Final Synthesis Report of Findings from ASPE "Leavers" Grants. Caseload and Characteristics of Welfare Leavers

11/27/2001

While welfare caseloads declined throughout the US during the 1990s, the magnitude of the decline varied from state to state. The average leaver from states with large caseload declines may come from deeper in the caseload and have more barriers to overcome in moving to work than the average leaver from other states. As such, these leavers may have less success in the labor market, face greater hardships, and may be more likely to return to welfare. Note, however, that recent research suggests that leavers are not becoming more disadvantaged over time (Loprest 2001).

Figure II.1 shows caseload declines across the fourteen states hosting leaver studies. The vertical line represents the date of TANF implementation, and the shaded areas denote the cohorts examined in the leaver studies. Caseload declines between

Figure II.1:
AFDC/TANF Caseload Changes in Welfare Leaver Study Sites: 1994-2000

AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000

AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000

AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000

AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000AFDC/TANF Caseload Changes in Welfare Leaver Study Sites:   1994-2000

Note: Shaded areas denote the cohorts of TANF leavers that were followed in the studies. August 1996 and December 1999 range from 30 percent in the District of Columbia to 68 percent in Florida. Caseloads fell by more than 50 percent in Georgia, Illinois, Ohio, South Carolina, and Wisconsin. In addition to DC, caseloads declined by less than 40 percent in California, Iowa, and New York.

Differences in the personal characteristics of welfare recipients and welfare leavers also must be considered when comparing findings across leaver studies. Indeed, part of any difference in outcomes across sites may be due to differences among leavers themselves. Further, states likely structure their welfare policies with their welfare populations in mindfor example, a state with a high proportion of high school drop outs may emphasize work readiness programsand this too may affect the status of leavers.

Table II.4 compares the characteristics of leavers across 12 studies which report such information. The table focuses on leavers ages, race/ethnicity, marital statuses, the number of children they have, and their educational attainment. Not all studies provide data on each of these characteristics, and they do not all report them in the same way. For example, Iowa and Massachusetts report the average age of leavers while other studies report a distribution of leavers ages. These differences make direct comparisons more challenging.

Table II.4
Selected Characteristics of Leavers by State

State/Study

Age of Unit Head1 Race/Ethnicity of Unit Head
<=20 21-30 31-40 40+ White Black Other2 Hispanic

Arizona

7 51 38a 4 42 10 14 35

District of Columbia

4 44 36 16 1 97 1 2

Florida

33c

42 40   17

Georgia #

34b 41b 21b 4b 31 67 2  

Illinois #

7 49 30 14 34 56 1 9

Iowa *

30c

81      

Massachusetts #

33c

60 20 20 29 8

Missouri

11 46 31 12 63 35 2 1

South Carolina *

20d 25d 38 17 22 78    

Washington *

3 46 35 16 70 8 23 13 8

California Bay Area *

14f 33f 35 18 28 12 16 44

Cuyahoga Co.

6 52 31 12 23 70 2 6
Table II.4:
Selected Characteristics of Leavers by State (Continued)

State/Study

Marital Status Number of Children Education
Never Married Married D/W/S3 0-1 4 2 3+ <HS HS HS+

Arizona

51 12 37       44 41 11.2 5

District of Columbia

87 5 9 50 29 22      

Florida

      2 6          

Georgia #

61 12 23 5 33 33 35 22 59 19

Illinois #

65 8 27 52 28 20 42 7 44 7 15 7

Iowa *

48 15         26    

Massachusetts #

59 14 27 35 31 35 27 40 33

Missouri

      51 30 19 39 47 10 5

South Carolina *

      40 30 30 44 40 16

Washington *

  15   2 6     29 38 9 33

Bay Area *

  10   49 30 21 48 26 26

Cuyahoga Co.

      45 32 24      

1Age breakdowns differ from headings as follows:
a- 31-45, 45+;
b- 18-24, 25-34, 35-44, 45+;
c- average age;
d- 18-24, 25-30;
e- 18-24;
f- 16-21, 22-29.
2 Other combines Asian, Native American, and Other.
3 D/W/S stands for divorced, widowed, or separated.
4 0 children can include pregnant recipients or families where children have been removed from the home.
5 Does not sum to 100 because the state includes another category, ie. cohabitating or did not respond.
6 Average number of children
7Administrative data reported for the survey cohort.
8 Race and ethnicity asked separately. Number represents percent Hispanic of all leavers.
9 Includes vocational technical school.
*Survey data.
#Administrative and survey data.

A priori, it is difficult to anticipate whether younger leavers will, on average, fare better or worse than older leavers. While younger leavers probably have fewer children and likely have shorter spells of receipt prior to exit than older leavers, they also probably have younger children and less work experience. Six of the studies under review report the proportion of leavers age 20 and younger. The share of leavers who are very young ranges from a low of 3 in Washington to a high of 11 percent in Missouri. If we consider data regarding the under 30 group, we can include the South Carolina and San Mateo county studies. In four studies (Arizona, Cuyahoga county, Illinois, and Missouri ) well over half of all leavers are 30 or younger; in DC, the Bay Area, South Carolina, and Washington, less than half of leavers are under age 30.

It is also difficult to anticipate how race/ethnic differences between may affect leavers outcomes because race is only one of many differences among the study areas' TANF caseloads and local population bases. In the District of Columbia, virtually all leavers are black, which is not surprising given the demographic make-up of the city and its caseload. The proportion of leavers who are black ranges from a low of 8 percent in Washington to a high of 97 percent in DC. The share of leavers who are Hispanic ranges from 1 percent in Missouri to 44 percent in the Bay Area. The Bay Area also has a very high proportion (16 percent) of others. Although the majority of "others" in the Bay Area study are Vietnamese, the "other" races and ethnicities may include Asian and Pacific Islanders and Native Americans.

Differences in marital status, number of children, and education all have a stronger theoretical link than race or age to the outcomes of welfare leavers. For example, married or previously married leavers may well have access to more sources of support (for example, child support) than never married leavers. Table II.4 shows that among the 6 leaver studies reporting this information, the share of leavers who are never married ranges from 51 percent in Arizona to 87 percent in DC.

Leavers with more children may have a harder time balancing work and child rearing than other leavers. We see that the share of leavers with one child or less ranges from about one in three in Georgia and Massachusetts to about one half in DC, Illinois and Missouri.14 Conversely, Georgia and Massachusetts are the two study sites with highest proportions of leavers with three or more children, again, about one in three. Missouri has the lowest proportion of leavers with three or more children (19 percent).

Finally, leavers with higher levels of educational attainment should have an easier time finding, keeping, and advancing in jobs than less educated leavers. In four study sites, over 40 percent of leavers had failed to complete high school: Arizona, Illinois, South Carolina, and the California Bay Area. In Massachusetts and Wisconsin, leavers tend to have more education, with about one in three having some schooling beyond high school.

Overall, there are some potentially important differences across leavers in the various ASPE-funded studies; however, these differences may have offsetting effects on outcomes. For example, Massachusetts leavers are more educated on average than other leavers but they also tend to have more children, and Illinois leavers are less educated but tend to have fewer children.