[ Main Page of Report | Contents of Report ]
As noted in the preceding chapter, employment rates for WtW enrollees in 9 of the 11 study sites were much higher one year after entering the program than at the time of entry. Nevertheless, fewer than half of enrollees were employed in eight of the sites. Fringe benefits were available to only a fraction of those who were employed, and for many the rate of pay was too low to have allowed them to escape poverty even if they had been working full time. Consequently, many enrollees needed assistance to make ends meet. Even with that assistance most remained in poverty, though relatively few experienced high levels of material distress.
This chapter presents a broad range of measures of the well-being of WtW enrollees and their families one year after they entered WtW. Much of the variation in the values of these measures across the 11 study sites can be attributed to differences among the enrollees at the time of entry.(56)This chapter simply describes how enrollees and their families were faring one year after program entry; it does not present estimates of the impacts of the WtW programs on their well-being. For reasons that were given in Section I.C, this evaluation did not attempt to estimate the impacts of the WtW programs that were studied.
[Go To Contents]
Notwithstanding the modest success that enrollees experienced in the labor market after entering WtW, most needed help from outside their households to make ends meet. This section documents the monetary and nonmonetary assistance that enrollees received from extended family members and friends, community organizations, and government programs.
1. Did Enrollees Receive Assistance from Family/Friends and Community Organizations?
Most WtW enrollees received assistance from extended family members and friends during the year following program entry, but only a minority received help from community organizations (Exhibit V.1).(57) Across the study sites, about 7 or 8 of every 10 enrollees received help from family and friends — most often in the form of transportation, which was received by half or more of enrollees in all sites other than Philadelphia (Appendix Exhibit B.15). The enrollees in Philadelphia had the lowest rate of receipt of any assistance from family and friends. In contrast, the male noncustodial parents on probation or parole who comprised the enrollees in the Milwaukee NOW program had the highest rate of receipt of this type of assistance, which was especially important given that community organizations were largely unresponsive to their needs and there were few government programs that they could turn to for support.
Assistance from community organizations was less pronounced among WtW enrollees. Exhibit V.1shows that between one-quarter and one-half of enrollees across the study sites received assistance from food pantries or soup kitchens, crisis hotlines or centers, thrift shops, or churches. Enrollees in Philadelphia and Milwaukee had the lowest rates of receipt of assistance from community organizations. The unique characteristics of the Milwaukee enrollees may have made it more difficult for them to access assistance from community organizations that could have addressed their particular needs.
2. Did Enrollees Receive Assistance from Government Programs?
Receipt of assistance from government programs was pervasive among WtW enrollees one year after program entry.(58) Only in Milwaukee were fewer than half of enrollees and their households receiving some form of government assistance (Exhibit V.2), which is further evidence of their limited options for assistance. In other sites, rates of receipt of such assistance ranged from 65 percent in Baltimore County to 87 percent in Chicago and Philadelphia. Among enrollees whose households were receiving some form of government assistance one year after enrollment, the average monthly amount ranged from about $550 to $750, except in the two JHU sites, where it was $508 and $441, respectively (Appendix Exhibit B.16).
Food stamps were the most common type of government assistance for WtW enrollees in every site one year after program entry, with rates of receipt as high as 82 percent (Appendix Exhibit B.16). TANF was the second most common form of government assistance in nine sites; in Baltimore County and Milwaukee the rate of receipt of Supplemental Security Income or Social Security Disability Insurance exceeded that of TANF. In every site, very small percentages — 8 percent or less — of enrollees received assistance from each of the other programs considered in this evaluation.(59)
In all of the study sites except Baltimore County, Milwaukee, and Nashville, WtW enrollees were far less likely to be receiving TANF one year later than they were at the time of enrollment.(60) Recipiency rates in many of the sites were roughly 50 percent lower at the end of the follow-up period than at the beginning (Exhibit V.3). While the WtW programs may have contributed to these differences, other factors may also have been instrumental. Consequently, we again caution the reader not to interpret the differences as impacts of the programs.
3. Did Enrollees Achieve Independence from TANF Through Employment?
The movement of families off welfare and into work was the fundamental objective of both PRWORA and the legislation that authorized the WtW grants program. Perhaps the best available measure of the success of the program in achieving this objective is the proportion of enrollees who were employed and off TANF at the end of the evaluation's one-year follow-up period. In all of the study sites except the two hosting the JHU program, no more than 40 percent of enrollees were employed and off TANF one year after program entry (Exhibit V.4). In six of these nine sites, the proportion of enrollees who were on TANF and not working was about as large or larger than the proportion who were employed and off TANF — meaning that heavy dependency on welfare was more common than self-sufficiency.(61)
Becoming employed and leaving TANF is a milestone for a WtW enrollee, but securing employment even without leaving TANF is a noteworthy partial success. One year after entering the program, roughly 10 percent of WtW enrollees in most of the study sites were employed and on TANF (Exhibit V.4). Nashville, West Virginia, Philadelphia, and Yakima were the best-performing sites according to this measure, with values of 15 percent or higher.
[Go To Contents]
The total income received by the households of WtW enrollees was very low or moderately low in all of the study sites one year after program entry. In all but one site, household incomes were below the poverty threshold for more than half of the enrollees. Nevertheless, they did not report having experienced high levels of material distress during the year following enrollment.
1. What Were the Amounts and Sources of Household Income for Enrollees?
The mean monthly total income received by the households in which WtW enrollees were residing one year after program entry ranged from $1,000 to $1,600 across the study sites (Exhibit V.5).(62) It was nearly $1,400 or more in only four sites: Baltimore County, Milwaukee, Yakima, and St. Lucie County.
Earnings by WtW enrollees and the persons with whom they were living were critical sources of household income in these four higher-income sites; the combined earnings from these two sources exceeded $1,000 per month, on average. Milwaukee provides a striking example of how important the earnings of other household members can be in determining total household income. In this site, earnings by other persons with whom WtW enrollees were living accounted for $686, or nearly half of the mean monthly total household income. This is consistent with the pattern of strong support from family and friends and weak support from other sources for the Milwaukee enrollees.
Low combined earnings by enrollees and persons with whom they were living characterized the household incomes of enrollees in the seven study sites where the mean total income was substantially less than $1,400 per month. The low combined earnings were only partially offset by relatively high amounts of government assistance, leaving mean total incomes well below those in the four higher-income sites. In West Virginia for example, the mean combined earnings of enrollees and others in their households was just $656 and the mean total income was only $1,186 (Exhibit V.5).
2. Did Enrollees Escape Poverty?
The incomes received by the households of WtW enrollees one year after they entered the program generally did not exceed the federal poverty threshold.(63)Baltimore County was the only site wherein a majority of enrollees had above-poverty incomes (Exhibit V.6). In contrast, more than two-thirds of enrollees in Boston, Chicago, Ft. Worth, Nashville, Philadelphia, Phoenix, West Virginia, and Yakima were living in poverty at the one-year follow-up point. And in six of these sites (all but West Virginia and Yakima), most enrollees were living in severe poverty, with total incomes below 50 percent of the federal poverty threshold.
3. Did Enrollees Experience Material Distress?
A high incidence of poverty did not necessarily mean that WtW enrollees experienced especially high levels of material distress. This evaluation measured five types of material distress that enrollees and their families may have experienced during the year after program entry: inability to fully pay the rent or mortgage, eviction, inability to fully pay a utility bill, termination of a utility, and disconnection of the telephone. Among these, inability to fully pay the rent or mortgage was most common in Milwaukee and St. Lucie County, whereas inability to pay a utility bill in its entirety was most common in the other sites (Appendix Exhibit B.22). An index created for this evaluation summarizes the five types of material distress on a 0-to-1 scale, with higher values indicating greater distress.(64), (65) Across the study sites, the mean value of the index ranged from 0.23 to 0.32, corresponding to the experience of between one and one-and-a-half types of distress (Exhibit V.7).
A relatively high incidence of poverty in a study site was not always accompanied by a relatively high mean value of the index of material distress. For example, WtW enrollees in Chicago, Philadelphia, and West Virginia had relatively low mean index values despite having the highest poverty rates among the study sites. In these same three sites, enrollees had very high rates of receipt of assistance from government programs (Exhibit V.2), which may have helped them to avoid the most extreme consequences of poverty.
[Go To Contents]
Enrollees who were employed one year after entering WtW typically had much higher total household incomes and were less likely to be living in poverty than those who were not employed. Nevertheless, the two groups experienced similar levels of material distress in most of the study sites.
Employed enrollees in all but the St. Lucie County site enjoyed significantly higher mean incomes than those who were not employed (Appendix Exhibit B.23). The average income differential associated with employment exceeded $500 per month in six of the sites. Given these differences in income, it is not surprising that household poverty rates were significantly lower for employed enrollees in all of the sites except St. Lucie County and West Virginia, as shown in Exhibit V.8. But the link between poverty and material distress again proved to be weak — in most of the sites where employed enrollees had significantly lower rates of poverty they did not have correspondingly lower mean values of the index of material distress.(66)
Although the household poverty rate at the one-year follow-up point was high for WtW enrollees in most of the study sites, the incidence of specific hardships was generally low. This section presents findings on homelessness and lack of health insurance. Due to the unique population served by the WtW program in Milwaukee, these two hardships were especially prevalent at that site.
1. Did Enrollees Experience Homelessness?
WtW enrollees in three of the study sites exhibited high rates of homelessness. Fifteen percent of enrollees in Boston lived in emergency or long-term shelters sometime during the post-enrollment year (Exhibit V.9). A more extreme form of homelessness was common among enrollees in Milwaukee; 17 percent of them lived on the streets sometime during the year. These two forms of homelessness were both prevalent at the same high rate of 12 percent among WtW enrollees in Phoenix.(67) In sharp contrast, the rates of these two forms of homelessness did not exceed 2 percent in St. Lucie County and West Virginia. And in the remaining study sites they were 7 percent or lower.(68)
WtW enrollees in Milwaukee were far more likely than enrollees in any of the other study sites to have experienced the more extreme of the two forms of homelessness — living on the streets. Furthermore, the ratio of the rate of living on the streets to the rate of living in shelters, was far larger in Milwaukee (3.4) than in any other site, the next highest being 1.8 in Baltimore County. This high ratio suggests that WtW enrollees in Milwaukee had needs for shelter that were not being met. This is further evidence of a point made earlier in this chapter (Section A.1) that the unique characteristics of WtW enrollees in Milwaukee made it more difficult for them to access needed assistance from community organizations.
2. Did Enrollees or Their Children Lack Health Insurance?
Almost all children of WtW enrollees were covered by public or private health insurance at the one-year follow-up point, but coverage for the enrollees themselves was markedly spottier. Coverage rates for children were 95 percent or higher in seven of the study sites and were no less than 85 percent in the other four sites (Exhibit V.10).(69) Most of the enrollees in Milwaukee were not receiving TANF and therefore were unlikely to qualify for Medicaid; consequently, only one-third of them were covered by health insurance. In two other sites — Baltimore County and Ft. Worth — somewhat less than three-fourths of enrollees were covered by health insurance one year after they entered WtW. In contrast, the coverage rate for enrollees exceeded 90 percent in Boston, Nashville, and Philadelphia and was just short of that threshold in Chicago.(70)
Exhibit V.1
Percentage of WtW Enrollees Who Received Assistance from Support Networks
During the Year after Program Entry
Exhibit V.2
Percentage of WtW Enrollees Who Were Receiving Assistance
from Government Programs One Year after Program Entry
Exhibit V.3
Percentages of WtW Enrollees Who Were Receiving TANF
at the Time of Program Entry and One Year Later
Exhibit V.4
Employment and Receipt of TANF by WtW Enrollees One Year after Program
Entry
Exhibit V.5
Mean Monthly Household Income of WtW Enrollees,
by Source, One Year after Program Entry
Exhibit V.6
Incidence of Household Poverty among WtW Enrollees One Year after Program
Entry
Exhibit V.7
Mean Value of Index of Material Distress Experienced by WtW Enrollees
And Their Households During the Year after Program Entry
Exhibit V.8
Incidence of Household Poverty among WtW Enrollees,
by Employment Status, One Year after Program Entry
Exhibit V.9
Percentages of WtW Enrollees Who Experienced Two Forms of Homelessness
During the Year after Program Entry
Exhibit V.10
Percentages of WtW Enrollees and Their Children Who Were Covered by Health
Insurance
One Year after Program Entry
[Go To Contents]
56. For example, the JHU sites in Baltimore County and St. Lucie County primarily served persons who were already working at the time of enrollment, whereas the other nine study sites targeted persons who were not working. Based on this difference in targeting strategies alone, we would expect enrollees in the JHU sites to be doing better, on average, than enrollees in the other sites one year after program entry.
57. The evaluation's 12-month follow-up survey gathered information on help received from family and friends in the form of transportation, use of a telephone, a place to stay, groceries or meals, children's things, and money. It also gathered information on help received from the following community organizations: food pantry or soup kitchen, crisis hotline or center, thrift shop, and church. See Appendix Exhibit B.15 for details.
58. The 12-month follow-up survey explicitly inquired about the receipt of income from seven government programs: food stamps, TANF, Supplemental Security Income or Disability Insurance, Social Security retirement or survivors benefits, Unemployment Insurance, General Assistance, and foster care or adoption assistance. Some respondents reported assistance from other government programs and that was also included in this analysis.
59. The analysis of assistance from government programs included programs for which we were able to conveniently measure both the receipt and the dollar value of benefits. Measurement of the value of public housing would have been problematic in the 12-month follow-up survey. Consequently, we did not attempt to gather information on the dollar value of either public housing or Section 8 housing subsidies in the survey. If housing assistance (public housing and Section 8 subsidies) had been included in the analysis of recipiency, it would have been the first or second most common form of government assistance in five sites and the third or fourth most common in six sites. Appendix Exhibit B.18 reports rates of receipt of housing assistance ranging from 5 percent in Milwaukee to 78 percent in Boston.
60. Many of the Milwaukee enrollees were incarcerated at the time of enrollment and consequently prohibited from receiving TANF. During the ensuing year, most left prison or jail and a few joined existing TANF units or formed new units, causing the TANF receipt rate to rise from 1 to 6 percent.
61. Milwaukee — where the NOW program had a distinctive focus on ex-offenders/noncustodial parents — presents the sharpest deviation from the pattern of greater welfare dependency than self-sufficiency. Also, in West Virginia and Yakima more enrollees were employed and off of TANF than were on TANF and not working.
62. The measure of total income reported here includes the value of food stamp benefits.
63. To be consistent with the standard methodology for determining poverty status, food stamps were excluded from the measure of household income for the poverty analysis.
64. The value of the index of material distress was computed for an enrollee by adding up the number of affirmative responses to the questions regarding the presence of the five types of material distress and dividing by the number of valid responses. If all five types of distress were experienced, the index took on its maximum value of 1; if only one type was experienced, it took on a value of 0.2 (assuming valid responses to all five questions); and if no type of distress was experienced, the index took on its minimum value of 0.
65. The design for the index of material distress closely resembles that for the "index of material hardship," which was used in several random-assignment evaluations of state welfare-reform initiatives in the 1990s (Bloom et al. 2002, Fraker et al. 2002, and Miller et al. 2000). However, two of seven specific types of distress, both reflecting failure to see a health care professional when needed, were omitted from the index in this study. Those two types of distress were instead included in this study's index of health-related distress (Appendix Exhibit B.22).
66. Appendix Exhibit B.23 shows that poverty status differed significantly by employment status, but the mean value of the index of material distress did not in five of the study sites (Ft. Worth, Milwaukee, Nashville, Philadelphia, and Phoenix). Both measures differed significantly by employment status in only four of the sites (Baltimore County, Boston, Chicago, and Yakima).
67. Enrollees who experienced one of the two forms of homelessness often experienced the other form as well. Consequently, in most sites the overall incidence of homelessness was substantially less than the sum of the two rates. The overall rates of homelessness are presented in Appendix Exhibit B.22.
68. Rates of homelessness were not consistently high in sites where rates of participation in government housing programs (receipt of housing subsidies or residence in public housing) were low. Exhibit B.18 shows that, across the 11 study sites, Boston had the highest rate of participation by WtW enrollees in government housing programs, while St. Lucie and West Virginia were among the four sites with the lowest rates of participation in government housing. Yet Exhibit V.9 shows high rates of homelessness in Boston and low rates in St. Lucie County and in West Virginia. On the other hand, enrollees in Milwaukee rarely participated in government housing programs but they had relatively high rates of homelessness.
69. The health insurance coverage rate for children was high over the entire one-year follow-up period, except in three sites — 2 in 10 enrollees in Baltimore County and Phoenix and 3 in 10 enrollees in Ft. Worth had at least one child for whom there was a lapse in health insurance coverage sometime during the follow-up period (Appendix Exhibit B.25).
70. Consistent with their generally high levels of health insurance coverage, WtW enrollees and their families did not experience much health-related distress over the evaluation's one-year follow-up period. The mean value of an index summarizing three types of health-related distress ranged from 0.20 to 0.29 in eight of the study sites (Appendix Exhibit B.22). This was indicative of less than one type of distress, on average. The mean value of the index was below 0.20 in Boston, Chicago, and Philadelphia, where health insurance coverage rates were especially high.
Main Page of Report | Contents of Report
Home Pages:
Human Services Policy
(HSP)
Assistant Secretary for Planning and Evaluation
(ASPE)
U.S. Department of Health and Human Services
(HHS)