANF was created. (See Figure 1.) In 2006, the year before the start of the Great Recession, 31 families received TANF for every 100 families in poverty. The TPR in 2014 was lower than in 2013 (23 vs. 26) because TANF cases continued to decline as poverty remained relatively high.
Through about 2000, we can attribute the decline in the TPR primarily to post-welfare reform drops in TANF caseloads. Even during TANF’s early years when the economy was very strong and employment among never-married mothers rose, the TANF caseload decline outpaced the drop in poverty. More recently, the drop in the TPR reflects growth in the number of families in poverty without a parallel rise in the number of families receiving TANF as well as absolute declines in the
TANF caseload while poverty was rising. Nationwide, the number of families in poverty was 17 percent higher in 2014 than in 2006, while the national TANF caseload was 14 percent lower.
The TANF caseload is lower than it was at the start of the recession even though poverty remains higher than it was then. In 2014, a little more than 1.6 million families received TANF nationally, down from 1.9 million in 2006. From its post-recession peak in 2010, when close to 2 million families received TANF nationwide, the TANF caseload has fallen by 17 percent.
Poverty among families rose quickly after the start of the recession, hitting a high of 7.4 million families in 2011, up from 6 million in 2006. The number of families in poverty has remained high in the recession’s aftermath, with the first significant drop — to 6.8 million families — occurring in 2013.
TANF Largely Disappearing as a Safety Net in More States
The national TANF-to-poverty ratio misses the extreme (and growing) variation among the states. In 2013-14 the TPR ranged from a high of 78 in Vermont to a low of 4 in Louisiana. (See Figure 2.) Increases in the number of families living in poverty, the failure of state TANF programs to meet increased need during the recession and, at least in some cases, state policy and administrative changes pushed the TANF-to-poverty ratios down in 45 states between 2005-06 and 2013-14. In 19 states, the decline was especially pronounced and the TPR dropped by more than 10 points over that period. In these states, either TANF caseloads fell while poverty rose, or caseloads increased by a relatively small amount in response to a greater increase in the number of families in poverty.
An especially troubling trend is the growing number of states with TPRs of 10 or less. In 2005-06, only two states, Idaho and Wyoming, had such low ratios. The list grew during the recession and has continued to grow in its aftermath. In 2013-14, 12 states — Arizona, Arkansas, Georgia, Idaho, Indiana, Louisiana, Mississippi, North Carolina, Oklahoma, Texas, Utah, and Wyoming — had TPRs of 10 or less. Two of these states had especially large drops in their TPRs since 2005-2006: Arizona’s dropped almost 20 points and Indiana’s dropped 27 points. (See Figure 3.)
State Policy and Process Changes Contribute to TANF’s Deteriorating Safety Net
In December 2014, 37 states’ caseloads had fallen below their December 2006 levels — in the case of Indiana, 77 percent below. Although many families’ need for cash assistance increased during and after the recession, a number of states have made major TANF policy and administrative changes that made TANF less available and significantly reduced their TANF caseloads. Some states shortened or otherwise changed their time limits. Others made administrative changes such as more stringent applicant requirements that made it harder for families to qualify for benefits. In some instances, states made these changes to free up funds to fill budget holes, while others made them for ideological reasons.
- Arizona reduced the TANF time limit from 60 to 36 months in 2010 and dropped it further in 2011 to 24 months. Also in 2010, Arizona terminated many child-only cases because it began to factor in the income of the relatives caring for children (even when these relatives were not seeking assistance for themselves) and instituted time limits on these child-only cases. These changes account for the majority of the two-thirds drop in Arizona’s TANF caseload between December 2006 and December 2014. (The state enacted a third time limit cut in 2015, and it will reduce the lifetime limit to 12 months effective in July 2016.)
- Indiana has changed its policies and procedures around work requirements, including changing sanction penalties and requiring applicants to complete 20 days of job searching before the state will approve benefits. Indiana’s caseload plummeted after these changes and dropped by 77 percent between December 2006 and December 2014.
- In November 2011, Kansas began to implement tougher sanction penalties and stricter eligibility requirements, such as counting the income of a cohabitating partner or spouse who is not the parent of the child in the household. The state also implemented an applicant job search requirement that imposed barriers at the start of the application process. Between 2011 and 2013, the state required applicants to make 20 job contacts a week before determining eligibility (with no child care or transportation assistance provided). State data show increased eligibility denials during this period. Kansas’ caseload fell by 60 percent between December 2006 and December 2014, with most of the drop coming since 2011.
- Maine established a 60-month time limit in May 2012 that applied retroactively to past months of TANF receipt and cut off otherwise eligible families. The state’s caseload fell by 53 percent between December 2006 and December 2014.
- Michigan made several time-limit changes in 2011. It tightened its 48-month time limit by eliminating certain exemptions or extensions. It also applied a 60-month time limit retroactively to many families who had been previously exempt from the state’s 48-month time limit. Michigan’s caseload fell by 65 percent between December 2006 and December 2014.
- New Hampshire made two policy changes that resulted in substantial caseload declines. In July 2011, the state ended its program for two-parent families. Then in February 2012, it made it harder for families receiving Supplemental Security Income (SSI) to qualify for TANF cash assistance. By December 2014, New Hampshire’s caseload had dropped by nearly half from its recession-related peak in April 2010 and by 39 percent since December 2006.
- The caseload decline in Ohio between 2011 and 2014 was partially a result of the strategy to improve the state’s work participation rate by reducing the number of families receiving aid (rather than by increasing the number of families participating in work activities), often by closing more cases due to sanction. Some counties also increased their work rates by implementing upfront work requirements for applicants, resulting in fewer families being able to access benefits. The state’s caseload fell by a fifth between December 2006 and December 2014 and by 42 percent from its June 2010 recession-related peak.
- In 2008, Rhode Island cut its lifetime time limits, including limiting an adult recipient to 48 months of assistance, down from its previous 60-month limit, and established an additional time limit that allows families to receive only 24 months of cash assistance within a five-year period. Additionally, the state started terminating the entire family after 60 months, where previously it only cut off benefits to the adult. Between December 2006 and December 2014, the state’s caseload fell 56 percent.
- Washington State made its time-limit policies more restrictive, starting in February 2011 by narrowing the policies under which time-limit extensions can be granted. The state also imposed time limits on child-only cases where the child resides with parents who aren’t eligible for TANF themselves (because they are ineligible immigrants, for example), cut benefit levels for all recipients by 15 percent, and started counting the income of relatives who are caring for the children in child-only cases but are not seeking assistance for themselves. The state’s caseload fell by 32 percent between December 2006 and December 2014 and by half from its recession-related peak in January 2011.
States rarely track the impact of major TANF policy changes, so data on how families fare when these changes are made are lacking. However, a recent study that Washington State conducted provides some important insights that are consistent with other research on families that have lost TANF benefits due to time limits. The Washington study, which compared the experiences of families that lost benefits due to the state’s time-limit changes with families that left for other reasons, found that those that lost benefits because of the time-limit changes were more likely to have barriers that prevented them from entering the labor market and leaving TANF on their own. For example, families cut off by the time limits were more likely to have significant mental and physical health issues: two-thirds of parents in that group had an indication of mental illness, 23 percent had a chronic illness, and 25 percent had an alcohol or drug treatment need. Three years after they lost their TANF benefits, fewer than half were employed and as a group, they had higher rates of homelessness than families who left for other reasons. In spite of their high rates of health issues, these families had not had been granted extensions based on disability and few transitioned to SSI.
TANF Lifts Far Fewer Children out of Deep Poverty Than AFDC Did
TANF’s predecessor, Aid to Families with Dependent Children (AFDC), played a significant role in reaching families, particularly those with children and those in deep poverty (defined as incomes below half the poverty line). TANF has failed to maintain that standard. TANF benefits are not sufficient to lift families out of poverty in any state, and TANF does far less than AFDC did to lift families out of deep poverty. While AFDC lifted more than 2 million children out of deep poverty in 1995, TANF lifted only 629,000 children out of deep poverty in 2010. (See Figure 4.) In 1994-95, only three states had more families living in deep poverty than receiving AFDC. By 2013-14, 46 states had more families living in deep poverty than receiving TANF.
New evidence shows that the drop in cash assistance receipt since welfare reform is one of the main drivers of rising “extreme poverty,” a measure used by the World Bank of the number households surviving on $2 or less per person per day and a more intense level of poverty than deep poverty. Research from Luke Shaefer and Kathryn Edin finds that the number of U.S. households living in extreme poverty in any given month more than doubled between 1996 and 2011, from 636,000 to 1.46 million; the number of children living in such households also doubled from 1.4 million to 2.8 million. Shaefer and Edin find that these households are “concentrated among those groups who were most affected by welfare reform.”  Even for families who receive TANF, benefits are so low that they do little to lift a family out of deep poverty. In Mississippi and Tennessee, families solely reliant on those benefits as income may still find themselves living in extreme poverty as benefit levels in those states are the lowest in the country for a family of three.
Two well-known poverty researchers, Greg J. Duncan and Katherine Magnuson, have shown that poverty among young children not only slows them in school but also shrinks their earnings as adults. Welfare-to-work programs and other anti-poverty experiments “suggest that income plays a causal role in boosting younger children’s achievement” in preschool and elementary school, they note. (TANF is often a critical source of income for the most vulnerable families with young children.) They also found that among families with incomes below $25,000, children whose families received a $3,000 annual income boost when the children were under age 6 earned 17 percent more as adults and worked 135 more hours per year after age 25 than otherwise-similar children whose families didn’t receive the income boost. The same effect was not found among families making over $25,000. This suggests that policy changes within TANF that cut income, such as establishing harsher sanctions or shorter time limits or significantly reducing benefits, could harm young children now and into the future.
States Should Be Held Accountable for Helping Families in Need
Policymakers created TANF 19 years ago with a two-pronged approach in mind — that our nation’s cash assistance system would be redesigned to create an expectation of work for able-bodied recipients and that states would maintain a safety net for parents who were unable to work due to a short-term crisis, a work-limiting disability, or a lack of available jobs. This is not what states have done. Using the flexibility afforded them, states have made policy, process, and budgetary decisions that have substantially weakened TANF’s role as a safety net. In fact, in a growing number of states, this cash safety net barely exists, leaving more families deeply poor and without the resources they need to meet their most basic needs through either employment or cash assistance.
Several features of the TANF block grant have effectively encouraged states to provide cash assistance to fewer families (and to provide less to those that get it). First, while TANF includes an accountability measure around work (the TANF work participation rate or WPR), there is no accountability measure related to providing assistance to families in need. In addition, when the U.S. Department of Health and Human Services evaluates a state’s performance on meeting the TANF WPR, it doesn’t consider the extent to which states are serving families in need. One of the easiest ways for states to meet the WPR is to deny assistance to families that may have trouble meeting the work requirement even though they are the very families that could benefit the most from employment assistance and work supports.
States are also rewarded for shrinking TANF caseloads with a “caseload reduction credit,” which allows them to do less to meet the program’s work participation rate requirements. Finally, because states can spend TANF funds for a broad range of activities, they can shift funds that they have previously used to provide assistance directly to families to plug state budget holes –— and many states have taken that option. And, after initial investments in the early years of TANF in helping parents connect to work, states have not used the dollars they have shifted away from providing a safety net to further TANF’s welfare-to-work goals.
As Congress considers TANF reauthorization, it should strengthen TANF by taking the following actions:
- Create a new accountability measure that encourages states to serve more families in need. States focus on what they are incentivized to do. To expand TANF’s reach, Congress should add an accountability measure that focuses on serving families in need (such as the TANF-to-poverty ratio). States that fail to meet a particular standard could be required to spend additional resources on cash assistance or work activities or denied access to additional funds like the TANF Contingency Fund.
- Eliminate the caseload reduction credit. The caseload reduction credit rewards states for denying or terminating aid to needy families without regard to whether the adult in the family is actually employed (or employable). States should be encouraged to serve families in need; instead, the design of the WPR and the caseload reduction credit gives them an incentive not to.
- Hold states accountable for employment outcomes, not participation in work activities. The primary measure of TANF’s success should be whether families leave the program with employment and on a path to earn enough to provide for their families, not simply whether they participate in a pre-defined set of activities that may or may not prepare them for employment and help them move out of poverty. The measure should capture employment and earnings outcomes and should align TANF with other workforce programs under the recently enacted Workforce Innovation and Opportunity Act.
- Require states to spend a specified share of federal and state resources on TANF’s core purposes. TANF’s purposes are broad, which has provided states with the flexibility to spread TANF funds throughout their state budgets. To direct more TANF resources to the program’s core purposes — cash assistance, employment assistance, and work supports — Congress could require states to spend a specific share, for example 50 percent, of their state and federal TANF funds on these core purposes.
- Increase the TANF block grant to account for its decline in value, and index it to inflation in future years. The TANF block grant is worth 30 percent less than when it was created in 1996. Without additional funds, states are unlikely to spend additional resources to provide a cash safety net for more families. Any additional funds should be restricted to TANF’s core purposes — cash assistance, employment assistance, and work supports.
Conservative policymakers often hail the TANF block grant as a model for reform for other safety net programs, such as Medicaid and SNAP. But shifting to a block grant with fixed federal funding has resulted in a substantially weakened safety net, leaving a growing number of families in very deep poverty, even as their needs increased during and after the Great Recession. Congress should now take steps to hold states accountable for helping poor families in need.
Methodology and Source Notes
TANF Caseload Data
In this analysis, TANF caseload data from January 1979 and August 2006 were collected from the U.S. Department of Health and Human Services (HHS). Beginning in September 2006, this analysis uses caseload data collected directly from the states rather than the official data reported by HHS, as the state data more consistently reflect the number of families with children receiving cash assistance in each state over time. These data differ from the official HHS TANF data in two important ways. First, they include cases from solely state-funded programs. In most instances, these families had been in state TANF programs but were shifted to a solely state-funded program on or after October 2006, when the Deficit Reduction Act of 2005 (DRA) went into effect, because states anticipated these families would not be able to meet TANF work participation requirements and would negatively affect the state’s work participation rate. These cases are not included in the data reported to HHS as no TANF or state Maintenance of Effort (MOE) funds are used. While not counted in the HHS TANF caseload numbers, these families generally are seen as part of the state’s cash assistance program and continue to receive the same or comparable benefits in these state-funded programs as they received when they were on TANF-funded benefits.
Second, unlike the HHS data, the state data we use exclude cases in worker supplement programs under which states provide modest TANF- or MOE-funded cash payments to working families. States generally created these programs after the passage of the DRA. Because these supplements make additional families eligible (or make current recipients eligible for a longer period of time), they result in increasing the TANF or MOE caseloads that states report to HHS. Often, states provide a very small cash grant to these families — as little as $8 to $10 per month. The main purpose of these small grants is to increase the percentage of TANF families who are meeting their work participation requirement and hence to help states meet their work-participation-rate requirement.
Including solely state-funded programs and excluding worker supplement programs in the caseload data used for our analysis allow us to have a more consistent trend of the number of families receiving cash assistance in each state over time. We decided to include families in solely state-funded programs because those families are receiving TANF-like services and would have been part of the regular TANF caseload if not for the new policies enacted in the wake of the DRA. We decided to exclude families in the worker supplement programs because the cash assistance those families receive is very small and not comparable to the assistance that regular TANF recipients receive.
Data on the Number of Families with Children in Poverty
The number of families with children in poverty was calculated using Current Population Survey (CPS) data and the official Census poverty thresholds. We counted related subfamilies and primary families in a single household as one family but counted and determined the poverty status of unrelated subfamilies separately. “Deep poverty” refers to families with incomes below half the poverty line, which in 2014 was about $12,000 for a family of four. Two years of CPS data were merged to improve reliability for state estimates.
Ratio of Families on TANF to Families in Poverty
Ratios are calculated by dividing the number of TANF cases (based on administrative data from HHS or, since late 2006, data collected from states by CBPP) by the number of families with children in poverty (CPS data). We use two-year averages for these calculations to improve reliability.
These ratios should not be interpreted as the percentage of families with children in poverty served by TANF because the number of families on TANF is not a perfect subset of the number of families in poverty. It is possible for a family to be above poverty and receiving TANF benefits, for example — because some families may be poor in the months they receive TANF assistance but have higher incomes the remainder of the year, or because states may choose to encourage work by permitting partial TANF benefits to continue for certain families who obtain earnings that put their total annual income slightly above the poverty line, or because in some households, large extended families may contain more than one eligible TANF case unit. For these reasons, it’s possible for a state to have more than 100 TANF families for every 100 families with children in poverty.
Using the Alabama ratio as an example, the data should be described as follows: In 1994-95, for every 100 Alabama families with children in poverty, the AFDC program served 34 families. In 2013-14, 12 families participated in TANF for every 100 families with children in poverty.
In Alaska and Hawaii, the TANF-to-poverty ratio is above 100 in 1994-95 because the HHS poverty guidelines used in determining program eligibility are significantly higher in these two states than the Census poverty thresholds used in determining the number of poor families. (This is not true for any of the other 48 states. HHS poverty thresholds are set higher in Alaska and Hawaii to allow for higher costs of living in these two states but do not vary elsewhere. The Census Bureau’s poverty thresholds do not vary for any state.)
|State TANF-to-Poverty Ratios Over Time|
|1994-95||2005-06||2008-09||2009-10||2010-11||2011-12||2012-13||2013-14||Ratio Change '05-06 to '13-14|
|National Single-Year TANF-to-Poverty Ratios|
|Number of families with children in poverty||Yearly average of number of families on AFDC/TANF||Ratio|
|TANF Caseloads Over Time|
|Dec-06||Dec-09||Dec-10||Dec-11||Dec-12||Dec-13||Dec-14||Percent Change ‘06-‘14|
 Based on administrative data from the U.S. Department of Health and Human Services and, since September 2006, data collected from states by CBPP.
 To improve the reliability of the poverty data at the state level, we created two-year averages of the poverty numbers. The caseload data are also transformed into two-year averages in order to calculate the TANF-to-poverty ratios. Therefore, the full effects of some of the state actions that occurred between 2013 and 2014 may not appear in the TPR analysis. We anticipate that those effects will become clear after another year of data is available.
 This paper’s analysis of state TANF-to-poverty ratios uses two-year averages of TANF caseload data so that the data are comparable to the two-year averages we use for state-level poverty data. Additionally, poverty data are not available on a monthly basis. In this section, because we focus only on a discussion of monthly TANF caseload trends, two-year averages are not needed.
 Legislation passed this year to reduce Kansas’ time limit from 48 months to 36 months. The state will implement the new policy on January 31, 2016.
 Some other research shows that those families reaching the TANF time limits have more barriers to employment than other families in the caseload. See Pamela Ovwigho et al., “The TANF Time Limit: Barriers & Outcomes among Families Reaching the Limit,” Family Welfare Research & Training Group, University of Maryland School of Social Work, November 2007, http://www.familywelfare.umaryland.edu/reports1/tl_barriers.pdf and
Kristin Seefeldt and Sean Orzol, “Watching the Clock Tick: Factors Associated with TANF Accumulation,” National Poverty Center, May 2005, http://www.npc.umich.edu/publications/workingpaper04/paper9/04-09.pdf.
 Deleena Patton et al., “TANF Caseload Decline: The Well-Being of Parents and Children Leaving WorkFirst in Washington State,” Washington State Department of Social and Health Services, Economic Services Administration, April 2015, https://www.dshs.wa.gov/sites/default/files/SESA/rda/documents/research-11-216_1.pdf.
 Ife Floyd and Liz Schott, “TANF Cash Benefits Have Fallen by More Than 20 Percent in Most States and Continue to Erode,” Center on Budget and Policy Priorities, updated October 15, 2015, http://www.cbpp.org/research/tanf-cash-benefits-have-fallen-by-more-than-20-percent-in-most-states-and-continue-to-erode.
 CBPP analysis of the Current Population Survey with additional data from the Department of Health and Human Services TRIM model. This analysis uses the most recent data available.
 H. Luke Shaefer and Kathryn Edin, “The Rise of Extreme Poverty in the United States,” Pathways, Summer 2014, http://web.stanford.edu/group/scspi/_media/pdf/pathways/summer_2014/Pathways_Summer_2014_ShaeferEdin.pdf. Also see Edin and Shaefer’s book, $2 a Day: Living on Almost Nothing in America, 2015.
 Floyd and Schott, “TANF Cash Benefits Have Fallen by More Than 20 Percent in Most States and Continue to Erode.”
 Greg J. Duncan and Katherine Magnuson, “The Long Reach of Early Childhood Poverty,” Pathways, Winter 2011, http://www.stanford.edu/group/scspi/_media/pdf/pathways/winter_2011/PathwaysWinter11_Duncan.pdf.
 Liz Schott, LaDonna Pavetti, and Ife Floyd, “How States Use Federal and State Funds Under the TANF Block Grant,” Center on Budget and Policy Priorities, updated October 15, 2015, http://www.cbpp.org/research/family-income-support/how-states-use-federal-and-state-funds-under-the-tanf-block-grant.
 The Workforce Innovation and Opportunity Act (WIOA) measures employment rates and earnings for the second quarter after a participant completes a training program, and employment rates for the fourth quarter after completion.