History of Welfare-To-Work Performance-Based Contracts in NYC
History of Welfare-to-Work Performance-based Contracts in NYC:
Lessons Learned
Swati Desai, Ph.D. Rockefeller Institute of Government, SUNY and Urban Institute
Lisa Garabedian, New York City Human Resources Administration
Karl Snyder, New York City Human Resources Administration
The paper was presented at the International APPAM conference, Moscow, Russia, June, 2011,
Introduction
This paper looks back over more than ten years of performance-based contracting in the Welfare to Work program of New York City’s Human Resources Administration/Department of Social Services. It documents the evolution of the program, the evolution of vendor-based contracting and looks at issues which were encountered resulting in modifications in program, contracting, and management. In the process of reviewing this history this paper details lessons learned from this experience of managing the program in partnership with the private sector, and specifically lessons learned in the effective use of contracts to achieve goals of welfare reform. The paper is the first description of the ten years of this contracting process begun at the beginning of Welfare Reform.
History of the Process in NYC
Contracting out social services to not-for-profit and faith-based organizations has been done since the turn of the 20th Century. But contracting out welfare (TANF) services, especially welfare-to-work services became an important practice after the enactment of the Personal Responsibility and Work Opportunities Act, (PRWOA), in 1996. A small number of states, Wisconsin, whose program was the best known among them, contracted out both case management and eligibility determination to the private sector. The majority of states kept the determination of eligibility in house and contracted out the employment-related services.[1][2] New York State and, specifically, NewYork City was a part of this group. In talking of these contracts for employment-related services, they could be divided in three groups: Pure-pay-for performance, cost reimbursement and fixed price for services. NYC’s Human Resources contracts, the topic at the heart of this paper, fall in the first category.
After the enactment of the TANF program, the emphasis of public assistance programs shifted from income maintenance to finding work for welfare recipients. Federal mandates for caseload engagement in work activities and job placement put a significant pressure on states and localities. This was especially true for New York City where 1.1 million of its 8 millionthe total population was on welfare. (Figure1). In NYC, Mayor Giuliani's campaign focused on the reduction in crime and a reduction in the welfare population. Jason Turner, who became Commissioner of the City’s Human Resources Administration in 1998, strongly believed that all welfare recipients should work or participate inactivities that lead them to employment and ultimately self-sufficiency. So the concept of “full engagement” was introduced. “Full engagement” required that all who manage welfare cases should know what activity each welfare recipient is involved in and whether these activities were helping them to move towards employment and self-sufficiency. Prior to the passage of PRWOA, HRA, like many other states and localities did not have strong employment components in their programs. So contracting out enabled NYC to expand capacity and restructure service delivery system quickly with flexibility. Contracting out for services fitted with Commissioner Turner’s philosophy that private sector could provide services more efficiently. According to the first Deputy Commissioner Mark Hoover, “Government is best at setting outcomes, designing policy and overseeing and supervising performance. It is not great at operational activities and service delivery. There are simply too many processes and inefficiencies. Contracting out is better, cheaper and more flexible, allowing you to add and contract when needed.” [3] NYC contracted out employment activities in 1999.
Initial contracts were divided in two parts, employment services for applicants for welfare, Skill Assessment and Job Placement, (SAJP) and employment services for welfare recipients who were considered “employable” or able to work, Employment Services and Placement (ESP). There were 5 SAJP contracts serving all applicants in all job centers and 12 ESP contractors. These contractors (also referred as vendors here) varied in size. They included large national for-profit companies and non-profit companies and small local for- profit and not-for-profit companies and local community colleges. A number of contractors also used local community-based organizations (CBO) as their subcontractors.
Figure 1: NYC Caseload, Employment Contracts and Other Important Events[4]
The contracts were performance-based and were awarded on a negotiated basis to each contractor. The unit price was negotiated between the contractor and HRA and did vary among contractors. The contracting method used is known as Negotiated acquisition. This method allowed the agency to negotiate combined milestone payments or the unit price with each contractors and it required less contract process time.
The average unit price for SAJP (interchangeably also used as SAP) contracts were $2,500. Payments were contingent upon meeting the performance milestones tied to participant’s assessment, engagement, job placement and retention in jobs. The unit price per contract for SAJP was distributed in the following milestones: Assessment - 10%, Engagement - 20%, Placement - 60%, 90 day Retention - 10% and a bonus milestone for a case closure due to earnings at 90th day. (See the summary table for contract history in Appendix I.) SAJP services were aimed at quickly assessing individual needs and interest, providing job readiness skills and attempting to attach applicants to jobs as quickly as possible before the application for welfare was approved. Some SAJP vendors located a staff at a job center to enroll applicants and conduct intake. On an average, the client stayed with the SAJP vendor for 4-6 weeks. If the individual was placed in a job, the welfare case could get rejected or closed and the vendor was expected to track the individual and to help him/her retain the job.
On the other hand average unit price for ESP contracts varied between $4,600 and $5,500 and the milestone payments once again were focused on the recipient’s job placement and retentions. The unit price was distributed as follows: Job Placement - 37%, 90 day retention - 36%, 180 day retention - 27% and bonus milestones for a job with high wage at 90 days and a case closure due to earnings at 180 days. Clients that did not get a job during the application phase and those who were already receiving assistance were randomly assigned to ESP to receive services for two days a week as a part of “full engagement Model”. Other three days a week the recipients were required to participate in the Work Experience Program (WEP). [5]In ESP contracts clients received more intensive mix of services compared toSAJP that included job search, job placement assistance, short-term training such as computer training and other trainings (home care aide, security guard, food service). After recipients were placed in jobs, in some contracts, the case managers checked with participants to see how they were adjusting to the work place and collect paychecks for retention milestones. Other contractors offered the bonus payments to those clients who brought paystubs. Employed participants in addition to transitional child care and Medicaid, also received bus/subway passes, training in financial literacy, and in some cases, ITA (Individual Training Account) vouchers to upgrade skill while working. Those who did not get a job in 6 months were reassigned to another vendor in the same borough.
At the inception of the contracts, annual value of SAP contracts was $29.9 mn. and of ESP $78.5 mn.
The contract design and milestones development went hand in hand. According to Commissioner Turner[6], design rationale for these initial contracts was:
- SAP allowed for a lower performance payment because the vendor had access to all new applicants, not just long term recipients who are presumably harder to place. We wanted to “cream” the easy to serve cheaply, leaving larger performance payments for ESP for those needing more help The result would be that overall costs to HRA would be lower and that the higher payments reserved for ESP would permit more intensive service, as opposed to a standard lower blended rate for both applicants and recipients.
- SAP had access to referrals for a fixed period, four to six weeks, after which they were referred to an ESP. Therefore SAPs had every incentive to work with all referrals to the max.
Also as HRA wanted the applicants to get jobs quickly, the higher proportion of unit price was allocated to job placement in both SAJP and ESP.
In 2002, these contracts were renewed with some modifications, especially the distribution between placement and retention milestones were changed. In SAP, the payment rate of placement milestone was decreased from 60% to 35% and the 90 day retention milestone was increased from 10% to 35%. Similarly in ESP, the rate of placement milestone decreased from 37% to 23%. But the 90 day retention rate was increased from 36% to 50%. The 180 day milestone percentages remained the same. Even though the budgeted value of renewed SAP contract went up, the value of renewed ESP contract declined. The overall value of renewed contracts declined from $108.4 mn. to $87.7 mn. due to the significant decline in the welfare caseload.
The focus on full engagement, emphasize on placement and successful implementation of employment contracts led to a big decline in the caseload (figure 1). At that time automated client tracking and accountability system (NYCWY), monitoring system (VendorStat) and automated payment and claiming system (PACS) were developed to manage the contracts efficiently That also allowed the Agency to focus on those with mental and physical barriers and the Agency developed WeCARE (Wellness, Comprehensive Assessment, Rehabilitation and Employment) program. The WeCARE contracts combined medical and employments serviceswere a mix of output and outcome milestone payments, not a pure performance-based contract. The ultimate goal of this contract was to obtain SSDI/SSI for disabled clients and to stabilize others enough to get them employment.
The 2nd generation of employment contracts which replaced ESP and SAP became operational in 2006 and were renewed in 2009. These contracts called Back to Work or BTW were almost 100% performance-based with some significant changes from the previous contracts. Under the BTW contracts, the same employment contractor was responsible for both applicants and recipients for a designated welfare office allowing each client to receive a continuous set of services. In addition, in these contracts the milestone distribution changed. In the previous contracts (SAP and ESP) the payment milestone for job placement was paid as long as recipients had a paystub even if it was only for a day’s work, In the BTW contracts, more emphasize was given to job retention and vendors were paid only if clients kept a job for 30 days. Also the placement milestone for both applicants and recipients increased from 23% in ESP to 35% in BTW.The 90 day retention payment milestone decreased to 29%, the 180 day retention milestone remained at 25% about the same as it was in ESP. In addition, there were two process milestones, one, that was the same as in the original contracts, Assessment - 5% and a new one, Job Retention and Career plan, -5%. In addition, the contracts included additional bonus milestones; a) job placement for sanctioned, 2) job placement for time-limited participants, 3) wage gain at 180 days and 4) case closure at 90 days. These payment points were a small fixed dollar amount. The unit price payment in BTW contract was an average of SAP and ESP unit prices. The total value of the contract declined from $97 mn, in ESP and SAP to $53 mn. in BTW. And the contracting method changed from the Negotiated Acquisition to the standard contracting model, Request for Proposal - RFP.
Vendors and their Perceptions
A performance-based contract requires a nimble organization that can operate within the constraints of performance payments, an organization that can survive with back end payments [7] and can handle a large flow of clients. Since,some CBOs might not be able to operate under either of these conditions; they might have difficult time succeeding under the conditions of the performance-based contracts. This became clear during the renewal of SAP and ESP contracts, one of the ESP contract could not perform under this conditions and withdrew. Of the original 15 SAP and ESP contracts[8], 9 were not-for- profits and 6 were for-profits. All but two not-for- profit organizations were NYC based, and half of for-profits were NY based small contractors. But the number of BTW contractors, mostly not-for-profits, declined significantly, only 7 of original 15 contractors were participating in the program.Of the 7 BTW contractors, 3 were for- profit and 4 were not- for profit. Since these contracts were very large, these contractors chose a number of small community-based organizations in the City as partners. These partnerships were forged out of necessity as HRA expected that these large contractors would not have the ability to serve a diverse client population residing in different neighborhoods and thus gave a priority in the selection to those contractors who brought these collaborations in their proposals[9]. Most of the contracted vendors (prime) had contracts with CBOs (subs) and used their services.
Interviews with these contractors show that vendors who have stayed with HRA, are mostly happy with the contract arrangement. Amelia Betanzos, president of Wildcat Corporationsaid in 2001;”We want to get paid for our results and will do well under this system. If you are good, competition is useful-compared with for profits, our motivations are different. This is our business-this is their opportunity.”[10]A current interview in 2011 with an executive of a large not for profit who has participated in HRA’s performance-based contract from the beginning also expressed the same sentiment; “a performance-based contract gives us autonomy and allows us to be creative." It also allows us to change the program and incorporate the best-practices from what we have learned.” [11] On the other hand at the onset of performance-based contracts, for-profit contractors who began to be significant in government contracts, especially in TANF, felt that they could provide these services more efficiently and at a lower cost than not-for-profit organizations and government. A number of for-profit organizations operated nationally and they brought their experience and management expertise in setting up and running programs. If needed, they leveraged local CBOs to provide services they didn’t have expertise in. Interviews with for- profit vendors also show satisfaction with performance-based contracts from the beginning to the present. Interviews conducted by Barnow and Trutko[12]in 2003 show that all contractors particularly, for-profit contractors were enthusiastic about performance-based contracts. “With performance-based contract you get paid for results, not process.” Recent interviews with two for profit organizations show [13]that they continue to be satisfied with performance-based contracts. According to one vendor, “performance-based contracts create a level playing field. “ They feel that those contractors, who can deliver services with innovative strategies as client-base changes, will be competitive and stay in the business. But another small for-profit contractor felt that the burden of paperwork in the Back to Work contracts has increased, especially after TANF reauthorization when states and localities were required to track attendance. According to this vendor, the contract should be hybrid,a line item for extra paperwork and a performance payment for outcomes such as placement and retention.
What milestones were achieved?
1)Both reported (direct) and unreported (indirect) job placements increased.
During the 10 year period, job placements reported by vendors doubled between 2001 and 2006 as a proportion of total placements and remained high through 2010.
The Agency’s welfare reform goals for all of its cash assistance clients have been “full engagement” in work and work-related activities and self-sufficiency through job placements. To emphasize the importance of job placements, HRA sets the placement goal for its cash assistance and non-cash assistance SNAP (originally known as Food Stamps) recipients every year. This goal includes not only placement by employment contractors,[14] but also placements reported at job centers by clients (not vendor reported) and placements obtained by matching the active client database with the New Hire data base.[15]. All three categories provide an unduplicated count of placements. The chart above shows that vendor placement as a proportion of the total placements have increased from 2001 to 2010 despite the severe recession of 2008 and 2009. Prior to the focus on full engagement and employment contracts, the total placement for the agency was in the range of 40,000-60,000. Since 1999, the total placements have continued to increase reaching over 100,000 between 1999 and 2003, Even during the 2008-2010 recession, placements have remained between 75,000 and 85,000. This table highlights the adage, “what you measure, will happen.”As the focus on employment increased with performance-based contracts, agency-wide job placements also increased.