Fixing to Change: A Best Practices Assessment of One-Stop Job Centers Working With Welfare Recipients. Major Differences



Several unique characteristics in each of the One-Stop systems underscore the diversity of local circumstances, capacities, and system designs. Differences in partnerships and integration, work requirements, data systems and funding reflect the variety of One-Stop responses to the welfare-to-work mandate. To some degree, the evolution of local One-Stop systems in the local economic, policy, and institutional environment is an ongoing, dynamic adjustment process, with different systems examined here illustrating different points on the learning curve. Major differences include:

  • Breadth of Partners: The number of state and local public and community-based agencies involved as partners in the different One-Stop systems ranges from 4 (Bellingham) to 18 (Kenosha). Both Tarrant County and Traverse City report 14 partners, while Marshalltown has 7 partner agencies. Those sites with high numbers of partners typically reflect the unique character and capacities of the site: in Traverse City, three different community colleges from the 10-county service area are included as partners, while in Tarrant County, several partners are providers of health care that are often not typically associated with One-Stop systems. Three of the sites (Marshalltown, Kenosha, and Tarrant County) have partnerships with community-based organizations, and only one site (Kenosha) has incorporated for-profit partners as part of their One-Stop system. Vocational rehabilitation agencies are partners in three of the five sites, and in some cases, this partnership has been reluctant and ultimately the result of state mandates.
  • Site and Service Integration: The level of integration of partners and services varies considerably. For example, in Bellingham, services are integrated by referral and coordination by system partners, but these services are often delivered in multiple locations by individual partners. Services for welfare clients in Bellingham, Marshalltown, Traverse City and Tarrant County all require referrals from economic assistance officials located in a separate facility. In the cases of Bellingham and Marshalltown, this separation is by design, while in the cases of Traverse City and Tarrant County, site changes currently in progress will bring welfare caseworkers on site. Kenosha offers the highest level of site and service integration, with TANF and other economic assistance services located on-site, and caseworkers from multiple partners working in the same space as a team. Several sites staff their resource rooms, available to all clients, with personnel from partners located on site.
  • Data Systems: The quality of data systems, and access to them, also varies among the One-Stop systems studied. Tarrant County's new, state of the art data system was developed by a private vendor, linking each of the 14 partners on its 92-acre campus. Intake can be accomplished via a user-friendly system at any computer terminal, and intake data is then automatically evaluated for potential eligibility for other programs on campus, allowing the caseworker to enable other partners to gain access to the appropriate client data (with confidentiality safeguards specific to each program). Kenosha has been perfecting its facility-wide data systems for 9 years, with an emphasis on using them to generate management level data to track client flows and program performance. Data sharing for client services is not an issue with the co-location of multi-agency teams. Traverse City's state Employment Security staff developed a common data system for case management that is being carefully introduced to other partner agencies, and adapted for use on a statewide basis. Bellingham has been using an earlier generation of the Tarrant County data system, but without state authorization to allow direct access to files across agency lines, diminishing the data system's usefulness as a tool for improving communications and coordination.
  • Work Requirements: Most of the states visited were tightening work requirements for TANF recipients. However, there were still a few variations that seemed to have significant implications. For example, in Texas, single parents with children under age 4 were exempt from work requirements, resulting in what appeared to be a significantly lower demand for employment services. By contrast, in Iowa, only parents with children under 6 months of age are exempt, while in Wisconsin, parents with children under 12 months were exempt at the time of our visit. The exemption window in Michigan was only 12 weeks, the same as what Wisconsin and Washington have moved to during the past year. Although it is difficult to draw a causal relationship based on observation, caseloads and the level of activity within the One-Stop central locations appeared to be higher in states with more stringent work requirements. Work requirements appeared to place greater strains on child care capacity in all sites, including Tarrant County where both child care capacity and work requirements were at lower levels.
  • Training Focus: Emphasis on providing training for welfare clients varied more than expected. Marshalltown places probably the greatest current emphasis on training, frequently using JTPA funds to provide transportation and child care services to recipients pursuing two-year degrees at the community college; some limited state funds and Pell grants are used to finance tuition at two- and four-year institutions. Kenosha focuses on GED or high school completion, employing a mix of intensive assessment and (for-profit) adult education services. Bellingham had probably the strongest training focus during the time frame for data collection, but at the time of our site visit they were rapidly shifting to a WorkFirst approach, rather than a "training first" approach, which is fairly consistent with the approach of Traverse City. Connecting welfare recipients with training on an ongoing basis after finding employment appears to be a challenge for all One-Stop systems, due largely to the demands on a working single parent's time and the lack of funding for child care, tuition and off-hours training opportunities.
  • Learning Curve: It is quite clear that each of the sites reviewed is at a somewhat different level of development. Kenosha has been developing its program and facility for 9 years, and still had not completed all of the space redesign when we conducted our site visit. Traverse City had been in operation for a few years, but still had yet to integrate all of its partners. Tarrant County was still fairly recent to its site, and has yet to incorporate their human services agency. Marshalltown has not been in operation long, but seems to have well established itself within the community. In most cases, One-Stop systems have moved beyond the old "gatekeeper" model for delivering employment services, and are beginning to provide self-service opportunities that improve access for more employers and job seekers and reserve labor intensive assistance for clients with special needs. Clearly, the development of a One-Stop can require anywhere from 3 to 5 years to bring partners together and to begin working through many of the coordination and joint operating challenges. It does not happen overnight, and even some of the best are still refining their approach after nearly 10 years.

Over time, it is likely that some of these differences will dissipate as innovations in data systems become more widespread and local centers move up the learning curve. However, differences in the breadth of partnerships and level of service integration are likely to persist, both within and across states. The differences among local areas within states will reflect the natural variation in the strength and depth of the institutional capacities and levels of need in each local area; they may also depend to some degree on the local flexibility allowed by different states. For example, one area may have special work and housing search services for farm workers while another may have special language-based services for legal immigrants. Some areas may have strong relationships with community colleges while others work more closely with the local K-12 school system. These differences might also not mean that different groups necessarily receive different levels of services, but they may receive their services in different ways.

Differences in partners may persist due to the unique characteristics of state and local policy and politics. In most cases, these differences may persist across states, due to differing levels of state control of local variation and contracting arrangements -- which are subject to a host of historical and political factors that reach far beyond just the issue of One-Stop Job Centers. For example, state control of welfare and employment services in one state such as Washington may result in different partners and levels of integration than in other states where welfare administration is subject to greater local control and variation, such as Wisconsin.

For these reasons, it is unrealistic to expect that a single One-Stop model will be appropriate for all localities. Similarly, it would be too simplistic to assume that all One-Stop Centers will be able to provide the same level or quality of services.


1.  Note that our goal was to highlight promising practices in One-Stop Centers and therefore the sites we selected and the people interviewed should not be considered a random or representative sample.

2.  Milwaukee is a good example of the first of these circumstances, Boston the second, and Renton (Seattle metropolitan area) the third.

3.  JOBS was the parallel employment and training program for AFDC clients.

4.  Note that the term "welfare" is used in this report as being synonymous with Aid to Families with Dependent Children (AFDC) or Temporary Aid to Needy Families (TANF).  Because the timeframe for this study encompassed the transition from AFDC to TANF, this term is used to avoid conceptual confusion.  Where a distinction between AFDC and TANF is necessary to articulate a change in policy, these more specific terms are used.

5.  In most cases, these employer focus groups were fairly representative of the employer community making the highest use of the One-Stop for hiring welfare clients.

6.  Robert Watrus, Emily Torkelson, and Elizabeth Flynn, One Stop Career Center Systems:  Lessons from the Field, Northwest Policy Center, Graduate School of Public Affairs, University of Washington, April 1996.

7.  The national review also included careful consideration of programs in Kentucky, Massachusetts, Michigan, Minnesota, and Texas.

8.  These included the Welfare Information Network, Corporation for a Skilled Workforce, Interstate Conference of Employment Security Agencies, American Public Welfare Association. and the National Center on Education and the Economy.

9.  Wagner-Peyser programs are those labor exchange and employment services mandated by federal law and funded by DoL. These include the posting of job openings and job counseling and referral for job seekers.

10.  For example, the Marshalltown Resource Center includes the American Indian Council JTPA Program, while the Kenosha Job Center includes the United Migrant Opportunity Service.