Analysis of the 2007-08 Budget Bill: General Government

Department of Finance (8860)

The Department of Finance (DOF) advises the Governor on the fiscal condition of the state and develops the Governor’s budget. The department also provides economic, financial, and demographic information. In addition, the department oversees the operation of the state’s accounting and fiscal reporting system and has a unit that assesses the operation of the state’s programs.

The Governor’s budget proposes to move DOF’s Office of Technology Review, Oversight and Security out of DOF. Under the administration’s proposal, technology review and oversight would move to the new Office of the Chief Information Officer (see Item 0502 for a discussion of this proposal). The DOF security staff would move to a new Office of Security and Privacy Protection within the State and Consumer Services Agency.

The Governor’s budget proposes expenditures of $84.4 million ($68.8 million from the General Fund and $15.6 million in reimbursements) to support the activities of DOF in 2007-08. This is an increase of $33 million, or 64 percent, above estimated current-year expenditures. As discussed below, this increase is due primarily to the continued development of a new computer system.

Financial Information System for California (FI$Cal)

In 2005 the Department of Finance (DOF) began an information technology (IT) project to replace the state’s budget system. After interviewing departments’ financial staff, DOF has concluded the project needs to be expanded because most state department automated financial systems are old and do not support modern financial reporting requirements. The Governor’s budget proposes a $1.3 billion IT project over the next decade to develop a statewide financial system that would be used by all departments. Our analysis discusses the primary components of this project proposal, key issues the Legislature should consider in evaluating the project, and recommends additional oversight tools if the Legislature decides the project should go forward.

Background

Original Project Scope. In the fall of 2005, DOF embarked on the planning phase of an information technology (IT) project to replace its internal budget system. The Budget Information System (BIS) project, as it was named, was planned to replace DOF’s budget development system. The existing budget system involves the collection of data in multiple systems. The objective of BIS was to implement a single, statewide budget data repository that would meet DOF’s budget development needs and the needs of individual state departments. Under the plan, departments would enter their budget requests into BIS and submit them to DOF online. This information would form the basis of the Governor’s budget proposal and would allow changes to be tracked throughout the legislative budget process.

Department Interviews Identified Broader Problems. During the first year of the project, BIS staff conducted workshops for budget staff in individual state departments. The purpose of these workshops was to define what the system would do to meet the needs of state departments. The BIS staff consistently heard in these workshops that departments needed more than just a new budget system. Many departments indicated that their automated financial systems were designed decades earlier and did not reflect modern financial reporting requirements. Departments have added subsystems, spreadsheets, and manual processes in order to meet reporting requirements. Problems attributed to these additions were data discrepancies, lack of a clear audit trail, and limited staff who could operate the systems.

Key Components of Proposal

Based on the feedback received from state departments, DOF now proposes to dramatically expand the BIS project. The new proposal is to implement a single, statewide financial system which would encompass budgeting, purchasing, cash management, and accounting. The expanded BIS project has been renamed FI$Cal. We discuss the key components of FI$Cal below.

Project Scope Expanded. The BIS project proposed a seven-year effort to purchase and implement a statewide budget system by 2012. Departments would have been responsible for interfacing their existing financial systems to the new budget system. In contrast, the FI$Cal project plan proposes full implementation by 2015 for over 100 state entities. The project would supply each participating department with a new integrated financial system. The Fi$Cal project cost is estimated to be $1.3 billion over the next decade with as many as 691 staff positions involved in project development, as shown in Figure 1. Figure 2 lists the major costs of the project by category. The ongoing costs, once the project is fully implemented, are expected to total $88 million annually. After full implementation, the primary ongoing costs are computer processing at the Department of Technology Services ($45 million), state staff ($16 million), and software licenses ($10 million).

 

Uses Commercial Off-the-Shelf Software. The DOF proposes to procure and implement enterprise resource planning (ERP) software. The ERP is an industry term for software that integrates processes to help a business better manage its activities. For instance, in the case of a financial system, the process of approving a purchase order will also create an accounting transaction that encumbers the funds in one step. The software assumes a set of common business processes and provides the framework to automate paperwork. The benefits from an ERP come from the standardization of the business processes and the automation of transactions. Figure 3 summarizes some of the system benefits as presented in the proposal.

Phased Implementation. Under the project proposal, the new system would be implemented in four phases:

• The first departments converted would be the control agencies whose financial systems are central to the state’s financial functions: DOF, Department of General Services, State Controller’s Office (SCO), and the State Treasurer’s Office. These agencies would have to significantly modify their operations to fit within the ERP framework. At this stage, SCO’s 21st Century project—an ERP for the state’s payroll and personnel management—would share data with FI$Cal.

• In the second phase, a cross section of departments with financial systems of various sizes and complexity would be converted.

• Most of the remaining departments would be converted in the third phase.

• Finally, several departments which have already replaced their financial systems with ERPs would be integrated into FI$Cal.

Existing Projects Would Go Forward. Two large financial system replacement projects are currently in process at the Department of Transportation (Caltrans) and Department of Corrections and Rehabilitation (CDCR). Both of these projects are in the procurement phase and are currently estimated to cost a combined total of $184 million. The FI$Cal project plan assumes that these projects will go forward because (1) the Caltrans and CDCR financial systems are complex and would not be good candidates for early FI$Cal project implementation and (2) postponing their replacement would create too much risk to these departments’ financial operations. Therefore, they would be allowed to proceed and would be integrated in the fourth phase of implementation.

Standardize Statewide Financial Processes. The project goal is to adjust departments’ business processes and not customize the software for individual departments. Thus, a key assumption underlying the project is that state departments will modify their operations to fit within the ERP framework. In many cases, these changes could be significant.

Bring Staff in Early. The project plans to hire a team of full-time state financial staff who have experience in the state’s existing financial processes. In order to backfill departments’ losses of experienced staff, the project plans to provide sufficient funding to departments so they can hire replacement staff a year ahead of the project taking experienced staff. This would allow time for the training and mentoring of the new staff in preparation for the departure of the experienced staff.

Training and User Support. Throughout the course of the system implementation, staff who gain experience in how to use the new system would then help convert subsequent departments to the new system. Ultimately, the project plans for a help desk and a training center using these experienced staff.

Savings Identification. The FI$Cal project proposal does not include a savings estimate from any efficiencies garnered by the system. Instead, a cost avoidance estimate is projected in the range of $3 billion to $5 billion over the next five years. This estimate was developed from information about the condition of existing state financial systems that are assumed will need replacement in the absence of the FI$Cal proposal.

Funding Plan. The project funding plan proposes to use General Fund support for planning and early development activities. During this time, DOF plans to work with the federal government to develop a methodology for allocating project costs to special and federal fund appropriations. Once the methodology is established, the special and federal funds will pay the majority of costs in the later years to reflect their minimal contributions during the early phases of the project. The plan currently reflects special and federal fund sources starting in 2011-12.

Key Considerations

As the Legislature contemplates the major investment of state resources, it will face a number of key questions, which we discuss below.

Scope, Schedule, and Cost

Unified Approach. The FI$Cal project plan treats the state’s financial infrastructure as a single system. In the 1980s, CALSTARS was California’s first attempt to achieve a single statewide financial system. The vision was to provide a statewide view of the state’s expenditures. However, CALSTARS was never meant to be a complete financial management solution as it was not designed to handle revenues or cash management. Implementing an ERP would allow the state to put all its financial processes in a single system. Such a unified approach offers several benefits­—­­like uniform reporting and easier training of financial staff. 

Is the Schedule Realistic? The project schedule reflects full implementation for over 100 state entities over the next decade. Even though a large number of staff resources would be provided under the plan, maintaining this schedule would require an aggressive pace. Each department’s financial processes would be reviewed and altered by the new system. Given that each department’s processes are unique, the project could encounter delays from a single department’s circumstances. In addition, recruiting the number of proposed project staff could create a challenge to staying on schedule. Contributing to this challenge are the Caltrans and CDCR financial ERP projects, which are planned to progress ahead of this project and will also need to add staff.

Is the Budget Realistic? The project proposal is realistic about the major financial commitment for this effort. Given the statewide involvement that will be necessary, the resources identified are representative of the investment that would be required to be successful. However, the project budget includes estimated contract costs that are 30 percent of the total budget. It is difficult to know the accuracy of the budget until after the procurement of these contracts.

Capturing Savings. As stated earlier, the project proposal only identifies savings from potential cost avoidances from future individual IT proposals. The administration did not assume any direct departmental savings from the improved efficiencies of an integrated fiscal IT system—such as reduced data entry, increased ease developing the budget, and less staff time spent gathering information. The project plan, however, commits the administration to capturing information related to departmental activities before and after system implementation. The Legislature, therefore, would be able to identify specific cost savings at a later date.

Role of Control Agencies

Rethinking Processes. The ERP software first introduced in the early 1990s was focused on private sector business processes. Over the past several years, public sector software versions have been released which focus on governmental accounting and fiscal practices. Several states have implemented ERPs and several more are in process. These states report that a major benefit of these systems is standardized processes across state organizations. Rethinking California’s most basic financial processes to achieve this benefit would be the FI$Cal project’s greatest challenge and biggest risk. An ERP system’s built-in business processes are based on industry and public-sector standards and need to be adopted in order to take advantage of its integrated features. Although several ERP systems have been implemented in California state departments, none have experienced these benefits. These departments were forced to make customized changes to the software in order to meet existing control agency processes. For instance, control agencies often mandate forms to be submitted in hard copy. An ERP system for a state entity, therefore, must be customized to generate printed copies even though the system is designed to use electronic transmissions. For example, SCO still requires that departments send hard copy bundled invoices (called claim schedules) for payments made by the state to vendors. These kinds of manual processes reduce the benefit of a modern online processing system (ERP). In addition, the longevity of the system’s usefulness is proportional to the extent that the software is implemented without customization. Upgrading the ERP software becomes much more difficult and expensive to the extent customization has occurred.

Defining Roles and Responsibilities. Under the project plan, a formal memorandum of understanding (MOU) between DOF and each of the other control agencies would be signed to provide a framework for their partnership. The MOU would lay out the roles and responsibilities for each department. Under the plan, each control agency would have “ownership” of their respective business areas in relationship to the system. Each partner, therefore, would have the authority to ultimately determine how the system will be developed and configured in relation to their business activities. While this will ensure each agency has a vested interest in the project, it also increases the potential for customization and creates added risk. It will be crucial, therefore, that executives in each control agency understand the importance of redesigning their work processes to fit the software.

Transitional Challenges. In addition, control agencies will have to manage many organizational challenges that may result from the redesigning of their processes. Particularly during the transition period to the new system, these challenges may include:

• Parts of the organization experiencing temporary disruptions.

• Functions shifting between units within the organization and changing reporting relationships.

• Staff having to acquire new skills and knowledge, which could affect classifications and salaries.

• Staff resisting the changes to long-time processes.

Project Leadership. Over the course of a decade-long project, there will inevitably be leadership changes within both the project and the control agencies. Throughout this time, project and control agency leaders would need to maintain their commitment to major change. It also would be important for FI$Cal leadership to be experienced in state administrative processes and place an emphasis on maintaining close communications among executives in the partner departments.

Weighing Potential Benefits and Tremendous Costs

The Legislature will need to weigh the potential benefits of a statewide system against its tremendous costs. Current financial systems across state departments are antiquated. Many current systems use old technologies for which maintenance will become increasingly difficult over the next few years as knowledgeable state staff retire. In addition, these older systems are inefficient and labor intensive to use. Many departments struggle to close their accounting books within regulatory time frames each year.

To the administration’s credit, it has developed a project plan that relies on various industry “best practices” for implementing large IT projects. For instance, the project has garnered executive support and has a defined governance structure. In addition, the project plan emphasizes departmental involvement, training, and ongoing support for the system’s users. Even so, a project of this complexity and statewide impact would be full of risks.

Key Issues. As discussed above, there are a number of issues that should be given careful consideration as the Legislature decides if this project should proceed. As the Legislature reviews the proposal during the spring budget process, we recommend the administration address key questions, including:

• Do the control agency executives understand their agencies’ responsibilities in the FI$Cal project? Do they understand the potential organizational impacts? Are they committed to leading their organizations through major upheaval and modernizing decades-old processes?

• What are the benefits of moving forward with the Caltrans and CDCR ERP systems while FI$Cal is under development? Are the extra costs and risks worth having these ERP systems in place for just a few years prior to FI$Cal? Could the administration instead just prioritize these departments to the early stages of FI$Cal implementation?

• Why are special funds not sharing in the early costs of the project? What assurances will the state have that the federal government will pay a share of the costs?

Project Would Need Special Oversight Tools. If the Legislature chooses to pursue the FI$Cal project, we recommend increased legislative oversight that is reflective of the project risk. In its oversight, the Legislature’s emphasis should be to ensure (1) that there is a clear understanding of roles and responsibilities and (2) that the appropriate controls are in place to maximize the potential for project success. The additional oversight might include:

• An increase in the standard 30-day legislative review time prior to contract signature. Additional review time, perhaps with hearings, would allow the Legislature time to review and understand the contract’s conditions.

• The use of independent reviews of key aspects of the procurement and the project implementation. For instance, the Legislature required the Bureau of State Audits to audit and report on the procurement phase of the California Child Support Automation System project.

• A requirement that DOF develop a succession plan for the project’s executive team to ensure competent leadership throughout the length of the project.

In the coming years, the state will be forced to take action to address its aging financial infrastructure. The administration has proposed a comprehensive, but costly, solution. Its implementation would require diligent management by the control agencies and targeted oversight by the Legislature.

---

Return to General Government Table of Contents, 2007-08 Budget Analysis