Analysis of the 2005-06 Budget Bill Legislative Analyst's Office February 2005

Water Policy Issues Facing the State

Introduction

Two of the most important water policy issues facing the state today are how to address what has been characterized by the administration as a "crisis" in flood management and how to finance the $8.1 billion CALFED Bay-Delta Program (CALFED).

The issue of flood management gained the attention of the Legislature last session due to a court decision that found the state liable for potentially several hundreds of millions of dollars as a result of a 1986 failure of a levee in the state's flood control system. The Legislature accordingly directed the Department of Water Resources (DWR) to report by January 2005 on its flood management expenditure priorities and options for funding. The department responded with a "White Paper"—Flood Warnings: Responding to California's Flood Crisis—that identified various factors leading to a flood management crisis. These factors include an aging flood control infrastructure with substantial deferred maintenance, escalating development in floodplains, declining resources available for flood management, and a number of recent court decisions that highlight the state's potential liability exposure from flood events.

In this write-up, we examine the problems identified by the White Paper, as well as some additional ones, assess the solutions proposed by the White Paper, and make recommendations for legislative action.

The CALFED is a consortium of 25 state and federal agencies created to address a number of interrelated water problems in the state's Bay-Delta region. The program is at a funding crossroads. This is because the program's traditional funding sources—General Fund monies and state bond funds—either are no longer available or have been substantially depleted. In addition, the program has delayed establishing fees to pay for the components of the program that directly benefit specific water users.

The state agency overseeing CALFED—the California Bay-Delta Authority (CBDA)—has recently approved a ten-year, $8.1 billion finance plan for CALFED. The plan provides that this cost be allocated among state public funds, federal funds, local matching funds, and new water user fees. Much of the plan requires legislative action to implement, and the Governor's January budget proposal indicates that a legislative package will be presented as part of the Governor's May Revision.

In this write-up, we discuss the Governor's 2005-06 budget proposal for CALFED and evaluate the proposed finance plan. We make recommendations on how the Legislature should respond to the plan.

Addressing the Flood Management Crisis

The Problem

As shown in Figure 1, DWR's White Paper identifies a number of factors leading to a crisis in flood management. We discuss each of these in turn, and raise additional issues for legislative consideration. To provide context for these issues, we first provide background on the state's responsibilities for flood management, as well as the roles of federal and local agencies.

Background: Responsibilities for Flood Management

Multiple Agencies Responsible for Flood Management. Multiple agencies at every level of government have some responsibility for flood management.

As far as federal agencies are concerned, the U.S. Army Corps of Engineers (Corps) is generally the lead agency on the construction of federally authorized flood control projects. (The construction costs are shared among federal, state, and local governments.) The Federal Emergency Management Agency (FEMA) administers the National Flood Insurance Program, which limits new development in the 100-year floodplain. A 100-year floodplain is defined as an area with a 1 percent chance of flooding per year, which translates to a one in four chance of flooding over the life of a 30-year mortgage.

Local agencies provide day-to-day maintenance and operation of the majority of flood control facilities in the state and have significant control over land use decisions in and around flood prone areas of the state, as discussed later.

As for the state, its roles in flood management fall into the following categories:

• Fund Flood Control Infrastructure. Along with the federal and local governments, the state shares in the funding of federally authorized flood control projects that are located throughout the state. These federal projects are sponsored by either (1) the state and are located within the Central Valley (commonly referred to as the Sacramento and San Joaquin flood control projects or the "Central Valley flood control system") or (2) local agencies that receive "subvention" funding from the state. In addition, the state provides financial assistance to local reclamation districts that operate levees in the Delta region that are outside of the federal-state-local cost sharing arrangement.
• Operate and Maintain Flood Control Infrastructure. The state has the responsibility to operate and maintain the Central Valley flood control system. This includes about 1,600 miles of levees, as well as other flood control works such as overflow weirs and channels. The location of these 1,600 miles of levees within the state-run Central Valley flood control system is shown in Figure 2. However, for about 1,300 (80 percent) of the levee miles, the state has turned over the operations and maintenance responsibilities (including the fiscal responsibility for such) to local reclamation districts. The remaining 300 or so levee miles are directly operated and maintained by DWR, with half of these miles paid by the state and the other half paid by local districts through reimbursements to DWR.
• Provide Oversight of Local Agencies. As mentioned above, the state oversees the locally performed operations and maintenance of about 1,300 miles of levees in the Central Valley flood control system.
• Provide Floodplain Management. The state provides floodplain management in the Central Valley by designating floodways and issuing permits for development within the designated floodways. These activities are carried out by the state Reclamation Board, which is part of DWR. Outside of the Central Valley, the state's role in floodplain management is relatively modest, consisting largely of providing technical assistance to local communities on complying with federal flood insurance requirements and mapping areas that are prone to flooding.

We now turn to the various flood management problems identified by the White Paper.

Aging Infrastructure and Deferred Maintenance

Inherent Design Deficiencies. As discussed in the White Paper, the state's Central Valley flood control system of levees, weirs, and channels is old. Some of the levees are well over 100 years old. There is much evidence that the system is deteriorating, in part due to deficiencies in the original design of the system. Specifically, many levees consist simply of dredged materials (such as mud) that were piled up on top of foundations that are subject to seepage and movement. Seepage is dangerous because it can undermine the levee's foundation, potentially causing it to collapse. In addition to seepage, erosion of levees and riverbanks is a major problem. For example, a recent study by the Corps found over 180 spots along the Sacramento River alone where levees have noticeably eroded.

Lack of Knowledge About State of Central Valley Flood Control System. While there is ample visible evidence of deterioration in the Central Valley flood control system, there is no program at the local, state, or federal level that assesses the structural integrity and the channel carrying capacity of the flood control projects on an ongoing basis. With sediment deposits and vegetation clogging channels, the department believes that some sections of the Central Valley flood control system have lost substantial capacity to carry the flow of water for which they were designed. However, there currently is no program (at any level of government) to verify that the components of the Central Valley flood control system are today meeting their "design flows" safely.

How much would it cost to complete a comprehensive assessment of the structural integrity and the channel carrying capacity of the Central Valley flood control system? According to estimates of the Corps, it would cost about $60,000 per levee mile for exploration, lab testing, and analysis. With 1,600 levee miles in the system, this translates to close to $100 million.

Inspection Program Is Modest. The state is responsible for overseeing the operations and maintenance work performed by local entities on the Central Valley flood control system. Our review finds that the department's current inspection program of the work performed by local reclamation districts is modest in scope. While the frequency of inspections is similar to federal inspection standards (four times per year), the inspections themselves generally only involve checking the status of maintenance practices and looking for any easily visible deterioration of the infrastructure. For example, levee inspections are done by driving on the levee crown roadway only, even though inspections from the water side by boat or by foot (which would be needed to identify damage caused by rodent burrows, for example) would provide much more information about the condition of the levee. The state does not conduct field studies to assess the structural integrity of the levees or their foundations.

Delta Levees Present Special Set of Problems. In the Delta, there are about 1,100 miles of levees, of which a majority (over 700 levee miles) are outside of the state flood control system and operated mostly by local reclamation districts. The department's role with respect to these 700-plus miles of levees is essentially to provide subvention funding for levee improvements and maintenance. The department's "inspections" of these levees are largely to ensure that subvention funds are being spent properly. The department indicates that if these 700-plus miles of levees were inspected to the same degree as levees of the Central Valley flood control system, it would cost an additional $1.4 million (and seven personnel-years) annually.

The 700-plus miles of Delta levees outside the state flood control system present their own set of challenges. Many of these levees were built on peat soil, a type of soil that is prone to sinking and erosion. Over the last 100 years, there have been over 140 levee failures in the Delta, including one as recently as June 2004 at Jones Tract. This recent levee break resulted in costs of close to $100 million—largely paid for by federal emergency and state public funds—for emergency response, levee repairs, and to cover property losses. The marginal condition of Delta levees is of substantial concern because these levees serve to protect a region of abundant resources, including major sources of the state's drinking water supply, fisheries, and agriculture.

The White Paper does not explicitly address the issue of the lack of state oversight over the operations and maintenance of flood control projects in the Delta that are outside the state Central Valley system. However, as the state recently found with the Jones Tract levee failure (a Delta levee outside the state system), the state can find itself paying significant emergency response and repair costs when these projects fail. Given the condition of other Delta levees outside of the state system, future Delta levee failures are possible. Over the long run, it may be more cost-effective for the state to assume an oversight role in the operations and maintenance of Delta levees. This is because it may be less costly to run an oversight program that serves to reduce the risk of levee failures than to pay for emergency response and other costs when the levees fail.

Substantial Backlog of Deferred Maintenance. The White Paper identifies a substantial backlog in deferred maintenance in the Central Valley flood control system, attributable to a significant reduction in resources for this activity since the mid-1980s. (Declining funding will be discussed in detail below.) For example, the White Paper notes that the number of maintenance staff members in DWR decreased by about one-third between 1986 and the present—from 81 to 53. Over this same period, the cost of performing the maintenance has increased substantially, in part reflecting environmental requirements (such as Endangered Species Act requirements). For example, in the 1980s, the cost to repair an erosion site was around $300 per linear foot; today, that repair cost has skyrocketed to $5,000 per linear foot. The department estimates that the cost to address the current backlog of 200 erosion sites alone would be around $600 million. When asked what it would cost to bring the flood control system for which the state is responsible to a safe level, the department provided a very rough estimate of $2 billion for capital improvements, remediation of deficiencies, and deferred maintenance.

Escalating Development in Floodplains

Land Use Decisions Made With Limited Flood Risk Information. The White Paper recognizes the challenge that is presented to the state's flood management system by new development that frequently occurs in areas that are susceptible to flooding. In this regard, DWR estimates that lands adjacent to at least 50,000 of the state's 200,000 miles of streams will likely see development over the next 20 years. In addition, some of the new development is occurring in areas where flood maps are out-of-date or in areas that have never been mapped.

The main floodplain mapping effort is conducted under FEMA's flood insurance program. While the department has supplemented the FEMA mapping efforts, in part by mapping areas outside the 100-year floodplain that may be at considerable risk of flooding, resources available for this activity have been reduced considerably in recent years. Accordingly, local governments often lack reliable information about the potential flood risks when making land use decisions.

Disconnect Between Who Makes Land Use Decisions and Who Pays the Related Fiscal Bills. The White Paper notes that while new development is being approved by local governments in flood-prone areas, it is often the state that bears the fiscal burden and the liability of these land use decisions after the floods have struck. The state, for example, often bears the burden of providing emergency response during floods to protect the development and has been found liable by a court for flood-related damages. We discuss the state's potential liability exposure from flood events in further detail later in this write-up.

Declining Fiscal Resources

Declining Funding for Flood Management. The White Paper identifies as an important issue the decline in funds for flood management at all levels of government. State funding for flood management supports capital outlay, operations, and maintenance expenditures for the Central Valley flood control system; subventions to local agencies for construction and upgrades of flood control projects; and floodplain management activities, such as floodplain mapping. As shown in Figure 3, state funding for flood management and DWR's flood management staffing peaked in 2000-01, largely reflecting the availability of General Fund and Proposition 13 bond monies to make one-time appropriations to pay for the state's share of federally authorized flood control projects. Appropriations for 2004-05 reflect a decrease of almost $200 million, or over 80 percent, from the 2000-01 peak. In addition, staffing for flood management has been reduced by about 30 percent from 1999-00 levels. General Fund resources for flood management have declined due to the weakened condition of the General Fund. In addition, state bond funds for these purposes have largely been depleted. These bond funds were used mainly for local assistance (flood control subventions, delta levee rehabilitation, and flood corridor projects).

State's Potential Liability Exposure

Court Decisions Highlight State's Liability Exposure. The White Paper highlights a couple of recent court decisions that expose both the state and local flood management agencies to major liability. First, in Paterno v. State of California, the court held the state liable for damages resulting from a 1986 levee break in Yuba County. The levee in question was constructed in the early 1900s by Yuba County and was turned over to the state in 1953 to become part of the state Central Valley flood control system, with the agreement that the local reclamation district would be responsible for maintenance and operations. Even though the state had not constructed the levee, the court held the state liable for the damages from the levee break. This was on the basis that the levee's original design deficiencies (the levee was built on a weak foundation, allowing water to seep through the levee) could have been discovered and should have been remedied by the state. (We discuss the budget's proposal to finance a pending $464 million settlement in the Paterno case with a judgment bond, in our write-up on DWR's budget in our companion document, Analysis of the 2005-06 Budget Bill.)

In another recent case, Arreola v. Monterey County, the court held a local flood management agency liable for flood damages resulting from its failure to properly maintain a flood channel. The court made this finding even though environmental requirements had impeded the agency's ability to remove vegetation that was clogging the channel. Specifically, the local agency was unable to secure the required permits from the state Department of Fish and Game.

Our review finds that in addition to the pending Paterno settlement, there have been at least two other major settlements of flood-related lawsuits against the state in recent years. These include a $20 million settlement in the Strawberry Manor case (relating to flood damage in the Sacramento area from the 1986 floods) and a recent $45 million settlement in the McMahan case (relating to flood damage in Yuba County from the 1997 floods). Based on these cases, there is a strong potential for the state being found liable for future flood events.

White Paper's Solution Strategies to Address Crisis

The White Paper recommends several solution strategies to address the many flood management challenges identified in the report. Basically, the strategies can be characterized as addressing one of two sets of issues—(1) those that address public safety issues and (2) those that seek to limit the state's liability exposure from flood events. Sometimes a strategy to address one of these sets of issues also addresses the other. For example, making upgrades to the flood control infrastructure serves to improve public safety and would likely also reduce the state's liability exposure. However, in other cases, a strategy to address one set of these issues does nothing to address the other. For example, a strategy to limit the state's liability exposure by requiring mandatory insurance for property owners in flood-prone areas may protect the General Fund, but it does nothing in and of itself to improve public safety.

The White Paper's main solution strategies are summarized in Figure 4.

LAO's Comments on White Paper's Recommended Strategies. In general, the White Paper recommends a set of rather aggressive strategies to address the challenges identified. We think that the White Paper appropriately recognizes the importance of evaluating the integrity and capability of existing flood control facilities as a fundamental first step. As discussed previously, there is an enormous information gap on this issue that needs to be filled. Second, the White Paper has a number of recommendations for creating reliable funding sources for flood management programs, including a recommendation for a fee (benefit assessment) on property owners who directly benefit from the Central Valley flood control system. There is a statutory precedent for such an assessment, and the concept merits the Legislature's consideration, as discussed later.

The White Paper is more limited in addressing the disconnect between agencies making land use decisions and the fiscal consequences of those decisions. We offer some additional recommendations to address this issue later in the write-up.

We summarize and comment on the White Paper's main strategies below.

Evaluate System Integrity, Rehabilitate as Needed, and Improve Maintenance. At the heart of this strategy is the objective of gathering the information needed to set priorities for the state's flood management expenditures. The strategy includes:

• Evaluating the state's system of levees using current Corps design standards and rehabilitating deficiencies in projects that are identified.
• Developing a state program to continuously evaluate flood system performance and capacity.
• Improving maintenance.
• Expanding the scope of routine inspections.

These strategies will be costly to implement fully. For example, the department's preliminary estimates of the capital improvements needed for the Central Valley flood control system are on the order of $2 billion, which would be spent over 10 to 15 years.

Create Reliable Funding Sources. The White Paper identifies two categories of funding requirements—funding to finance flood management activities and funding to provide reimbursement for flood damages. In addition to funding from the General Fund, bond funds, and federal funds, the White Paper proposes two new funding sources—(1) benefit assessments from a yet-to-be-established Central Valley Flood Control Assessment District and (2) insurance as a replacement for public funds to compensate property owners for damages from floods.

The strategy to enact a flood control benefit assessment in the Central Valley is intended to distribute the state's costs of flood control measures among those that directly benefit from such protection. This is an application of the "beneficiary pays" principle that is a statutory policy guiding the funding of many of the state's environmental protection and resources programs. Specifically, the White Paper envisions a state-levied benefit assessment that would be levied on property parcels in the Central Valley that are within floodplains or in upland areas draining into floodplains. The entire valley could be one assessment district, or it could be split into multiple districts, with the Delta being a single district, for example. Funds from the assessments would be used largely for the operations, maintenance, and rehabilitation of the Central Valley flood control system, but could also be used for floodplain mapping and potentially to compensate property owners for flood damages. A systemwide assessment makes sense because of the interconnectedness of the Central Valley system and the need to evaluate the design and capacity of the system from a systemwide perspective.

Our review finds that there is a statutory precedent for such a benefit assessment to fund flood management in the Central Valley. Specifically, current statute (dating from the early 1900s) establishes the Sacramento-San Joaquin Drainage District, with the authority to fund levee construc tion. Although such a district is currently inactive, these statutory provisions could serve as a starting point for establishing the assessment district envisioned by the White Paper. Importantly, the current statutory provisions would need to be revised to authorize the district's assessment to cover flood control costs beyond construction, such as maintenance.

As far as the insurance-related funding strategy is concerned, the White Paper recommends that the state require all homes and businesses in flood-prone areas have some form of flood insurance. In discussions with the department, it appears that this strategy is also viewed as a means to reduce the likelihood of ill-advised development approvals in flood-prone areas. The White Paper envisions that such an insurance requirement would encompass property owners beyond those which are required to have insurance under the federal program (which is limited to property owners with less than 100-year floodplain protection). The state requirement, which presumably would be integrated with the federal program, could be implemented by establishing a statewide insurance fund (as is done for earthquake insurance) or by requiring those at risk to purchase private insurance. From the state's fiscal perspective, the latter option would probably involve less risk.

The insurance strategy raises issues about how a state insurance program would be integrated with the current federal program as well as with any system of local flood assessments. In addition, the White Paper assumes that by implementing the insurance strategy, the state would be reducing its liability exposure from flood events. However, for this to be the case, the insurance would have to be structured in a way that includes the state as an additional insured party on any privately issued insurance policy or waives the right to sue the state if a state insurance fund were to be established.

Improve Floodplain Mapping and Outreach on Flood Risks. The White Paper recommends increasing floodplain mapping efforts as a tool to provide more comprehensive and up-to-date flood risk information to local agencies that authorize development as well as the public. Local agencies have expressed the need for better information about flood risks; in general, local land use agencies do not have the expertise to evaluate flood risks, so they defer to the state to provide them with this information.

This is a good step in an effort to ensure that local land use decision makers are aware of flood risks. However, we do not think that it goes far enough, as it leaves largely unaddressed the disconnect between agencies making land use decisions and the fiscal burden and liability resulting from such decisions. We offer our recommendations in this regard later.

Reduce State's Liability Exposure. Finally, in addition to the insurance strategy discussed above, the White Paper recommends a few statutory and constitutional changes to reduce the state's exposure for funding flood disaster claims. First, the White Paper recommends that the state Tort Claims Act be amended to preclude recovery of tort-based damages from the state due to flooding and to add a specific immunity for flood protection activities, similar to that which is provided currently for police, correctional, and fire protection activities. Second, the White Paper recommends that the State Constitution be amended to exempt flood control projects from the type of liability—"inverse condemnation"—that was the basis for the court's decision in Paterno. Inverse condemnation claims have their basis in both the State Constitution (Article 1, Section 19) and the United States Constitution (Fifth Amendment). Simply stated, these constitutional provisions require the just compensation of property owners whose property is "taken" or severely damaged by governmental action.

The Legislature may wish to evaluate the pros and cons of enacting the statutory change related to tort claims recommended by the White Paper, given the statutory precedent to grant tort-related liability immunity to other public safety activities. As far as the White Paper's recommended change to the State Constitution is concerned, the Legislature may wish to ask Legislative Counsel to evaluate the legal issues raised by this proposal.

Governor's Budget: First Steps in Addressing Crisis

Budget Proposes Increases for Flood Management. The Governor's budget proposes an increase of $9.7 million (General Fund) for state support and $21.1 million ($16.7 million General Fund and $4.4 million reimbursements) for capital outlay in DWR's flood management program in 2005-06. The budget request includes an increase of 27 personnel-years. With these increases, DWR's total flood management budget for 2005-06 will be about $73 million, an increase of about $31 million, or 72 percent, above the current-year appropriations. The increase in expenditures requested by the Governor's budget request is summarized in Figure 5.

First Steps to Addressing Crisis. According to DWR, the requested increase reflects the first steps under a three-year budget plan to begin addressing the flood management crisis by improving public safety and reducing the liability threat to the General Fund. (The $9.7 million increase from the General Fund for state support is part of a series of General Fund increases for state support planned over the next three years, totaling cumulatively $43 million.) Of the requested increase for the budget year, about $5.1 million (19 personnel-years) is for maintenance work on levees and flood channels. This is about a doubling of current expenditures for this purpose. In addition, $835,000 is proposed to begin a systematic evaluation of levees and channels to identify deficiencies. This is a minimal start for an effort that is projected to cost about $100 million for a full systemwide evaluation. The department indicates that it will seek funding from the Corps to share in the total costs of a systemwide evaluation, perhaps up to 75 percent. The requested increase also includes $2 million for floodplain management (leveraging federal funds for floodplain mapping) and $1.7 million for emergency response (including flood forecasting).

The requested increases for flood management are well justified in light of the challenges identified in the White Paper. However, the increases reflect a small fraction of the funding requirements identified. Finally, it is important to note that the budget does not reflect the establishment of new funding sources recommended by the White Paper. Rather, the department has indicated that a legislative package to implement such funding strategies is currently under consideration by the administration.

Recommended Next Steps

We think that the White Paper does a good job of identifying the major flood management challenges facing the state today. The Governor's budget proposal takes some initial steps in addressing these challenges. Below, we enumerate what we consider to be the most important initial actions that the Legislature should take to begin addressing the problems. We highlight the recommended next steps in Figure 6.

Develop Multiyear Plan to Assess Systemwide Structural Integrity and Carrying Capacity.While the budget proposes a modest effort to begin evaluating the structural integrity and channel carrying capacity of the Central Valley flood control system ($835,000 in the budget year), it is clear that a comprehensive systemwide evaluation is required. This will require a multiyear effort and a substantial funding commitment. We therefore recommend the Legislature direct the department to develop and submit to the Legislature for its review a plan that: (1) schedules over time an evaluation of the complete system, based on a clear set of priorities to guide the timing of the work; (2) estimates the costs of such an evaluation; and (3) identifies funding sources to support the effort, including federal funds and flood control benefit assessments.

Enact Central Valley Flood Control Benefit Assessment. Based on the application of the beneficiary pays funding principle, we recommend the enactment of legislation to establish a systemwide benefit assessment as recommended by the White Paper. Potentially, this could be done by revising existing statutory provisions governing the Sacramento-San Joaquin Drainage District.

Re-Evaluate State's Role With Respect to Delta Levees. As discussed, a majority of the levees in the Delta are not subject to Corps construction standards or state oversight of their operations and maintenance. (These are referred to as "nonproject" levees.) Yet, there is a statewide interest in ensuring the performance of these levees, particularly given much of the state's dependence on the Delta for water supplies. Given the substantial public costs that can result when a nonproject levee fails (such as the recent levee failure at Jones Tract), it may be more cost-effective for the state in the long run to expand its oversight role over these levees. However, such an expanded oversight role would have to be carefully structured so as avoid exposing the state to additional liability due to its new role.

Improve Connection Between Land Use Decision Making and Resulting Flood-Related Fiscal Consequences.Finally, as noted, the White Paper recommends improved floodplain mapping as well as mandatory flood insurance requirements as strategies to reduce the likelihood of ill-advised development approvals in flood-prone areas.

We think that additional strategies are warranted. First, the Legislature could provide that local agencies are ineligible for flood subvention funding from the state in cases where local land use decisions result in substantial flood risks. Such eligibility criteria could be used to encourage land use decision makers to give greater consideration to the potential costs and benefits of their decisions.

Second, the Legislature might also consider enacting a floodplain development fee that would fund the state's additional flood-related costs resulting from new development in floodplains. The rationale for this fee would be "growth funding growth." As with the Central Valley systemwide benefit assessment discussed earlier, such a fee would be justified based on the beneficiary pays principle. However, while the systemwide benefit assessment would be used to pay for the rehabilitation, operations, and maintenance of the existing system, the floodplain development fee would pay for the additional costs that are imposed on the flood control system because of new development.

Financing the Calfed Bay-Delta Program

The Governor's January budget document indicates that elements of a ten-year CALFED finance plan recently approved by CBDA will be incorporated in the Governor's May Revision, along with a package of legislation necessary to implement the plan. The finance plan, as currently developed, is a framework to guide the financing of CALFED from 2004-05 through 2013-14, with a total funding target of $8.1 billion. As noted in the budget document, the plan calls for new revenue sources, including water user fees.

In the sections that follow, we first provide background information on CALFED, including legislative and other direction on how CALFED should be financed; CALFED's funding history; and why CALFED is at a funding crossroads. We then summarize the Governor's January budget proposal for CALFED for 2005-06. Finally, we address the ten-year finance plan, where we discuss the uncertainty underlying some of the plan's assumptions and make recommendations on how the Legislature should respond to the plan.

In order to give the Legislature sufficient time to evaluate the finance plan (which will be further refined and detailed from the version approved by CBDA in December 2004) and the legislation proposed to implement it, we recommend the administration provide the Legislature with the implementing legislation proposal by April 1.

Background

What Is CALFED? Pursuant to a federal-state accord signed in 1994, CALFED was administratively created as a consortium of state and federal agencies that have regulatory authority over water and resource management responsibilities in the Bay-Delta region. The CALFED program now encompasses 12 state and 13 federal agencies, overseen by a new state agency—CBDA—created by statute in 2002. The objectives of the program are to:

• Provide good water quality for all uses.
• Improve fish and wildlife habitat.
• Reduce the gap between water supplies and projected demand.
• Reduce the risks from deteriorating levees.

After five years of planning, CALFED began to implement programs and construct projects in 2000. The program's implementation—which is anticipated to last 30 years—is guided by the "Record of Decision" (ROD). The ROD represents the approval of the lead CALFED agencies of the final environmental review documents for the CALFED "plan." Among other things, the ROD lays out the roles and responsibilities of each participating agency, sets goals for the program and types of projects to be pursued, and includes an estimate of the program's costs—$8.5 billion—for its first seven years (2000-01 through 2006-07). As discussed below, the program's actual funding for its first five years has been significantly lower than envisioned by the ROD. In addition, the relative contribution of the various funding sources has differed significantly from that which was envisioned by the ROD.

The ROD Adopts the "Beneficiary Pays" Funding Principle. The ROD states that "a fundamental philosophy of the CALFED Program is that costs should, to the extent possible, be paid by the beneficiaries of the program actions." The ROD, however, provides few details as to how this principle would be implemented. One exception where specific guidance was provided is the ROD's direction that a user fee be developed—to raise $35 million annually—to support ecosystem restoration activities that benefit Bay-Delta water users.

Legislative Direction Regarding CALFED Financing. While neither the CALFED governance legislation (Chapter 812, Statutes of 2002 [SB 1653, Costa]) nor any other legislation lays out a comprehensive framework for how CALFED should be financed over the long term, the Legislature on a number of occasions has stated its intent regarding CALFED financing. These include budget control language in the 1999-00 and 2000-01 Budget Acts stating that beneficiaries of surface water storage projects that proceed to construction should reimburse all prior planning expenditures made from the General Fund. Similarly, in the Supplemental Report of the 2002-03 Budget Act, the Legislature directed CALFED to draft a financing plan for potential surface storage facilities consistent with the beneficiary pays principle. Finally, the 2003-04 Budget Act includes a statement of legislative intent that CBDA submit a broad-based user fee proposal for inclusion in the 2004-05 Governor's Budget, consistent with the beneficiary pays principle specified in the ROD. However, such a fee proposal was not submitted to the Legislature.

State Funds Have Contributed Most to CALFED. Although the ROD envisioned CALFED being financed over time by roughly equal contributions of federal, state, and local/user funding, the state has been by far the major funding source for the program's first five years, providing about $1.9 billion, or close to 60 percent, of funding. Figure 7 shows the imbalance of the contributions from these three funding sources.

Almost all of the state funds supporting CALFED have been taxpayer-supported "general-purpose" funds, namely monies from the General Fund and bond funds. Apart from a relatively small contribution from the State Water Project and Central Valley Project contractor revenues, no user fees have supported the program. The local funding support for the program, while significant, largely reflects a local match for state bond funds, mainly for water use efficiency projects.

CALFED Is at a Funding Crossroads. The CALFED program is clearly at a funding crossroads. This is for a number of reasons. First, funding sources that the program has traditionally relied on—such as the General Fund and state bond funds—are either essentially unavailable or have been substantially depleted. There will be a significant drop in available funding beginning in 2006-07, particularly due to the depletion of available bond funds. Second, the program's funding requirements are likely to increase as major projects that have been in the study stage for a number of years move toward funding. Third, as will be discussed below, the ten-year CALFED finance plan projects a substantial funding gap between its ten-year funding targets and currently available funding. The funding shortfall absent new revenue sources—$6.3 billion—is close to 80 percent of the funding targets.

Governor's Budget Proposal

Figure 8 shows the breakdown of CALFED expenditures from state funds in the current year and as proposed for 2005-06, among the program's 12 elements.

Current-Year Expenditures. As shown in the figure, the budget estimates CALFED-related expenditures from state funds of $397.9 million in 2004-05. Of this amount, $11.9 million is from the General Fund, with the balance mainly from Proposition 50 bond funds ($194.4 million), Proposition 13 bond funds ($147.9 million), and State Water Project funds ($40 million).

For the current year, the largest state expenditures are in the ecosystem restoration ($101 million) and water storage ($92.4 million) programs.

Budget Proposes $240.6 Million of State Funds for 2005-06. As shown in Figure 8, the budget proposes $240.6 million of state funds for various departments to carry out CALFED in 2005-06, a decrease of $157.3 million, or 40 percent, from the current year. Of this amount, $12 million is proposed from the General Fund, with the balance mainly from Proposition 50 bond funds ($137.3 million), Proposition 13 bond funds ($57.1 million), and State Water Project funds ($25.4 million).

As Figure 8 indicates, CALFED expenditures are spread among seven departments. The largest expenditures are found in DWR ($203.1 million) and CBDA ($19.7 million). The largest state expenditures are proposed for water use efficiency ($75.8 million), water conveyance ($44.7 million), and ecosystem restoration ($30.5 million).

Fee Proposals Will Come at May Revision. As previously noted, the budget proposal does not reflect any new revenue source, such as new water user fees. Rather, these will be incorporated in the Governor's May Revision. As discussed above, we recommend that the proposed legislation to implement the finance plan be submitted to the Legislature for its review by April 1.

The Ten-Year Finance Plan: A Summary

Finance Plan Revises Funding Targets in ROD Downward. The finance plan adopted by CBDA in December 2004 provides a funding framework for CALFED from 2004-05 through 2013-14. The plan is based on a ten-year funding target of about $8.1 billion. The financing plan reflects a significant reduction ($4.5 billion, or 36 percent) from the funding targets envisioned in the ROD for this time period. According to the finance plan, the ROD funding targets were updated based on a review of several factors, including program actions needed to meet program objectives, program priorities, and the "fiscal realities" of the next ten years.

Finance Plan Allocates Funding Among Beneficiaries. Using the $8.1 billion ten-year funding target, the finance plan allocates this cost among state taxpayers, federal taxpayers, water users, and local grant matching sources. It does this based on the plan's evaluation of who benefits from the programs and projects encompassed by the $8.1 billion. Figure 9 shows the finance plan's allocation of the $8.1 billion total cost, by beneficiary category.

Finance Plan Shifts Costs Away From State. Figure 9 highlights a significant reallocation of funding contributions for the program, compared to the allocations in the program's first five years (as shown in Figure 7). Specifically, in the program's first five years, state taxpayer funds supported close to 60 percent of the program's costs. However, the ten-year finance plan allocates about 30 percent of costs to the state, with increased shares to be assumed by the federal government (from 7 percent to 21 percent), water users (from 6 percent to 9 percent), and local matching fund sources (from 27 percent to 40 percent).

Finance Plan Anticipates Substantial But Undefined New Revenues. As noted previously, revenue sources that CALFED has traditionally relied on are running out. In order to meet the $8.1 billion funding target, substantial new sources of revenue—totaling over $6.3 billion, or 78 percent of the funding target—will need to be identified. Figure 10 shows the finance plan's funding target versus currently available funding for each of the broad fund source categories that support CALFED.

Specifically, the finance plan target anticipates a federal funding contribution that is many times greater than federal funding received to date. We discuss in the next section whether this is realistic.

In addition, the finance plan includes new fee revenues from water users to account for benefits to these users in four program elements: ecosystem restoration, Environmental Water Account, levees, and science. Water users benefit from these four programs since each of these programs serves in part to make the supply of water to these users more reliable. The finance plan does not include specific proposals for these new fees. Rather, CBDA staff is currently developing fee options. It is anticipated that the structure for a water user fee that would raise $25 million annually (in addition to the $20 million paid currently by Central Valley Project water users) for the ecosystem restoration program will be ready for legislative evaluation at the May Revision. The authority's staff is considering a number of fee structures, including fees based on the amount of Bay-Delta water diverted by the water user from the system, the storage capacity in Bay-Delta system reservoirs, or some combination of these two options.

Finally, the finance plan anticipates a substantial amount of new state funds (about $1.5 billion). These could include funds from a yet-to-be-proposed state water bond, the General Fund, or a new statewide water surcharge. Apart from stating a need for new sources of state funding, the finance plan includes no recommendations or proposals for new state funding sources.

Uncertainty Underlying Finance Plan

Two major sources of uncertainty underlying the finance plan are the plan's assumption of greatly increased levels of federal funding and new sources of state public funds.

Federal Funding Uncertainty. As shown in Figure 7, the federal government has lagged significantly in its funding contribution to CALFED. Since 2000-01, only about $240 million, or 7 percent, of the program's funding has come from federal funds, in spite of the ROD envisioning that that the federal government would cover roughly one-third of the program's costs. The finance plan assumes that the federal share would increase to 21 percent over the next ten years, with about $1.7 billion of new federal funding. While this is certainly more realistic than a one-third share, it is still many times greater than the federal contribution to date.

While a recent federal authorization bill signed by the President includes $389 million for CALFED, it is risky to assume that all of those funds actually will be appropriated over the six years of the authorization. Under prior legislation authorizing $430 million for CALFED, only about one-half was actually appropriated.

State Funding Uncertainty. As mentioned previously, the plan assumes that $1.5 billion of the program's costs will be funded by unidentified new sources of state public funds. This amount reflects the funding shortfall between the funding target for state funds ($2.4 billion) and "available" state funding ($885 million). It is risky to assume such a high level of funding from unknown sources.

How Should the Legislature Respond to Finance Plan?

We think that there are a number of issues for the Legislature to consider when it evaluates the finance plan.

Are Funding Targets Unrealistic? We think CBDA has made a good effort to incorporate the beneficiary pays principle into the finance plan, consistent with legislative direction and the ROD. However, we are concerned that the funding targets may be unrealistic, given that they assume high levels of highly uncertain federal funding and unspecified sources of new state funds.

Unrealistic funding targets can lead to implementation problems if fee structures are set too rigidly. Here is a simple example to illustrate this. Suppose that a funding target of $100 million is set for a program, and the application of the beneficiary pays principle would allocate this cost based on who benefits to water user fees (50 percent) and federal funds (50 percent). Suppose further that the water user fee is structured rigidly to raise $50 million (50 percent of $100 million), but only $10 million of federal funds materialize. Then the actual funding for this program would deviate substantially from the beneficiary pays principle and water users would be paying for a share of the program's costs that does not benefit them directly.

In order to avoid this problem while holding true to the beneficiary pays principle, it would be important for any CALFED funding mechanism approved by the Legislature, such as new water user fees, to be flexible so that it will adjust accordingly when changes are made to the fund ing target that drives the amount of revenues to be contributed by a particular funding source. If funding targets are realistically set, frequent adjustments to the fee structure should not be required.

Need Statutory Parameters to Guide Beneficiary Pays Principle Application. As we discussed in our Analysis of the 2004-05 Budget Bill (see page B-28), we recommend the enactment of legislation that adopts the beneficiary pays principle for funding CALFED and provides guidance regarding its application. We think that providing this guidance will be important for two reasons. First, if this funding principle is not defined, there is a substantial risk that stakeholder gridlock would result when CALFED attempts to apply it on its own, due to inevitable disagreements among program beneficiaries about the extent to which costs should be allocated to them. Second, as noted by the California Business Roundtable in a report on financing water infrastructure, there is a tendency to overallocate the costs of water projects to the broad public benefit, perhaps because it is easier to do and it avoids difficult decisions about allocating costs to specific beneficiaries.

Since adjustments to CALFED's funding targets are likely to be made over time, as the targets are brought in line with realistic revenue assumptions, legislative direction is needed to guide the resulting reallocation of costs among beneficiaries. Specifically, we think that there should be a statutory definition of "public benefit" and "user benefit," so as to provide objective guidance when public funding and fee-based water user funding, respectively, are appropriate.

Setting Expenditure Priorities for CALFED. Finally, given the significant uncertainty underlying the funding targets in the finance plan, we think that the Legislature should play a role in setting expenditure priorities for CALFED. To this end, we recommend the Legislature hold joint hearings of the water and natural resources policy committees and budget subcommittees in each house on the finance plan. At the hearings, we recommend that CBDA inform the Legislature of the programmatic implications and CALFED's expenditure priorities if the assumed level of funding in the finance plan does not materialize. To the extent that those priorities do not coincide with the Legislature's priorities, the Legislature should provide clear direction to guide CALFED expenditures at reduced funding levels.

Conclusion

Addressing the state's flood management problems and the financing requirements of CALFED present major fiscal challenges for the state. In both cases, there are opportunities to apply the beneficiary pays funding principle to shift costs away from the state to those who directly benefit from the state's programs.

The crisis in flood management, reflected by an aging, deteriorating state flood control infrastructure, grows every day that the problems go unaddressed. As a first step, the state needs to perform a comprehensive evaluation of the structural integrity and the capacity of the state's flood control system, to serve as the basis for developing a multiyear plan to rehabilitate the system.

The Legislature has a major role to play in guiding the financing of CALFED. The program's funding targets should be based on realistic revenue assumptions, and the Legislature will need to establish expenditure priorities so that the program can be "right sized" consistent with those priorities. Statutory guidance regarding the application of the beneficiary pays principle in financing CALFED would significantly facilitate the implementation of this funding principle.