Analysis of the 2004-05 Budget Bill Legislative Analyst's Office February 2004

Department of Water Resources (3860)

The Department of Water Resources (DWR) protects and manages California's water resources. In this capacity, the department maintains the State Water Resources Development System, including the State Water Project. The department also maintains public safety and prevents damage through flood control operations, supervision of dams, and water projects. The department is also a major implementing agency for the CALFED Bay-Delta Program, which is putting in place a long-term solution to water supply reliability, water quality, flood control, and fish and wildlife problems in the San Francisco Bay/Sacramento-San Joaquin Delta Estuary (the "Bay-Delta"). (Please see the discussion on the "CALFED Bay-Delta Program" in the "Crosscutting Issues" section of this chapter.)

Additionally, the department's California Energy Resources Scheduling (CERS) division manages billions of dollars of long-term electricity contracts. The CERS division was created in 2001 during the state's energy crisis to procure electricity on behalf of the state's three largest investor owned utilities (IOUs). The CERS division continues to be financially responsible for the long-term contracts entered into by the department. (Funding for the contracts comes from ratepayer-supported bonds.) However, the IOUs manage the receipt and delivery of the energy procured by the contracts.

The budget proposes total expenditures of about $6.3 billion in 2004-05, a reduction of about $1.9 billion, or 23 percent, below estimated expenditures in the current year. This reduction is partly a result of the administration's decision to defer its submittal of most of its resources bond proposals to later in the spring. (Please see the discussion on "Re sources Bonds" in the "Crosscutting Issues" section of this chapter.) It also reflects a decrease of $1.4 billion for the energy contracts entered into on behalf of the IOUs during the energy crisis. This reflects a reduction in the amount of electricity purchased under contract for the budget year, as well as lower prices on the electricity currently under contract.

Major budget proposals include $16.1 million from the General Fund for transfer to the Colorado River Management Account to fund the lining of the All American Canal project. The budget also proposes a reduction of $1.6 million from the General Fund for various water management activities.

State's Approach to Flood Management Should Be More Strategic

In the sections that follow, we find that the state's investments in flood management activities should be more strategic, including an increased focus on critical floodplain management activities. We recommend the enactment of legislation that would reduce the state's share of funding for federally authorized flood control projects. This would create future savings that would provide an opportunity for the state to make more strategic investments in flood management, while at the same time benefiting the state's General Fund.

Responsibilities and Funding for Flood Management

The state shares responsibility for flood management with federal and local governments. Its role in flood management varies depending on geographic area and it generally has a relatively small role in floodplain management outside of the Central Valley.

What Is Flood Management? Flood management is a term that encompasses both structural flood control projects and floodplain management.

Structural flood control projects are projects that modify the natural flow of floodwaters to reduce losses to human resources and/or protect natural resources during floods. Structural flood control projects include conveying flows via levees and managing flows through reservoir operations.

Floodplain management includes activities in and around the state's floodplains that reduce losses to human resources within the floodplain and/or protect natural resources associated with floodplains during floods. Floodplain management includes minimizing the impacts of flows through flood proofing (raising structures in the floodplain so they are at levels less likely to be flooded), implementation of the federal flood insurance program, and relocation of at-risk structures in the floodplain. It also encompasses actions that restrict development in the floodplain through mapping and land-use decisions.

Who Is Responsible for Flood Management? Multiple agencies at every level of government have some responsibility for flood management. Figure 1 provides a summary of the many agencies involved in flood management and their responsibilities.

Several federal agencies are involved in flood management activities. The U.S. Army Corps of Engineers (Corps) is generally the lead agency on the construction of federally authorized flood control projects. (Local agencies typically seek federal authorization of major flood control projects, making them eligible for funding from the federal and state governments. However, there are a large number of smaller flood control projects that do not gain this designation and are largely constructed and maintained by local reclamation districts.) The Federal Emergency Management Agency (FEMA) plays an important role in floodplain management through its administration of the National Flood Insurance Program (NFIP), which limits new development in the 100-year floodplain (areas that have a one in four chance of flooding over the life of a 30-year mortgage).

The DWR is the primary state agency responsible for flood management, with the State Reclamation Board (SRB) specifically responsible for flood management in the Central Valley (along the Sacramento and San Joaquin rivers). The Central Valley has the most extensive flood management system in the state since the entire valley floor regularly flooded before its development. Although SRB is technically independent from DWR and has an appointed board, DWR employees carry out its activities. These activities include maintaining a portion of the levees in the Central Valley that are located in designated state maintenance areas. Both DWR and SRB have responsibility for implementing the state's floodplain management laws. In this regard, DWR mainly assists communities in complying with NFIP standards and mapping flood-prone areas outside of FEMA's 100-year floodplain, while the SRB administers a permit and enforcement program for development within the Central Valley's floodplains.

Local entities are also critical to flood management since they provide day-to-day maintenance and operation of the majority of the flood control facilities in the state. They also have considerable control over land-use decisions in and around flood prone areas of the state.

Costs of Federally Authorized Flood Control Projects Are Shared. The costs of federally authorized flood control projects are shared among federal, state, and local governments. Under federal law, the federal government contributes 65 percent of the planning and construction costs of federally authorized flood control projects. (The federal government does not typically pay for land acquisition costs associated with a flood control project, unless they exceed 50 percent of the total cost of the project.)

Historically, the state has contributed 70 percent of the nonfederal share of the costs, with local governments contributing the remaining 30 percent. However, Chapter 1071, Statutes of 2000 (AB 1147, Honda), modified the state-local cost-sharing formula for participation in federally authorized flood control projects. Specifically, the state's funding share was reduced to 50 percent of the nonfederal share of the total costs, but this share can rise to 70 percent if DWR finds that a project provides multiple benefits, including habitat, open space, and/or recreation. Accordingly, local governments are responsible for at most 50 percent of the nonfederal share of the total project costs.

Given SRB's responsibility, the state has considerably more involvement in the development of the flood management projects in the Central Valley. However, Chapter 1071 improves state oversight of the locally sponsored, federally authorized flood control projects it helps to fund outside the Central Valley by requiring the projects to meet various criteria in order to receive the maximum state contribution.

State's Role in Floodplain Management Is Relatively Modest Outside the Central Valley. Under current law, the state's role in floodplain management is relatively modest and varies considerably depending on the geographic area of the state. Specifically, the state has a greater role in the Central Valley where the SRB has the authority to designate floodways. This authority can significantly reduce allowable development within designated areas since any development within the designated floodways requires a permit from the board. Outside of the Central Valley, the state's role in floodplain management is relatively modest, consisting mainly of a small program in DWR focused on promoting good floodplain management throughout the state. This program provides technical assistance to communities to improve compliance with NFIP standards and expands the mapped areas prone to flooding that are outside of the 100-year floodplain. These activities are focused on providing local land-use managers with more information to promote good floodplain management.

Funding Levels for Flood Management Have Varied Significantly. As shown in Figure 2 (see next page), state funding for flood management has varied considerably on a year-by-year basis over the past five years. Expenditures peaked in 2000-01, reflecting the availability of General Fund monies to make one-time appropriations to pay for the state's share of federally authorized flood control projects, including past arrearages. After that year, General Fund expenditures for flood management programs have been reduced considerably, resulting in continued increases in the amounts owed locals for the state's share of flood control projects.

Governor's Budget Proposal. The budget proposes $33.6 million for flood management programs for 2004-05. This is a slight increase from the current year due to the redirection of unallocated capital outlay funds to perform needed maintenance on a Central Valley flood control project. (The budgeted amount for flood management is a reduction of over 70 percent from the 2000-01 level.)

Risk of Flood-Related Losses Increasing But State's Efforts Reduced

Development in and around floodplains has contributed to increased losses due to floods. This is largely a result of inadequacies with the primary floodplain management tool, the federal National Flood Insurance Program. California's current pattern of development is likely to result in an increased risk of flood losses if good floodplain management is not exercised. Despite the increased risk of flood losses, the state's efforts to improve floodplain management have been reduced significantly.

Flood Losses Continue Despite Flood Management Efforts. Historical data on flood damage collected by the National Weather Service show that average annual flood damages in California (adjusted for inflation) have increased over the last three decades. These damages have continued despite significant efforts to reduce them. The costliest flood in California since 1955, occurred early in 1997 when flood waters caused 120,000 people to be evacuated from their homes in the Central Valley and resulted in approximately $2 billion in property damage. This flood was the result of unprecedented water flows from rainfall and melting snow pack that resulted in numerous breeched levees.

There are many reasons why flood losses continue to increase despite the array of federal, state, and local agencies responsible for flood management. These include the fundamental fact that there are simply more people living in and around the state's rivers and coastal areas subject to flooding.

Deficiencies in Federal Efforts Increase Importance of State Role in Floodplain Management. At present, the major floodplain management tool is FEMA's NFIP program. The NFIP program prohibits new development in areas it designates as special flood hazard areas (more commonly referred to as the 100-year floodplain) and requires owners of existing structures in such areas to purchase federally subsidized flood insurance. To support the NFIP, FEMA is involved in mapping the nation's floodplains. However, FEMA's mapping activities have not kept pace with changing conditions, which is of critical concern in areas such as the Central Valley where development is occurring relatively fast. The department estimates that at least 50,000 of the state's 200,000 miles of streams will likely see development over the next 20 years. However, only 15,000 miles of the state's streams have been mapped by FEMA during the past 30 years, leaving considerable gaps in information on potential flood prone areas of the state. Although FEMA plans to update and expand its mapping efforts, it will likely be a considerable time before FEMA fills in the current gaps.

To address this problem, the state created the Awareness Floodplain Mapping Program with support from Proposition 13 bond funds and the General Fund. This program performs less detailed studies (as compared to FEMA mapping) and maps areas outside the 100-year floodplain that nevertheless may be at considerable risk of flood losses. These maps are especially important for advising local governments about the risks of development in flood prone areas that are outside of FEMA maps.

Although floodplain mapping provides useful information for local planners, we find that there are significant risks in relying too heavily on the 100-year floodplain designations made under the NFIP program. First, the boundaries of a 100-year floodplain are based on estimates and are relatively arbitrary since flood events do not have a natural break at the boundary of the 100-year floodplain. That is, people living ten feet outside of the 100-year floodplain are not necessarily any safer than people living ten feet inside of the floodplain. In addition, California has a relatively short historical record of flooding events, which means that the data used to draw the boundaries for the 100-year floodplain are based on statistical best guesses that often have large margins of error. Given these factors, it is very unlikely that the federal maps drawn for NFIP correctly represent the actual 100-year floodplain boundaries.

Despite these problems, NFIP standards are the basis for most local land-use decisions in and around the state's floodplains. As a consequence, the inaccuracies in the standards may allow development to occur in areas that may have a relatively high likelihood of flooding. The deficiencies in FEMA's NFIP program therefore increase the importance of the state's role in providing local governments with accurate information and maps on flood-prone areas that are being considered for development by local governments.

Population Increases in California's Flood Prone Areas Raises the Importance of Floodplain Management. Floodplain management is becoming increasingly more important in California because of the increased population pressures along the state's rivers and coastal areas. For example, population growth in the Central Valley along the Sacramento and San Joaquin River systems has been consistently higher than in California as a whole over the past two decades. The growth in the Central Valley is projected to continue so that by 2010 one in four Californians will be living in the Central Valley (only one in five persons lived in the Central Valley in 2000). This is likely to result in pressures for more development on the fringe of the 100-year floodplain, thereby potentially increasing the risk of losses from future flood events. Given this, land-use decisions being made could have significant impacts on flood management costs, including the need to build additional structural flood control projects.

State Funding for Floodplain Management Has Been Reduced. As mentioned previously, the state's role in floodplain management is relatively modest—mainly limited to providing technical information and mapping. However, for reasons discussed above, the value of state efforts in floodplain management have become increasingly important. However, at the same time, DWR's budget for floodplain management has been reduced significantly—to $1.2 million, which is a 70 percent reduction from funding levels in 2000-01. This reduction is a result of fewer available bond funds and significant reductions in General Fund support, which have been the primary funding source for these programs. These reductions have reduced the rate at which the department can complete its mapping of flood-prone areas slated for development and jeopardizes opportunities to partner with FEMA in modernizing federal floodplain maps. The overall effect of these reductions is to reduce the state's ability to promote good floodplain management by local entities.

State's Flood Management Expenditures Should Be More Strategic

The state currently owes approximately $404 million to local governments for its share of the costs for federally authorized, locally sponsored flood control projects. The Governor's budget provides no funding for these subventions in 2004-05. We recommend the enactment of legislation that reduces the state's share of the nonfederal costs for federally authorized flood control projects, saving the state between $115 million and $230 million in future budget years. Furthermore, we find that the future-year savings created by this recommendation would provide an opportunity for the state to be more strategic in its approach to funding its flood management programs, while at the same time saving General Fund monies.

State Unable to Pay Locals Its Full Share of Costs. The state owes approximately $404 million to local governments for its share of federally authorized, locally sponsored projects. In past years, the General Fund has contributed the majority of support for these subventions. However, some bond funds have also been used. The 2003-04 Budget Act included $116 million from the General Fund to fund the state's share of federally authorized, locally sponsored flood control projects. However, the administration has proposed to revert $105 million of this amount back to the General Fund, as part of its mid-year expenditure savings proposals. In addition, the 2004-05 Governor's Budget does not provide an appropriation for these flood subventions. Despite the inability of the state to pay locals its share of the costs associated with these flood control projects, construction of these projects will generally not be affected. This is because the state funds are generally used to reimburse locals for costs they have already incurred in constructing a flood control project.

Local Beneficiaries of Flood Control Projects Should Share Costs More Equitably. As mentioned previously, up until recently, the state has generally contributed 70 percent of the nonfederal share of the costs of federally authorized flood control projects. This resulted in local governments funding 30 percent of the nonfederal share of the total project costs which likely understated the benefits derived by localities. We find that flood control projects provide direct benefits to local communities. For example, in addition to the direct public safety benefits, the projects often allow for new development and/or exemption from NFIP requirements which provide significant economic benefits to local communities. Given the benefits to local communities, we think that it would be more equitable if local governments paid a greater share of the costs to build federally authorized flood control projects. We believe that this is consistent with recent legislative direction to fund resources programs under the "beneficiary pays" principle. For example, recent legislation initiated a fee for certain landowners to defray some of the state's costs for providing fire protection to their land.

Recommend Legislation to Reduce State's Funding Share. As mentioned previously, Chapter 1071 provides that the state will fund between 50 percent and 70 percent of the nonfederal costs of a federally authorized flood control project. As discussed above, we think that local governments should assume more of the cost burden for flood control projects based on the application of the beneficiary pays principle. Given this, we recommend the enactment of legislation that reduces the state's share of the nonfederal costs of federally authorized flood control projects to 30 percent. However, the state would reimburse up to 50 percent if the project satisfies existing statutory criteria for a higher level of reimbursement for projects with multiple benefits. We also recommend, consistent with current law, that the cost sharing provisions apply equally to state sponsored and locally sponsored projects. We further recommend that the new cost sharing be applied to all future flood control projects as well as those projects that have been authorized, but have not yet received payment from the state.

We estimate that the state could save between $115 million (50 percent state share of costs) and $230 million (30 percent state share of costs) by applying the new cost-sharing provisions to authorized projects where a state share of costs is currently owing. (The range of savings depends on the level of state funding the projects would qualify for based on multiobjective criteria established in Chapter 1071.) As there is no funding for local flood control subventions in the budget, these savings would accrue in future years as the state would appropriate funds at the lower share of project cost.

We find that the new cost-sharing arrangement can be applied to projects that have already been authorized by the state, but have not been funded. This is because the authorization process (by which a local project becomes eligible for state funding) is separate and distinct from the process that funds these locally sponsored projects, generally the annual budget appropriation process. Therefore, savings could be achieved by reducing the state's cost share of projects already authorized by the state. As mentioned previously, the state's funding share is typically used to reimburse locals for costs already incurred. Therefore, reducing the state's funding share should not affect the delivery of these projects.

Savings Present Opportunity for State to Fund Flood Management Programs More Strategically. We think that the future savings created by the funding change recommended above provides an opportunity for the state to be more strategic in how it funds flood management. Specifically, we think the state's flood management activities would be more effective if investments were made for (1) increased oversight and review of local flood control projects outside of the Central Valley and (2) floodplain management. We think it would take a relatively small amount of future savings—less than $10 million annually—to improve the state's current approach to flood management activities. We discuss below the particular investments that we recommend be made using some of these future savings. The balance of the future savings—which would be potentially over $200 million—could be used for other legislative priorities.

As mentioned previously, DWR's oversight and review of flood management projects outside the Central Valley is limited. Chapter 1071 did improve the state's oversight over the flood control projects funded outside of the Central Valley by making full funding contingent on the project satisfying certain multipurpose criteria. However, DWR currently has no staff budgeted to review these projects. Therefore, we recommend that when savings accrue in the future associated with reducing the state's share of costs for federally authorized flood control projects, the Legislature consider using a portion of these savings to improve oversight and analysis of flood management projects outside of the Central Valley. An investment of $1 million to $2 million would allow DWR to have input into the development of local flood projects, thereby enabling the department to assess whether projects meet the criteria set forward in Chapter 1071. This will allow for more strategic investment of state funds in local flood management projects.

As mentioned previously, floodplain management is a cost-effective way to avoid flood damages. The state's current investments in floodplain management are minimal and we think improvement in this area would be a strategic investment, especially given the rapid development of flood prone areas of the state. Therefore, we recommend that the Legislature consider allocating some of the future savings—$2 million to $3 million annually—to leverage federal funds over the next several years by participating in FEMA's forthcoming nationwide map modernization program. Without a funding commitment by the state, California could lose tens of millions of dollars in federal funds over several years to improve the state's floodplain maps. Also, the Legislature should consider using some of the future savings—$3 million to $4 million annually—to provide ongoing funding to the state's floodplain management programs, including the department's Awareness Floodplain Mapping Program and its Proactive Floodplain Management Program (formerly funded by Proposition 13) that maps potentially flood prone areas and provides training and workshops for local communities on complying with NFIP requirements.

State's Ability to Recover Maintenance Costs From Locals Needs Improvement

The state maintains a portion of levees and other flood control projects in the Central Valley. Current law does not allow the Department of Water Resources to recover all of its costs when it takes over the maintenance of a flood control project from the locals. We recommend the enactment of legislation to allow the state to recover all of its costs associated with establishing and operating a state maintenance area.

Maintenance Versus Improvements: Who Is Responsible? While the SRB oversees the maintenance of the over 1,600 miles of levees as well as other flood control projects in the Central Valley, these projects are for the most part directly maintained by local reclamation districts. (Numerous additional levees are built and maintained by individual landowners but are not overseen by the SRB.) However, the state is directly responsible for maintenance of around 150 miles of levees in ten "state maintenance areas" located in the Central Valley and significant portions of the Sacramento River Flood Control Project. In the former case, the local reclamation districts are responsible for the costs of maintaining the projects, while in the latter case the state is responsible for maintenance and operation. Project improvements or upgrades, however, are cost-shared among federal, state, and local governments. Improvement and upgrades to these projects in the Central Valley are prioritized according to plans developed jointly by the Corps and SRB.

Costs Incurred in Forming New State Maintenance Areas Are Not Reimbursable. Current law allows DWR to take over maintenance of a flood control project (such as levees) from a local reclamation district if the local district is not maintaining the structure in accordance with federal regulations. The law also allows a local reclamation district to voluntarily transfer its obligation to maintain any flood control structure to the state if the local district declares that it no longer desires to operate and maintain the structure. In either case, DWR is authorized to collect assessments from the local reclamation district to cover the costs associated with maintaining the local flood control structures. However, the law does not allow the department's start-up costs associated with forming a new maintenance area to be reimbursed from the local district.

Recommend Legislation Authorizing the State to Recover Start-Up Costs of Maintaining Local Flood Control Projects. We think the state should be reimbursed for all of its costs associated with maintaining local flood control projects, including the start-up cost associated with forming a new maintenance area. Without this authority, the state bears the entire costs of establishing new maintenance areas. We do not think that these costs should be treated differently than the ongoing maintenance costs which are cost-recoverable under current law. Therefore, we recommend the enactment of legislation that authorizes the state to recover all the costs of maintaining local flood control projects from the local reclamation districts, including the start-up costs associated with setting up a new state maintenance area. This will allocate costs more appropriately to the local districts.