March 6, 2007
GEORGIA’S INCREASED TANF WORK PARTICIPATION RATE IS DRIVEN BY SHARP CASELOAD DECLINE:
Available Data Raise Questions About Whether Georgia
Should Be Labeled as a Model for the Nation
by Liz Schott
Georgia’s success at increasing its TANF work participation rate has been touted as a welfare reform model for other states as they seek ways to meet the new work participation requirements associated with the Deficit Reduction Act changes to the Temporary Assistance for Needy Families (TANF) block grant. An examination of available data on Georgia’s TANF program and of the state’s policies and procedures, however, raises serious questions about whether the state has achieved its higher participation rate in recent years by doing a better job to help parents move from welfare to work or by restricting poor families' access to assistance.
Under federal law, states must meet a specified “work participation rate” in its TANF assistance programs. The work participation rate is simply the ratio of the number of adult TANF recipients who are working or in specified work-related activities to the number of families with adults receiving cash assistance through TANF-related programs. Georgia’s work participation rate has increased dramatically in recent years – from 11 percent in 2003 to over 65 percent in 2006. But the increased work participation rate is primarily a factor of fewer families receiving assistance. In fact, despite the dramatic increase in the rate of work participation, there were fewer adult TANF recipients participating in work or work-related activities in May of 2006 than there were in an average month in 2003. It is the significant decline in the TANF adult caseload — more than 80 percent since January 2004 — that has produced the state's high work participation rate.
If the number of families receiving assistance had fallen because the number of families that need assistance was falling, these data would not necessarily be cause for concern. At this point, however, there is no evidence that the number of very poor families (to be eligible for TANF in Georgia families typically must have incomes well below the poverty line) in Georgia has been cut during this period. In fact, Census data show an increase in the number of children living in single-parent families with incomes below one-half of the poverty line over the period when Georgia's TANF caseload began to fall sharply.
While sharply reduced need does not appear to be driving the decline in the caseloads, there is strong evidence that since 2004, a series of policies and procedures in the state’s TANF program have been put in place that likely have had the effect of discouraging needy families from applying for TANF assistance or remaining assistance recipients.
- Data show, for example, that most families that leave Georgia’s TANF program are not working: less than one-third of TANF case closures between October 2004 and April 2006 are due to the family having earnings and no longer needing assistance.
- Georgia’s own research shows that the proportion of former recipients who are working in the quarter they exit the rolls has fallen since 2001. The report by the state’s Department of Human Resources that covers 2004 exits states that the reduction in employment among TANF leavers in 2004 suggests that the state’s policies initiated during that year were "more powerful in motivating recipients to leave TANF without employment than in moving recipients from welfare to employment."
- Data on new applications show a sharp rise in the number of applications that are denied for procedural reasons — one-third of Georgia’s TANF denials are due to withdrawal of application and another third are due to failure to cooperate in new application procedures. Only 7 percent of denials are due to the family having too much income and not being in need.
Over the past twenty years, much has been learned about how best to help families move from welfare to work. Models have been developed through careful evaluations and research. New strategies certainly are needed to improve the shortcomings in TANF and other programs designed to help families succeed in the labor market. At this time, there is no data or research to suggest that Georgia provides a model to other states for how to engage more recipients in work activities, help recipients overcome barriers to employment, or assist parents secure jobs. Before national leaders tout the state's approach as one to be emulated across the nation, more information is needed about whether the states' caseload decline has been caused by a reduction in the number of needy families or a policy and program changes that restrict needy families' access to poor families.
Work Participation Rate Driven By Declining Caseloads
During Federal Fiscal Year 2003, Georgia's average monthly work participation rate was 11 percent, with, on average, 3,194 families participating in countable work activities each month. In 2004, Georgia's participation rate increased significantly — to a monthly average of 25 percent. This increase was due to two main factors: an increase in the number of families participating in work activities and significant decline in the number of TANF cases that include adults. (See Figure 2.)  Since the end of 2004, however, the number of adults participating in work activities has decreased. As of May 2006, only 3,001 families met the work requirement standards — fewer than had participated in an average month in 2003 — yet the state achieved a work participation rate of 67 percent. (See Figure 2.) The state’s federal work participation rate increased because during this same time period the number of families with an adult receiving assistance plummeted. Between January 2004 and May, 2006, the number of families with an adult receiving TANF assistance has declined by 80 percent — from more than 33,488 to just 6,587 families. The number of TANF cases that include an adult recipient continued to decline for the rest of 2006 — down to 3,856 families — so that, all tolled, Georgia's TANF adult caseload declined by 88 percent during the three-year period of 2004 through 2006.
At the same time, the state's "child-only" caseload declined only slightly. (In addition to families in which adults — generally parents — and children receive assistance, there are some families in Georgia, as in all states, that receive assistance only for the children in the family.) These so-called “child-only cases” are typically cases in which children are cared for by relatives other than their parents, by parents who receive SSI, or by parents who are ineligible for TANF based on their immigration status. In May 2006, there were roughly 23,000 child-only cases in Georgia, only a slight decline from the number of such cases in 2004. (See Figure 2.) Child-only cases now represent more than three of every four TANF cases in Georgia.
What is Causing Georgia's Caseload Decline?
The caseload decline appears to be due to a range of factors, including changes in statewide policies and new procedures individual counties have developed in response to statewide directives to reduce the number of families receiving assistance and to increase work participation rates. Georgia’s TANF program is county-administered, which means that each county determines eligibility and manages the program subject to state-set policies and directives. It is not always clear what means any given county used to achieve its caseload reduction or work rate results and the extent to which a practice in one county is found in others as well.
Based on the state's directives to counties, changes to its policy manual in 2006, and available data on application denials and case closures, the following appear to be factors in the decline:
- Work rate performance mandate and goals: In early 2004, the state sent a strong directive to counties to achieve a 50 percent work participation rate. The statewide average work participation rate at the time was about 20 percent. Counties subsequently have been told that they now are expected to achieve a 70 percent work rate. Counties' work participation rates have been measured and monitored each month against these goals, with special recognition of the counties that have achieved 100 percent participation.
- Culture and message change: The agency touts culture and message change as a big part of the caseload drop, with the oft-repeated message from DHR Commissioner B. J. Walker that “Welfare is not good enough for any family.”  There is evidence, discussed below, that at a minimum raises questions about whether this message has been translated at the local office level into caseworkers actively discouraging families from following through on their applications for TANF assistance and encouraging families to leave the program.
- Changes in the application process and application approvals: Under new statewide rules, families must attend an orientation and develop an “employment plan” before the family’s application can be approved. This can mean that families are required to participate in multiple meetings and perform various tasks — which can include job search — before their benefits can be approved.
Data show a sharp decline — 25 percent — in the number of applications submitted and an even sharper drop in the number of filed applications that are approved. In April 2006, only 20 percent of applications were approved, down from 40 percent in January 2004.
A large share of TANF applications in Georgia are denied for reasons unrelated to whether the family was poor enough to qualify for TANF and whether they met other basic eligibility criteria (such as having a child in the family).
- Of the Georgia TANF applications denied in the first third of 2006, about one in three was denied because the family chose to withdraw its application before the state made an eligibility determination. In 2002 and 2003, by contrast, just one in five applications was denied for this reason. As is discussed below, this increase in “voluntary withdrawals” appears related to a new emphasis on discouraging applicants from following through on their application for aid.
- Another one-third of all applications denied in the first third of 2006 were denied because, according to the state, the family failed to cooperate with the eligibility process; this too represents an increase from 2002 and 2003 when one if four applications were denied for non-cooperation. Again, this increase in application denials appears related to policy and procedural changes that have increased the number of “hoops” families must go through before their application is approved. These hoops — such as attending multiple meetings with caseworkers and completing job search activities — can make it particularly difficult for families with significant problems to complete the application process.
- Only 7 percent of applications that were denied in the first third of 2006 due to the family having too much income, down from 11 percent of denials in 2002 and 2003.
- Benefit terminations have been a major factor in declining caseload: A significant portion of the caseload decline also comes from the increased rate at which families’ benefits are terminated. A recent study of families leaving TANF conducted by the Georgia Department of Human Resources notes that while the number of families applying for and opening new TANF cases has fallen significantly, a majority of the caseload decline in 2004 and 2005 was attributable to the increase in the number of families leaving the TANF program. According to the state's study of families leaving TANF, 52 percent of adults who left TANF in 2004 had some earnings at some point in the calendar quarter in which their TANF benefits ended. The share of TANF leavers with earnings in the quarter in which they left TANF has declined each year and is down from 61 percent of 2001 TANF leavers having earnings in the quarter they left TANF.
What is less clear is precisely why the number of terminations has increased. Data from the state TANF agency covering the period from October 2004 through April 2006 shows that most families that leave TANF do not exit because they have earnings to meet their needs and no longer need assistance. During this period:
About 16 percent of TANF case closures (among cases that include an adult) were families requesting that their TANF benefits be terminated but that their other benefits, such as food stamps and Medicaid, continue. These data suggest that poor families may be encouraged to withdraw from the TANF program even if they remain eligible and wish to continue receiving other forms of assistance.
- The economy does not appear to be the major reason for decline: The recent report on Georgia families that left TANF in 2004 issued by the Georgia DHR notes that despite some job growth in the economy since 2004, "persistent losses in manufacturing and information have muted Georgia's economic recovery and limited employment opportunities for TANF recipients and leavers." The report concludes that the pattern of caseload decline in light of the labor market circumstances "suggests that something other than the economy was contributing to the decreasing caseload." The report further notes that the program changes appeared to be effective in keeping families that left TANF without work from returning to the program, as there was no corresponding increase in employment to suggest that the families no longer needed assistance. In addition, another recent study notes that other poverty indicators – including poverty and unemployment rates as well as Georgia's food stamp and Medicaid caseloads increased since 2002, also suggesting that the state was not experiencing a sharp reduction in the number of poor families with children.
Moreover, data from the Census Bureau's American Community Survey shows that the number of Georgia children living in single-parent families with incomes below half the poverty line increased between 2003 and 2005. (This analysis focuses on children in single-parent families because these are the families most likely to participate in TANF assistance programs.) The Census data show that in 2005, 162,200 children in Georgia lived in single-parent families with incomes below one-half of the poverty line. (One-half of the poverty line in 2005 amounts to just $649 per month for a family of three.) This is a statistically significant increase from 2003, when 143,200 children lived in such families. These data strongly suggest that the reduction in the number of families receiving TANF assistance over this period was not driven by a reduction in the number of very poor families, but rather by policy and procedural choices the state made in its TANF program.
Survey of Caseworkers from Domestic Violence Shelters Provides Further Evidence that Families are Discouraged from Applying and Encouraged to Leave the TANF Program
The Georgia Coalition Against Domestic Violence conducted a survey in February 2006 of "TANF Assessors" — domestic violence shelter employees who typically are stationed at county welfare offices to provide on-site assistance to victims of domestic violence. The survey was designed to learn more about recent changes in application procedures in county welfare offices in response to concerns by shelters that families were finding it increasingly difficult to apply for and receive TANF. The survey generated more than three dozen responses, providing information on patterns and practice inside Georgia welfare offices that go beyond anecdotal reports. The survey — while not definitive — points to the need for more research and analysis of Georgia’s policies and procedures before concluding that they should serve as a national model.
Some of the key points about the application process that the survey results reveal include:
- More than two-thirds of respondents indicated that they had seen “indicators that something is happening during orientations which results in applicants leaving before the TANF application is completed.” Some respondents reported that county staff or directors talk with clients before the orientation with the result that applications are dropped and the client does not attend the orientation. (Cases in which a family decides not to continue with the application are likely coded as “voluntary withdrawal of application,” which accounted for one-third of application denials in April 2006.)
- More than two-fifths (44 percent) of respondents report that they have seen cases in which applicants were assigned to work activities that were difficult or impossible because of the individual’s disability. One noted that clients who claim they have a disability are encouraged to withdraw their applications and generally do. Another reported that a client (with medical documentation of inability to work) was told that she could not apply for TANF because she could not work.
- Nearly two-fifths (39 percent) report that they have seen cases in which applicants were assigned to work activities that were difficult or impossible because of pregnancy. Several reported seeing cases in which pregnant women were told (incorrectly) they could not get TANF or should come back and apply after the baby is born.
- In response to open-ended questions, some respondents reported seeing cases in which individuals that came to a welfare office were dissuaded from filing an application and circumstances where the director of the welfare office personally talked with families to discourage them from applying. A few even reported that some families were told that if they applied for TANF assistance, child protective services might investigate their ability to meet their children's needs.
The TANF changes enacted in the Deficit Reduction Act of 2005 require states to meet much higher work participation requirements. Many states (and others) are concerned that it will be difficult to meet increased work participation rates and that significantly increasing work participation rates will require increased resources and creative new strategies that will take time to develop. To assuage these concerns, some tout the Georgia experience as evidence that states can quickly increase their work participation rates. These officials tell a simple story — Georgia more than doubled its work participation rate in under two years, so other states can and should do so as well.
At this time, however, far too little is known about the impact of Georgia's policies and procedures on important welfare reform goals — including helping parents prepare for and secure employment and reducing hardship among families — to conclude that Georgia’s program should serve as a national model. In fact, data showing sharply increased application denials, high rates of benefit terminations among families that are not working, and an increasing share of families leaving TANF when they do not have jobs all suggest that more research — and healthy skepticism — are called for.
 The work participation rate calculation typically includes cases in which both children and the adults in the family (typically the parents) receive assistance and does not generally child-only cases in which only the children receive assistance. So-called “child-only” cases are typically cases in which children are being cared for by relatives other than their parents, parents who receive SSI, or parents who are ineligible for TANF based on their immigration status. (While the DRA changes alter somewhat which families are included in the calculation, this change is not likely to have a significant impact on Georgia and most of the data discussed in this report arise prior to the effective date of the DRA changes.)
 See discussion on p. 4.
 “TANF Adult Closures: Closures Due to Income,” October 2004-April 2006. Georgia Department of Human Resources.
 "Employment, Earnings, and Recidivism among Georgia's TANF Leavers: Findings from the TANF Follow-Up System," Georgia Department of Human Resources, May 2006 at http://dfcs.dhr.georgia.gov/DHR-DFCS/DHR-DFCS_CommonFiles/Leavers%202004.pdf. The report, while useful, covers a period of significant change in Georgia's TANF program and, thus, does not provide a complete picture of the ultimate and more recent impact of the new policies that were just being put in place. This recent report looks at the circumstances of families that have left TANF through 2004; Georgia's adult TANF caseload has declined by more than 80 percent since 2004.
 U.S. Department of Health and Human Services, Administration for Children and Families, TANF Work Participation data, Tables 1 and 4A, http://www.acf.hhs.gov//programs/ofa/particip/indexparticip.htm#2003. For this period, state data reflecting varying monthly participation data is not available to author, so the average annualized data reported to HHS by the state is cited.
 For FY 2004, Georgia's average monthly participation rate was 24.8 percent and the average number of adults counting as participating in work activities was 6,052. The number of families participating in work activities each month increased in late 2003 and 2004, reaching its high in December 2004. By 2005, however, the number participating began a steep decline falling from 8,968 participating in December 2004 to 3,001 in May 2006, the most recent month for which data are available. The sharp decline in the TANF adult caseload began during 2004.
 The data shown here are the participation rates reported by the state. http://dfcs.dhr.georgia.gov/DHR-DFCS/DHR-DFCS-publication/12_2006_OM_FI_State.pdf . Federal statistics are not available for years after 2004.
 http://www.gov.state.ga.us/press/2006/press1173.shtml; see also, Georgia Department of Human Resource’ TANF Program Manual, Section 1801.
 TANF Application Denials, 2002-2006YTD, Georgia Department of Human Resources.
 "Employment, Earnings, and Recidivism among Georgia's TANF Leavers: Findings from the TANF Follow-Up System," Georgia Department of Human Resources, May 2006,at http://dfcs.dhr.georgia.gov/DHR-DFCS/DHR-DFCS_CommonFiles/Leavers%202004.pdf.
 "Employment, Earnings, and Recidivism among Georgia's TANF Leavers: Findings from the TANF Follow-Up System," Georgia Department of Human Resources, May 2006,at http://dfcs.dhr.georgia.gov/DHR-DFCS/DHR-DFCS_CommonFiles/Leavers%202004.pdf. There are several reasons why the state's case closure data shows less than one-third of adult closures due to income and the leaver study reflects about half of leavers with some earnings in the quarter of closure. The leaver study includes earnings that might be from a period after the TANF case has closed but within the same calendar quarter. In addition, coding of reasons for closure do not always capture all of a family's circumstances; a case might be coded with another reason for closure (such as failing to provide information or complete a renewal) when a family with earnings discontinues TANF.
 TANF Adult Closures: Closures Due to Income, October 2004-April 2006. Georgia Department of Human Resources.
 Mandatory Adult Closure Reasons, January 2006, Georgia Department of Human Resources.
 "Employment, Earnings, and Recidivism among Georgia's TANF Leavers: Findings from the TANF Follow-Up System," Georgia Department of Human Resources, May 2006, at http://dfcs.dhr.georgia.gov/DHR-DFCS/DHR-DFCS_CommonFiles/Leavers%202004.pdf.
 Robert Z. Welsh, "Georgia's Declining Welfare Caseload: Sign of Success or Failure?, Georgia Budget and Policy Institute, January 2007, http://www.gbpi.org/pubs/specialreport/20070119.pdf.
 These data are from the American Community Survey which surveys families over a 12 month period about their circumstances over the prior twelve months. Thus, the data for 2003 actually represent families' circumstances over a period that spans from January 2002 to November 2003. The data for 2005 represent families' circumstances from January 2004 to November 2005. The 2005 data are the most recent data available.
 Survey on Current TANF Benefit Access in Georgia.