Jacqueline Kauff and Michelle K. Derr
Mathematica Policy Research
About This Project and Brief
This report is available on the Internet at:
Printer Friendly Version in PDF Format (16 pages)
How to Obtain a Printed Copy
Suggested Further Readings
Both Maryland and Utah have more than doubled their state Temporary Assistance for Needy Families (TANF) work participation rates within a one-year period. Maryland's rate increased from 20.5 percent in fiscal year (FY) 2005 to 44.5 percent in FY 2006, and Utah's rate increased from 16 percent in October 2006 to 45.5 percent in October 2007. They accomplished these results through a combination of strategies aimed at increasing the number of TANF recipients engaged in federally countable program activities for the requisite number of hours per week. Both adopted a philosophy of universal engagement, a performance monitoring system, work experience and subsidized employment programs, and solely state-funded programs for subgroups of recipients. In addition, Maryland uses designated staff to help counties increase their participation rates and pay-for-performance contracts with employment service vendors, while Utah uses a transitional benefits program for those who become employed and comprehensive sanction policies and procedures. This brief describes each of the strategies these states implemented.
Recent legislation has increased pressure on states to engage more TANF recipients in work or work-related activities. States that do not engage a minimum percentage of welfare recipients in specified activities for a specified number of hours per week face financial penalties. Some states have attempted to increase their work participation rate by closing cases that are not meeting or are not likely to meet their work requirements, thereby reducing the size of the state's caseloads. Others have implemented new service strategies to increase the likelihood that recipients on the caseload can participate in the required activities for the requisite number of hours.
Maryland and Utah are two examples of states that have taken a multipronged approach to increasing their state's work participation rates, and both have more than doubled their rates within one year of implementing most of their reforms. Both states adopted a universal engagement philosophy; implemented a system that uses program data to monitor program performance and increase participation in work activities; increased their use of work experience and subsidized employment programs; and created solely state-funded programs for subgroups of recipients. In addition, each implemented several strategies not adopted by the other state. Maryland hired designated staff to help counties increase their participation rates, and several larger counties implemented pay-for-performance contracts with their employment service vendors. Utah implemented a transitional benefits program that continues to pay a cash grant to people who have moved into competitive employment, and has overhauled its sanction policies and procedures.
This practice brief based on data collected during telephone interviews with state program administrators, in-person site visits to three localities in Maryland (Montgomery County, Wicomico County, and Baltimore City), and in-person site visits to Salt Lake City in Utah describes each state's strategy. It highlights the experiences and successes of two states that adopted a variety of strategies in their efforts to achieve their work participation goals.
[ Go to Contents ]
Historically, the TANF work participation rates in Maryland and Utah have been low. In the years prior to the Deficit Reduction Act of 2005 (DRA), the rates in both states were substantially below the national average, which remained relatively constant at 32-34 percent between FY 2000 and FY 2004. Based upon data submitted to the Department of Health and Human Services (HHS) Administration for Children and Families (ACF), Maryland's rate ranged from 6 percent in FY 2000 to 16 percent in FY 2004 (see Figure 1). The rate the state was required to meet also was low, however, largely because the state received a substantial credit for caseload declines from FY 1995 levels. Thus, despite its low participation rate, the state was able to meet federal requirements each year. Based on data submitted to ACF, the rate in Utah also was lower than the national average in the years prior to the DRA, but was much closer to it than was Maryland, ranging from 26 to 31 percent between 2000 and 2004 (see Figure 2). These data from Utah were, however, based on state reporting practices that were not consistent with the new requirements created under the DRA. In October 2006, Utah changed its management information system practices so that the work participation rate calculation was based on the DRA reporting requirements. If the state had used these new reporting practices earlier, Utah's participation rate likely would have been half of what was reported for 2000 to 2004.
TANF Work Participation Rate Maryland and U.S.
Note: The required rate is the adjusted work participation rate, based on the state's caseload reduction credit.
TANF Work Participation Rate Utah and US
Notes: Utah actual data based on new reporting system are for November 2006 and November 2007.
These data are based on preliminary state calculations and may differ from the official ACF participation data when they become available.
The required rate is the adjusted work participation rate, based on the state's caseload reduction credit.
The debate leading up to the reauthorization of the TANF program led administrators in both states to conclude that they needed to substantially increase their work participation rates. Administrators in Maryland anticipated that welfare reauthorization legislation would make changes to the participation rate calculation which would make it difficult for the state to meet required rates going forward without substantial TANF policy, programmatic, or administrative changes. Thus, in FY 2004, in anticipation of the DRA, Maryland set into motion a series of new efforts designed to increase the state's work participation rate.
Similarly, anticipated changes concerning TANF's reauthorization coupled with concerns about their dwindling caseload reduction credit prompted state and regional TANF administrators in Utah to rethink the state's approach to work participation requirements. Just prior to and shortly after the DRA, administrators instituted an aggressive statewide planning effort to revise the TANF program and created a steering committee and nine subcommittees to explore strategies for increasing the work participation rate. This planning effort led to the implementation of several policy and procedural changes.
The policy, programmatic, and administrative changes each state made have met with success. Maryland's work participation rate more than doubled between FY 2003 and FY 2005 (from 9.1 to 20.5 percent) and then again between FY 2005 and FY 2006 (from 20.5 to 44.5 percent). Utah's participation rate increased from 30.3 percent in FY 2005 to 42.5 percent in FY 2006 based on the state's old data reporting system, and from 16 percent in October 2006 to 45.5 percent in October 2007 based on the new system.
[ Go to Contents ]
While Utah had begun experimenting with a universal engagement model as early as 1993, Maryland implemented such a model in September 2003 in anticipation of TANF reauthorization. The universal engagement model views self-sufficiency as a continuum. The concept is that everybody can engage in some type of activity for some period of time, regardless of his or her personal or family circumstances. Recipients who can engage in 40 hours of federally countable activities should do so, but those who cannot meet that standard because of physical, mental, or intellectual limitations should be expected to do what they can. Some of their activities might not be related to work directly, but are necessary in order to prepare for work in the near future. For instance, a recipient who is temporarily unable to work due to a medical incapacity might be expected to spend one hour per week attending medical appointments or rehabilitative treatment, while a recipient who is temporarily unable to work because of a recent birth might be expected to spend time caring for her newborn and researching child care options. Thus, in Maryland and Utah, universal engagement does not necessarily mean that every recipient is engaged in federally countable activities. Tasks such as attending medical appointments or caring for a young child are considered state-defined activities that do not count toward the federal work participation rate calculation.
The theory behind the universal engagement model is that engaging recipients in any activity and progressively increasing the number of hours engaged eventually helps to move recipients into federally countable activities for the federally required hours. Embedded within the theory are two key precepts: (1) in order to meet the federal work requirements, some TANF recipients first need to focus on removing personal barriers to employment, and participation in activities focused on mitigating barriers will prepare recipients to take part in countable work activities in the future; and (2) participation begets more participation; that is, once a recipient starts participating in an activity at least minimally, the number of hours the recipient spends in activities in general, and countable activities in particular, is likely to increase over time. Thus, universal engagement is a means to achieving the federal work participation requirement.
Maryland and Utah implement the universal engagement model somewhat differently. Utah provides no exemptions from the work requirements; all recipients are subject to the same requirements and expected to participate to their maximum ability. The state holds program staff accountable to one participation measure the federal work participation rate and expects counties to achieve a 50 percent rate (and each case manager to achieve a 50 percent rate for their caseload).
Maryland, on the other hand, distinguishes its universal engagement requirement from its work participation requirement. All TANF recipients are subject to the state's universal engagement requirement that is, they are required to participate in at least one federally countable or state-defined activity for at least one hour per week. Only some recipients, however, are subject to its work participation requirement that is, single-parents are required to participate in activities for at least 40 hours per week (30 of which must be federally countable core activities if they do not have children under age 6, and 20 of which must be federally countable core activities if they do); adults in two-parent families in which one parent is providing child care to their own child must participate in activities for a combined total of 35 hours per week (30 of which must be in federally countable core activities); and adults in two-parent families that receive federally-funded child care assistance must participate in activities for a combined total of 55 hours per week (50 of which must be in federally countable core activities). Maryland exempts recipients from this requirement if they are (1) caring for a child under one year of age (limited to 12 months in a lifetime); (2) a parent caring for an ill or disabled household member who does not attend school full time; (3) long-term disabled (that is, disabled for 12 months or more) and applying for or receiving Supplemental Security Income (SSI); (4) subject to a work-related sanction for not more than 3 months in a 12-month period; (5) a needy caretaker relative not receiving assistance; or (6) under age 16 or age 16-18 and in school full time. While Maryland's work participation rate goal is 50 percent of all families, its universal engagement rate goal is 100 percent.
To help achieve the universal engagement goal, both states require all TANF applicants to engage in federally countable or state-defined activities. In Utah, state TANF administrators implemented a statewide requirement for completion of a standardized program orientation during which applicants are informed about the program requirements, available resources to help them meet those requirements, and consequences of nonparticipation. Before financial benefits are issued, applicants also must review and sign a participation agreement that specifies the activities in which they will participate as well as their commitment to participate. In the counties we visited in Maryland, required activities during application included a TANF program orientation, job search, a job readiness workshop, skills training, and/or work experience. County staff responsible for eligibility determinations could not open a TANF case until receiving confirmation that the applicant had a personal responsibility plan (similar to the participation agreement in Utah), was enrolled in an activity, and was adhering to the plan. In both states the applicant requirements serve the purpose of reinforcing the message of mandatory participation and work, thus helping to ensure that recipients enter the rolls knowing how and prepared to comply. In Maryland, making the development of, and cooperation with, a personal responsibility plan a condition of TANF eligibility also ensures that all applicants who enter the rolls do so already engaged in activities. In state fiscal year (SFY) 2007, for example, about 22 percent of TANF applicants in Montgomery County and 45 percent in Wicomico County actually became recipients and virtually all who did (99 percent in each county) entered TANF already engaged. While there are no data on the percent of applicants who do not become TANF recipients because of employment or because of failure to cooperate with work requirements, more than one-quarter of total TANF job placements in Montgomery County in SFY 2007 were attributable to applicants, although these did not count toward the work participation rate.
[ Go to Contents ]
In anticipation of TANF reauthorization legislation, both Maryland and Utah began using data more effectively to monitor and improve program performance. This strategy did not require the development of new data collection systems but did require an integrated method of using and reporting existing data. Both states developed new data reports to accomplish three specific objectives: (1) to clarify program expectations for staff; (2) to hold staff accountable for their performance and contribution (or lack thereof) toward participation goals; and (3) to help staff at all levels (case management, supervisory, and administrative) identify problems quickly and brainstorm ways to improve the participation rate. Maryland's tools and procedures are based on a system developed in New York City. New York City's system and Utah's system both are described in detail in Brief #2 of this series, "Using Data to Monitor and Improve the Work Participation of TANF Recipients: Examples from New York City and Utah."
Utah uses two data reports to track case management activity and monitor clients' activities as well as clients' progress in meeting participation requirements. Both were derived from the state's Your Online Data Access system, which contains detailed client- and program-level information for a variety of programs (including the TANF, food stamp, and Workforce Investment Act programs).
Case managers and supervisors in Utah use the Case Management Customer Report (CMCR) to identify potential participation problems. The report provides a real-time snapshot of the assigned activities and hours for each TANF recipient, by case manager. The CMCR summarizes recent case manager activity for each case and identifies recipients not meeting participation requirements. In addition, the report includes warning flags on recipients whose assigned hours are insufficient to meet participation requirements or who have been assigned to job search activities for more than the maximum number of countable weeks in the current year.
The Participation Report provides information on the participation rate at all levels, from the state to individual case managers. The report is formatted like the dashboard of a car, with one gauge in the state-level report for each region of the state. The four dials on each gauge represent different participation rate calculations: (1) the year-to-date rate based on planned hours, (2) the year-to-date rate based on verified hours, (3) the current monthly rate based on verified hours, and (4) the rate for the remainder of the year that would be needed to meet the required 50 percent annual rate. If a user selects a region, the report displays a set of dashboard gauges for each office within the region. Selecting an office provides a set of dashboard gauges for each team of case managers, and selecting a team will display gauges for each case manager. At the case manager level, the Participation Report provides easy access to detailed participation information for each recipient and complements the CMCR by tracking both actual and planned activities and hours.
In April 2004, Maryland began using the JobStat system, a management system designed to monitor each county's performance relative to the state's performance, particularly with respect to the state's work participation and universal engagement goals. The heart of the JobStat system is a monthly report that summarizes monthly and year-to-date outcomes on key measures including the timeliness of application processing, benefit payment accuracy, fiscal spending, and "client independence" (including the caseload reduction rate; the employment placement, job retention, and earnings gain rates; and the universal engagement and work participation rates). Much like a school report card, the report scores counties on their performance on each indicator and gives a combined score as well. The report summarizes the data in table format on one page but is supplemented with additional graphic displays of the data.
The state developed the template for the JobStat report and produces it each month using data from various state management information systems. Virtually all of the statistics in the report exist in data systems or in other data reports; JobStat simply makes existing data more accessible. Producing the JobStat report requires no effort on the part of the counties. Reports are available on the state TANF agency's website and counties merely are required to download, print, and distribute them to relevant staff each month. A separate report exists for each county in the state and any county may access any report.
The JobStat system incorporates a formal process to review the report and discuss strategies for improvement. Local managers are required to meet monthly with state staff to discuss their performance on each JobStat indicator. The meetings are intended to celebrate accomplishments and encourage local staff to maintain positive results, to hold local staff accountable for shortcomings in their performance, and to consider and subsequently address factors that may be impeding their performance. The objective is not to criticize local managers, but rather to judge them on their awareness of potential problems and their strategies to improve conditions. The team approach creates a forum for analyzing problems by combining the talents of administrators and line managers and helps local staff to make informed decisions. State staff can offer suggestions to local staff based on experiences in other counties or best practices in other states. An Action Item Report identifies tasks to be accomplished by the county going forward, persons who are responsible for each task, deadlines for each task, and results from these team meetings.
While some counties like Montgomery and Wicomico were monitoring outcomes on their own prior to JobStat, the system has helped them to examine outcomes that they could not measure before because of lack of access to either the appropriate data or staff resources. JobStat also has served as a quality control check of their pre-existing efforts to monitor out-comes and as a means for putting local outcomes into context (by comparing local performance to the performance of other counties and to the state as a whole). In addition, the report card aspect results in a process of continuous improvement by encouraging local staff to ask why they did not receive the maximum possible score and to devise strategies for increasing their score the following month.
Baltimore City uses a more detailed management information system, called CenterStat, to monitor its performance. Like JobStat, the system is comprised of monthly reports and meetings, but the outcome measures are more refined and specific than those in JobStat. For instance, while JobStat reports the universal engagement and work participation rates, CenterStat provides all of the elements that feed into these calculations. These elements include the total number of TANF cases; the percentage that were engaged under the universal engagement provision (and among those, the percentage that were engaged in federally countable work activities, the percentage that were engaged in state-defined activities, and the percentage whose cases were closed in the month); the percentage of cases subject to and exempt from the state's work participation rate requirement; the percentage that were meeting their federal work requirements; and, among those that weren't, the percentage that were not participating at all, the percentages that were participating in activities for one to 20 hours per week, 21 to 24 hours per week, 25 to 29 hours per week, and for 30 or more hours per week but not in enough federally countable activities. While JobStat provides data for the city as a whole, CenterStat provides data for each of the city's eight local service centers.
Baltimore City began using CenterStat reports in January 2004 and began conducting monthly meetings with each local service center in September 2005. Center managers participate in each meeting and other staff participate on a rotating basis. Involving front-line staff in these meetings focuses staff at all levels on the same goals. Because case managers have an essential role in providing the supports and services to TANF recipients that can enable their participation in work and work-related activities, it is critical that case managers understand the connection between the work they do everyday and the overall participation rate. The monthly CenterStat reports and employment case management meetings contribute to this goal.
[ Go to Contents ]
Both Maryland and Utah increased their use of programs that provide work opportunities for TANF recipients that can count toward meeting federal requirements specifically, work experience and subsidized employment programs. In both states, administrators provided guidance to localities in making more extensive use of work experience and provided additional funding that enabled the creation of new subsidized employment programs.
Work Experience Programs. In 2005, Maryland began encouraging counties to expand the use of work experience as a means toward achieving universal engagement and meeting federal work participation rates. The state issued several memoranda providing technical assistance to counties to ensure that work placements are responsive to the Fair Labor Standards Act requirements while maximizing the hours recipients may spend in work experience. In response, some counties created new work experience programs and others increased the number of available work experience slots in existing programs by partnering with additional employers for placement sites. For example, Baltimore City has had a long-term partnership with the Mayor's Office of Employment Development for work experience positions, but in recent years also has conducted an extensive marketing campaign to develop new work-site agreements with other public and private for-profit and non-profit employers. The unit of job developers from the city's Department of Social Services called itself the Baltimore Employment Exchange (BEE) to present a more business-like, rather than a social services, image to employers. Currently, the city has worksite agreements with 50 to 60 employers, the majority of which are in private for-profit organizations. Between 200 and 250 TANF recipients are in BEE work experience positions such as clerical, hospitality, or custodial positions at any point in time.
In addition to expanding the number of work experience slots, some counties in Maryland have used work experience in unique ways. Wicomico County provides a good example. First, the county contracted with a vendor to provide 15 drop-in slots in addition to 30 traditional slots. Drop-in slots are intended for use by clients who are waiting for another program activity to begin and who otherwise would be inactive during that time. A client who is slated to begin a training program in two weeks, for example, can participate in a work experience position in the interim, and thereby continue to count toward the state participation requirements (and the federal requirement if the client participates for sufficient hours) without using up time in time-limited activities such as job search. Second, the county uses work experience to round out hours that clients spend in other activities such as GED classes or training programs. Clients who participate in 20 hours of education or training can supplement these hours with work experience so that they may be counted toward the federal work participation rate (assuming that they can be required to work at least 10 hours based on Fair Labor Standards Act criteria). And third, the county uses work experience during the TANF application process so that those who become recipients enter the rolls already engaged in a non-time-limited federally countable activity.
In Utah, the state TANF administrative entity shifted responsibility for developing work experience sites from employment case managers, who had few business contacts and were not trained to do employer outreach, to existing business consultants. Business consultants are state employees located in workforce centers whose roles are consistent with those of a job developer. In addition, the state encouraged case managers to refer more recipients to this activity. In the past about six percent of recipients were assigned to work experience; now that number is about one-third. To ensure there were enough slots for all of these clients, the state began encouraging their education and training providers to offer work experience in addition to education and training services.
Subsidized Supported Employment Programs. Beginning in SFY 2006, Maryland allocated funding to counties to design subsidized employment programs as an additional means of moving toward universal engagement and meeting the federal work participation rate. The state created a standardized contract for counties to use with employers that established the responsibilities and scope of services to be provided by the counties and the employers. It required that employers hire TANF recipients for a minimum of 30 and a maximum of 40 hours per week. It also specified that the counties would reimburse employers 100 percent of gross salary costs for each of the first two 4-week periods of the participant's employment, 50 percent for the third 4-week period, and 25 percent for the fourth 4-week period (up to $8,000 for each participant). If an employer permanently hires the client and the client completes a fifth 4-week period of employment, the employer would be reimbursed by the county for the balance of the gross salary costs for the third and fourth 4-week periods (i.e., 125 percent of gross costs for a 4-week period, derived by adding 50 percent from the third 4-week period and 75 percent from the fourth 4-week period).
Baltimore City is one of only a few counties in the state that has taken advantage of the additional funding to create a subsidized employment program. Since the early 1990s, the city has been operating a grant diversion program that reimburses an employer for up to $300 of total wage costs per month. The money was intended to cover the costs of training as an incentive for employers to hire TANF recipients who may not be completely work ready or do not have the requisite skills for a particular job. Baltimore City took advantage of the new funding from the state to build on its successes with the grant diversion program and expand its capacity to provide subsidized employment opportunities for its recipients. Only recipients with recent and/or substantial work history may participate in the program. Currently, Baltimore City has subsidized employment agreements with approximately 35 employers.
Utah allocated $1 million to the local regions to develop innovative strategies to increase work participation rates. Two of the five regions used the funds to create supported employment programs for TANF recipients with mental health disabilities. For example, the northern region began referring TANF recipients to the Diversified Employment Opportunity (DEO) program operated by a large county mental health treatment provider, Davis Behavioral Health (DBH). DBH hires TANF recipients living with mental health disabilities as permanent employees and pays them $6.50 per hour to perform jobs within the agency such as janitorial work, food service, clerical, or landscaping. Work hours are based on clients' mental health status. Typically, clients start out working a small number of hours and gradually increase them as their conditions improve and their capacity to work increases. While recipients' wages are paid by DBH, supportive services such as child care are paid for with TANF funds. Currently, the program is small, but plans exist to replicate it in other parts of the region and another program based on the DEO model is in the works.
[ Go to Contents ]
Both Maryland and Utah recently created new, solely state-funded programs. These programs help the states avoid fiscal penalties for failing to meet work participation rate requirements by removing families from the TANF caseload that they believe have very little chance of meeting their work participation requirements either individually or as a group. Both states pay assistance to families in the program with state general funds that are not counted toward meeting their Maintenance of Effort (MOE) requirement. Maryland's program has been in operation since 2003 using MOE funds, but was converted to a solely state-funded program in 2006. Utah created its program in 2006.
Both Maryland and Utah created solely state funded programs for two-parent families. Both maintain the same work requirements for these families that were in place through each state's TANF program; in Utah, parents have to participate for a total of 60 hours per week and in Maryland, for a total of 35 hours (30 of which must be in federally countable core activities) unless the family receives federally funded child care in which case the requirement increases to 55 hours per week (50 of which must be in federally countable core activities). One major difference between the two programs is that in Maryland, families receive the same standard cash assistance grant they would have received under TANF, while in Utah clients receive a grant based on the hours they participate following a pay-for-performance model.
In addition to its two-parent program, Maryland created a solely state-funded program targeted to TANF recipients who are living with a disability expected to last longer than 12 months and who are applying for SSI. By state law, TANF applicants or recipients living with a long-term disability cannot be required to participate in work activities. They are, however, required to submit an application for SSI and cooperate with the Disability Entitlement Advocacy Program (DEAP), which was created to help individuals with disabilities through the SSI application process. Program participants are required to authorize the Social Security Administration to reimburse the Maryland Department of Human Resources for any benefits they receive during the application period. If recipients fail to submit an SSI application or otherwise do not cooperate with the DEAP program, they are required to participate in work activities and are subject to the same sanctions as all other TANF recipients. Administrators and program guides instruct case managers and vendors to be very careful when referring a recipient with a long-term disability to a work activity. Maryland's TANF manual, developed to provide assistance to program staff in understanding and implementing all aspects of the state's TANF work participation and education requirements, instructs staff to "evaluate the program available to ensure that the customer is in a program that he or she can succeed at and gain employment skills and knowledge" before referring a recipient with a long-term disability to an educational or vocational program. In so doing, the state hopes to ensure that (1) customers are not denied SSI because they are employed, (2) customers are not set up for failure by being assigned to activities they cannot do, and (3) customers are not "warehoused" by being assigned to an activity that does not move them toward independence.
[ Go to Contents ]
Hiring Work Participation Specialists. In SFY 2005, Maryland created a new position within the state TANF agency called a Work Participation Specialist. According to the state's official job description, "The main purpose of this position is to assist local department administration and case management staff in achieving the 50 percent federal work participation rate and maintaining the 100 percent universal engagement rate. Critical to this purpose are the monitoring and tracking of the local department's success in having TANF recipients participate in work activities, identifying the systemic and individual reasons why participation is not occurring and the recommendation to senior management of effective strategies (such as operational changes and/or procedures) to ensure an ongoing work participation rate of 50 percent or greater." The Governor's supplemental budget for that year provided for 24 such positions to be placed in local offices throughout the state. While there are 24 counties in the state of Maryland, some larger counties have multiple specialists while some specialists are responsible for multiple smaller counties. Requirements for the position include experience with TANF and a college degree.
Work Participation Specialists use several strategies and sources of information to accomplish their goal of ensuring that counties meet state and federal participation requirements. First, they review primary data and data reports from management information systems to examine participation patterns and identify clients who are not participating to the maximum extent possible or in accordance with their personal responsibility plans. Second, they regularly interact with program staff (including county or city staff and staff from contracted service providers) through formal case reviews as well as informal communication to discuss and address participation issues on a case-by-case basis, as well as more broadly. And third, they meet directly with clients to identify and attempt to resolve barriers to participation or, alternatively, to suggest amendments to the personal responsibility plans that include more appropriate activities or hour requirements.
Shifting to Performance-Based Contracts. Many counties in Maryland rely on vendors that is private non- or for-profit agencies in the community to provide services to TANF applicants and recipients. In July 2004, Montgomery County became the first county in the state (and the County Department of Health and Human Services became the first agency in the county) to shift from cost reimbursement contracts with vendors to pay-for-performance contracts. Baltimore City did so in February 2006.
In the pay-for-performance structure, vendors are paid a set amount for each client who accomplishes specified milestones. Vendors are reimbursed based on a system of payment points designed to motivate the vendors to focus on achieving desired outcomes for each recipient. For example, there are three types of payment points in Baltimore City's contracts: (1) for each client engaged in 30 hours of activities per week (and at least 20 in federally countable core activities) throughout the month; (2) for each client who remains in an unsubsidized job paying minimum wage or more for at least 30 hours per week for four consecutive weeks; and (3) for each client who remains employed for 17 weeks within a 7-month period and for whom the vendor achieved the second payment point. Because vendors do not receive payment for providing services to clients who do not meet the federal work participation requirements, they have an incentive to place recipients into countable work activities for the requisite hours as quickly as possible.
Montgomery County uses a system of graduated payment points to motivate its one employment services vendor to focus on encouraging recipients to achieve maximum participation. The contract specifies a series of milestones for clients to accomplish and the closer the milestone is to the goal of meeting the federal work participation rate requirement, the higher the vendor's payment. For example, in November 2004, the state instructed counties to require recipients not exempt from the federal participation rate requirement to participate in activities for 40 hours per week. The hope was that even if clients missed some hours of activities because of circumstances such as seeing a doctor or caring for sick children, they could still meet the minimum federally acceptable level of participation of 30 hours per week (or 20 hours per week for single parents with a child under age 6). To encourage the vendor to focus on the 40-hour requirement, Montgomery County pays its vendor the highest amount per client ($500) for those who participate for 40 hours per week throughout the month. For each client the vendor engages in less than 40 hours per week (even if the client meets the federal work participation rate requirement by participating for 30 hours per week), the county pays less than half that amount $205 per client.
In addition to payment points, bonuses can keep vendors and clients focused on meeting state and federal participation requirements. In Montgomery County, for instance, if the county achieves a greater than 50 percent federal work participation rate, the vendor receives a bonus equivalent to the difference between the $205 payment and the $500 payment for each client who is counted toward the federal work participation rate but participating for less than 40 hours per week. If a client participates for 40 hours per week for an entire month and is enrolled in a core activity that counts toward the federal work participation rate calculation, that client receives a bonus of $100. In addition, the first time the client becomes employed, the client is paid a bonus of $150. The vendor issues client bonuses and is reimbursed by the county.
While Baltimore City's pay-for-performance contracts do not use bonuses, they use other tools to encourage vendors to reach specified participation goals. The contracts establish only one goal for vendors that at least 70 percent of their clients must be counted toward the federal work participation rate. While there are no additional financial incentives provided for reaching this goal, vendors that fail to reach the goal must write a corrective action plan identifying how they will succeed in the future. The goal of 70 percent is substantially higher than the federal work participation rate required of states (50 percent of all families) because only TANF recipients who are not exempt from the federal work participation rate requirement are referred to a vendor. Together, these recipients must achieve a participation rate that is high enough to compensate for the low participation rate among other TANF recipients who are exempt from the work requirement and never referred to a vendor but are still included in the federal work participation rate calculation.
[ Go to Contents ]
Providing Income Supplements to Working Families. In May 2007, Utah implemented a transitional benefits program that continues to pay a cash grant for a period of three months to those recipients who have moved into unsubsidized employment and become ineligible for TANF benefits because their earnings are too high. The transitional grant encourages job retention and allows those who are no longer eligible for TANF due to earnings to continue to count toward the work participation rate as long as they receive the grant.
Eligible individuals consist of those who (1) have received TANF benefits for at least one month; (2) are employed for an average of 30 hours a week per month; and (3) have verified their income and hours in the month following the closure of their TANF case. Self-employed individuals are eligible as long as their earnings average the equivalent of 30 hours per week for the month at the federal minimum wage. Two-parent families also are eligible as long as each parent meets the 30-hour work requirement.
Eligible families receive their full TANF benefit ($474 for a family of three) for two months and 50 percent of their benefit in the third month. Clients are eligible for transitional benefits once in a 24-month period. The state uses MOE funds for the program so that the months of cash assistance do not count toward the federal TANF or Utah's lifetime limit on welfare cash assistance receipt. Families that receive a transitional benefit also may receive 24 months of case management and 6 months of transitional child care, supported with MOE funds. They also may be eligible for discretionary grants to cover specific work expenses (such as uniforms, tools, or car repairs).
Revising Sanction Policies and Procedures. In 2005, Utah overhauled its sanction policies and procedures to implement sanctions more quickly and to increase the penalty for repeat sanctions. Specifically, the state streamlined the conciliation process, required a minimum of two weeks of participation before a sanction could be cured, increased the wait time before a second or subsequent sanction could be cured, and shifted from a gradual to an immediate full-family sanction for second and subsequent sanctions. (In the past, all sanctions were subject to a $100 grant reduction for two months followed by cessation of cash assistance for continued noncompliance; now, the first sanction results in a $100 grant reduction for one month followed by cessation of cash assistance, and all subsequent sanctions result in immediate cessation of cash assistance.)
As a protection for clients, Utah also implemented additional safeguards and supports to increase the likelihood that a sanction is implemented appropriately. Specifically, the state is making greater use of in-house licensed clinical therapists stationed in employment centers across the state. These therapists have always provided social work services to clients, but now are being incorporated into the sanction conciliation process. At a minimum, they are required to serve as a consultant to the case manager, recommending resources or approaches for dealing with difficult cases. In addition, they are required to be involved in formal case reviews and conferences and may conduct in-depth clinical assessments, short-term crisis counseling, or clinical case management for clients who appear to have mental health, substance abuse, or domestic violence issues. Moreover, licensed clinical therapists now are being used more extensively to help determine how to keep TANF recipients progressing and counting toward the participation rate.
It is impossible to determine what effect each of these strategies has had on each state's participation rate. First, it is difficult to isolate the effect of one particular strategy from another that was implemented around the same time and from other variables. Second, the increases in the participation rates in Maryland and Utah did not occur suddenly in response to a particular policy, programmatic or administrative change, but rather changed steadily over time. If a spike had occurred immediately after a particular strategy was implemented, we might have been able to speculate about the effect of that strategy.
What can be determined with available data, however, is that the increase in each state's participation rate was a result of both an increase in the numerator of the work participation rate (i.e., the number of people participating in countable activities for the requisite hours) and a decrease in the denominator (i.e., the number of people receiving TANF or MOE-funded assistance). The denominator may have decreased as a result of some of the policy changes (i.e., the applicant engagement requirements in both states or the sanction reforms in Utah), and/or other factors such as the economy. By the last few months of FY 2007, the denominator in each of the three counties we visited in Maryland was about half of what it was in the first few months of FY 2003. In Baltimore City and Wicomico County, however, increases in the numerator outweigh the decrease in the denominator. In Baltimore City, the numerator increased almost seven fold between early FY 2003 and late FY 2007, and in Wicomico County, the numerator increased almost six fold over that period. In Montgomery County, where the participation rate was relatively higher to begin with the numerator increased between FY 2003 and FY 2004, reached a plateau in early FY 2005, and has been slowly declining since. Similarly in Utah, the denominator of the participation rate decreased by about 16 percent between October 2006 and October 2007, but the numerator increased more than two fold over that period. Figures 3 and 4 illustrate the change in the numerator and denominator in each locality over time.
TANF Cases in the Participation Rate: Utah
Note: Utah figures are based on the state's new reporting system and are only available for FY 2007.
These are preliminary state figures and may differ from the official ACF participation data when they become available.
TANF Cases in the Participation Rate: Maryland Sites
Cases in the Participation Rate Baltimore City
Cases in the Participation Rate Wicomico County
Cases in the Participation Rate Montgomery County
Note: These are preliminary state figures and may differ from the official ACF participation data when they become available.
[ Go to Contents ]
In recent years, Maryland and Utah both have employed several strategies that together have contributed to a substantial increase in their federal work participation rates. Because it is impossible to isolate the effect of one particular strategy from another or from other factors, it is difficult to determine which strategies have been the most effective. To the extent that each of the strategies has had an effect on the participation rates, three key ingredients may have facilitated these effects-each state's proactive approach to change, willingness to learn from past experience, and management of the change process.
Proactive approach to change. Administrators in both Maryland and Utah have been extremely proactive in recognizing the need to increase their federal work participation rates and in taking appropriate measures to that end. Maryland began the process of planning for anticipated changes from TANF reauthorization in 2003, fully two years before the DRA was passed, and between 2003 and 2005 put into effect several policies it anticipated would be consistent with the new requirements. Utah began experimenting with the universal engagement model as early as 1993, but began its aggressive statewide effort to increase its work participation rate in 2005.
Willingness to learn from past experience. In addition to being proactive, state administrators in both Maryland and Utah have been eager to learn from past experience about strategies that have the potential to increase participation. Maryland has drawn heavily on best practices from other states. For instance, the concept of JobStat originated in New York City and the concept of the Work Participation Specialist began in Georgia. In both cases, Maryland sought out strategies that were working well in other states, identified the critical elements of success within each, and adapted the strategies to best meet its own needs. State administrators are open to new ideas and state welfare legislation provides a foundation for embracing changes that will further the goals of encouraging self-sufficiency while maintaining a safety net; to date, no legislative changes have been required to implement the strategies highlighted in this brief. Utah drew from its past experience with universal engagement and also sought input from frontline staff using an online survey that asked about their perceptions of which policies and procedures seemed to work well and which did not, in addition to forming workgroups comprised of both frontline workers and supervisors.
Management of the change process. State policies and directives are meaningless unless embraced and implemented at the local level. State administrators in both Maryland and Utah went to great lengths to ensure that county managers and staff understood the rationale for, and key elements of, each new state directive. Administrators in both states conducted extensive training. In Utah, the TANF executive director and other administrative staff attended interactive "listening" sessions with employees statewide, and state program staff developed and delivered reauthorization training in offices in all regions to support frontline and supervisory staff in understanding and implementing new policies and strategies. In Maryland, action transmittals and policy memoranda accompanied each directive, and state administrators convened regular work participation and universal engagement conferences with local agencies to impart information about national policy changes, address concerns about new state directives, and provide technical assistance regarding implementation. The state TANF director spent substantial time in the field emphasizing the importance of meeting federal and state participation requirements, each directive's intended contribution toward that goal, and the potential consequences of failing to meet established federal work participation and universal engagement rates. In both states, the state TANF director and other state administrators acknowledge that achieving the required federal work participation rate is a formidable challenge, but they believe it is doable and they enthusiastically encourage counties to face the challenge squarely.
[ Go to Contents ]
1. Federally countable core activities include: unsubsidized employment; subsidized private sector employment; subsidized public sector employment; work experience; on-the-job training; job search and job readiness; community service programs; vocational education; and child care for an individual participating in community service. In Maryland, non-exempt recipients are required to participate in activities for 40 hours per week to ensure that they actually participate for the required 30 hours to be counted toward the federal work participation rate (recipients are sanctioned only for failing to participate for 30 hours total and at least 20 hours in federally countable core activities).
2. In state fiscal year (SFY) 2007, for example, all of the 3,506 adult single-headed cases in Montgomery County were subject to the universal engagement requirement and 1,583 were subject to the work participation requirement. In Wicomico County, all of the 196 adult single-headed cases in the month of December 2007 were subject to the universal engagement requirement and 136 were subject to the work participation requirement.
3. Baltimore City changed its TANF application requirements in 2004 in order to get applicants into a work program more quickly. Prior to 2004, all applicants had to participate in a three-week job readiness workshop and then attend an employment and training program orientation at a vendor the following week. Because state regulations stipulated that TANF applications be approved within 30 days, this practice left no time to engage applicants in an ongoing work activity, such as work experience or subsidized employment, before they became recipients. Thus, many entered the rolls unengaged. In 2004, the city shortened the workshop to two weeks and required that applicants attend an orientation at a vendor the Monday of the following week. This provided vendors ample time to engage individuals in ongoing work activities while they were still applicants.
4. In Maryland, time spent in job search activities during the application process does not count against the time limit for this activity (six weeks in the preceding 12-month period, of which no more than four weeks may be consecutive) for purposes of calculating the federal work participation rate. Thus, individuals who conduct job search activities while applying for TANF may enter the rolls prepared to engage in activities that have no time limits (such as work experience). Other states pay benefits from the date of application and may count individuals as participants during the application process, thus invoking the job search time limits. Note that in states that qualify as a "needy state" for contingency fund purposes or that have higher than average unemployment, participation in job search activities can be counted for a maximum of 12 weeks, not more than 4 of which can be consecutive.
5. The remaining one percent is not engaged because of administrative issues. For instance, if a denied application is overturned on appeal, an individual may be entitled to retroactive assistance for months during the appeal process in which he/she would not have been engaged in activities.
6. These figures do not include the number of employers with whom the city's three employment services vendors have worksite agreements or the number of recipients who are engaged in work experience through a vendor rather than directly through the city's BEE. Only recipients with recent and/or substantial work history are referred to BEE; all others are referred to a vendor.
7. Other solely state-funded programs are described in the Summary Report of this series, which can be accessed at http://aspe.hhs.gov/hsp/08/TANFWPR.
8. The vendor can collect on the first payment point for up to six months per client and on the second and third payment points only once per client. While the specific payment amounts differ across vendors, the first payment point represents 40 percent of each contract, the second payment point represents 25 percent of each contract, and the third represents 35 percent of each contract.
9. Other payment points and amounts in the county's contract with its vendor include (1) $120 for completing an orientation and assessment; (2) $70 for issuing a correct child care voucher; (3) $580 for employment placement; (4) $303 for 30-day employment retention; (5) $311 for 180-dayemployment retention; and (6) $180 for access to health insurance through an employer.
[ Go to Contents ]
The Deficit Reduction Act (Title VII, Subtitle A), http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=109_cong_bills&docid=f:s1932enr.txt.pdf
Final TANF Regulations, http://www.acf.hhs.gov/programs/ofa/law-reg/finalrule/finalru.htm
General Strategies to Increase Work Participation
Gardiner, Karen, Mike Fishman, Mark Ragan, and Tom Gais. "Local Implementation of TANF in Five Sites: Changes Related to the Deficit Reduction Act." Fairfax, VA: The Lewin Group; and Albany, NY: The Nelson A. Rockefeller Institute of Government, March 2008. http://www.acf.hhs.gov/programs/opre/welfare_employ/local_impl/reports/five_sites_reduction/five_sites_reduction.pdf
Martinson, Karin, and Pamela A. Holcomb. "Innovative Employment Approaches and Programs for Low-Income Families." Washington, DC: The Urban Institute, February 2007. http://www.acf.hhs.gov/programs/opre/welfare_employ/inno_employ/index.html
Strategies for TANF Recipients Living With a Disability
Derr, Michelle K. "Providing Specialized Personal and Work Support." Washington, DC: Mathematica Policy Research, Inc., February 2008. http://www.acf.hhs.gov/programs/opre/welfare_employ/identify_promise_prac/reports/personal_employ_support/personal_employ_support.pdf
Derr, Michelle K., and LaDonna Pavetti. "Creating Work Opportunities." Washington, DC: Mathematica Policy Research, Inc., February 2008. http://www.acf.hhs.gov/programs/opre/welfare_employ/identify_promise_prac/reports/creating_wk_opportunities/creating_wk_opportunities.pdf
Pavetti, LaDonna, Michelle K. Derr, and Emily Sama Martin. "Conducting In-Depth Assessments." Washington, DC: Mathematica Policy Research, Inc., February 2008. http://www.acf.hhs.gov/programs/opre/welfare_employ/identify_promise_prac/reports/conducting_in_depth/conducting_in_depth.pdf
Sama Martin, Emily, LaDonna Pavetti, and Jacqueline Kauff. "Creating TANF and Vocational Rehabilitation Agency Partnerships." Washington, DC: Mathematica Policy Research, Inc., February 2008. http://www.acf.hhs.gov/programs/opre/welfare_employ/identify_promise_prac/reports/creating_tanf_vocational/creating_tanf_vocational.pdf
Using TANF Sanctions to Increase Work Participation Rates
Kauff, Jacqueline, Michelle K. Derr, LaDonna Pavetti, and Emily Sama Martin. "Using Work-Oriented Sanctions to Increase TANF Program Participation." Washington, DC: Mathematica Policy Research, Inc., September 2007. http://www.acf.hhs.gov/programs/opre/welfare_employ/sanction_pol/reports/sanction_pol/sanction_pol.pdf
Universal Engagement and Working With the Hard-to-Employ
Kauff, Jacqueline, Michelle K. Derr, and LaDonna Pavetti. "A Study of Work Participation and Full Engagement Strategies." Washington, DC: Mathematica Policy Research, Inc., September 2004. http://aspe.hhs.gov/hsp/full-engagement04
LoPrest, Pamela J., Pamela A. Holcomb, Karin Martinson, and Shelia R. Zedlewski. "TANF Policies for the Hard to Employ: Understanding State Approaches and Future Directions." Washington, D.C.: The Urban Institute, July 2007. http://www.urban.org/publications/411501.html
[ Go to Contents ]
This practice brief is one of a series describing state and local Strategies for Increasing TANF Work Participation Rates. The Deficit Reduction Act of 2005 (DRA) resulted in significant increases in the effective work participation rates that states must achieve. The series of briefs is designed to assist state and local officials in thinking about strategies that might aid them in meeting federal work participation requirements in their TANF programs.
The briefs in this series draw on information gathered from case studies of nine programs and describe approaches adopted by selected states and/or local offices that might be of interest to other program administrators. None of these programs has been rigorously evaluated, so their effectiveness is unknown. The U.S. Department of Health and Human Services does not specifically endorse any of the approaches described in this series. All briefs in the series can be accessed at http://aspe.hhs.gov/hsp/08/TANFWPR.
This brief was prepared by Mathematica Policy Research, Inc., (MPR) under contract to the U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation and the Administration for Children and Families.
For further information on this practice brief, contact LaDonna Pavetti at 202-484-4697 or at firstname.lastname@example.org.
To receive a printed copy of this report, send the full title to email@example.com with your name and mailing information.
We are interested in knowing whether readers of this brief found it useful and how it has informed their work and interests in this topic. You are invited to send comments on how you found out about the brief and whether it contributed to considerations concerning policy implementation. Please email your comments to firstname.lastname@example.org and include the title of the brief in the subject line of your email.
Top of Page | Contents
Human Services Policy (HSP)
Assistant Secretary for Planning and Evaluation (ASPE)
U.S. Department of Health and Human Services (HHS)
Last updated: 04/13/2009