Analysis of the 2008-09 Budget Bill: Health and Social Services

County Administration and Automation Projects (5180)

The budget appropriates funds for the state and federal share of the costs incurred by the counties for administering the following programs: Food Stamps, California Food Assistance Program, Foster Care, and Refugee Cash Assistance. In addition, the budget provides funds for the ongoing maintenance and development of county welfare automation systems.

For 2008–09, the budget proposes an appropriation of $429 million from the General Fund for county administration and automation systems. This represents a reduction of $20.8 million, primarily attributable to proposed budget balancing reductions which (1) cancel the Interim Statewide Automated Welfare System (ISAWS) Migration Project and (2) reduce Food Stamps administrative funding by 10 percent.

The Future of County Welfare Automation Consortia

Each county uses one of four automated systems to administer California’s human services programs. To reduce costs and increase efficiency, we recommend enactment of legislation establishing a goal of standardizing the state’s human services programs on no more than two automated systems. In addition, we recommend increasing legislative oversight of information technology consortia contracts that support these systems.

Background

The Department of Social Services oversees the administration of California’s social services programs. The actual delivery of services at the local level is carried out by 58 separate county welfare departments. Since the 1970s, the state has made various efforts to develop a single, statewide automated welfare system.

Establishment of the County Consortia Structure

In the 1990s, the state was working with certain counties to develop an automation system which came to be known as ISAWS. At the same time, Los Angles County was pursuing its own system called the Los Angeles Eligibility Automated Determination, Evaluation, and Reporting System (LEADER). Meanwhile other counties came together to pursue their own automated systems. Each group was attempting to demonstrate that its system could be the one statewide system.

There was active discussion about this in the 1995 budget hearings and the Legislature ultimately decided that one statewide system was not feasible. The 1995 Budget Act instructed the Health and Welfare Data Center (which is now called the Office of System Integration [OSI]) to collaborate with the County Welfare Directors’ Association (CWDA) on a consortia strategy for statewide welfare automation. Specifically, the Legislature required that there be “…no more than four county consortia, including ISAWS and LEADER.”

During the fall of 1995, OSI worked with CWDA and the counties to develop an agreement on the consortia systems and their member counties. They decided there would be two more consortia in addition to ISAWS and LEADER. An existing system, which included Bay Area counties, would be renamed CalWIN and the Merced County system would be renamed Consortium IV (C–IV). The remaining, unaligned counties selected the consortium they each wanted to join and the four county consortia were formed. Figure 1 shows the relative size of each consortium.

Consortia Systems Technology

The technology used to develop large automated systems has evolved rapidly over the past 20 years. Several evolutionary cycles have greatly changed the way these systems function. Systems of the size and complexity of the consortia take years to complete and cannot be redesigned midstream in order to take advantage of evolving technology. Therefore, the technology employed to develop each consortia system reflects the time period during which the system was designed. The older systems do not have the ease of function and support commonly available with more current technology. Below we summarize the technology status of each consortium.

ISAWS. The ISAWS was designed in the late 1980s and uses hardware and software that is nearing the end of vendor support. The programmers needed to support the software are not readily available because the programming language is not commonly used today. Therefore, programmers must be trained specifically for this purpose. In addition, the software must reside on hardware that is available from only one vendor and so it cannot be competitively replaced. The state enters into “sole source” contracts for this ISAWS support.

LEADER. The Los Angeles County LEADER system uses the same technology as the ISAWS system. Over the years, Los Angeles County has entered into a number of sole source contracts to maintain and update its system.

CalWIN. The technology used to develop CalWIN is referred to as client/server. With this technology, the data is stored in a database on a large mainframe. This data interacts with an application on the desktop personal computer (PC). For client/server systems, as the amount of software on the PC grows, the PC must also grow. Therefore, the PC’s capacity must be increased periodically via an upgrade or replacement. This drives up the cost of maintaining client/server systems.

C–IV. As use of the Internet increased, vendors began to develop applications that could be accessed over the web, referred to as “web enabled.” Web enabled applications do not require special software on a PC to access the application like client/server applications. At the time C–IV was being formulated, vendors also changed the way they develop large systems. Now a series of smaller applications are developed and each performs a discreet function or “service.” This is referred to as “service–oriented architecture” and it allows for system changes to be accomplished more quickly. The C–IV system takes advantage of these more current technologies. This makes it easier to maintain and less expensive to adapt the C–IV system to process and regulatory changes.

Recent Re–Procurement Decisions

ISAWS Migration to C–IV. With respect to the 35 ISAWS counties, the Legislature concluded that it was more efficient to consolidate ISAWS counties into the existing C–IV system, rather than procure a new system. This consolidation, approved by the Legislature in 2006, is known as the ISAWS Migration Project and has an estimated cost of $245 million over four years. In light of California’s budget deficit, the 2008–09 Governor’s Budget proposes to cancel the ISAWS Migration Project. The administration has stated that it plans to resume this project when it can be accommodated within the state budget. The outcome of this budget proposal is unknown at this time.

LEADER: A New Procurement. As LEADER was approaching the end of its useful life, the initial (2005) procurement strategy was for Los Angeles to receive a replacement system based on either C–IV or CalWIN. In 2007 the county and the administration changed this approach to open the procurement to all viable vendor proposals. The Legislature approved this change, thus allowing Los Angeles to procure a new system.

Where We Stand Today. California has four disparate welfare automation systems. We view the proposed cancellation of the ISAWS migration to C–IV as a temporary delay on a path toward potentially three systems. Each of these systems processes caseload using different business processes, even though they each adhere to the same laws and program regulations. In addition, the consortia systems don’t talk to each other; meaning they do not share data, and caseload information cannot be transferred among consortia systems. These siloed business operations have further divided county human services operations across the state.

How Many Consortia Systems in the Future?

The 1995–96 Budget Act stated that there would be “no more than four consortia.” With the decision to move ISAWS to C–IV, the Legislature previously expressed a preference for reducing the number to three: C–IV, CalWIN, and Los Angeles.

Benefits of Further Consolidation. Reducing the number of consortia reduces maintenance costs that are incurred because there are fewer systems that must be modified for regulatory and legislative changes. In addition, there are other administrative savings. Currently, when a client moves to another county with a different system, client information must be recreated. This increases workload and the opportunity for fraud. Having fewer systems reduces the frequency of this occurrence. While it is difficult to quantify total savings, reducing the number of consortia will result in ongoing annual savings for system changes that are currently costing between $10 million and $20 million per system.

Setting a Consolidation Goal. By setting a goal for reducing the number of consortia systems, the Legislature would provide clear guidance for future consortia system proposals. The administration could then make the appropriate plans for current consortia systems as they come to the end of their useful life. This could reduce the cost of future consortia planning activities.

Legislative Oversight of Consortia Contracts

Under Budget Control Section 11.00, state–managed information technology (IT) projects must provide legislative notification 30 days prior to entering into a contract that will increase the project budget by 10 percent, or $500,000, whichever is less. This provides the Legislature an opportunity to review proposed contract terms and conditions. For some state IT projects, vendor contract terms have been renegotiated because of concerns expressed by the Legislature under Control Section 11.00 reviews. However, consortia procurements are conducted at the county level and, while the resulting contracts undergo OSI review, they can be entered into without any legislative notification and review. These county consortia contracts can exceed $100 million and have very limited legislative oversight. Given the substantial state investment in these consortia systems, we believe the Legislature should increase its oversight of consortia contracts.

Analyst’s Recommendation

Establish a Goal of Only Two Welfare Consortia Systems. We recommend enactment of legislation which sets a goal to further standardize California’s welfare operations by ultimately reducing the number of consortia to two systems. As we discuss above, further consolidation can produce efficiencies and reduce system support costs. By moving in this direction, one–time development costs of $80 million (based on recent state experience) could be saved for each consortia system that is consolidated rather than replaced. Similarly, for each system that is consolidated, there are annual savings in the tens of millions of dollars for ongoing application maintenance.

Enhance Legislative Oversight of County Consortia. Legislative review of consortia contracts should be consistent with Control Section 11.00 requirements to provide 30–day legislative notification prior to contract signature. County consortia contracts are funded, in total, with state and federal funds. Accordingly, the Legislature should be afforded the opportunity to review the contractual arrangements that obligate those funds, consistent with state IT contracting procedures. Specifically we recommend amending Control Section 11.00 notification requirements to include county welfare consortia contracts.