[ Main page of Report | Contents of Report ]
Researchers and policymakers are interested not only in which strategies programs use to engage recipients but also in the extent to which the strategies have been successful. There is special interest in the extent to which TANF recipients are engaged in activities beyond those that are captured in the data reported by states to the federal government for purposes of calculating federal participation rates. This chapter presents findings from analyses of administrative data on program participation in two study sites that have management information systems (MIS) that contain substantially more data than used to calculate federal participation rates — El Paso County, Colorado, and Utah.(1) These sites offer the broadest range of activities to the broadest group of TANF recipients and, as such, are not representative of the seven study sites or of TANF programs nationally.
In El Paso County, we conducted analyses of adult recipients on TANF in August 2003. In Utah, we conducted analyses of adult recipients on TANF in May 2003. Both months are typical in each site. The analyses focus on four research questions:
[ Go to Contents ]
In identifying the activities to which recipients are assigned, we looked at the following: how often recipients are assigned to activities that are considered in the federal participation rate calculation relative to those that are not; the range of activities to which recipients are assigned; the activities to which case managers most and least often assign recipients; and the mix of activities to which case managers assign recipients (i.e., how often and in what ways they combine activities). For purposes of this discussion, we grouped all program activities into three categories:
In the two sites where adequate data were available, El Paso County and Utah, the vast majority of all TANF recipients are assigned to participate in program activities (90 percent and 82 percent, respectively).(2) However, in each site, a sizeable portion of the caseload is assigned to activities that are not considered in the federal participation rate calculation — 44 percent in a typical month in El Paso County and 62 percent in a typical month in Utah (see Figure V.1 and V.2). Many clients participate exclusively in nonfederal activities, while others combine nonfederal with federally countable activities. Despite the high number of assignments to nonfederal activities, a substantial portion of the caseload is assigned to at least some activities considered in the federal participation rate, including core and other federally countable activities — just above and just below 60 percent in El Paso County and Utah, respectively. In fact, 46 percent of recipients in El Paso County and 19 percent in Utah are assigned exclusively to federally countable activities, and the majority of these are core activities.
In any given month, some portion of the caseload is not assigned to any program activities — 10 percent in a typical month in El Paso County and 17 to 18 percent in a typical month in Utah. The primary explanation for this is that a substantial portion of unassigned recipients either has recently entered the caseload and therefore has not yet been assigned to activities, or is about to exit the caseload and likely already completed their activities altogether. Excluding these individuals, only a small percentage of recipients — roughly 4 percent in El Paso County and 5 percent in Utah — remain on TANF without being assigned to any activities (see Table V.1). It is possible that these recipients are in transition between activities, are waiting for activity slots to become available, are particularly difficult to engage in activities, or have lost contact with their case managers.
Figure V.1.
Assigned Activities In A Typical Month
(El Paso County)
Figure V.2.
Assigned Activities In A Typical Month
(Utah)
| Recipients With No Activities in | Percentage | |||
|---|---|---|---|---|
| El Paso County | Utah | |||
| Among All Recipients | Among Recipients with No Activities | Among All Recipients | Among Recipients with No Activities | |
| Cases that opened or closed in typical month* | 5.5 | 54.1 | 6.6 | 39.8 |
| Cases that opened in previous month | — | — | 4.2 | 25.3 |
| Cases that closed in following month | 0.2 | 2.5 | 0.6 | 3.8 |
| Other cases | 4.4 | 43.4 | 5.2 | 31.0 |
| Sample Size | 1204 | 122 | 6187 | 1109 |
| * In El Paso County, most cases that "opened" were reinstated — that is, the cases were closed, but reopened within 30 days. Cases that were reinstated may have been reinstated in typical month or before. Data on case openings prior to typical month are not available for El Paso County. | ||||
As noted in Chapter III, the list of activities to which recipients can be assigned is extremely extensive. El Paso County has more than 27 activities, 17 of which are considered in the federal participation rate calculation and the remainder of which are not. Utah has more than 70 activities to which recipients can be assigned, 37 of which are considered in the federal participation rate calculation and the remainder of which are not.
By far, the prevailing activities in which El Paso County and Utah TANF recipients participate are job search/job readiness activities and nonfederal activities. In El Paso County, the largest percentage of recipients assigned to any activities (47 percent) is assigned to job search/job readiness activities followed by nonfederal activities (38 percent). In Utah, the largest percentage is assigned to nonfederal activities (62 percent) followed by job search/job readiness activities (32 percent) (see Table V.2). In both sites, unsubsidized employment is the next most common activity, though only about 14 percent of recipients are assigned to employment.
| El Paso County | Utah | |
|---|---|---|
| Job search / job readiness | 46.9% | 61.9% |
| Nonfederal activities | 38.0 | 31.6 |
| Employment (full-time in El Paso) | 13.8 | 14.5 |
| Case management | 12.0 | — |
| Education with no HS degree | 7.8 | 13.6 |
| Job skills training | 5.8 | 13.5 |
Unlike the El Paso County data, the Utah data contain codes identifying specific activities within the broad category of nonfederal activities. The most common nonfederal activities in Utah are related to issues in three areas: child care — for instance, looking for child care or resolving child care problems — child support enforcement, and physical health treatment (see Table V.3). Other common nonfederal activities include mental health treatment, activities related to other support services — such as life skills activities not considered to be job readiness activities — and pursuing SSI benefits. These two groups of activities account for two-thirds of all nonfederal activities in Utah.
| Working on child care issues | 12.7% |
| Working on child support enforcement | 12.4 |
| Physical health treatment | 12.2 |
| Mental health treatment | 10.3 |
| Working on other support services | 8.3 |
| Other life skills activities | 4.9 |
| Pursuing disability income | 4.8 |
The majority of recipients assigned to activities are assigned to multiple activities (see Table V.4). Some combine multiple activities within the same category — for instance, job search with work experience, which are both core federally countable activities, or domestic violence counseling with resolving child care issues, which are both nonfederal activities; others combine activities across categories — for instance, job search with mental health counseling. Employed recipients combine work with other activities in different ways in the two sites. In El Paso County, employed recipients, particularly part-timers, combine work primarily with job search; relatively few are assigned to nonfederal activities as well. In Utah, however, the majority of employed recipients combine work with nonfederal activities.
| Number of Activities | Percentage | |
|---|---|---|
| El Paso County | Utah | |
| 0 | 9.9 | 16.6 |
| 1 | 42.9 | 13.4 |
| 2 | 31.6 | 23.6 |
| 3 | 11.5 | 23.3 |
| 4 | 4.0 | 14.6 |
| 5+ | 0.1 | 8.4 |
[ Go to Contents ]
Our analysis of assigned and actual hours of participation is based on MIS data. Most states and counties, including El Paso County and Utah, use different methods to record TANF program participation hours in their MIS. In El Paso County, case managers record assigned hours of participation and actual hours of participation in separate fields, but are required to do so only for federally countable activities; as a result, data on hours of participation in nonfederal activities are incomplete or nonexistent in El Paso County's MIS. In Utah, case managers record assigned hours of participation in each and every activity — federally countable and nonfederal activities — in the state MIS and are instructed to overwrite assigned hours with actual hours if the two vary. The extent to which they make this change, however, varies with how often they meet with recipients, their ability to closely track participation, and competing caseload demands. While it is therefore difficult to compare data on or draw conclusions about hours of participation across the two sites, the data do suggest a number of basic findings.
First, recipients are assigned to participate in activities for a substantial number of hours per week but not necessarily the 30 hours required of single parents under the current federal participation rate definition. In El Paso County, recipients are assigned for an average of 24 hours per week to each federally countable, and for an average of 36 hours per week in total across all of these activities (see Table V.5). However, more than one-third of recipients assigned to federally countable activities are assigned for fewer than 30 hours per week. In Utah, recipients are assigned to all types of activities for a combined average of 22 hours per week, and almost two-thirds are assigned to activities — including federally countable and nonfederal activities — for fewer than 30 hours per week.(3)
| Percentage or Average | ||
|---|---|---|
| El Paso County (in federally countable activities only) |
Utah (in all types of activities) |
|
| 1-19 | 12.7% | 39.7% |
| 20-29 | 22.5% | 21.3% |
| 30-39 | 26.0% | 27.8% |
| 40+ | 38.8% | 11.2% |
| Average total | 36.1 | 22.0 |
| Average per activity | 24.4 | 7.9 |
Second, even though a substantial portion of the caseload is assigned to nonfederal activities, these activities account for a relatively small share of a recipient's weekly participation time. In Utah, weekly assigned hours in federally countable activities are substantially higher than weekly assigned hours in nonfederal activities (recall that there are no data on hours in nonfederal activities in El Paso County). On average, recipients in nonfederal activities are assigned to those activities for 4 to 5 hours per week, compared with 24 hours for employment, 21 hours for work experience, 25 hours for on-the-job training, 15 hours for job search, 18 hours for vocational education, 11 hours for education with no high school diploma, and 13 hours for satisfactory school (see Table V.6).
| Average | |
|---|---|
| Employment | 23.6 |
| Work experience | 21.1 |
| On-the-job training | 25.9 |
| Job search | 14.8 |
| Vocational education | 17.9 |
| Job skills training | 11.5 |
| Education with no high school degree | 11.4 |
| Satisfactory school | 12.9 |
| Non-federal activities | 4.5 |
Finally, recipients are actually participating in federally countable activities for the majority of time they are assigned. On average, across all of these activities, TANF recipients in El Paso County actually participate for 70 percent of the time they are assigned (recall that there are no data distinguishing actual and assigned hours of participation in Utah) (see Table V.7). Still, however, about one-quarter of the caseload is still not actively participating in any of the federally countable activities to which they are assigned.
Recipients spend the most time relative to assigned hours in education-related activities and even more time than they are assigned in job skills training and basic education. This may be because job skills training and education courses provide more hours of instruction than required by TANF programs.(4) In contrast, they spend the least amount of time relative to assigned hours in job search and job readiness activities. However, it is often more difficult to track hours of actual participation in job search and job readiness activities for two reasons. First, many of these activities are self-directed, and second, the calculation of hours may correspond not to the actual time devoted to the activity but to the number and types of employer contacts or other accomplishments such as the development of a resume. For instance, one telephone contact may be considered equivalent to one hour of participation, or one face-to-face contact equivalent to four hours of participation. In addition, TANF recipients who find jobs might stop participating in assigned job search/job readiness activities pending the start of their employment. Therefore, the extent to which the data on actual hours in job search and job readiness reflect true levels of participation as opposed to limitations of data collection or definitions is not clear.
| Ratio of Actual to Assigned Hours | |
|---|---|
| Any activity | 0.70 |
| Job search/job readiness | 0.35 |
| Post-secondary education | 0.92 |
| Job skills training | 1.10 |
| Basic education | 1.42 |
| Percent with no actual hours | 24.6% |
[ Go to Contents ]
Most TANF programs strive to build recipients' capacity for employment and self-sufficiency. Though they may offer a range of activities to meet recipients' needs, most programs want recipients to progress from activities that address their barriers and supportive service needs to ones that provide more direct work experience and job search support (and finally, to unsubsidized employment). In fact, PRWORA includes provisions intended to prevent recipients from languishing on the caseload, including time limits on assistance, requirements for states to engage recipients in work (as defined by states) after two years of welfare receipt, and limits on the amount of time job search/job readiness activities and vocational education count toward the federal participation rate.
However, the majority of recipients tend to remain in a given set of activities for several months. Two-thirds of recipients in nonfederal activities in El Paso County remain in those activities for at least three to five months, and on average, recipients in nonfederal activities in Utah remain in those activities for over eight months (see Table V.8).(5) However, recipients may move more often from one activity to the next within the set of all nonfederal activities. For instance, a recipient in nonfederal activities for six months may participate in domestic violence counseling for four months and life skills training for two months. Similarly, two-thirds of recipients in job search and job readiness activities in El Paso County remain in those activities for at least three to five months, and on average, recipients in these activities in Utah remain in them for over five months. In fact, among all adult recipients in El Paso County, almost one-third are assigned to job search/job readiness activities or education activities beyond the maximum time that those activities count toward the federal participation rate (similar data are not available for Utah). However, many of them may be participating in other activities along with extended job search or education; recall that 47 percent in El Paso and 70 percent in Utah are assigned to two or more activities concurrently.
| Percentage in El Paso County | Average in Utah | ||
|---|---|---|---|
| 1-2 months | 3-5 months | — | |
| Nonfederal activities | 35.2% | 64.8% | 8.4 |
| Job search/job readiness | 35.8% | 64.2% | 5.2 |
| Employment | 49.1-53.8% | 46.2-50.9% | 6.9 |
Moreover, an increasing percentage of recipients remaining on TANF is not assigned to any activities at all as time goes on. Among a cohort of TANF recipients in a typical month, the percentage assigned to no activities doubles within five to six months — from 10 to 20 percent in El Paso County within five months and from 18 to 38 percent in Utah within six months (see Figure V.3 and Figure V.4). The proportion of recipients remaining on TANF who are in federally countable or allowable activities shrinks over the same period, though at a somewhat slower pace. These data suggest that it may be difficult to keep recipients engaged in activities over time and that programs may need to develop better strategies for working with those who do not find employment or leave the rolls quickly.
Finally, the majority of recipients who stay on the caseload for five to six months do not, during that time, progress from nonfederal to federally countable activities, or from other federally countable to core federally countable activities. In El Paso County, more than 70 percent of recipients remained in the same category of activities for five months, while 20 percent made forward progress at some point and 9 percent actually moved backward at some point (see Figure V.5). In Utah, 56 percent remained in the same category of activities for six months, while 23 percent made progress at some point and 21 percent moved backward at some point.
Figure 3.
Assigned Activities Over Time Among August 2003 Cases in El Paso
County
Figure 4.
Assigned Activities Over Time Among May 2003 Cases in Utah
[ Go to Contents ]
Differences in engagement among different subgroups of TANF recipients are policy relevant because they can inform decisions about the availability of services and the allocation of resources, especially in programs striving to engage all or most recipients in activities. For instance, if Group A represents one-quarter of the caseload in El Paso County and Group B represents three-quarters of the caseload, and if substantially more recipients in Group A than in Group B are not assigned to any program activities, then it might be more efficient for El Paso County to target additional resources to services and activities that would engage recipients in Group A rather than spreading resources across the caseload as a whole. This section describes the results of subgroup analyses in each site.
El Paso County divides its caseload into hard-to-employ recipients and job-ready recipients, assigning activities according to the needs unique to each group.(6) Highly skilled and trained case managers determine which recipients are hard-to-employ and which are job-ready on the basis of comprehensive individualized assessments conducted at intake. In a typical month, about one-third of the caseload consists of hard-to-employ recipients, and two-thirds consists of job-ready recipients. Results of an analysis comparing hard-to-employ and job-ready cases are shown in Table V.9.
Differences in the types of activities to which hard-to-employ and job-ready recipients are reflected in the data. The percentage of job-ready recipients that is assigned to federally countable activities is almost double the percentage of hard-to-employ recipients assigned to these activities — 80 percent compared to 43 percent. Conversely, the percentage of job-ready recipients that is assigned to nonfederal activities is substantially lower than the percentage of hard-to-employ recipients assigned to these activities — 14 percent compared to 45 percent. In fact, while job search and job readiness are the most common activities among job-ready recipients, nonfederal activities are the most common among hard-to-employ recipients.
Although hard-to-employ and job-ready recipients are assigned to different kinds of activities, they participate in about the same fraction of their assigned hours in federally countable activities. On average, hard-to-employ recipients participate in these activities about 72 percent of the time they are assigned to participate (33 hours per week on average), and job-ready recipients participate in these activities about 70 percent of the time they are assigned to participate (37 hours per week, on average).
Finally, job-ready recipients progress from activities that address their barriers and supportive service needs to activities that provide more direct work experience and job search support more often than do hard-to-employ recipients. At any point in the five-month period over which we tracked recipients, 22 percent of job-ready recipients progressed from nonfederal only to federally countable activities or from other federally countable to core federally countable activities. By comparison, 16 percent of hard-to-employ recipients made forward progress.
| Percentage | ||
|---|---|---|
| Hard-to-Employ | Job-Ready | |
| Types of Assigned Activities | ||
| Some federally countable | 43 | 80 |
| Nonfederal only | 45 | 14 |
| None | 12 | 6 |
| Most Common Assigned Activities | ||
| Job search/job readiness | 23 | 58 |
| Nonfederal | 34 | 14 |
| Full-time employment | 10 | 15 |
| Case management | 11 | 11 |
| Part-time employment | 7 | 14 |
| Average Weekly Assigned Hours in Federally Countable Activities | 33.1 | 37.7 |
| Average Ratio of Actual to Assigned Hours | 0.72 | 0.70 |
| Progress over Time | ||
| Moved forward | 16 | 22 |
| Moved backward | 10 | 9 |
| Did not move | 74 | 69 |
The data for Utah do not allow us to distinguish between hard-to-employ and job-ready recipients, but they do contain information that can be used to make another interesting distinction between recipients in the state — total time on TANF. We divided the caseload in Utah into two groups — long-term recipients, or those who had been on TANF for a total of 36 months or more, and shorter-term recipients, or those who had been on TANF for a total of fewer than 36 months. In a typical month, only 10 percent of the caseload consists of long-term recipients, and 90 percent consists of shorter-term recipients. Results of an analysis comparing long-term and shorter-term recipients are shown in Table V.10.
| Percentage | ||
|---|---|---|
| Long-Term | Shorter-Term | |
| Types of Assigned Activities | ||
| Some federally countable | 54 | 56 |
| Nonfederal only | 29 | 26 |
| None | 17 | 18 |
| Most Common Assigned Activities | ||
| Nonfederal | 63 | 62 |
| Job search/job readiness | 30 | 32 |
| Employment | 13 | 15 |
| Education with no high school degree | 12 | 14 |
| Job skills training | 12 | 14 |
| Average Weekly Assigned Hours in Activities | 18.2 | 18.3 |
| Progress Over Time | ||
| Moved forward | 23 | 23 |
| Moved backward | 21 | 21 |
| Did not move | 56 | 56 |
There are few, if any, differences between long-term and shorter-term recipients in terms of the types of activities and number of hours to which they are assigned and progress to activities that increasingly build their capacity to work and become self-sufficient. Slightly more than half of the recipients in each group are assigned to at least some federally countable activities in a typical month, and one-quarter to one-third are assigned to nonfederal activities only. In both groups, the most common activities are nonfederal activities, followed by job search or job readiness activities, which are federally countable. Both groups are assigned to participate in activities for an average of slightly more than 18 hours per week in total (including recipients with no assigned hours). In each group, 23 percent of those who were on TANF in all of the six months we tracked progressed from nonfederal only to federally countable activities or from other federally countable to core federally countable activities at some point during that period.
[ Go to Contents ]
In the two study sites that offer the broadest range of activities to all TANF recipients, virtually all recipients are included in the denominator of the federal participation rate, but the majority of recipients are not included in the numerator.(7) As noted previously, the federal TANF legislation specifies 12 activities in which TANF recipients must participate for at least 30 hours per week to be included in the numerator of the federal participation rate. At least 20 of those hours must be spent in at least one of nine core countable activities and up to 10 hours may be spent in three other countable activities. However, administrative data in El Paso County and Utah reveal that a substantial percentage of the caseload in those locales is in activities other than the 12 specified in the legislation — nonfederal activities — or participates in activities for fewer than 30 hours per week. While 95 percent of the caseload in El Paso County and 97 percent of the caseload in Utah is included in the denominator of the federal participation rate, only 20 and 24 percent, respectively, is included in the numerator (see Figure V.6).(8)
Reasons for relatively low federal participation rates extend beyond frequent assignment to nonfederal activities. Many recipients are not included in the numerator because they are participating in specified activities for fewer hours than PRWORA requires. While one-third of recipients not included in the numerator of the federal participation rate are assigned only to nonfederal activities, 53 percent in El Paso County and 46 percent in Utah are assigned to at least some countable activities (see Figure V.7). Among recipients assigned only to core countable activities, only one-third in El Paso County and one-half in Utah are included in the numerator.
Figure V.6.
Federal Participant Rate Status among All Tanf Recipients in El Paso County
and Utah
Figure V.7.
Types of Recipients Excluded from Numerator in El Paso County and
Utah
[ Go to Contents ]
The analysis of administrative data on program participation in El Paso County, Colorado, and Utah reveals that more TANF recipients are engaged in activities than federal participation rates suggest. A substantial portion of the caseload is engaged only in nonfederal activities or combines nonfederal with federally countable activities. Currently the vast majority of these recipients are not counted in the numerator of the federal participation rate calculation either because they are not engaged in one of the 12 activities considered in the calculation or because they are not engaged in one or more of those 12 activities for the minimum number of hours required in the calculation. If these recipients were included in the numerator, it is likely that federal participation would be much higher than they are now.
In addition, TANF recipients seem to be engaged in activities for a substantial number of hours per week, though not necessarily for the 30 hours required of single parents under the federal law. Recipients actually participate in federally countable activities for the majority of time they are assigned, and even though a substantial portion of the caseload is assigned to nonfederal activities, these activities account for a relatively small proportion of recipients' weekly activity time.
Finally, it is difficult to keep TANF recipients engaged in activities over time, and progress up the activity ladder is slow. The longer a recipient remains on TANF, the more likely she is to receive assistance without being assigned to participate in any program activities. And recipients who are assigned to program activities tend to remain in the same kinds of activities for extended periods of time — including activities which count toward the federal participation rate for only limited periods of time, such as job search or job readiness activities — without progressing to activities that are higher on the list of federal priorities. This evidence suggests that sites striving to engage all or most TANF recipients in work or work-related activities may have to demonstrate patience and a willingness to "stay the course" with recipients who move relatively slowly toward self-sufficiency or develop improved strategies for addressing their needs.
[ Go to Contents ]
(1) A description of the administrative data in each site appears in Appendix B.
(2) Data supporting all findings and statistics referenced in the report may be found in the tables in Appendix C, regardless of whether the findings and statistics are presented in tables or figures throughout the body of the report. Some of the tables in Appendix C present more results than are discussed in the body of the report.
(3) We assume that the data on hours of participation in Utah more accurately reflect assigned hours than scheduled hours given the inconsistency with which case managers overwrite assigned hours with actual hours of participation.
(4) Note that analyses do not include recipients who participated in an activity to which they were not assigned.
(5) In El Paso County, data on number of months in activities are based on monthly confirmations of activity assignments. In Utah, data on number of months in activities are based on activity start and end dates.
(6) Hard-to-employ recipients receive case management services from county staff and job-ready recipients receive case management services from staff at a contracted service provider. For each TANF recipient, the data in El Paso County's MIS distinguishes the type of staff — county or contracted — providing the case management services.
(7) In El Paso County, all recipients should be included in the federal participation rate calculation, but they are exempt from participation requirements if they are disabled, caring for a severely disabled child, or under a federally recognized good cause domestic violence waiver. In Utah, all recipients should be included in the calculation except those subject to sanction for no more than three months within the preceding twelve months, and those who are disabled are exempt from participation requirements.
(8) Recall that analyses are for adult recipients only; child-only cases are not subject to participation requirements and are also excluded from the denominator of the participation rate calculation.
Top of Page | Contents
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)